It’s been nearly a year since we announced the Original Prusa MK4. Lately, I feel like time is flying by much faster than ever before, and reading the previous sentence only confirms this feeling. It all seems like yesterday. However, looking back at what we achieved in the past twelve months, it’s clear we didn’t stop for a single moment.

The MK4 quickly became our fastest-selling machine ever. Its popularity, however, has been bit of a double-edged sword, as more demand drove up the lead times. The same thing goes for the XL. Most of the year, we were exploring new ways to expand our production capacity to push the lead times down and finally ship a massive portion of the Original Prusa XL preorders. It took a lot of effort but, eventually, we made it – and now production runs smoother than ever before, and we’re well on our way to ship out the remaining XL preorders during spring. And that’s not all! Check out the box below where I summarized some of the most important things we achieved last year.

2023 Summary

  • MK4 has been receiving improvements since the day it started shipping. Not only has MK4 become much faster with Input Shaper support but we also brought in quality of life improvements like the Cancel object feature or faster network speeds. And we also backported the Input Shaper to the MINI, a printer released 4 years ago – making it feel like a brand new printer.
  • In the second half of 2023, we bumped up the production capacity to nearly 200%. This allows us to push the lead times as low as possible. We’re very close to the point where we will have MK4 in stock with no lead time at all.
  • We have started shipping all XL versions. The increased production capacity enables us to fulfill all preorders during this spring. In addition, XL upgrade packages are now ready and available to order in our eshop.
  • Two new PrusaSlicer versions were released, full of important new features – such as text embossing on curved surfaces, organic supports, plane cutting with automatic pin generation, built-in dimension measuring, and more.
  • The MK4 and MINI have been added to Prusa Connect. The recent updates to our remote print management system brought a massive speed boost to network printing. The print starts mere seconds after you send it to Prusa Connect thanks to the new G-code streaming option.
  • Printables.com saw the launch of two major features: support for creators via Clubs and the option to open a Store. Printables is now more than twice as big compared to last year. Check out the infographics below!
  • We won two awards in the 3D Printing Industry Awards! The first one was for the Company of the Year, and the second one was given to me by Dr. Adrian Bowyer, the father of the RepRap project, for Outstanding Contribution to 3D Printing – a great honor and honestly, it means a lot to me!
  • With new printer models, variants and record numbers of sold units, our 24/7 tech support had to answer more questions than ever before, so they had a record year, too. And while chatbots and automation have improved substantially in just the last year, direct contact with a real person is still proving to be the best approach for customer care, so we’re sticking to it. We have also prioritized the development and maintenance of our extensive Prusa Knowledge Base.

Every morning, and every evening when I come back from work, I try to read as many of your comments on Twitter, Reddit, Facebook and other channels as possible. Truth be told, I probably spend a lot more time on social media than I should. It’s hard to resist, though. Our 3D printers became what they are thanks to the 3D printing community. As cliché as it sounds, it’s true: we would be nothing without our community.

And of course, I’m not talking just about positive comments, awesome prints and huge 3D printed projects. I read about your concerns and frustrations, too. Your feedback often leads to long internal Slack conversations where we’re discussing how to improve this or that. Honestly, sometimes I’m also frustrated – not only when I see we could have done something better but also when some of our actions are misunderstood. And this brings me to this blog post.

Usually, my articles are about announcements of new products, or they cover the latest updates and shed light on what’s in the pipeline. But lately, I’ve been thinking that it might be good to revisit why we’re here and why we do things the way we do. Many of you have “grown up” with us, and your first printer might have been the MK1 or MK2. We probably even met in person at various events, or maybe I personally helped you build a printer ten years ago. Since then, however, a lot of water has flowed under the bridge; the market has changed, and so has the mood in 3D printing communities. We’re evolving too, constantly seeking the right balance between remaining that enthusiastic group of garage-born inventors, while also being capable of shipping over a hundred thousand 3D printers a year.

So, let’s take a look at some of the challenges that we need to tackle when we want to continue manufacturing 3D printers the way we’ve done over the last decade, and answer some of the most frequently asked questions on social media.

 

“Why don’t you produce printers faster?”

The MK4 is selling faster than any of our previous models. In fact, if we had a larger production capacity, I think it would sell even better because the current lead times are likely putting some people off.

We made plenty of changes to our manufacturing process throughout the year. Their effects are, of course, not instantaneous – the improvements come into view gradually. From the outside perspective, it may seem like we’re not doing enough and I see where that’s coming from. Although we still hire new colleagues (over 150 in 2023 and we’re not stopping), rent new manufacturing spaces and optimize the manufacturing process, the lead times are still there – however, they are much shorter than in summer 2023. The last few months have been absolutely record-breaking for us in terms of the number of printers manufactured and shipped. We’re getting close to the moment when the MK4 will be in stock. However, some factors will continue to affect the availability of our printers.

Our printers usually have more than one version. You can buy an assembly kit or a fully assembled printer, you can configure the number of toolheads on the XL and so on. Each of the variations requires a different workflow. A factory-assembled printer is prepared and tested differently than an assembly kit. To have more XL configurations with a different number of tools means longer testing. And the manufacturing process for a single-head printer and a five-tool printer is completely different.

It would have been way more efficient for us to abandon the kit version of the MK4 and only sell the fully assembled version. But I cannot imagine doing that. Every event I go to, every online community, you all keep reminding me that all the effort we put into the kit version is worth it. Not only is it a fun project that kids can build with their parents, but it’s also a great way to learn how the machine works from the inside out, so you can easily perform maintenance and repairs on it. One of the most rewarding feelings is when I hear that building the kit sparked an engineering interest in someone young.

Helene Virolan on Twittter, Tim Searles on Twitter, Kaliber 2020 on Twitter

The same goes for our upgrade kits. With each new model, we try to offer a way to upgrade from the previous one, which is quite unusual in the world of home electronics or tools. One of the things our upgrades allow you to do is to choose the upgrade path – whether you want to go all-in and get the full upgrade, or just a partial upgrade that brings a lot of nice-to-have things at a more affordable cost. That’s why we brought the MK2.5, MK3.5 and MK3.9 upgrades to the market.

Honestly, it takes a lot of effort to design the upgrades in a way that makes sense both for you and for us – we need to integrate them into our manufacturing process, allocate enough components to them (components that would otherwise go into fully assembled printers) and we also need to consider various upgrade paths the users can potentially take – can the MK3 (non S+ version) be upgraded directly to MK3.5? Should it be possible? What components would have to be added to make it compatible? And then we need to prepare all the assembly instructions, etc. It’s a lot. However, it was one of the first things we established here: upgrades are a part of our DNA and we’re going to push them as hard as we can, even though it may not be the best course of action from the economic standpoint.

We don’t want to churn out new types of printers every few months and then phase them out in a year or two. We keep a long-term stock of spare parts even for models that we stopped producing a few years ago. For example, we still have a stock of MK2 Mini Rambo boards, and we still provide firmware updates for the original MK3, a model that was released seven years ago.

We design our printers so that they can be assembled at home. This means that everything is easily accessible and repairable. You won’t have to take apart hugely complicated assemblies to change a bearing with uncertain results or have to break layers of glue. You can perhaps think of it as making classic desktop PCs. Sure, they might not be as sleek and stylish as consoles, but you can easily replace any component.

Thanks to this approach our old printers do not end up in landfills but are still in service. I have a lot of friends who are still happily printing on their old MK2s today, and they don’t forget to remind me of this all the time. 🙂

“I can get the same quality printer for half the price”

We manufacture our printers, filaments and resins directly in our HQ in Prague, Czech Republic, in the heart of the European Union – meaning that there are components that need to be imported from other countries. After the COVID-19 pandemic, we fully realized that being so dependent on faraway suppliers is too risky, so we decided to change things up.

Our in-house SMT line is used both for prototyping, but also full scale production of MK4/XL/MINI PCBs.

We started finding ways to get components locally and even produce them in-house (printed circuit boards, for example). We’re also looking for more suppliers in the EU and USA – they usually provide higher quality parts. However, they are usually also more expensive and they often have lower manufacturing capacity, which may cause situations like a sudden shortage of important components. Some can be sourced from other suppliers (which takes time), other parts may be so specific that we have no other option but to wait until the supply returns back to normal. Very often, the problem isn’t in the quantity, but in the quality.

It is quite common for a supplier to change some seemingly insignificant manufacturing process and even parts with the same specs suddenly show different results in our tests. This can negatively affect the printer’s performance.

To give you an example: some MK4 printers from the initial batches had stepper motors causing unexpected noise, contradicting our claim about the extremely quiet operation, something our printers are well known for. It was hard to find these noisy motors because they were mixed randomly in different batches. It took our team two weeks to create a special device that could quickly identify them. We used this device to check motors in our factory and also built another one and sent it to our motor supplier. Once we fixed the quality issue, we had to return many motors we already had in our warehouse. This situation has made our production a bit slower and more complicated.

Similar problems happen when a supplier suddenly changes any of the used materials or a step in the manufacturing process. This one case was unusually complex, but similar (although less severe) issues are something that our purchasing and quality departments need to deal with several times a month. Dealing with such problems in the supply chain happens often and can cause unexpected delays, which is why we sometimes inform you of these issues at the last moment.

A dedicated department focuses on the development and upkeep of custom testing machines

From time to time, I see a simple suggestion: “Well, forget the local suppliers, I want my 3D printer and I don’t care where you source your parts.” In light of the problems described above, it’s not a universal solution. You will find parts sourced from China in our printers because some components are manufactured in sufficient quantities only there. And things are more complicated than just finding the supplier. Government subsidies, tax breaks and strategic efforts to dominate the 3D printing industry (similar to what happened in the market of drones, pharmaceuticals, or currently, electric cars) have resulted in such situations as diverting ordered and already paid-for components to Chinese companies.

I’m trying to state the facts with as little emotion as possible. Simply to give you a bit of background on what it means to produce something in the EU. It’s obvious that when you’re choosing between various brands, the questions of “morality” or “fate of the industry” are definitely not anywhere near the top of the priority list. And that’s ok. But as perhaps the last western manufacturer of desktop 3D printers, we have to fight back. Because we believe that in the long run, our approach will pay off for both us and you.

“You have stopped innovating!”

Comments like this are probably the most upsetting. I see Prusa Research as a company that innovates and comes up with new ideas to give the 3D printing community new options and new tools. We have always been like this and will continue to be. The Original Prusa XL is still a one-of-a-kind printer with its tool changer, segmented bed, minimum waste printing, and automatic offset calibration for each toolhead. It allows for combining multiple polymers together in a single print and it has a ton of other cool features.

The MK4 is just as packed with new features that may not be obvious at first glance but it clearly pushes the user experience and print quality forward. We’ve come up with embedding a Load Cell sensor into the heatsink for fully automatic first layer calibration. We’re developing our own Buddy platform with a host of security features, we’ve come up with our own implementation of Input Shaper and Pressure Advance written directly for the bare metal STM32, and much more. We have dedicated teams developing features such as pressure turbine cooling on the HT90, or image recognition on the AFS, things that will inevitably trickle down to our future desktop 3D printers.

Our polymer team developed the Prusament PETG V0, which passed the strictest level of the UL certification for self-extinguishing. We have also released the extremely unique PETG Tungsten. This filament is filled with tungsten powder by 75 % in mass, and it can be used for 3D printing not just (otherwise very expensive) radiation-shielding components, but as our community discovered, also for printing fishing lures! And we have we created a new line of 3D printing materials by using recycled materials & biobased pigments from food and pharma byproducts, a project 2 years in the making.

And then of course there’s PrusaSlicer. Try running a year (or more) old version. There’s no end to the amount of features our 13-person team has added. Things like the (automatic) support painting, text embossing on curved surfaces, organic supports, plane cutting with automatic pin generation, built-in dimension measuring, smart brush for MMU painting, changing the orientation of the model by clicking on the surface, modifiers… I could go on like this for several more paragraphs. In addition, a lot of the new features affect the very core of G-code generation. These changes are more “under the hood” and less visible, no new icon appears in the UI. Actually, something may even disappear from the UI because it is no longer needed. However, the direct impact on print quality is usually far more significant. Whether it’s anchoring the inner fill to the perimeters, dynamic deceleration on overhangs, variable extrusion width, or perhaps nozzle pressure equalization, we carefully pick what to add to our slicer. With all this in mind, I would go as far as saying that there has been a much bigger leap forward in 3D printing in the last few years just because of the new features in PrusaSlicer (and in Cura!) than in terms of hardware.

These innovations benefit the entire 3D printing community, regardless of whether you have a Prusa 3D printer. You don’t even have to use PrusaSlicer, the innovations are quickly appearing in all the derivative versions, from SuperSlicer, OrcaSlicer, Bambu Studio, to Pantheon Slicer, Anycubic Slicer and more. It would definitely be easier and cheaper for us to use one of the existing solutions and just “rebrand” it to our colors.

We don’t close our software and firmware to others, but that doesn’t mean we won’t tailor it to our printers. Our firmware team consists of 21 people who are constantly working on new features while also keeping security in mind. That’s one of the benefits of using our own custom-made embedded platform. We take into account the design of our printers and their strengths, and make sure our firmware is simply perfectly tailored for them. There’s another thing that goes along with that – we don’t want to release products without knowing exactly what’s under the hood. And most importantly, where YOU don’t know what’s under the hood. Our printers are primarily designed as tools that must work flawlessly offline in the first place. Online features should be a convenient extra, not a necessity, without which half of the features do not work at all.

3D-printed parts are a bottleneck

Every once in a while, I see a comment suggesting that we should stop using 3D printed parts on our printers, as it is slowing our production down. While I see where the logic behind such comments is coming from, it couldn’t be further from reality. Printed parts were never a bottleneck in our production. After all, we can always add more printers to the farm, optimize the G-code or use injection molding for the parts that take the longest time to print.

Yes, we do actually have injection molding lines in-house. However, molds are really expensive and take weeks to get right. And you are limited with the geometry of the parts – there are shapes that injection molding simply cannot do. On the other hand with 3D printing, we iterate the parts and constantly improve them.

Upgrading the MK3S+ to MK4 on our print farm basically doubled our production capacity of printed parts

Last but not least, printing the parts on our farm means that WE ARE USING OUR OWN PRINTERS all the time! 🙂 Our manufacturing is literally depending on our own product, which is great! It pushes us to make the printers better and if there is ever something that could use an improvement, we’re often the first to discover it. That said, we are using injection molding where it makes sense. Typically bigger parts that rarely need updates – the spool holder, MINI front panel, etc. We can even reuse our plastic waste for injection molding, which is nice.

Surprisingly, I’ve even encountered a few posts where people have actually criticized us for using printed parts in our printers at all. But here, I feel I don’t need to convince you much about the countless benefits of this approach. Every school, business, or organization we talk to quickly mentions how great it is to have the option to print spare parts on-premise, download and print an improved part, or customize it to their liking.

Having competition is healthy

At the beginning of this blog post, I wrote that the MK4 is selling faster than the MK3 did back in its day. Thanks to that, and also thanks to XL and Prusament, last year was a record year for us. Either way, sometimes it’s hard to skip snide remarks about how the competition will destroy us and that it only serves us right because we did a thing (or didn’t do another thing), so we had it coming!

To this, I can only say that competition isn’t bad. It’s, in fact, very good and beneficial! We’re talking about a market, not a deathmatch. 🙂 We have never been a dominant player (according to independent estimates, we have a share of around 10 percent in the home 3D printer segment) and I sincerely hope that no manufacturer will dominate the market in the future. Without competition, the world of 3D printing will slow down, independent communities will gradually disappear, users will lose their choices, and there will be no push to innovate.

I do not intend to make light of our mistakes and shortcomings, or even hide them. I know we have them. But I want you to know that we are doing everything we can to eliminate them and bring you better and better printers as well as all the services around them. I hope that this text gave you a better understanding of how we do things here and why we do them the way we do them. We have a lot of exciting new things planned and I can’t wait to tell you more about them in future articles.