Yesterday I showed you the first issue of China Watch. Today we’re launching the second part of the new Atomic Layers format: Western Report. This is its inaugural edition.

The report is divided into four sections:

1. Mainstream Report covers the week’s most important developments among leading AM companies.

2. AM Software Report focuses on software.

3. CEE Report covers Central Europe.

4. And finally, AM HR Corner, dedicated to personnel moves. In this last section, I’m also starting something that I hope will eventually become a value of its own: tracking and analyzing the job market.

We’re right in the middle of the holiday season in Europe, so there’s much less happening than in China (as you could read yesterday). But stay with me. Things are about to pick up.

Mainstream Report

I’ll start with the story that has been heating up the open-source community over the past few days. Bondtech - the Swedish manufacturer of accessories for FFF 3D printers, has officially confirmed that the nozzles for its INDX system are not hardened, despite being marketed as such. Their hardness is 30–32 HRC, while genuinely hardened nozzles reach 55–60 HRC.

When printing with carbon fiber, glass fiber, or glow-in-the-dark filaments, they wear out extremely quickly. The issue affects the Founders Edition as well as the first production batches of kits for the Prusa CORE One+.

Prusa Research’s role here is significant. As the distributor of the assembly kits, the company has the right of recourse against Bondtech over customer claims. The market expects compliance with the relevant EU directive: either free replacement with proper hardened nozzles or a proportional price reduction. Instead, Bondtech is offering only vague “upgrade options” for an unspecified surcharge at some unspecified point in the future.

Customers who spent around $1,000 don’t want to pay a second time for something they were already promised.

The company’s financial situation helps explain its nervousness. Producing properly hardened nozzles will take months. That’s a serious blow, as INDX was supposed to become the company’s main revenue driver in 2027.

Adding to that is disastrous communication, including reports from users that critics are being blocked on Discord. This story is still very much developing...

At the same time, we’re staying with Prusa, this time on the software side. The company has introduced subscription plans for its EasyPrint cloud slicer. The tool runs on the Printables platform and allows users to slice models directly in the browser, including from smartphones and Chromebooks.

The controversy began when a user named Dustin shared a screenshot showing a message encouraging him to upgrade to a higher-tier plan in order to process a more complex model. The free tier includes 1 GB of storage and 60 seconds of computation time. The Basic plan costs $8.99 per month (20 GB, 120 seconds), Premium costs $29.99 (100 GB, 240 seconds), with custom Enterprise plans also available.

Josef Prusa responded directly. He explained that the supposedly “simple” model actually contained hundreds of thousands of triangles and placed a significant load on the servers. He reminded users that PrusaSlicer remains the foundation: free, open-source, local, and subscription-free. And that this will not change, and if it ever did, the community could immediately fork the project.

He added that EasyPrint is simply a convenient tool for occasional use, available even to people who aren’t Prusa customers. He also acknowledged that the upgrade prompt was poorly worded and promised improvements.

Now, industry.

EOS has ordered 155 Dual Setup Stations for its M4 ONYX system from Grenzebach. Forty units were ordered before Formnext 2025, with another 115 added after the system’s launch. The stations separate material preparation from the actual printing process, reducing job changeover time to just 30 minutes.

The second EOS story is about hydrogen. The company has joined the MAT-H2 consortium, which focuses on materials and technologies for the hydrogen economy. The consortium brings together industrial and research partners, including Wärtsilä, to address the challenges hydrogen environments pose to metals. EOS will contribute its expertise in metal additive manufacturing and material characterization. Hydrogen is extremely demanding for metallic materials, and additive manufacturing makes it possible to develop and test new solutions much faster.

A nice piece of news from the medical sector. 3D Systems has helped the 3D Medical Applications Center at Walter Reed Hospital secure FDA premarket clearance for its titanium cranial plate system. This is the first implant in history to receive such approval for a point-of-care facility - that is, one manufacturing implants directly at the place of treatment. The personalized implants are intended for active-duty U.S. military personnel suffering head injuries, wherever they are deployed around the world.

Materialise has commissioned a photovoltaic park at its headquarters in Leuven, Belgium. The €1.56 million installation has been fully operational since January 2026. Its 3,200 solar panels are expected to generate around 1,700 MWh of electricity annually - roughly the annual consumption of 600 Flemish households, while reducing CO2e emissions by more than 40 tonnes per year. The project was carried out together with Argona and Induzz3. Its goal is to increase the share of renewable energy and reduce the company’s dependence on external electricity suppliers.

To wrap up this section, a few numbers. VoxelMatters has released its latest report on the polymer 3D printing market for 2025. According to the report, the segment reached $9.45 billion, up 18.7% year-over-year. Hardware accounted for the largest share at $4.36 billion (+20.9%), followed by services at $3.21 billion (+17.3%) and materials at $1.89 billion (+15.9%).

Among individual companies, the report highlights Bambu Lab, which alone generates more revenue than all of the leading industrial polymer AM players combined. HP has returned to growth (+25%), while Stratasys remains the industrial market leader, growing steadily and consistently. Formlabs, Carbon, Axtra3D, and Caracol are also on the rise, as are service providers Materialise, Prototal, and Xometry.

On the other side of the market, the picture is much less optimistic. 3D Systems is facing serious difficulties, BigRep is heading toward liquidation, Nano Dimension has largely withdrawn from the market, and brands such as Anycubic, Elegoo, Arkema, and Photocentric are all reporting declines.

The report covers more than 800 companies across the entire polymer AM ecosystem.

AM Software Report

There’s quite a lot happening in software around simulation and automation. Aibuild has validated its FETS thermomechanical solver using data from Additure’s experiment: a 15-layer wall manufactured using the WAAM process. The GPU-accelerated simulation accurately predicted both the thermal history and the resulting distortions. The calculations took 6.2 minutes on a laptop GPU (NVIDIA RTX 5090), making it possible to generate toolpaths with distortion compensation directly within the CAM process.

Aibuild is also hosting an online event dedicated to Aibuild OS on August 4, 2026. It’s a platform where the user describes their requirements in plain language, while the system automatically operates CAD and CAM software to deliver the final result. Registration is open, and a recording will be made available afterward.

AMIS has unveiled Runtime, a print preparation automation engine that brings programmable control to AMIS Pro. It allows users to script the entire workflow - from import, through nesting and slicing, all the way to export. The goal is repeatability, whether you’re producing a batch of five parts or five hundred. The approach combines the visual interface of AMIS Pro with Runtime’s scripting layer, reducing both the total cost of ownership and the cost per part.

An interesting story from the open-source world. A GitHub user has created FreeCAD MCP, an extension that allows FreeCAD to be controlled directly through Claude AI using plain language. The tool is built on the MCP (Model Context Protocol) and enables geometry creation, editing, and even finite element analysis (FEA). The project was featured two days ago in the RECODE.AM series.

CEE Report

The second edition of the AM Connect conference will take place on September 23–24, 2026, at the Faculty of Polymer Technology of the University of Primorska in Slovenj Gradec, Slovenia. It’s a cross-border event organized as part of the ADDCIRCLES project and co-funded by the INTERREG Slovenia–Austria programme. Attendance is free of charge and on-site only, with organizers expecting between 80 and 120 participants.

From Poland comes news from ROSA3D. The company is launching a filament dropshipping program aimed at online stores, e-commerce sellers, and startups. The model allows businesses to enter the filament market without maintaining inventory or making an upfront investment: the customer handles the sale, while ROSA3D takes care of packing and shipping. System integration is designed to automatically forward orders for fulfillment. The offer is aimed at companies that want to sell filament without tying up capital or spending time on logistics.

AM HR Corner

Elementum 3D has a new CEO. Andrew Crickenberger has taken over the role from founder Dr. Jacob Nuechterlein. Nuechterlein is moving into the position of Chief Strategy Officer while remaining Executive Chairman of the Board. Crickenberger brings more than 30 years of experience in the aerospace and defense industries. Most recently, he served as Chief Growth Officer at NIOA Group, and before that as Vice President of Strategy for the Aerojet Rocketdyne division of L3Harris. The leadership change is the result of a planned succession process.

Zortrax has also announced changes to its management board. On July 24, 2026, Angelina Stokłosa stepped down as CEO, citing other professional commitments. On the same day, the Supervisory Board appointed Rafał Tomasiak, one of the company’s co-founders from 2013, as the new CEO. Tomasiak returns after spending several years away from the company’s day-to-day operations. Zortrax has been undergoing restructuring since 2024 and continues to struggle with delayed financial reports, suspended stock trading, and the controversial issuance of blockchain tokens.

Finally, something that’s only just getting started but has every chance of becoming my favorite part of this series. I’m beginning to track the 3D printing job market.

Below you’ll find charts based exclusively on data from the past week. In future editions, I’ll update them and compare them over time so we can see trends rather than just isolated snapshots. The project is only just beginning - it’s still under construction, so stay with me.

For now, the analysis covers 18 companies, but that number will continue to grow.

This is my debut in The 3D Printing Journal, right in the middle of the summer holidays. I know, not much is happening on our side of the world right now. So let’s make a deal: give this project a chance and stick with me at least until mid-September, when the market usually starts picking up again.

Then again... this is China. Those 14 stories below all happened within just the last five or six days. Nobody takes a break over there.

This is the first version of the format, and it will probably evolve over time. I’m curious myself to see what it’ll look like in two or three months. But for now - China Watch #1.

Bambu Lab: Apple Stores and education

Since the beginning of July, Bambu Lab has entered more than 60 authorized Apple stores across China. According to the company, it is the first consumer 3D printing brand to achieve this. The initial lineup includes the P2S model and PLA filament. Apple is putting 3D printers directly in front of customers who may never have considered them before.

At the same time, the company has launched Bambu EDU, a collaborative educational program focused on developing learning courses for children and families. Content creators will build complete educational programs around models previously published on MakerWorld. The courses will include parent-and-child sessions for children aged 4-9, as well as STEAM classes for ages 9-14. Bambu’s education team will help refine the teaching materials, and the finished courses will be published on the Handy EDU platform with global availability. Creators retain full credit and authorship of their work.

Kickstarter: full-color and record-breaking campaigns

HeyGears has raised more than $10 million on Kickstarter during the first week of its campaign for the G1 series. The company describes it as the world’s first desktop 3D printer combining full-color 3D printing with UV printing. Within the first 24 hours, the campaign surpassed $7.22 million, equivalent to around 50 million yuan. There are still 43 days left before the campaign ends.

Another campaign making headlines comes from Sovol. Its M1D multicolor desktop printer has already attracted more than $2.7 million from 1,710 backers, far exceeding its original $190,000 goal. At the heart of the machine is the hybrid DualX architecture: one fixed nozzle combined with six interchangeable print heads, allowing color changes in about five seconds while minimizing material waste. The Super Early Bird price starts at $1,299.

New consumer printers and hardware

Creality has unveiled the Ender-3 V3 Mega, priced at 3,399 yuan (~447,8 EUR). The printer offers a build volume of 420 × 420 × 420 mm and supports flexible TPU filaments with Shore hardness between 85A and 95A. Features include 81-point automatic bed leveling, vibration compensation, filament runout detection, and power-loss recovery. Remote monitoring is available through the Creality Cloud app.

ELEGOO has launched the H1 HT filament dryer. It accommodates standard 1 kg spools and reaches temperatures of up to 85°C, making it suitable for moisture-sensitive materials such as TPU and nylon. The dryer connects directly to the printer, features a 3.5-inch touchscreen, and includes 11 preconfigured material profiles. It is priced at $70, with shipping scheduled to begin in late August.

Materials

eSUN, the filament brand owned by Guanghua Weiye accelerated its product development during the first half of 2026. New releases include PLA-Coffee, lightweight PEBA-LW foam, photochromic PETG, upgraded ABS+, and medical-grade PLLA. The company generated 712 million yuan (~93,8 milion EUR) in revenue in 2025, an increase of +31.05% year over year, while sales of its 3D printing products grew by +40.25%. It has filed more than 100 patent applications and has been granted 96 patents.

Another company entering the consumer market is Shenzhen-based Shengjialun, a well-known manufacturer of TPE materials. The company is introducing TPE filaments designed for toys and everyday consumer products. For more than 20 years, it has supplied food-grade and child-safe soft materials to brands including Hasbro and McDonald’s, and is now expanding into the FFF 3D printing market. Two dedicated brands have been introduced: Xiantiao Bodian for children’s filaments and Shengjialun for flexible materials.

Exports and macro data

Guangdong province has released its economic results for the first half of 2026. Regional GDP grew by +4.5%, reaching 7.23 trillion yuan. Production of 3D printers in the province increased by +51.8%. Nationwide, output grew by +48.5%, while exports reached 3.62 million units, up +90.2% year over year.

It’s worth noting that the 90.2% figure refers to the number of units exported - the shipment volume. The monetary value of those exports is a separate metric and typically grows at a slower pace than unit sales. Shenzhen remains the country’s primary manufacturing hub, serving as home to companies including Creality, Bambu Lab, ELEGOO, and Anycubic.

Industrial segment

Construction continues on Shangpin Huayun’s new facility in the Fengjing Industrial Park, Huyi District, Xi’an. The 350 million yuan investment covers 16,735 m² of land and 29,240 m² of floor space. The facility will provide ceramic composite forming services, develop ceramic 3D printing equipment, and offer metal additive manufacturing services. Structural work has already been completed. Construction is expected to finish in October 2026, with full-scale production planned for December.

Xinjinghe Laser Technology Group, a manufacturer of additively produced aerospace metal components, has put a 3.2334% equity stake up for sale. The minimum asking price is 170 million yuan (approximately €22.4 million), implying a total company valuation of roughly 5.26 billion yuan (around €693 million). The stake is being sold by Yuhuan State-owned Investment. Founded in 2015, the company generated 533 million yuan in revenue and 39.45 million yuan in net profit during 2025. In December 2025, it officially began preparations for an IPO on China’s A-share market.

Finally, two noteworthy patent applications

Zhejiang Feiyue Group has filed patent CN122425892A covering a multi-nozzle switching system. The design enables automatic nozzle changes through a movable printhead base and independently mounted nozzles, simplifying multi-material printing and on-demand nozzle replacement.

Meanwhile, Xiaomi together with Anhui Zhongti New Materials, has filed patent CN122466309A for a titanium alloy powder intended for additive manufacturing. The material combines aluminum, vanadium, nitrogen, and titanium. The application was submitted in June 2026, and the project itself suggests Xiaomi is exploring the use of metal 3D printing in automotive component production.

That’s all for the first edition.

Let me know what you think. Is this the right direction, or is there something you’d like to see changed?

Something that would have sounded like yet another overhyped startup pitch destined to go bankrupt, has now become reality: you can control FreeCAD directly from Claude. You open a chat, describe what you want to create, and the software builds it.

It’s called FreeCAD MCP, created by GitHub user neka-nat. And yes - it actually works.

MCP stands for Model Context Protocol: a standardized way for AI to communicate with external applications. Think of it as a common language that lets a model issue commands to software.

There are countless MCP bridges appearing these days. One of the newest connects directly to FreeCAD.

Installation is surprisingly simple. Copy a folder into the /Mod directory, restart FreeCAD, and a new workbench called MCP appears in the interface. One click enables it, another starts the RPC server. Claude immediately recognizes the application and lists everything it can do.

And that’s where things get interesting, because the list includes much more than just “create an object.” It can also launch finite element simulations. I’ll come back to that, because it’s easily the strongest feature of the entire project.

But before anyone declares that CAD designers are no longer needed - hold on!

For simple parts, it works remarkably well. For complex designs, you still need someone who actually knows FreeCAD and can step in when the AI gets lost.

Still, the direction is unmistakable. The barrier to entry for CAD has just become significantly lower.

So how do you get it running?

FreeCAD is a free, open-source CAD package. It has always been incredibly powerful, and unfortunately, notoriously unfriendly to beginners.

Get FreeCAD

Everything is built around MCP (Model Context Protocol). It’s a standard that gives AI a predefined “grammar” for interacting with software. The model knows which commands exist and how to call them. FreeCAD gains this capability through an add-on created by GitHub user neka-nat.

Get FreeCAD MCP

To get everything working, you need to:

1. Copy the addon/FreeCADMCP folder into the /Mod directory (the exact location depends on your operating system).

2. Restart FreeCAD.

3. Select the MCP Addon workbench.

4. Click Start RPC Server in the toolbar. You can also enable Auto-Start Server, which stores the setting in freecad_mcp_settings.json and automatically starts the server whenever FreeCAD launches.

5. On the Claude Desktop side, you’ll first need to install uvx and make a single change to the claude_desktop_config.json file.

That’s it.

After restarting Claude, it detects FreeCAD automatically and lists all available commands. And there are plenty of them.

It can create and edit documents and objects, delete them, insert parts from libraries, capture screenshots of the active viewport, inspect the scene, and much more. It can also execute arbitrary Python code inside FreeCAD - which, in practice, means the ceiling is as high as FreeCAD itself.

I saved the best part for last…

The command is called run_fem_analysis.

It launches the CalculiX solver on an existing finite element analysis and returns actual engineering results: maximum von Mises stress, maximum displacement, and the number of mesh nodes.

If the analysis doesn’t already include a solver, the add-on inserts one automatically. Which means the AI doesn’t just draw geometry, but it can actually verify whether that geometry is likely to survive the intended load.

Not long ago, this required two separate applications and half a day of clicking.

So how well does it actually work?

Early public tests reveal a clear pattern. Simple geometries are almost instantaneous.

A cube? No problem.
A shelf or a simple mounting bracket? Also easy.

Things become more challenging as complexity increases. The AI may omit details - holes that were supposed to be there simply disappear, or invent features that nobody asked for.

The good news is that it improves when you point out the mistakes.

The catch is that someone still has to recognize those mistakes in the first place, and ideally that person knows FreeCAD.

Which brings me to conclusion:

Why would anyone want CAD controlled with plain English? Because we’re watching the design interface evolve into natural language.

A simple sentence typed into a chat window can now accomplish what used to require a mouse, dozens of toolbars, and layers of menus.

I’ve been saying this for years: in additive manufacturing, hardware alone is no longer enough. Competitive advantage is shifting toward software.

FreeCAD MCP is a small but tangible example of exactly that. Nothing about FreeCAD itself had to change. The innovation lives entirely in a software layer added on top of an existing application.

And thanks to that layer, a mature and incredibly capable CAD package suddenly becomes accessible to people who’ve never touched a sketching tool before.

P.S. This isn’t limited to Claude. You can connect the same MCP server to ChatGPT (via Codex) as well. Kerry Stevenson from Fabbaloo did exactly that and documented the results. It’s well worth reading.

Last Friday, on July 24, Angelina Stokłosa resigned from the Management Board of Zortrax S.A., citing other professional obligations.

On the same day, the Supervisory Board appointed Rafał Tomasiak as CEO, while simultaneously terminating his previous independent commercial proxy authority.

His term runs until the end of 2026.

In 2013 Tomasiak co-founded the company and the Zortrax brand. After spending the past few years largely behind the scenes, he recently returned to lead the newly established Warsaw-based company, Zortrax Defence.

In 2017, Tomasiak brought Mariusz Babula into the company as Head Of Investor Relations, and Babula later replaced him as CEO in 2021. Babula stepped down in August 2025, and now, after Stokłosa’s brief tenure, the position returns to the removed co-founder.

In parallel, Zortrax is reorganizing its capital structure. On February 24, 2026, an extraordinary general meeting approved a 10-to-1 share consolidation, and on March 18 the registry court entered the amendment to the company’s articles of association: the share capital amounts to PLN 22,021,025 and is divided into 22,021,025 shares with a nominal value of PLN 1.00 each.

The operation is still awaiting its technical entry in the National Depository for Securities, so the share price remains at the old level.

On July 28, 2026, Zortrax shares traded at PLN 0.0485, or just under five groszy.

At the current exchange rate, that is approximately EUR 0.011 and USD 0.013, meaning in both currencies the share is worth only a little over one cent.

No change in nominal value will improve that valuation.

Since 2024, the company has been undergoing restructuring, has experienced delays in publishing its annual financial report, and its shares have been suspended from trading as a result. It has also sought financial rescue through various measures, including the controversial issuance of blockchain-based tokens.

The last in-house FFF printer, the Endureal, was unveiled in 2019; the last printer of any kind, the resin-based Inkspire 2, appeared in 2021. Both launches came before the rise of Bambu Lab and the market shift that has since left the consumer segment dominated by Chinese manufacturers.

In the background are two episodes from the company’s past. In 2014, Zortrax announced a contract for 5,000 printers for Dell. More than two years later, it turned out that no binding agreement existed, only a preliminary arrangement with Dell’s Indian branch, which the company did not have the capacity to fulfill anyway.

Five years later, in 2019 Formlabs attempted to buy Zortrax. The deal was organized by Babula, and according to his account, Tomasiak ultimately refused.

Today Zortrax, once the largest and most profitable Polish additive manufacturing company, is still fighting for survival.

In less than a month, three American 3D printing companies have disappeared from the market, each in its own way.

California-based 3DEO entered insolvency proceedings. Its patents, software, and metal sintering systems are being sold as a single package, while the rest of its manufacturing equipment is headed to the auction block.

North Carolina’s Fusion3 has been liquidated. Its printers and spare parts are being auctioned off by a Greensboro auction house: 15% buyer’s premium, credit card pre-authorization required to register.

Würth Additive Group announced the “orderly wind-down of all operations” in a plain LinkedIn post. Around fifty words. No explanation beyond a “carefully considered business decision.”

Three companies. Three states. One direction.

The United States is the country that invented 3D printing. In 1986, Chuck Hull and Ray Freed founded 3D Systems, the world’s first commercial 3D printing company. Soon afterward came Stratasys, DTM, Z Corporation, Solidscape, and many others.

Now, forty years later, America can count the manufacturers that are truly doing well on one hand. The rest are either niche businesses or forgotten entries in the registry of dissolved companies.

Yet despite the lamentations of commentators mourning America’s defeat to China and claiming it has surrendered the entire industry, I’m nowhere near writing the obituary for American additive manufacturing.

Because what we’re witnessing is nothing more than the next stage of market evolution: consolidation. The exact same process that transformed the computer industry, the paper printer industry, and the mobile phone industry.

At the same time, this is nothing more than a reality check for a country. Sometimes, despite the best intentions, effort, and talent, you simply cannot build something people aren’t ready for - or don’t actually need or want.

A few years ago, I went through exactly the same process in Poland.

Yes, it was painful. Pretty disappointing. And sad.

And no, there was nothing anyone could do to stop it.

American AM

Let’s start with the ones actually doing well:

The two pillars are standing where they’ve always stood. In Q1 2026, 3D Systems posted roughly $95 million in revenue, with double-digit growth in dental, medtech, and defense. Stratasys closed 2025 at $533 million. Neither is near its old peak. Neither is close to bankruptcy. They’re not going anywhere.

Then two giants who keep dabbling in 3D printing: HP, with MJF and Metal Jet, and GE, running its metal arm as Colibrium Additive. Both are serious names on the AM scene and rounding errors on their parents’ balance sheets.

Beyond those four, two companies are actually thriving. Formlabs, the only one that walked out of both the consumer boom and the “mass production” era intact, holds around 55% of the global SLS market. Carbon prints over 1 million end-use parts a week and reached positive cash flow without raising fresh money.

Velo3D survived, but with an asterisk: it came within touching distance of bankruptcy, restructured, and today runs on defense contracts.

After that, the niche: B9Creations in dental and casting resins, Vision Miner in high-temp PEEK, Xact Metal in low-cost metal LPBF, and LulzBot, still shipping out of Fargo.

And below them, a long bench of specialists quietly moving hardware: Mantle, Optomec, and Sciaky in metal; Thermwood and re:3D in large-format polymer; Boston Micro Fabrication, Fabric8Labs, Impossible Objects, Intrepid Automation, Inkbit, and Axtra3D in the specialty corners. Real companies, real machines, none of them big enough to change the story.

Now the second list…

MakerBot, the company that put 3D printing on newspaper front pages between 2010 and 2013, got absorbed into Ultimaker and is quietly fading out. Printrbot died in 2016. Solidoodle, Type A Machines, Robo3D, M3D, Airwolf3D: gone, or shrunk to a garage and a website.

Desktop Metal, Markforged, Nexa3D. Please don’t get me started again 😡

These were the companies that drove the growth of 3D printing in the US and worldwide.

Today they’re footnotes.

Why the market keeps growing while companies keep dying

This is where one of my Three Laws of the AM Market comes into play once again:

The 3D printing market continues to grow. Most forecasts expect annual growth in the double digits for years to come. And yet companies continue to disappear.

That’s the rule, not the exception.

There have always been far more players than the market could realistically support.

At any given moment, only a handful make real money. The rest hover around break-even or below it. Every five or ten years the deck gets reshuffled. Leaders become laggards. Some disappear entirely.

The market keeps growing anyway.

We’ve already seen exactly the same movie elsewhere.

In the 1980s and 1990s, dozens of companies built personal computers. Today, the meaningful players can be counted on one hand. Paper printers? Once crowded with brands. Today, a handful divide the global market. Mobile phones? Same story.

Markets mature. The crowd thins out. The elite survive, along with a few niche specialists.

And yet, nobody writes an obituary for “computer technology.”

All the companies that survived all had one thing in common: they had a reason to exist. Customers kept coming back. They had a moat.

In 3D printing, that moat is now software and low manufacturing costs. The survivors sell an ecosystem or a unique, often certified solution.

The companies that failed sold a printer. And that was the end of the story.

That’s why I look at America’s growing list of casualties without much emotion, because I know this is the natural order of things.

Bloated ambitions. Founders with oversized egos. Financial projections and sales forecasts that made the universe of Peppa Pig look brutally honest and true.

These companies collapsed because they had to. For years they survived on borrowed money, feeding themselves on fantasies straight out of the most optimistic investor pitch decks.

Then reality arrived.

Business models built entirely on someone else’s patience and someone else’s money died one after another.

That’s all there is to it. There’s nothing new about it. Nothing surprising.

I watched exactly the same thing happen in Poland a few years ago.

Did you know that we once had more than thirty manufacturers of 3D printers?

Poland: the fallen “King of 3D Printers”

Between 2014 and 2018, Poland was home to more than 30 manufacturers of 3D printers (in fact, it was closer to forty).

Ready?

Zortrax, Zmorph, Sinterit, 3DGence, Omni3D, Hbot3D, Verashape, Monkeyfab, UBOT 3D, 3D Proto, 3D Universal, 3DDDBot, 3dimension, 3DKreator, 3NOVATICA, 7H7, Atmat, Aye Aye Labs, CB Printer, CORO Technology, Deus, Dexer, Dragon 3D, Endivio, Golem3D, Graften, Infinum 3D, Idea Lab, Jelwek, Lume 3D, Paramid 3D, Pirx3D, Sandmade, Totem 3D, Tytan 3D, Urbicum.

And I know there were at least another three that I’ve simply forgotten.

Now add more than 10 filament manufacturers and 2 producers of 3D scanners.

We were supposed to become Europe’s 3D printing powerhouse.

Why were there so many?

For exactly the same reason as everywhere else: RepRap, expiring FDM patents, cheap hardware, and the belief that if you could assemble a printer in your garage, you could build a company around it.

And you could. For a year or two.

Right until the first serious wave of warranty claims, or the moment you had to scale beyond hobby-level production.

Most of those thirty-plus companies were selling exactly the same thing: a local variation of an open-source design, with no proprietary materials and no market beyond Poland.

Unlike the Americans, we didn’t have deep pockets, so our collapse simply happened faster. Today, only a handful remain.

Sinterit in Kraków continues to push forward in SLS - although it seems to require another financial lifeline from Germany that owns it, every few years. Omni3D remains focused on large-format systems. Ubot3D quietly operates in its niche.

And then there’s the curious case of 3DGence, which now sells HP printers under its own brand, even though neither company has publicly explained exactly how that arrangement works.

Everyone else on that list is either gone or has reinvented themselves into something completely different.

It’s the same curve America is drawing today - just on a larger scale.

The Polish aftermath

Alright, so Poland once had all those manufacturers. And we lost almost all of them.

Truth is, most of them were failures anyway. Maybe five companies ever reached any meaningful scale, and only two - Zortrax and Sinterit, left a real mark on the global industry.

So what happened next? Did 3D printing in Poland die? No domestic manufacturers = no 3D printing?

Well, quite the opposite.

After years of barely scraping by, it’s only now that you can honestly say there’s real money to be made in Polish additive manufacturing.

Companies stopped building 3D printers and they started using them instead.

Yes, those printers come from China. They keep coming from China.

But additive manufacturing itself is booming.

Poland has never sold as many 3D printers as it does today. Every segment is growing: consumer, industrial, medical, education... you name it.

Whether that’s a good thing that the growth is based on Chinese technology, is a completely different debate. Sure, in theory it would be better if Polish companies were using Polish-built 3D printers.

But the moment you start grounding that idea in reality - looking at the history of individual manufacturers, the business decisions they made over the years, and why they ultimately failed - you quickly realize you’re no longer talking about reality.

You’re talking about wishful thinking. About what we’d like to be true, not what actually is.

The uncomfortable truth is that every failed Polish 3D printer company was ultimately brought down by its own founder or CEO.

I write that with genuine sadness. I know those stories. Every single one of them. But I also write it because it’s true.

And despite all of that, the industry kept growing. It still is. Nothing is stopping it.

I realize that, for a proud American engineer, this probably sounds close to blasphemy. It will probably make some people hit “unsubscribe” before it makes them stop and think.

But then again... You don’t subscribe to this newsletter because you want to be told everything is great, do you?

Here’s the reality. This is simply how markets work. And there’s very little anyone can do to change it.

The market is doing what it has always done: thinning the herd, rewarding companies with products that genuinely matter, and lowering the barriers to entry for the next generation.

America is still learning that lesson. We learned it years ago in Poland.

We paid our tuition in hard currency - thirty-something companies.

Only then were we able to reach the point where we are today.

That’s the way it is...

In an industry that announces a revolution roughly every two weeks, changing the format of a newsletter is hardly headline news. But if you’ve stuck with this series through 330 editions, you deserve an honest explanation of why we’re making the change.

So here’s the honest answer: Atomic Layers stopped being fun for me in its current format. It needed a refresh.

Alongside 3DP War Journal and RECODE.AM, Atomic Layers has been the third pillar of The 3D Printing Journal. It launched in August 2024 and ran every single day for the next 300 days.

Then came a six-month hiatus, forced by my contractual commitments to Bambu Lab Europe, followed by its return earlier this year as a weekly publication.

The structure remained the same: a main story of the day (later, of the week), a historical note, and a roundup of other news and industry gossip.

Back in 2024, this was something new in the AM industry, and it worked.

It also had one flaw that only becomes apparent after dozens of editions: publishing every day slowly eats you alive, while the 3D printing industry simply doesn’t produce enough meaningful news to justify sending an email every single day.

Switching to a weekly schedule seemed like a sensible compromise, and for several months, it was.

But then... well... uhm…

Freshness gradually gave way to routine, and routine is where media projects begin to die. I went through exactly the same process between 2021 and 2023 with my Polish publication Centrum Druku 3D, which I eventually sold.

I hadn’t reached that point here yet.

But I could already see it on the horizon.

So I’m changing everything at once.

What’s changing

is taking over the stewardship of the series.

Many of you already know my wife from The 3D Printing World Guide and several other projects. She has spent well over a decade in additive manufacturing and helped me build the foundations of the Polish AM market back when the phrase “3D printer” in Poland was still a curiosity rather than a machine on a factory floor.

We’re splitting the newsletter into two publication days and several dedicated sections.

Every Thursday: China Watch

This is an addition the industry needs the most.

Think of it as Atomic Layers focused exclusively on the Chinese market.

Pay attention to the word exclusively, because we’re not talking about Chinese stories that already have global relevance. Those will eventually reach Western media anyway - usually with a two-week delay.

We’re talking about local news.

What’s happening inside China’s domestic AM market. The companies, technologies, and market segments nobody in Europe is talking about yet.

We have good sources. We’ve decided it’s time to start sharing them.

Every Friday: Western Report

An expanded version of today’s Atomic Layers, divided into dedicated sections.

• Mainstream Report - the biggest news of the week, just as before.

• AM Software Report - a sector where over the past two years, more has happened than in hardware, despite generating only a fraction of the noise.

• CEE Report - Central Europe is a region consistently ignored in global industry coverage, despite the fact that an extraordinary amount is happening here.

• AM HR Corner - this in my opinion, is the single most important part of this entire overhaul.

AM HR Corner will cover recruitment, layoffs, and executive moves across the Western side of the industry. Our industry loves talking about technology, yet stays remarkably quiet about who just lost their job - even though, over the past three years, workforce movements have been the most honest indicator of a company’s actual health.

Layoffs in the R&D department tell you more about a company’s future than its roadmap.

A departing CTO tells you more than a quarterly earnings report.

So we’re going to start writing about it.

But be a little patient, ok… 🥺

The first few editions will inevitably involve a bit of trial and error. We’re launching in the middle of the summer holiday season, which is such a poorly timed decision that it’s almost refreshingly honest.

But waiting until September would mean introducing the change when it was convenient - not when the decision was actually made.

So I ask for a little patience.

By autumn, everything should be running like a well-oiled machine.

And if it isn’t...

You’ll read about it in AM HR Corner 😅

This is the last Atomic Layers of its kind.

Starting next Thursday, we’ll see each other in its new form.

Atomic Layer from the Past:

7 years ago, BASF launched Ultrafuse 316L: steel powder in a polymer binder, spooled for ordinary FDM machines. IGO3D, MatterHackers and Ultimaker distributed first; Prusa, Raise3D, 3DGence and BCN3D followed. You print it normally - then you need catalytic debinding and a sintering furnace, neither of which fits on a desk.

The product outlived its maker’s name: BASF 3D Printing Solutions became Forward AM, which went independent in July 2024. But only to declare insolvency a year later, and got acquired by Stratasys.

Read all:

News & Gossip:

#1

This week’s top news is once more Nano Dimension and - damn, you won’t believe it - another CEO swap. Murchinson - damn, you won’t believe it - takes control again and installs its own people on the board.

My daughter Roksana wrote about it here; now the update: Nano Dimension has appointed Moshe Rozenbaum as Interim Chief Executive Officer.

I’d like to write something more, offer congratulations and stuff, but I’m afraid it’ll all be out of date again in a moment.

#2

Beehive has announced a $70 million Southwest Ohio expansion, unveiled at Farnborough with Governor DeWine, JobsOhio and REDI Cincinnati in attendance.

The project is expected to create more than 200 jobs. But what makes it even more interesting is the sequence of investments: 30 EOS M 400-4 ONYX systems last month (believed to be the largest single metal AM purchase from EOS ever), followed by two Nikon SLM NXG 600E machines, and then the acquisition of the assets of Able Tool and Planet Products to establish production-ready machining capabilities.

#3

And here are the winners of the great AI gold rush!

Meshy has raised nearly $400 million in Series B, valuing the generative 3D model company at $1.5 billion. Text, image or sketch goes in - FBX, OBJ, GLB and STL come out.

The claim is that up to 97% of generated models process successfully in slicing software.

Which slicers, which model classes, under what test conditions, unspecified

Also, Meshy reports twelve million registered users, 100 million models, and ARR up roughly twelvefold year-on-year.

Yeah. $400 mil… Nice one.

#4

HeyGears G1 Series launched on Kickstarter! Full-color 3D, UV texture reliefs up to 5 mm, and 2D direct-to-object printing at 1440 x 2400 DPI across 400+ substrates, in one box.

The G1X uses an Epson i3200 head, water-soluble supports, a 420 x 330 x 130 mm build volume, and starts at $3299 against a $4999 MSRP. The hardware looks serious. If it really works as advertised, SSYS may have another another worry to add to the list…

#5

Sinterit starts reselling DyeMansion's VX1 vapour smoothing system and the DM60 colouring unit alongside its BIANCO2 SLS platform. Good and safe choice. I only wish it could help…

#6

And finally, some tough drama from the wonderful world of trüe makers and open-source fundamentalists. I won’t get into the details here - just click the links below and enjoy the read...

• https://www.reddit.com/r/prusa3d/comments/1uzdl5p/so_uhh_bondtech_outright_lied_and_just_did_a_bait/

• https://www.reddit.com/r/prusa3d/comments/1v45o7l/indx_update_from_bondtech/

Creating a texture on 3D model isn’t an easy task...

Parametric CAD won’t do this. Blender can, but it comes with its own language, its own logic, and UV unwrapping kills motivation faster than the tutorial finishes.

In April, well-known content creator Stefan Hermann (CNC Kitchen) addressed this problem by creating BumpMesh App, and making this browser-based tool available for free.

Now a second app from the same family has joined it.

PunchForm was created by Ozarkexpeditions, a MakerWorld model designer with a background in design and software engineering.

The app is free, open-source, runs entirely in a browser window, and provides roughly 100 ready-made textures.

And it has a feature everyone needed most: a separate texture for each surface of the model within a single session.

Brick on one wall, plaster on another, a smooth engraving area on a third. One file, one output to the slicer.

CLICK: PunchForm

In practice, that one capability changes what a surface means in a project.

PunchForm’s geometry engine comes from BumpMesh. What’s new is everything around it: the interface and the entire workflow for interacting with the model. Most of the texture library comes from ambientCG materials.

You drop in an STL, OBJ, or 3MF file. Pick a texture from the list, paint the area that should receive it with a brush, and see the result immediately in the preview. Then move on to the next surface and the next texture.

All processing happens locally inside the browser. The model never leaves your drive, and you can even self-host the application if you want.

BumpMesh offered 24 patterns and one texture per cycle. Want a second texture? Export the STL, re-import it, apply the next texture, and export again. Hermann himself identified that loop as the tool’s main inconvenience.

PunchForm closes that loop in a single session, with a hundred textures available!

What is it useful for?

Trophies and awards: rough body, smooth area for engraving or a nameplate.

Architectural models: brick, plaster, roof tiles, and concrete on a single structure, each wall treated independently.

Tabletop and wargaming terrain: rock, wood planks, dirt, and sand without sculpting in Blender.

Handles and grips: knurling where you hold the object, smooth where you don’t.

Displays and promotional gadgets: textured background with a clean area for a client logo.

Lampshades and light diffusers: a pattern on the wall changes how light is scattered.

Molds and formwork: the texture is later transferred into concrete or resin.

The common denominator across that list is simple: in every one of those cases, a single texture across the entire object would be a compromise.

Texture in 3D printing is often treated as decoration applied at the very end. The ability to assign it separately to each wall moves it into the set of core design decisions, right alongside wall thickness and build orientation.

Because every surface has a different job.

Where you grip the object, knurling or a coarser grain helps. Where engraving, printing, or a label will go, you want a smooth finish. On the sides where seams meet and layer lines are visible, a fine texture hides both more effectively than almost any slicer setting. On the bottom, you usually want nothing at all, because bed adhesion matters more than appearance.

Stefan Hermann demonstrated something less obvious in his BumpMesh video: a properly chosen texture can actually improve the strength of a print.

The surface is therefore working on several levels at once, and PunchForm gives you control over that wall by wall.

Working with the program is extremely intuitive, and honestly it’s hard to justify writing a tutorial for it. Everything feels smooth, with one minor issue I noticed: when rotating the model excessively and moving it around the screen to reach mesh recesses, the application occasionally jumps back to the start screen 😖

Fortunately, returning to the interrupted edit is available with a single button, so I’d call it annoying rather than disqualifying.

Besides, come on... it’s a free browser-based app. Respect! 🫡

Directors and executives Pons, Rosensweig, Sriubas, and Stehlin resigned from both the board and all positions within the company.

They were replaced by three candidates previously nominated by Murchinson for the same shareholder meeting: Fruchthandler, Rozenbaum, and Tarlow. Murchinson currently controls approximately 7.4% of Nano Dimension’s outstanding shares.

David Stehlin’s resignation deserves particular attention, as he was serving as the company’s CEO and stepped down from all of his positions.

The announcement does not identify a successor or an interim CEO, nor does it disclose the full composition of the newly restructured board.

As a result, it remains unclear who is currently leading Nano Dimension or even how many directors now serve on its board.

What is clear is that this marks yet another leadership change since Murchinson removed Yoav Stern from the board on December 26, 2024. Stern was followed by interim CEO Julien Lederman and later by Ofir Baharav before Stehlin assumed the role.

The immediate catalyst for the dispute was Nano Dimension’s proposed transaction with Infinite Epigenetics, a company developing an early disease detection platform that owns the certified laboratory TruDiagnostic and Tally Health, as well as a database of more than 120,000 biological samples and a platform capable of extracting over one million epigenetic signals from a single test.

Nano signed a non-binding term sheet valued at $890 million, proposing the creation of a combined company called Infinite Epigenetics under the Nasdaq ticker **IEAI**. Under the proposed structure, existing Nano shareholders would receive a 20% premium over the company’s net cash value while retaining a minority stake in the combined business. Nano’s shares fell 15% on the day the announcement was made.

Murchinson described the proposal as “a SPAC in disguise,” arguing that shareholders would not even receive the standard right to redeem their shares for cash. The fund also questioned what expertise Nano’s board possessed to evaluate an epigenetics company, what financial benefits board members stood to receive following the closing of the transaction, and why the company agreed to a termination fee of up to $10 million, along with specific performance and best efforts clauses.

Murchinson also pointed out that retired General Michael X. Garrett served as an advisor to Infinite Epigenetics while simultaneously sitting on Nano Dimension’s board when it unanimously approved the acquisition of Desktop Metal.

That connection may be the most significant element of the entire dispute.

Desktop Metal filed for bankruptcy just 117 days after the acquisition closed, and its key assets were subsequently sold for only a fraction of Nano’s purchase price.

For fiscal year 2025, Nano Dimension reported $139 million in impairment charges and a net loss of $293 million. In April 2026, the company sold AME and Fabrica to Inspira Technologies for up to $12.5 million. On May 27, it also sold Markforged to Stratasys for $42.5 million, representing just 36% of the price it had paid one year earlier.

The July 17 settlement agreement makes no reference whatsoever to the future of the Infinite Epigenetics transaction.

Formally, it remains unknown whether the deal has been cancelled, suspended, or is still moving forward.

What is known, however, is who now controls its fate: the very individuals who only weeks ago publicly described it as value-destructive.

Nano Dimension is therefore entering yet another chapter with a board appointed by a minority shareholder, no CEO, approximately $500 million in cash, and its Nasdaq listing as its only truly significant remaining assets.

If the new board decides to terminate the Infinite Epigenetics term sheet, it will first need to determine the cost of exiting an agreement its predecessors signed with a termination fee that could reach $10 million.

Source: www.nano-di.com

At every AM trade show, there was always at least one booth like this.

Behind it stood an odd-looking guy who had spent the last four years working on something nobody had managed to do before. His machine could process a material with properties no competitor could match. Or it offered a process that finally bypassed a fundamental mechanical limitation.

The sample prints sat inside generic IKEA display cubes, lit from below, on black backgrounds. The machine itself wasn’t there (because there was only one, and it was sitting in a lab).

People stopped and took pictures, and after a short conversation, they were genuinely impressed.

Someone would say it was going to change the entire industry. Someone else would ask for a sample to show their boss.

Right after the trade show, the LinkedIn recap post collected 200 reactions and 40 comments (half of them replies from the exhibitor himself).

And that was it.

The inbox was still waiting for RFQs. The meeting calendar was still wide open.

The inventor explained it to himself by saying the market wasn’t mature yet, and customers first had to be educated.

„Just wait another two years. Then everyone will get it” – he repeated to himself...

The problem was that two years later the company was gone, and the trade show samples had either ended up in a landfill or in a box of old souvenirs.

All because the inventor had made one fundamental mistake from the very beginning.

He answered a question nobody had asked.

In 3D printing, that single mistake has killed more companies than price pressure and Chinese competition combined.

Hundreds of projects that impressed everyone while solving nothing.

Technologies with value locked inside themselves, adding no value to anyone outside.

Admiration is a currency you can’t book

The additive manufacturing industry is made up mostly of people who invent and develop technologies.

Alongside them is a small group of people who know how to turn someone else’s invention into an actual business. And an even smaller group who managed to do both.

That last group consists of just a handful of names everybody knows.

Those proportions have financial consequences, because the first group tends to think in a very particular way.

The inventor’s world is the workshop, the laboratory and the factory floor. His product is the sum of his work, his expertise and several years of his life.

And because it is the sum of all those things, its value seems obvious to him. Not just to him, but to the entire world.

His product is both his vision and an advertisement for himself. His product is supposed to be an argument.

The problem is that every single time, it becomes a pretentious monologue.

But the end customer - the one who is supposed to buy the technology - doesn’t have time to listen to things that sounds to him like mumbling, geeky nonsense.

Because the end customer is judged by delivery dates, unit costs and the number of customer complaints.

And yes, he already has a supplier who gets the job done. That supplier doesn’t have to be exceptional. Average is good enough, as long as he’s predictable. And in manufacturing, predictability is a harder currency than quality or innovation.

A new technological capability is a cost. It has to be tested, documented, built into procedures, approved and defended in front of management.

And if something falls apart during the tenth production batch, it’ll be the customer explaining himself - not the inventor.

The technology may be innovative, and be impressive.

But the admiration is free. Implementation often costs more than the machine itself.

Innovators die in silence

So the innovator sits there, waiting for the breakthrough that never comes. The grant money runs out. The investment dries up. Eventually, he shuts the business down.

Usually, there is no farewell post. No press release. There is, however, plenty of bitterness and frustration aimed at the world.

After every failure, the same explanation comes up:

The market wasn’t ready.

It’s a convenient explanation, because it makes the end user of the technology the one to blame.

The real story is much simpler.

The product simply wasn’t needed by anyone, for any reason.

The inventor’s dilemma boils down to a choice between inventing and listening to the customer.

You can have objectively the best technology on the market and spend years waiting for the world to appreciate it.

Or you can simply ask someone what hurts today and build the answer.

The first path brings awards and recognition.
The second brings wire transfers.

There is, however, a narrow path that this industry has walked only a handful of times in its entire history.

It’s worth looking at what those cases have in common, because the conclusion is uncomfortable for everyone.

What actually made it into mass production

There are only a handful of 3D printing applications that have genuinely reached industrial scale:

• patient-specific medical devices; prosthetics and orthodontics

• production tooling

• parts for aerospace, automotive, maritime and rail

• defence

• energy.

And don’t forhet millions of colorful toys made on giant farms of desktop 3D printers.

Every one of those applications came to life the same way. Someone found a single use case that made financial sense, then repeated it hundreds of thousands of times, improving small details with every iteration.

None of them won a Nobel Prize. They won something else instead: financial stability and growing revenue.

Funny enough, everything behind those success stories sounds like a punishment when you write it down:

• process qualification

• documentation

• material validation

• training operators - including the ones who will leave after a year and have to be replaced

• handling complaints

• audits

• building a customer’s trust over eight years, and losing it because of one bad production batch.

This is the part inventors hand over to somebody else, or skip altogether, because this isn’t what they went to engineering school for. And yet this is exactly the part that decides whether a technology makes money.

Customers don’t care how ingenious their supplier is - they care about the outcome.

Shorter lead times. Lower costs. Fewer phone calls in the middle of the night.

Innovation means nothing to them until it cuts delivery times by days or weeks.

The conclusion is uncomfortable for anyone who fell in love with this industry because it was fascinating.

It has to make ordinary, completely unremarkable everyday products. The kind nobody picks up and asks, “How was this made?”

The product should be remarkable - the manufacturing process should remain the manufacturer’s private business, preferably a closely guarded one.

The less you see the words ”3D printed” in the sales message, the better the result.

This technology is still moving forward, faster than most people realize. Its future, however, fits into one rather unimpressive sentence:

Make it ordinary.

Three weeks ago, I included this note in NEWS & GOSSIP:

“And finally, a hint at a bigger story still developing. Two prominent Würth Additive figures have announced they are looking for new jobs, while letting you read between the lines that the company itself is done. Nobody, however, wants to confirm any of this on the record yet. So for now it sits somewhere between rumor and an unconfirmed fact. Keep watching.”

Well, I kept watching. And here it is…

Würth Additive Group, Inc. has officially confirmed it is shutting down operations.

The announcement came from the company’s headquarters in Greenwood, Indiana, and appeared on LinkedIn as a brief post.

About 50 words 😣

The company says it is conducting an “orderly wind-down of all operations” and directs anyone with an active account or unresolved business to a Closing Office at co@wurthadditive.com.

No explanation was given, other than calling it a “carefully considered business decision.”

Anyone who ever worked with or spoke to the team at Würth Additive Group knows they knew exactly what they were doing.

The real story happened one floor above them, inside the corporation that created the division - and eventually stopped caring about it.

Würth entered the 3D printing business in 2017 and spun it into a dedicated unit in 2021 under the name Würth Additive Group. The concept actually made a lot of sense: Digital Inventory Services - storing spare parts as printable digital files instead of physical inventory, combined with distributing third-party hardware.

In the spring of 2026, at AMUG, Würth Additive Group integrated B9Creations printers into its system. At RAPID + TCT 2026, it announced a partnership with HP focused on on-demand part manufacturing.

The team kept delivering. The problem was that the key decisions weren’t theirs to make.

Once again, 3D printing was treated as just another line in a spreadsheet. One that, at some point, no longer balanced or simply wasn’t green enough.

Würth has been streamlining its global operations for some time. In March 2026, it exited Kenya after nearly 30 years in East Africa. The company has not linked the two decisions, but the pattern looks familiar: review the portfolio, cut whatever no longer fits.

So why did this happen?

Because a large industrial distributor lives on volume and catalog sales. You order 10,000 identical bolts, they ship them, the margins work, everyone is happy. The company’s people, systems, incentives, and instincts are all optimized for repeatability and scale.

3D printing works in the opposite direction.

Small batches. Application-specific parts. Constant iteration. Quality data collected over years before a process is truly qualified. It operates on a timeline that corporate spreadsheets struggle to tolerate because, in the short term, it often looks like pure cost with very little volume.

Würth tried to bridge those two worlds with Digital Inventory Services, and for a while it genuinely looked like a smart strategy.

The problem is that bridges like that take longer to build than a single budgeting cycle.

The team understood that. Apparently, the floor above them didn’t.

So yes, once again, someone has switched off the lights in the 3D printing department of a large global corporation.

But don’t mistake that for a verdict on the technology itself… 3D printing keeps moving forward and it keeps delivering.

Just perhaps not on someone else’s spreadsheet.

Atomic Layer from the Past:

8 years ago, Prodways Group acquired wax 3D printing pioneer Solidscape from Stratasys, hoping to strengthen its position in the dental and jewelry markets.

Founded in 1994, Solidscape had sold over 5,000 printers worldwide and was a leading name in investment casting. The acquisition failed to meet expectations, however.

After disappointing results, Prodways exited the jewelry printer business, and in April 2024, Solidscape was sold to a new investor, regaining its independence.

Read all:

News & Gossip:

#1

I'll start by apologizing for the late publication. I’ve just returned from a week-long business trip to Athens, Greece, where together with the Bambu Lab Inspire team I put in a tremendous amount of work - intense, but incredibly productive. You'll start seeing the results on LinkedIn before long.

Also, a little teaser… This is one of the very last editions of Atomic Layers in its current format. As I mentioned two weeks ago, my wife Anna and daughter Roksana have joined The 3D Printing Journal. You'll soon see why 😉

#2

AM Research puts the 3D printing market at $4.35 billion for Q1 2026, up 13.1% year over year. Metal climbed to $1.76 billion, polymer to $2.59 billion, services to $2.42 billion.

However, whenm you look closer, the quarter itself barely moved: $4.29 billion in Q4 2025 crawled to $4.35 billion now. Scott Dunham - AM Research Executive Vice President, credits defense and supply chain reshuffling, and the numbers back him. Drone AM alone hit $140 million in 2025 and may near $900 million by 2034.

#3

SWISSto12 just closed a $70m Series C. The Swiss outfit 3D prints radio frequency hardware for satellites: over 1000 products in orbit, $140 million in 2025 revenue, contracts north of $500 million, positive EBITDA coming in 2026.

The company uses 3D printing to make its own RF products, so nobody calls it a 3D printing company. It gets called a space company. That framing is exactly why the round happened...

#4

EOS is retiring the FORMIGA P110 after almost 20 years. It was one of the most capable budget SLS machines ever built, historically about a quarter the price of the new P3 NEXT.

#5

6K Additive has named AMPERE Alloys its European distributor. The powders on offer run from titanium to nickel superalloys to refractory metals, aimed at aerospace, defence, medical, and energy.

#6

And to wrap things up, a weekend read for those of you buried in metal 3D printing.

The US National Institute of Standards and Technology (NIST) found a way to print high-entropy alloys, the five-plus-element mixes that refuse to blend evenly in a melt pool.

The fix is almost insultingly simple: instead of straight scan lines, run the laser in tiny elliptical loops so it stirs the molten metal before it freezes. No new hardware, just custom toolpath software, since commercial slicers can't draw these patterns.

They proved it by welding a dense refractory alloy to lightweight titanium and watching it mix under Argonne's synchrotron. The prize down the line is a "color printer" for metal, blending simple powders in-machine instead of stocking 12 pre-alloyed ones.

Still a lab demo, though. Read all here: www.nist.gov.

Swiss company Spherene has launched SphereneNXT, software focused on everything you can’t see with the naked eye: the inside of a 3D printed part. The entire application runs in a web browser, performs its calculations in the cloud, and is available in plans ranging from students to large engineering teams.

Anyone who has ever printed a part is familiar with the infill slider in a slicer: honeycomb, grid, gyroid. These are repeating patterns applied uniformly throughout the model, regardless of where the part is actually subjected to stress. Hollow on the inside, equally stiff everywhere.

Spherene takes a different approach.

Its ADMS (Adaptive Density Minimal Surface) technology creates smooth, curved internal walls - somewhat like a soap film stretched inside the part - that distribute loads across continuous surfaces rather than concentrating them in the thin beams of a lattice.

More importantly, those walls are adaptive. They become denser where stresses increase and thinner where little or no load exists.

The closest analogy in nature is bone.

Bones are not solid inside. Instead, they contain a porous internal structure that becomes denser precisely where it carries the greatest loads. Millions of years of evolution arrived at the same conclusion as the Swiss engineers: material only makes sense where it actually does work.

How does it work?

The software focuses exclusively on the internal structure of a part.

The external geometry remains entirely yours, meaning it doesn’t matter which CAD software you use. You upload your model, Spherene computes the internal architecture, and you either export the optimized geometry or send it directly to a 3D printer. Plugins are also available for nTop, Fusion, Rhino, and Grasshopper for users who prefer to stay within their existing design environment.

Alongside ADMS, the software also offers traditional TPMS (Triply Periodic Minimal Surface) structures, with density, wall thickness, and orientation controlled through scalar fields applied directly to the part’s surface.

Another feature is Flow ADMS, designed specifically for fluid flow and heat transfer applications.

Instead of producing a uniform internal structure, it generates geometry that guides fluids along predefined paths while delivering lower pressure losses than a conventional gyroid structure at comparable thermal performance. For engineers designing heat exchangers, that can make the difference between a theoretical exercise and a component that performs effectively in real-world operation.

Why does this matter for engineers?

Until now, designing advanced internal structures has largely been the domain of specialists working with expensive software licenses and highly complex workflows.

SphereneNXT removes several of those barriers.

Engineers can produce lighter, stronger parts without manually designing lattice structures. The algorithm automatically distributes material according to expected loads, meaning that when the geometry or material changes, there is no need to redesign the internal architecture from scratch. You simply adjust the parameters and run the calculation again.

Fewer design iterations and more consistent results. The generated structures are also self-supporting.

There are no internal support structures that must later be removed - a task that is often one of the most frustrating parts of post-processing.

Users also don’t need a powerful workstation or complicated software installation.

All computations are performed in the cloud. You simply open a browser, while the finished result is exported as a compact, error-free mesh ready for additive manufacturing.

This may not be a tool for every engineer, despite what the marketing brochure might suggest.

But for engineers developing lightweight brackets, heat exchangers, orthopedic insoles, or other performance-critical components, it removes a genuine bottleneck from the design process.

SphereneNXT became available on June 30, 2026, through spherene.io, with subscription plans tailored to everyone from students and independent designers to professional engineers and enterprise teams.

Just look at this lineup...

Desktop Metal debuted on December 10, 2020 through the merger with Trine Acquisition Corp. A post-transaction equity value of up to $2.5 billion and roughly $580 million in fresh capital. Revenue in its debut year: $16.5 million. Net loss: $90.4 million.

Markforged debuted on July 15, 2021. Merged with the SPAC “one” (NYSE: AONE), sponsored by Kevin Hartz, at a $2.1 billion valuation, bringing in about $361 million in gross proceeds.

Velo3D debuted on September 30, 2021. The merger with JAWS Spitfire Acquisition Corporation, delivered $274 million in net proceeds at a pro forma enterprise value of $1.6 billion.

Shapeways debuted the very same day on September 30, 2021 at a $605 million valuation. The business combination was run through Galileo Acquisition Corp.

Fathom debuted on December 27, 2021 by the combination with Altimar Acquisition Corp. II. The deal announced at a $1.5 billion valuation.

Fast Radius debuted on February 7, 2022 through the merger with ECP Environmental Growth Opportunities Corp.

And finally, BigRep which debuted on July 31, 2024. Merged with SMG Technology Acceleration SE, a SPAC listed in Frankfurt since October 2023. The opening price of €11.20 implied a market capitalization of around €163 million.

Every one of these companies reached the stock market the same way: by merging with a shell company, because none of them qualified for a traditional IPO.

On September 30, 2024, I published an article titled „SPAC merger means death”. My argument was simple: AM companies entering the public markets through the back door were never truly ready for it, and a sudden influx of investor cash would solve none of their structural problems.

That thesis has now been tested over 21 months.

I was right in every single case.

Not one of these companies is still trading where it started. Each of them has either ceased to exist, been absorbed into another, privately held entity, or been rebuilt from scratch on the ruins of the original shareholder base.

Velo3D, the only one back on an exchange today, got there by handing roughly 95% of the company to a creditor and wiping out everyone who bought the story in 2021.

At the end of June, the list finally closed, proving the point once and for all.

BigRep - the last AM company to take this route, announced its delisting and voluntary liquidation, less than two years after ringing the opening bell in Frankfurt.

Below is the full scorecard: who survived, who didn’t, how much it cost shareholders, and why this mechanism could only have ended exactly the way it did.

The Class of 2020-2021

Desktop Metal was first.

Ric Fulop wasn’t selling printers - he was selling the market itself. Investor presentation stated, in black and white, that additive manufacturing would grow from a $12 billion industry in 2019 to $146 billion by 2030.

Today, in mid-2026, the entire market, even under the broadest definition, is worth between $26 billion and $31 billion.

A company generating $16.5 million in annual revenue promised investors a market that never existed and never will.

Everything that followed was simply a long, bumpy descent.

Desktop Metal was struggling with both its finances and its ability to raise additional capital. The acquisition by Stratasys failed, only for the company to be acquired a year later by Nano Dimension.

Nano Dimension completed its acquisition on April 2, 2025, under a court order, for $179.3 million. 117 days later, Desktop Metal filed for Chapter 11 bankruptcy protection. The liquidation plan was approved by the court on September 30, 2025. The core assets were ultimately sold for roughly $7 million, while subsidiaries such as ExOne, EnvisionTEC, and AIDRO were acquired by Anzu Partners for another $10 million.

A $2.5 billion valuation. 7 million dollars at the finish line.

Desktop Metal remains the most disgraceful symbol of the industry’s excesses - one that will remain with additive manufacturing for years to come.

Shapeways, the creator marketplace once valued at $605 million, shut down overnight on July 3, 2024. The brand, the Eindhoven factory, and the remaining assets were later bought back from the bankruptcy trustee by the founders and the Dutch team.

The company still exists today, only at the scale it should have operated from the very beginning.

Fast Radius went public in February 2022 and filed for bankruptcy just nine months later. An industry record no one wants to break.

Fathom was taken private by CORE Industrial Partners in 2024 and is doing just fine. It operates across 25 manufacturing processes, with 3D printing being only one of them. It survived because it was never a pure-play AM company, and because it left the stock market before the stock market destroyed it.

Markforged: a $2.1 billion valuation in 2021. $70 million in revenue in 2025. Sold for $42.5 million in 2026. All assets ended up in the hands of direct competitor Stratasys. The technology was real. Continuous carbon fiber reinforcement remains one of the company’s genuine engineering achievements. The valuation, however, turned out to be nothing more than a costume.

Velo3D is the only exception on this list.

First quarter of 2026: $13.8 million in revenue, up 48%. Gross margin improved to 17.2%, compared to negative 73.6% a year earlier. Debt cut by 70%. A contract with the Defense Logistics Agency. The stock price up several hundred percent in six months.

Today, the company is finally doing what it should have done all along: metal printing, defense, and a narrow industrial specialization.

In April, it raised another $50 million through a share offering, meaning the public market remains a viable source of capital. The difference is that this time investors are financing a business with a positive gross margin.

Velo3D recovered only after the SPAC nearly killed it, and after a restructuring that erased almost everything contained in the original promise.

The Polish market had its own miniature version of the same phenomenon.

Zortrax entered the market through Corelens. Sygnis did the same through MODE. Formally, these were reverse mergers rather than SPACs, but the mechanism was essentially identical. Both companies remain listed on NewConnect. Both now have only marginal exposure to 3D printing, while Sygnis derives most of its business from drones.

And finally, BigRep, which I covered in details this past Friday.

Why the back door always leads to the same place

A company that cannot enter the stock market through the front door usually cannot do so for one simple reason: it has nothing convincing to present.

A traditional IPO requires financial results, an operating history, and the confidence of institutional investors capable of reading a balance sheet.

None of these companies had the full package. And SPAC allows all of that to be bypassed.

But bypassing the process comes at a price.

The day the merger closes, the company’s product changes. Yesterday it sold machines, materials, and services - today it sells quarterly growth. The machines merely become evidence supporting that story.

Being a public company is expensive, and those costs are fixed.

Auditors, law firms, transaction advisors, investor relations, quarterly and annual reporting. Legal work for every refinancing attempt, every capital raise, every prospectus.

You pay exactly the same whether you sold two machines, twenty, or two hundred that quarter.

A healthy business covers those costs from its operating margins. The SPAC companies on this list covered them with money raised from the stock market, because the market had become their only meaningful source of cash.

To preserve access to that cash, they had to keep the story alive. And to keep the story alive, they had to promise growth.

To promise growth, they had to keep acquiring companies, announcing products, launching new initiatives, and appearing at trade shows.

Which, in every single case, meant spending more and more money without generating a matching increase in machine or service sales. It never stopped until the companies ran out of resources entirely.

So what actually survived?

Go through the list once again and look at what remains.

• Shapeways’ machines are still printing in Eindhoven.

• Fathom still manufactures in ten production facilities.

• ExOne continues operating under Anzu Partners.

• Markforged’s technology lives on inside Stratasys.

• BigRep’s large-format printers will continue to be built in Berlin.

Every time, the same thing goes into liquidation: the listed company, the former valuation, an investor presentation, a forecasts, and a stock ticker.

If there’s one thing we can learn from all this, it’s this:

That is the entire thing. As boring as 3D printing itself.

For an AM company, merging with a SPAC means death. The listed entity dies. The workshop survives. And the bill remains on the table.

Oh, and the people who prepared the investor presentation are never the ones who pay it.

On July 31, 2024, BigRep rang the opening bell at the Frankfurt Stock Exchange. Standing on the trading floor was NERA, the company’s fully 3D-printed motorcycle, alongside its latest industrial printer.

Management spoke about building a European 3D printing champion through acquisitions. It was, in every sense, a celebration of the company’s bright future.

Less than two years later, on June 29, 2026, the same company announced its voluntary liquidation.

What makes the story even more interesting is that as recently as spring 2026, BigRep was still unveiling new pellet extrusion technology and a machine featuring a build chamber exceeding one cubic meter.

Now is heading into liquidation.

How is it even possible?

What was actually announced

The announcement published on June 29 is surprisingly specific:

The transaction will be completed at fair market value based on an independent valuation.

Neither the purchase price nor the transaction timetable has been disclosed.

Any remaining proceeds after the company’s assets are liquidated will be distributed to shareholders if anything remains.

That final reservation says more than the rest of the announcement combined…

Founded in Berlin in 2014, BigRep spent a decade doing one thing exceptionally well: building large-format FFF 3D printers capable of producing meter-scale parts.

In November 2023, the company “acquired” Austrian manufacturer HAGE3D, known for its high-temperature systems.

In reality, it was more of an equal transaction rather than one company buying the other, but that was irrelevant at the time…

Few weeks later, it went public.

And here’s where problems started…

BigRep entered the stock market through the back door via a merger with SPAC company SMG Technology Acceleration SE, which had been listed in Frankfurt since autumn 2023.

The strategy sounded ambitious: consolidate the fragmented additive manufacturing market and create a European AM champion.

But instead of building a champion, BigRep delivered annual losses followed by several unsuccessful refinancing attempts.

Just take a look…

The snake eats its own tail

The most fascinating part is the list of buyers.

HAGE now returns as both co-owner and creditor, buying back the company that once absorbed it.

De Krassny and Koehler, former shareholders and would-be rescuers whose refinancing efforts ultimately failed, are acquiring the operating business from what remains.

All three have been part of the story from the beginning.

Now they are simply taking the business that still works and leaving the publicly listed shell to the liquidator.

Which raises an obvious question:

The uncomfortable answer is that the journey to the stock market is precisely what pushed the company into this position.

Large-format 3D printing is a solid niche business. Companies can thrive in that business for decades.

On one condition: they do not promise investors tenfold growth.

A publicly listed growth company is expected to keep expanding. It has to acquire competitors, consolidate markets, meet quarterly expectations, and finance a buy-and-build strategy with capital that a healthy niche business simply does not generate.

1. BigRep GmbH built large-format industrial printers.

2. BigRep SE sold the story of a future European champion.

The first survives - the second is being liquidated.

And that is the real conclusion - far less dramatic than the word “liquidation” suggests.

Because throughout this entire process, the technology remains intact.

Only the costume disappears: the listed holding company and the narrative built around it.

The new owners - who have been involved with the business from the very beginning - will most likely do what should have happened two years ago.

They will remove the burden of stock-market expectations from the machines, streamline the portfolio, and focus on the products that actually generate profits.

At last, someone is cleaning up after years in which everyone wanted to become a publicly traded legend…

And as I wrote about it two years ago: for every AM company SPAC merger means death…

Atomic Layer from the Past:

11 years ago, Polish 3D printer manufacturer Omni3D introduced the Factory 2.0, the country’s first industrial-class FFF system with a 500 × 500 × 500 mm build volume. Featuring advanced monitoring and control systems, it marked the company’s shift from unreliable desktop printers to industrial manufacturing.

The transformation proved decisive: after early struggles with the desktop RapCraft 2, Omni3D successfully reinvented itself into a globally recognized supplier of large-format industrial 3D printers.

Read all:

News & Gossip:

#1

While BigRep exits the German stock market, China writes Snapmaker a €130 million check. That's what 1 billion RMB buys: the biggest consumer-3D-printing raise in 2 years, a Series C led by Cathay Capital, with Meituan, Hillhouse and Shunwei piling back in. Six years ago Snapmaker was considered a declining brand. Today it's the best-funded name recently in desktop printing, and the check says one thing plainly: the money in this industry no longer speaks German.

#2

Revopoint, the Chinese 3D-scanning maker out of Xi'an and Shenzhen, has launched a new scanner: the Trackit SR. It's a marker-free optical-tracking rig with 0.02 mm single-frame accuracy, three laser modes, Wi-Fi 6 with no cables, up to 2.5 hours on battery, and a carbon-fiber build rated to work from 100,000-lux sunlight to low-light sites.

A CMM Edition adds a metrology-grade ball plate for on-site accuracy checks. Scan data runs into Revopoint's software: Revo Track for capture, Revo Measure for inspection, Revo Design for CAD. Note the fine print: peak accuracy depends on your PC, and the software is Windows-only.

#3

EOS has partnered with aluminum developer Constellium to widen its metal-AM materials lineup. The deal adds Constellium’s Aheadd CP1 alloy as EOS Aluminium Constellium CP1, available from August 2026, and renames the existing Al5X1 grade to EOS Aluminium Constellium Al5X1.

CP1 targets heat exchangers and semiconductor parts, with simplified heat treatment that skips quenching and enough strength to stand in for many AlSi10Mg jobs. Al5X1 stays pointed at aerospace and motorsport. The wider aim is co-developing more AM aluminum alloys with defined parameters, so customers qualify parts for production faster.

#4

New Zealand’s Zenith Tecnica, a contract manufacturer running titanium EBM for satellites and orthopedic implants, has been acquired by Andrew Burgess and Blair Jordan, who bring backgrounds from Boeing, Amazon’s satellite work, and regulated food production.

The buyers are adding capacity immediately, taking Zenith from six EBM machines to eight, with demand already booking out every system.

Founded in 2014, the company grew revenue 490% since FY2020 and has printed parts for NASA’s Psyche mission.

#5

Now this is interesting, because it’s not every day someone announces they’ve made a “new technology”…

Inkbit, known for Vision Controlled Jetting, has filed a patent application (WO2025041082A1) for a chemistry that drops UV curing. Two fluids jet separately: one carries a monomer and activator, the other a ROMP pre-catalyst.

They mix only on the print, like a printed two-part epoxy, so polymerization can’t clog the nozzles. The fit with their vision-based process control is convenient: a delayed reaction like this needs tight concentration control.

#6

And to close out today's roundup: Formlabs has announced three new printers coming over the next 12 months, though it won't yet say which technologies or price points they'll target.

The news lands on the back of its best-ever quarter: Q2 2026 set records for total, hardware, non-hardware, services, and SLS sales, the last driven by the new Fuse X1. Dental non-hardware notched a fourth straight growth quarter, and Formlabs closed several June Fuse 1+ deals of eight units or more.

Joel Telling is one of the best-known creators on YouTube. What many people don’t realize, however, is that the creator of the 3D Printing Nerd channel also runs his own 3D print farm, where he claims to have completed as many as 3735 production jobs since April this year.

The farm consists of multiple 3D printer models from different manufacturers, all managed by software that Telling wrote himself.

And he released it for free under an open-source license.

The code is available on GitHub under the MIT License, allowing anyone to clone, modify, and improve it.

Called Print Farm Manager (not the most original name) the software is a self-hosted web application designed to manage print farms built from printers made by different manufacturers. It provides a single dashboard showing every machine, its status, and print progress.

The reason Telling decided to write it was simple: existing tools didn’t fit his printer fleet.

Prusa Connect couldn’t do what he needed. Bambu Lab has its own farm management software. Other printers basically have none. What Telling wanted was a single interface that worked across multiple brands, with one goal in mind: maximize the farm’s throughput without drowning in manual bookkeeping.

Previously, his workflow looked like that of most small print farms. The operator copied files to USB drives, loaded them onto printers one by one, and manually tracked how many parts had been produced.

Print Farm Manager automates all of that. The central dashboard distributes G-code files to available printers while automatically tracking production counts.

At the heart of the system is its production model. A project consists of parts, each with a required quantity. Those quantities are fulfilled by assigning G-code files across multiple printers. The operator simply tells the software what needs to be produced and how many pieces are required. The software decides which printers should do the work.

One feature that even YouTube viewers highlighted in the comments is support for multiple G-code files per part, each prepared for a different printer model.

If a production run requires 50 chassis, the application automatically combines whatever compatible printers are available - in Telling’s case, Prusa MK4S and Core One machines, to complete the order. Adding another printer model is as simple as attaching another G-code file for that machine.

The dashboard can simultaneously manage printers from several manufacturers, including Prusa printers through PrusaLink, Elegoo Centauri Carbon and Carbon 2 machines, Bambu Lab printers with AMS support, Klipper-based printers and Vorons via Moonraker, as well as anything running OctoPrint.

Job completion is handled deliberately. The application assumes a print has succeeded but never confirms it automatically.

Instead, a human operator does. Clicking the green “Set Ready” button marks a successful print, while the red “Bad Print” button immediately sends the machine into a decommissioning workflow. Every completed part is confirmed by the operator.

The rest of the software focuses on the day-to-day realities of running a print farm.

Status colors indicate whether a completed print is waiting for removal, a print is currently running, or a machine requires attention, typically because it has run out of filament.

Additional features include filtering by material and color, grouping printers by shelves or rows, tracking partial production runs such as “24 out of 25 completed,” importing up to 50 printers from a CSV file, a TV dashboard for workshop displays, and backing up the entire farm into a single JSON file.

Print jobs are uploaded five at a time because, according to Telling, Prusa’s networking can become unstable when too many simultaneous transfers occur.

The software is clearly being used in real production.

One machine named Applejack had already completed 71 jobs, produced 257 parts, and maintained a 96% success rate over 311 printing hours since early April.

Telling is very clear about who he built the software for.

It runs entirely locally, requires no cloud services, no user accounts, and no subscription fees.

His target user is the hobbyist with two or three printers in a garage – eg. a Prusa MK4, a Bambu P1S, and an A1 - who wants to turn that setup into a real manufacturing operation using free software.

Print Farm Manager is a solid, thoughtfully designed tool. It doesn’t claim to revolutionize print farm management, and Telling himself makes no such promises.

Its real value becomes obvious when looking at how small mixed-brand print farms actually operate. People buy printers from different manufacturers because each excels at different tasks. A large-format Prusa XL handles oversized flat parts, while faster Bambu machines are ideal for smaller components.

The problem is that every manufacturer provides software designed exclusively for its own hardware. Telling tries to solve this with a single unified dashboard.

Telling also openly acknowledges that users should approach the project with reasonable expectations. It remains a one-person project, tested primarily in his own production environment, so rough edges are inevitable.

He admits that data visualization is still a work in progress, while future plans include automatic STL and 3MF plate arrangement as well as maintenance reminders based on accumulated print hours.

There is also one important warning in the repository documentation.

Ironically, these characteristics are exactly what many of its intended users are looking for.

The code is open, multi-brand, and easy to customize.

Anyone who enjoys tinkering with their own print farm gets a foundation they can modify, extend, and adapt to their own workflow.

Eight years ago, on July 2, 2018, German inventor Kai Parthy introduced the world to GROWLAY. A filament on which plants could grow. And fungi. And mold - the same kind that gives blue cheese its distinctive character.

The material was absolutely brilliant. Its microcapillary structure absorbed water and nutrients, allowing moss, grass, or even bacterial cultures for pharmaceutical applications to grow directly on the printed part. Nobody had ever created anything like it before.

And nobody ever used it for anything.

Today, GROWLAY survives only in the memories of a handful of people. It has existed in memory longer than it was ever actually used. It is the purest example of what 3D printing did best for most of its existence: creating things that amazed everyone and that nobody needed.

On July 1, a day before this „remarkable” anniversary, Rajeev Kulkarni, an AM industry veteran, published a LinkedIn article posing a controversial question: what if additive manufacturing needs less innovation?

I read it and even commented on it, praising the arguments he made. Because I am probably the last person entitled to disagree with him.

I perfected the art of creating fascinating things that nobody needed.

I did it professionally for nearly four years in my own startup, GREENFILL3D. A startup that no longer exists and has gone to the same place as GROWLAY. Along with hundreds of other 3D printing projects, products, and startups.

Into oblivion.

So when I say that 3D printing spent years producing things everyone talked about but nobody bought, I say it as a repeat offender - with a complete record of my own offenses.

What is a “made-up” product?

Let’s start with something I call “made-up” products. Here’s my own definition:

It’s the perfect product to receive for free. You put it on a shelf to collect dust, and a year later you throw it away.

I have my own collection of such products.

Exactly five of them.

The first was Branfill3D - a genuinely biodegradable filament made from wheat bran, a waste product from the production of pasta. During printing, it smelled like freshly baked bread straight out of the oven. It decomposed in soil, but it also degraded naturally over time when exposed to sufficient moisture. A beautiful zero-waste concept wrapped onto a single spool.

The problem was that it was a nightmare to print. It clogged 0.4 mm nozzles. It snapped easily. Retraction was the enemy. Stringing was terrible. Removing supports left surface scars just as ugly as TPU does. Honestly, the best thing you could print with it was a vase. Damn, you really needed a very good reason to use it.

The second product was a series of advertising displays made from the same filament. And here’s the best part: a display made from wheat bran was promoting pasta produced from that very same wheat bran in stores. The perfect circular economy story every sustainability presentation dreams of. The displays amazed the world, won awards, and GREENFILL3D became the only Polish company nominated for the 2022 3D Printing Industry Awards.

But manufacturing the displays was simply too expensive and too slow. Printing 200 units took longer than producing 2000 conventional cardboard displays. Not to mention the unit cost. And a large retail chain orders 2000 displays - not 200. Nor is it willing to pay the price of 2000 cardboard displays for just 200 printed ones. The math was brutal and unforgiving.

The third product was decorative protective covers for light switches and electrical outlets, printed in 3D and finished with UV printing. Colorful. Personalized. Available with QR codes or Braille descriptions. They looked great.

The only problem was that nobody actually needed them. Nobody bought them. On the other hand, everyone loved getting one for free at trade shows and conferences. Then they would even install it at home. But never returned to get another one...

The fourth product was a collection of lampshades made from fully compostable materials: Branfill3D or Tarfuse Envi-1 from Grupa Azoty, based on potato starch. Shapes you couldn’t find anywhere else. Named in Esperanto, because why not? Brano Vortico or Ora Arĝento.

It turned out, that nobody wanted plastic lampshades that couldn’t even be washed in water because they would begin biodegrading. Also, eco-friendly decorative lampshades ranked surprisingly low on most people’s home shopping lists.

Finally, there were accessories for the stretch ceiling industry: protective rings, profile connectors, and installation tools. These actually made practical sense.

The problem was that they were too precise for the quality of the metal profiles and tools they were supposed to work with. Precision was simply too good for a world that didn’t need it.

Why the printer was perfect until you tried to make money

For all of this, 3D printing was the perfect tool. Fast, inexpensive, and ideal for producing a single part. You come up with an idea in the morning, and by the evening you’re holding a prototype in your hand. No other technology can do that.

The problem begins when you scale up.

Print time and unit cost, which looked perfectly reasonable at ten units, became devastating at a thousand. Additive manufacturing has a built-in ceiling. It’s a method for producing fast, inexpensive, and unique products - but only in small quantities (unless you do it in China - though the economics are different there).

The real mistake, however, came much earlier.

For years, I kept asking myself the wrong question:

“What cool thing can I make with a 3D printer?”

While the right question was:

“What is a genuinely good product (that just happens to be suitable for 3D printing)?”

I was looking for products made with a printer instead of simply looking for good products.

That is a subtle difference with enormous consequences for the bottom line.

This is exactly what Kulkarni argues:

Kulkarni draws a comparison with nuclear power. France and China standardized reactor designs and kept building the same ones over and over again. Costs came down, and every project made the next one better.

Meanwhile, the West redesigned every reactor from scratch and ended up coining the term negative learning - a situation where, despite accumulating experience, every new project becomes more expensive than the last.

Customers don’t care about innovation for its own sake. They care about outcomes. Shorter lead times. Lower costs.

A new photopolymer chemistry means absolutely nothing to them unless it gets their parts delivered faster.

The next decade belongs to those who execute - not those who endlessly invent.

And that brings us back to GROWLAY. The eight-year-old filament that could grow plants is Exhibit A.

Eight years in people’s memories. Zero years in serial production. A brilliant invention that never found a place because there wasn’t one to find. A made-up product in its purest form.

Just with moss.

3D printing needs to become boring

The conclusion is an uncomfortable one for everyone who loves this technology because it’s fascinating.

3D printing needs to become boring.

It should produce ordinary, unremarkable, everyday products. It should manufacture objects that nobody ever looks at and asks “How was this made?” Because that’s when it starts generating real profits.

The product should be exceptional. The manufacturing process shouldn’t matter. In fact, it’s better if it remains the manufacturer’s little secret.

The less “3D printing” appears in the marketing message, the better the product will sell.

The First Law of the 3D Printing Market states this very clearly: a company that uses 3D printing to manufacture its own products is perceived as a company from that industry - not as a 3D printing company.

Reverse that equation, and the business model falls apart. The market expects a 3D printing company to sell printers, materials, or printing services - not finished products.

My advertising displays failed, among other reasons, because I was selling them as made in 3D. The market, meanwhile, wanted my filament and printing services from me - which was the exact opposite of my original idea.

The product itself, that entire circular economy miracle, was irrelevant to customers. Exactly as the theory predicts.

The only difference is that I didn’t know the theory yet. I was still testing it against my own balance sheet.

So let me leave you with one warning:

If a product is advertised with the slogan “3D printed” there’s a very good chance you’re looking at a made-up product.

A genuinely good product doesn’t need that kind of marketing. It sells itself through its price, its function, and the simple fact that people genuinely want to buy it.

3D printing is still moving forward. But its entire future can be summed up in one rather unglamorous sentence:

it has to become so good that people stop noticing it.

P.S. The flower pots in the cover image were made from a biodegradable material. Which meant that the more often you watered the plant inside, the faster the pot itself began to break down.

Clever, wasn’t it? 🤣

VulcanForms has just made two major moves, and both are bigger than the press release suggests.

First, the company appointed Michael Kenworthy as its Chief Technology Officer - a leader who until recently was building rocket engines at Relativity Space, and before that was responsible for technology at Seurat Technologies and Divergent (among others).

Second, the company secured $21.26 million in tax incentives from the Commonwealth of Massachusetts to support a manufacturing facility of up to one million square feet, expected to create 1063 jobs!

In an industry where most “breakthrough” announcements are promises of things yet to be built, this is something real.

Let’s start with the money, because the real story lies in how the incentives work. These tax credits are performance-based. Massachusetts only grants them once the factory is operating and the promised jobs have actually been created.

In this funding round, the state awarded $52 million in tax incentives across 11 projects, with VulcanForms receiving the largest individual allocation.

In return, the participating companies are expected to invest more than $1.4 billion of their own capital and create 2793 jobs.

That is worth remembering, because we’ve heard the phrase “rethinking how products are made” before in this industry. The last company making that promise was Desktop Metal - shortly before filing for Chapter 11 bankruptcy protection.

The key difference is the market VulcanForms serves: aerospace, defense, and medical manufacturing. These are sectors where certification is part of the product itself, and low-cost competition from Asia carries far less weight because nobody is going to install an uncertified component inside a jet engine.

The new Devens facility will become the company’s third manufacturing site in Massachusetts, joining its existing operations in Devens and Newburyport.

Back in January, VulcanForms raised $220 million in a Series D funding round. The picture that emerges is of a company scaling methodically rather than chasing headline-grabbing leaps.

The company has also strengthened its leadership by bringing in Michael Kenworthy.

He joins from Relativity Space, where he served as Vice President of Engineering. Before that, he was Chief Product Officer at Seurat Technologies. Earlier still, he spent more than six years as CTO at Divergent, where he built and deployed a digital manufacturing system - the very layer that metal additive manufacturing has so often struggled to develop.

His background also includes GE Aviation, Honeywell, and a mechanical engineering degree from MIT.

This is someone who has repeatedly taken complex metal manufacturing processes and turned them into production systems that consistently deliver working parts.

Taken together, VulcanForms can count this as an exceptionally successful week: a substantial performance-based incentive package from the state, a third manufacturing facility, and a CTO with a proven track record of transforming fleets of 3D printers into true production systems.

Companies like VulcanForms are building something more durable than many of the stars of the previous era. Their competitive advantage lies in vertical integration, certifications, and experienced people who have already taken difficult manufacturing operations from concept to production.

There are no guarantees, and the company itself acknowledges that.

But if you’re looking for a place in this industry where the noise of press releases meets genuine manufacturing engineering, this is where you should be looking.

Atomic Layer from the Past:

12 years ago, colorFabb launched sales of 3D printing pellets for FFF technology, targeting hobbyists and DIY enthusiasts who owned home filament extruders like Filabot.

The company offered test kits with four jars of pellets in any color combination, each weighing 1.2 kg.

The move was driven by high filament costs, which discouraged many potential users. Pellets were about twice as cheap as traditional spools. The offer included PLA/PHA, PLA, woodFill, and XT-copolyester.

However, the concept of home filament production never became widespread. Despite widespread praise, few users actually adopted it due to low production efficiency and poor final quality.

Read all:

News & Gossip:

#1

Velo3D, the company that flirted with death back in 2024, is putting up a 288,747-square-foot plant in Livermore, California. 40-plus large-format metal printers at launch, infrastructure for over 100. Fremont keeps R&D, Livermore takes production.

Former head of marketing Katelyn Spuhler is back too, publicly glad to rejoin the team and hinting at more to come. When people return to a company they left, that says more than any press release.

#2

John Kawola, who recently stepped down as CEO of BMF, is joining Harvard’s Office of Technology Development. He’ll help identify early-stage technologies, shape their path to market, and turn research into startups.

One notably candid line from his post: seven years at BMF taught him what to do-and what not to do. In an industry that rarely admits the second part, that honesty is worth more than the title.

#3

AMUG has its new board, effective July 1, 2026.

Dallas Martin (Toyota) takes the presidency, with Daniel Landgraf (3D Spark) as VP, David Leigh (UT Austin) on treasury, and Daniel Braley (V2X) running membership.

The rest of the slate spans Stratasys, Oak Ridge, and GoEngineer, a decent spread across industry, academia, and national labs. Their job now is building the 2027 conference in Atlantic City, March 14-18.

Congratulations!

#4

My dear colleagues from VoxelMatters just announced the new AM market report about polymers. According to it’s worth is $9.45 billion for 2025, up 18.7% year over year. Hardware took the biggest slice at $4.36 billion, ahead of services at $3.21 billion and materials at $1.89 billion.

The dataset is serious: 239 hardware makers, 249 material suppliers, 465 service providers, nearly 200,000 data points.

Most of that growth came from desktop and consumer printing, the corner generalist media buried years ago as a dead fad. Bambu Lab alone matched the top three industrial polymer players combined. APAC overtook North America in hardware and grew fastest on Chinese brands, while North American hardware revenue went essentially flat.

Industrial systems will cycle back as buyers digest earlier capacity, but the shape of the market has changed in plain sight.

The full 2026 Polymer AM Market Report drops July 14 and is open for pre-order now on the VoxelMatters website.

#5

Formnext is launching its first Defence Summit on November 17, 2026, kicking off the Frankfurt show’s main run through November 20.

Organized with AMDefNet, the summit puts industrial 3D printing squarely inside the EU and NATO supply-chain conversation: resilient sourcing, shorter response times, decentralized production.

The panels cover procurement, standardization, cybersecurity, and data. Access is limited to EU and NATO member states, and pre-registration is open at formnext.com/defencesummit.

#6

ASTM’s AM Center of Excellence, working with the UK Ministry of Defence under Project TAMPA, has published a strategic guide for certifying additively manufactured parts in defense, free to download from the AM CoE site.

The core idea is a four-tier classification, Class A through D, that ties the certification evidence you need to how badly a part failing would hurt. Certification is the actual product in defense AM, so it’s nice that’s someone finally wrote it down.

Do you still remember what it’s like to write an email to a client completely from scratch? Let alone in another language. Seriously, give it a try. Then compare it to the last email you sent.

Not so great, is it?

AI seeped into our daily lives so quietly that nobody noticed the exact moment we stopped doing ordinary tasks on our own.

We ask ChatGPT things we would have figured out ourselves just a few years ago. We use Claude to make important business and life decisions in much the same way we drive everywhere with Google Maps.

No idea where we actually are, as long as the arrow keeps pointing the way.

Now imagine it disappears. Imagine waking up one morning and free AI is simply gone.

Why?

Because the companies giving away intelligence for free today, go bankrupt.

Impossible?

OpenAI, if the numbers are to be believed, burns money at a historic pace. The “give it away now, figure out the economics later” model has an expiration date. And that date is approaching.

When the money runs out, free access will be the first thing to disappear. At that point there will only be two options: bankruptcy of AI giants, or a price for the service that most people simply won’t pay.

That’s when things get interesting. Because we’ll finally find out how many of us can still do anything without a prompt.

The free lunch someone else paid for

Free AI has been subsidized since day one. The investor picked up the bill.

Will Lockett has become my main guide to the financial reality of this sector (though not my only one). For months, he has been breaking the numbers apart and exposing a truth that is obvious once you see it, yet almost everyone ignores.

Will Lockett's Newsletter

The AI Industry Is Panicking

Have you noticed that the AI bros are starting to panic? All of a sudden, they have dramatically U-turned on the notion that AI will take vast swathes of the job market despite that claim being critical to AI’s hype and the entire point of this ‘revolution’, they are…

Read more

a month ago · 781 likes · 7 comments · Will Lockett

Subscriptions were sold below cost. The goal wasn’t profitability - it was our addiction.

OpenAI itself admitted that, if priced honestly, its $200 plan would cost well over $2,000.

Imagine renting someone a Ferrari for $10 a month. Of course they’ll fall in love with it. The problem begins the day you ask them to pay what it’s actually worth.

And that day is coming.

Copilot’s pricing has increased several-fold in some cases. Uber reportedly burned through its annual AI budget within months and publicly concluded that AI can be more expensive than human workers. Microsoft reportedly canceled a Claude Code license because it was too expensive. One company generated a monthly Anthropic bill approaching half a billion dollars.

A tool that once felt “free” suddenly became too expensive - even for an NBA star-player.

Beneath the surface, there’s a genuine cash bonfire. According to Lockett, OpenAI posted an operating margin of negative 122% in the first quarter of 2026. Anthropic looks slightly healthier, but largely thanks to accounting gymnastics.

Funding sources are drying up. Venture capital isn’t keeping pace. Big Tech is nearing the limits of its willingness to fund losses. Debt markets are becoming saturated with risky AI-related paper.

That leaves one exit. The stock market.

Will Lockett's Newsletter

AI Losses Are About To Spiral Out Of Control

It seems like everyone, including the AI bros, now know that the AI industry is one of the biggest bubbles humanity has ever created. Yet, rather than actually do something about it, like regulate Big Tech, enforce anti-trust laws, crack down on round-trip investments, or have the stewards of our delicate financial systems tell these morons to f**king stop it, we collectively seem to be enabling these idiots and letting this entire circus self-implode. The trouble is, when something is this detached from reality, devoid of logic, and utterly irrational, like the AI bubble, all the typical forces that would cause said self-implosion achieve nothing. They can seemingly just choose to ignore something that would be a fatal blow to any other business/industry and carry on unscathed. It is going to take a mind-altering tsunami of reality to break this cult-like hysteria and pop the bubble. Well, that sucker punch of reality is just over the horizon…

Read more

a month ago · 78 likes · Will Lockett

Bring in retail investors and pension funds so they can throw more fuel onto the fire. In this situation, an IPO becomes a search for someone willing to grab the hot potato before it cools down.

We’ve seen this movie before in additive manufacturing - just with far fewer zeros.

During the SPAC boom in 3D printing, valuations reached billions while revenues barely reached millions. The story ended with Desktop Metal collapsing.

Markets can keep rising even while companies are falling apart. The mechanism is exactly the same. Only the numbers are bigger.

So if that’s the case - it has to end the same.

The post-AI hangover

Let’s assume Lockett is right. Free AI disappears, or becomes so expensive that it might as well disappear.

What happens then?

We'll discover who spent the last two years outsourcing their thinking.

Today everyone is a software developer. Together with Claude.
Everyone is a content creator. Together with ChatGPT or Gemini.

As long as you have another browser tab open, you’re competent.

Close it, and you’re left with the uncomfortable truth about what you can actually do yourself.

Science is beginning to measure this, and the results aren’t encouraging.

MIT Media Lab connected EEG electrodes to 54 participants and asked them to write essays. One group used ChatGPT, another used Google, and the third relied solely on their own minds.

The AI group showed the weakest neural connectivity of all three. They remembered less of what they had “written.” Many couldn’t even quote sentences they had produced just minutes earlier.

Michael Gerlich studied 666 people and reached essentially the same conclusion from another direction. The more frequently people relied on AI, the weaker their critical thinking became.

Young people suffered the strongest effects, particularly those who had never learned to work without AI in the first place.

It’s brain fog that we’ve engineered ourselves. A muscle that isn’t used becomes weak. Everyone understands that in the gym. The brain works the same way.

Then there’s the second layer. Withdrawal.

Addiction clinics are already documenting AI withdrawal symptoms: irritability, anxiety, fractured concentration, and the constant urge to return to the chatbot just to ask one more question.

In a joint MIT/OpenAI study, participants agreed with statements like:

If I lost access to my chatbot, I wouldn’t be able to cope.

That’s language straight out of an addiction therapist’s office.

Now combine the two.

You take away the tool people have become dependent on, and at the very same moment expose the fact that they’ve forgotten how to work without it. Withdrawal and incompetence arrive together.

It will be a long hangover...

Months of relearning how to sit in front of a blank page with nothing except your own slowly warming-up brain.

Because for the past two years we’ve lived under the illusion that intelligence was free.

The currency we paid with was our ability to think

Personally, I’m not particularly worried. For one thing, I seem to have a trait I call “immunity to addiction.” Maybe I’ll write about it one day.

At some point I simply realize that something is bad for me, and I quit. Immediately. Permanently.

No more alcohol, no more nicotine, I’ve never used drugs, I’ve never gambled, no social media (never really on Facebook or Instagram, never on Reddit or TikTok; LinkedIn is mostly a publishing platform for me rather than a place to comment on everyone else’s life, and X is a sewer - that’s the single one “I needed to quit”).

So even if part of me could theoretically become addicted to something, I don’t see any difficulty in walking away from it in matter of hours.

Yes, I’m not gonna lie - I use AI extensively in my work. But I don’t depend on it.

The words you’re reading now are mine. They weren’t generated by ChatGPT, Claude, or Gemini.

Yes, I rely on AI to translate my writing - Polish is my native language. So the worst-case scenario is that my English becomes raw, with more stylistic and grammatical mistakes.

But the core ideas won’t change.

If you’re genuinely good at something, you’ll remain good at it.

If your primary skill is writing prompts for AI, you’ll need a hell of a lot of money to keep doing it. This might be a good time to start buying lottery tickets. Or take part in some basketball camp and hope for a lucrative contract. Just for you to know - NBA free agency just opened, so this could be your moment.

And if you can’t afford a real, full-price AI subscription then here’s my advice, as someone who genuinely wishes you well:

Hello everyone,

This isn’t another article - it’s a special announcement! Three things, and I’d love it if you read to the end.

First: in under two weeks, The 3D Printing Journal turns two. It’s surreal. I never imagined this project would come this far, or that my articles would reach some of the most important people in the AM industry. But here we are, and I’m deeply grateful for it.

Second: given how many topics we now cover, two people are officially joining the editorial team - my wife Anna and my daughter Roksana.

Anna has been in the AM industry almost as long as I have (since mid-2014) and is a veteran of the fight against hype and overpromising in 3D printing. She won’t publish much herself, but she’ll take an enormous amount off my plate on the organisational side. She’s already been deeply involved in building The 3D Printing World Guide, it’s genuinely our shared project, and she’ll now drive it much further.

Roksana is newer to the industry but already brings real experience: since late 2025 she’s been a contributor at VoxelMatters, the best AM news outlet in the world. Her work here will be different and won’t overlap with Davide Sher’s platform. I have a lot of ideas for her, and I think you’ll like the results.

Third - and please read this part carefully. From tomorrow, July 1, you’ll see a paid-subscription option appear.

What we’re turning on is simply an option to support the work, a kind of voluntary tip. It commits you to nothing. You lose nothing by not paying, and you gain nothing by paying. We gain, but we’re realistic about that.

Mostly, we’re just grateful you’re here and reading.

That’s it. I’m not asking anyone for anything. It’s an entirely optional button, and our common content stays free: 3DP War Journal, Atomic Layers, and RECODE.AM.

Thank you, again, for being here. I hope the articles ahead turn out to be even more interesting than anything I’ve published so far.

I read a great article over the weekend. Sam Davies from TCT Magazine decided to answer a simple question: are industrial 3D printing machines too expensive?

At first glance, it’s a rather trivial question. Of course they are. Show me someone who thinks paying €1 million for a machine is cheap.

But that’s not really the point. The article raises a much more interesting issue - one that aligns perfectly with something I’ve been patiently arguing in my own writing for quite some time.

Two of Davies’ interviewees: Joe Calmese from ADDMAN and Christina Perla from Makelab, shared their thoughts on machine productivity.

Because that’s really the biggest obstacle holding back metal AM adoption: Productivity. Repeatability. Reliability.

Poor reliability in metal AM is one of the industry’s dirty little secrets. People either whisper about it or don’t talk about it at all. Everybody knows. Nobody wants to say it out loud in front of a prospective customer who’s just taken out their wallet.

If I’m being honest, metal AM today is roughly where low-cost desktop FFF was between 2018 and 2022.

The Ender 3 era.

Well, that’s quite expensive for what you actually get.

Fortunately, the solution is already visible on the horizon. And it will come from exactly the same place it did before.

Software.

The article

Sam Davies set himself an ambitious goal: to break down, line by line, exactly what makes up the price of an industrial AM machine. About half of the CEOs he approached politely declined to answer - which, in itself, is an answer.

One of them, however, agreed. Avi Reichental.

As some regular readers may have noticed, I’m not exactly what you’d call a fan of Mr. Reichental. But credit where credit is due - he was remarkably open. Maybe that’s because he’s reached a point in his career where he simply doesn’t care anymore. I don’t know. Either way, this time I genuinely appreciate his transparency.

According to Reichental, roughly 20-25% of the price reflects amortized R&D, 30-35% covers components and manufacturing, and another 15-20% goes toward sales and distribution. The rest is spent on material development, hidden costs, and profit, which, in reality, quickly disappears into investors’ pockets.

His conclusion is refreshingly honest: the price of the machine reflects “absorbing risk and complexity.” At today’s production volumes, nobody can significantly lower costs because there simply isn’t an application base large enough to generate economies of scale.

Others are singing the same gospel.

Rich Garrity from Stratasys and Sam O’Leary from Nikon SLM both argue that what really matters is the cost per part over the machine’s entire lifecycle. Cutting costs isn’t an option because reliability would suffer. In their view, an AM machine costs roughly what you’d expect to pay for a high-end CNC machine tool.

And you know what? They’re right.

The argument is logical, measurable, and difficult to dispute.

However, Filomeno Martina, former CEO of WAAM3D, probably summarized it best:

The industry needs a better economic language, not just better machines.

And yeah, it could have ended there as another fairly standard industry article...

…but Davies expanded his list of interviewees to include actual users.

And that’s where things get really interesting!

Joe Calmese from ADDMAN - a company operating a serious fleet of metal AM systems, states it bluntly:

For us, hardware cost isn’t the real issue.

Christina Perla from Makelab adds an observation that should send chills down every machine manufacturer’s spine. Every company she visits starts by talking about machine problems - not machine prices.

It’s much more about effectiveness and throughput.

Jacob Rindler from The Exploration Company is even less diplomatic:

The machines are too expensive - and I’ll keep saying it until my last breath.

Machine vendors want to discuss purchase price. The people who actually run these machines want to discuss downtime.

To illustrate the point, Davies quotes this brilliant analogy (I quote it in full because it’s simply too good):

In the 1980s, UK sitcom The Blackadder was cancelled after its first series, largely because there was a greater focus on production value than actual comedy. As Ben Elton, the co-writer of the show’s subsequent three series, jokes in his 2026 memoir, it looked a million dollars, but it cost two.

The same logic applies here.

If a $1 million machine only delivers 50% utilization, it might as well cost $2 million, because the customer will ultimately need a second one just to achieve the required throughput.

Sixty-five percent

ADDMAN guarantees a maximum machine utilization rate of 65%. They often exceed that number, but above it, they make no promises.

Calmese says their metal AM systems experience roughly 30% more downtime than CNC machines, and describes CNC machines as bricks: they sit there, they run continuously, and they only require scheduled maintenance.

Well, yeah... Sixty-five percent.

A machine costing €1-2 million, sold to aerospace or defense manufacturers, is only confident about performing as intended for roughly two-thirds of the time.

The remaining third is spent doing something every owner of a cheap desktop printer from a decade ago remembers all too well:

Breaking down.
Being repaired.
Producing failed builds.

This leads to a conclusion Davies never explicitly states, but one that follows directly from the numbers.

Today’s €1 million metal AM machine has roughly the same reliability profile as a $200 desktop printer from 2018.

It’s a metal Ender 3.

The only difference is that an Ender 3 cost about as much as a decent bicycle.

For AM veterans, that’s hardly surprising. For a new customer who’s just signed a seven-figure leasing agreement, it’s a punch straight to the solar plexus.

And the purchase price is only the beginning. As Davies writes:

The point-of-purchase price is not just a challenge for smaller operations, but also for larger corporations just getting started, and even experienced users looking to unlock a new application. Even when the machine advancements do pertain to the business case – such as the productivity improvement a larger laser configuration can provide – that only serves to help existing applications. Because of the cost hike a machine with multiple lasers will demand, the business case for a new application often doesn’t make sense.

In plain English:

Even when manufacturers genuinely improve their machines - for example by adding more lasers to increase productivity, it only benefits applications that already exist.

For new applications, the business case still falls apart because multi-laser systems become prohibitively expensive.

The machine gets bigger. The investment threshold rises with it. The cycle repeats itself.

Calmese estimates that reducing machine price by 10% lowers part cost by only around 3%. Increasing utilization from 65% to 85%, on the other hand, would fundamentally change the economics of the entire industry.

The real question is: where will that improvement come from?

Well… There’s no shocking plot twist here.

The answer, once again, is software.

Hardware alone is not enough

Let’s go back to 2022–2023, to what completely transformed low-cost desktop FFF.

That revolution wasn’t driven by hardware. It was driven by software. Firmware. Automatic calibration. Slicers capable of selecting parameters automatically. Integrated ecosystems of validated print profiles and ready-to-print models.

Suddenly, a $500 printer became repeatable and predictable. A cheap machine became something you could actually trust.

That’s exactly the transformation metal AM has yet to experience. Because all the fundamental 3D printing technologies have already been invented. Hardware can be copied.

What’s left is process preparation, validation, process monitoring and post-processing.

And every one of those areas is fundamentally a software problem.

Yes folks, hardware alone is no longer enough.

The manufacturing process stops being a collection of settings stored inside an engineer’s head and becomes a formal, programmable description. Something that can be version-controlled, compared, validated, and automated.

That’s what moves reliability from 65% toward 85% without adding a single extra laser.

Companies like Dyndrite are pushing metal AM in exactly that direction. And that’s where my cautious optimism comes from.

Real progress in metal AM is happening. Quietly. Away from the spotlight.

It’s happening inside the code that simply makes machines work.

Metal AM is still waiting for its own 2022 moment. From where I stand, it’s no longer a question of „if”, only „when”.

“You can’t see it yet, but below the surface the fire is growing.”

China is driving down the cost of technologies that, for years, have justified their premium primarily through price. The question is whether a desk is really the right place for them?

Last week I wrote that it had been an exceptionally good week for metal 3D printing. I also said we’d wait for more details before drawing conclusions.

Well, the details arrived sooner than I expected - along with the price tags.

Glorious Platinum has unveiled the price of the DP-C1, its “consumer” metal SLM printer: $4,888. Meanwhile, Guangzhou-based HeyGears has opened pre-orders for its G1 series of full-color UV material jetting printers, starting at $1,699, with a Kickstarter launch scheduled for July.

Two launches. One strategy.

China is no longer content with dominating consumer FFF and SLA. It is moving into metal and full-color UV printing - and doing what it always does: driving prices down.

That leaves us with two simple questions:

What exactly do you get for $4,888?

The DP-C1 measures 500 × 515 × 830 mm and features a cylindrical build volume of Ø108 × 100 mm. It is equipped with a single 300 W air-cooled fiber laser and currently supports just one material: stainless steel.

Glorious Platinum was founded in 2023 and claims to have sold more than 1000 machines across 60 countries. The company already offers a desktop metal printer for the dental industry, so I’m not assuming anyone is making things up.

I’d simply like to see what an air-cooled 300 W laser does to stainless steel over the course of a full build.

The $4,888 price tag certainly grabs attention, but there’s an important footnote.

It only applies to export orders of at least 20 units.

In other words, it’s a dealer price and does not include shipping, taxes, or import duties.

The domestic Chinese version - the upgraded DP-C2 - costs around $8,779, with a minimum order quantity of 10 units.

So yes, the “desktop printer” is sold in batches of twenty.

At the same time, the company is following exactly the same playbook we’ve already seen from consumer FFF manufacturers. The hardware is only part of the package. Around it comes AI, a model library, and an online ecosystem. The DP-C1 can generate models from photos, text, and voice prompts, while the IronNova platform includes a library of 10,000 free designs.

So no worries about designs. The problem is that metal printing is difficult for entirely different reasons.

With SLM, the finished part comes out welded to the build plate and buried in metal powder. Someone still has to remove the loose powder, cut away the supports, and often perform stress-relief heat treatment.

And having fine metal powder in a home office is, frankly, an absurd idea for which there is no simple solution.

Anyone with even a basic understanding of metal additive manufacturing knows that the real work begins after the print is finished.

That hasn’t changed.

HeyGears takes a different approach

HeyGears is attacking a different segment - and doing so more sensibly.

The G1 is a full-color UV material jetting printer. Parts come off the platform already in full color, and a single print can combine opaque and transparent elements.

The printer will be offered on Kickstarter. Early bird prices are going to be $1,699, $2,999, and $3,299.

Shipping is scheduled to begin in 2026. For now, customers can reserve a unit with a fully refundable $50 deposit.

Of course, tariffs, taxes, and shipping costs are still unknown.

The G1X version features an Epson i3200 printhead, eight ink channels, support for more than 10 million colors, water-soluble support material, and compatibility with more than 400 substrates, including metal, acrylic, and textiles.

The company is also building the same kind of ecosystem around the machine: HeyVerse software, Blueprint Studio, a library of more than 500 ready-to-use assets, and AI-powered model generation from text and images.

Full-color printing could genuinely be useful for small creative studios, making this launch sound considerably more practical than the idea of a metal printer sitting on your desk.

The more interesting question is who HeyGears actually wants to compete with.

Is it Stratasys and Mimaki? Or is it companies like Anker and EufyMake, whose Kickstarter campaign broke funding records last year?

Of course, the HeyGears G1 is not a 2.5D printer like the EufyMake - it is allegedly true 3D printing platform. But knowing Chinese hardware companies, I wouldn’t even be surprised if they had never heard of “Objet Geometries” 😉

Does any of this make sense?

The pricing certainly does.

China’s cost-reduction machine has finally reached technologies that, for years, defended themselves with premium pricing: metal additive manufacturing and full-color PolyJet-style printing.

The direction is obvious.

The first wave of pricing pressure will hit the segments where every thousand dollars on the invoice matters most - small studios, service bureaus, and the mid-market.

Putting a metal SLM printer on a desk is another story.

At least for now, that sounds much more like clickbait than a sensible solution.

Atomic Layer from the Past:

13 years ago, Microsoft announced native 3D printer support in Windows 8.1, enabling direct printing from third-party apps via a simple “print” command. A demo showed a vase printed on a Makerbot Replicator 2 without extra drivers.

Microsoft partnered with Makerbot, 3D Systems, Formlabs, Autodesk, and Netfabb. The Type A Machines Series 1 became the first open-source printer with direct support, followed by Fabbster and UP! 3D.

Although the initiative ultimately faded, it sparked two key developments: the 3MF file format (to replace .STL) and the 3D Builder app, later succeeded by Paint 3D.

Read all:

News & Gossip:

#1

Congratulations to Jigar Patel from the University of Waterloo, who took first place for best oral presentation at HI-AM 2026. The topic: vector-level toolpath optimization for tighter LPBF process control.

Sounds dry, matters a lot.

A warm nod to Dyndrite too, because the work was built on Dyndrite LPBF Pro, and the Dyndrite team fielded all 124,534 of his questions without flinching. Climbing that learning curve clearly paid off. Nicely done.

#2

ExOne is back with the S-Print Pro, a compact sand binder jetting system aimed at the small foundries and pattern shops that everyone else skips.

Under 12m2 of floor space, a 1200 x 750 x 500 mm build box, 400 dpi CoreBoost printhead, and a full job box per shift. The pitch is affordable to buy, install, and run, which is exactly what small shops have been asking for.

Worth noting this is the first real product since the ExOne and voxeljet merger, so it carries some weight. Deliveries start in the second half of 2026, beta partners first. We will wait for the parts.

#3

Phase3D, a Chicago metrology outfit, just closed a $2.9 million oversubscribed round led by Quest Venture Partners.

Its product, Fringe Inspection, uses structured-light metrology to capture layer-by-layer height data during metal powder bed fusion, so you can catch recoater issues, spatter, and powder bed problems while the build is happening, not after you cut it open.

The money goes to manufacturing capacity and more deployments. This is the unglamorous software-and-inspection layer that metal AM actually needs to grow up. Small round, sensible direction.

#4

Dr. Dieter Schwarze, co-inventor of Selective Laser Melting and Chief Engineer at Nikon SLM Solutions, has retired after a career spanning more than 40 years.

He had a hand in every generation of the company’s technology and holds over 100 patents, which is the kind of number that quietly underwrites a whole industry.

Roughly 30 years ago he helped turn SLM from a concept into something foundries and aerospace shops now treat as routine. The company sent him off with a guard of honour and a 3D printed trophy, which feels exactly right.

#Respect 🙏

#5

TPM, the South Carolina solutions provider (not the Chinese TPM3D who manufactures SLS 3D printers), has sold its HP 3D printing and Advanced Manufacturing division to Master Graphics, an HP-only specialist.

All contracts and service relationships move over, and the account manager goes with them. TPM frames it as sharpening focus, keeping its 3D Systems, Artec, and SOLIDWORKS lines.

#6

And finally, a hint at a bigger story still developing. Two prominent Würth Additive figures have announced they are looking for new jobs, while letting you read between the lines that the company itself is done. Nobody, however, wants to confirm any of this on the record yet. So for now it sits somewhere between rumor and an unconfirmed fact. Keep watching.