Consumer 3D Printing in 2026

Within the hobby/DIY community 3D printing has become practically an absolute necessity for makers of all kinds. Within the world of hobby robots and Combat Robotics, 3D printing has changed everything about how robots are designed and built today. This shift happened slowly at first, but has massively accelerated since 2022. I thought it may be a good time to reflect on the state of things as of September 2026. I'll cover why you likely already own, or should own a 3D printer. I’ll also touch on why I am discontinuing my own 3D printing service, despite this technological advancement. TL;DR: As of today my own 3D printing service is ending. If you want a reliable 3D printing service within the combat robotics community, check out Team Malice!

Here is an old video from 2021 I made about 3D printing materials for robots. Keep in mind it is fairly outdated though still relevant in broad strokes. 

Part 1: Before 2017, 3D Printers All Used to Suck

I began my 3D printing journey all the way back in 2015 when I first entered high school. I joined the high school science olympiad team and gained access to a donated MakerBot Replicator 2. This 3D printer was, by modern standards, an utter piece of garbage. It had major bed adhesion issues, one of the most poorly designed extruder systems imaginable, could essentially only print PLA, and required the bed to constantly be covered with blue painters tape which was replaced after every couple of prints. It jammed frequently and was a pain to service. Also, at retail it cost around $2400 USD. The print volume was 285 x 153 x 155mm and if you ever made an attempt to print something nearly the size of the bed, god help you. It would almost certainly warp to hell and pop off mid print. However, it sometimes did make it possible to 3D print a functional part, which for my build projects at the time was pretty amazing. My school didn't have a lathe or milling machine, but we did have this 3D printer that worked some of the time. I taught myself CAD in Autodesk Inventor and learned to use and fix this printer so that my whole team could utilize its capabilities.

After I graduated from high school I attended the University of Rochester. While I was there I had access to and utilized a number of other 3D printers. There were a few XYZ Davinci machines for public paid use, Printrbot Metal printers in one lab, a number of Replicator 2X machines (Replicator 2 but with dual extruder, even more expensive, and still total crap). I came across and revived an old Afinia that was only marginally better than the Replicator 2 in functionality.

After founding the UofR Combat Robotics Team in 2017, I also managed to convince the school to buy a Creality CR-10. At the time, this was the best 3D printer I had regular access to by far. By today's standards it was still utter trash, but it was the only one of these printers that had a smooth glass bed which made bed adhesion much more reliable, and it had an enormous print volume of nearly 1ft cubed (300 x 300 x 400mm). It was painfully slow but it did actually print objects successfully more often than not, more than I could say about most of the other printers. 

The real star of the show however, was in Gleason - a sort of makerspace area with mostly a bunch of manual lathes, milling machines, and other machine shop tools. There, two enormous and expensive industrial 3D printers were proudly on display right at the entrance. One we won't discuss as it was not a filament printer and barely worked, a ZCorp ZPrinter 650. The other was more notable: a refrigerator sized Stratasys uPrint SE Plus. This machine actually had a smaller print volume than the CR-10, and printed significantly slower typically. But it had a unique dual extruder system that was locked into proprietary Stratasys ABS-like filaments on one nozzle and a dissolvable support on the other. This allowed significantly trickier geometries to be printed successfully with really solid feeling parts. Too bad the filament was $200 per pound (but I made good friends with Jim the shop manager and usually got parts for free anyhow). The downside to this printer was that we only had one. It was slow, occasionally broke down and needed servicing, many other people needed to use it, and the machine was around $20,000 brand new so it wasn't exactly like I could use it whenever. So the $350 CR-10 became the primary printer for me at the time we received it.

This was pretty much the state of consumer 3D printers back in early 2017 and before. Machines were slow, ugly, unreliable. Bowden extruders were the norm, and it was not a guarantee you would get a heated bed with a lot of the cheaper models. The only filaments you could expect to use were PLA or PETG most of the time, and while many printers claimed to print ABS it would warp off the bed and crack apart with these unenclosed machines. You would be extremely lucky to get anything with a direct drive extruder that could print TPU, like, at all. I had one research project to test out if we could 3D print artificial skin with soft TPE for prosthetic limbs using the Printrbot Metal printers, and we were struggling to get reliable prints at speeds like 5-10mm/s. However, I was aware there was this very well regarded company called Prusa making waves in late 2017 with the release of the Prusa Mk3.

Part 2: 2017-2021, Creality Clones, and Prusa Mk3

Right around the time I graduated college in May 2019, I wanted to make my own combat robots and lost all support from my university. I joined a local makerspace and there found a few Prusa machines which were actually pretty functional for the time. The ease of use of the hardware/screen/dial system was a lot better than I was used to. And PrusaSlicer was years ahead of the other slicers I had used to date. I was originally stuck with proprietary slicers from MakerBot, Davinci or Afinia, and the research labs all used the paid slicer Simplify3D. I also dabbled with Cura to use the Creality CR10. But PrusaSlicer was different. I found it far more powerful, flexible, easy to use and configure than Cura. In 2019 I couldn't really afford a $750 Prusa Mk3S kit for myself though - I used the makerspace Prusas a bit, but eventually decided buying my own printer was the right choice for how much I needed to use it for my bots. I decided on the Artillery Sidewinder X1 - this machine was extremely hyped at the time and with some good reasons. This was still an 8 bit Marlin firmware machine and it lacked any sort of self levelling function like a BL-Touch clone. That came later with the Sidewinder X2 which I eventually upgraded to. But it had an E3D Volcano clone hotend allowing it to print big parts much faster than other machines, and a direct drive extruder that was far more reliable than most bowden setups common at the time. It also had a unique AC powered heated bed which would get to temp way faster than the DC 12V or 24V bed heaters common in this era. Bed levelling was still a fiddly manual affair but once it was dialled in, I got pretty consistent PLA prints off it, and this printer was what allowed me to make the very first version of Division.

When I started actually saving up money from my mechanical engineering job, I was looking at printers that might be able to handle higher temperature materials like Alloy 910 Nylon which was all the rage back then. This was long before every 3lb combat robot was all TPU, in fact flexible filaments were practically a novelty back in 2019/2020. This brought me quickly to the massive hype around Prusa with the Mk3S. The Mk3 series offered a few headline features that made it stand out from the rest. First of all, Prusa was pretty legendary for making a quality, reliable product back then, and the consensus online seemed to be that they were the only company making a machine you could count on to work without much fuss or setup. Part of this came to the unique PINDA inductive bed probing and the magnetic steel flexible bed - again, this was by no means a common or guaranteed feature at the time. I ended up purchasing a Mk3S kit in October 2019 and built it myself over a weekend. I also used the makerspace laser cutter and some scrap acrylic to fabricate a cheap enclosure around this printer so I could attempt some high temp nylon printing. 

Full disclosure - I recently partnered with Prusa for an Open Sauce project for which they gave me a Prusa Core One + for free. No money changed hands and I have not signed any contract with Prusa restricting anything I say about their products. I never had any relationship with Prusa, or any other 3D printing company, prior to 2026, and paid cash at full price for every other printer discussed in this article.

The Prusa Mk3S was what I can comfortably describe as good but not great. It definitely had its faults. It was extremely annoying and time consuming to repair or replace parts of the extruder or hotend, for instance, and while it used standard E3D V6 nozzles and other parts, the wires running through an umbilical back to the control box made it a time consuming process to fix a broken thermistor for instance. Blob of Death was a serious issue that took hours to fix. But, overall, it worked much more consistently than my Sidewinder printers, and it allowed me to focus more on building robots and less on fiddling with my 3D printer. This was the point where I’d say we really began to see a shift from a 3D printer becoming a hobby, to becoming a tool. 

Interestingly, the best selling 3D printers up to this point were all bedslingers which for the most part just looked exactly like the same aluminum extrusion frames as Creality had been producing for years now. They were all basically “Ender 3 but with X added” or “CR10 but with Y added”. The Artillery machines were among those - CR10 clones with some added features, all at a very low sub $500 price point. Prusa had some clones as well, but the way they were built was more expensive. So most of the off the shelf parts available on amazon or aliexpress were still very much aimed at upgrading aluminum extrusion, bowden extruder machines. But, this was about to change.

Part 3: 2022-2024, Enclosed CoreXY Revolution

So begins the Bambu Lab Era of 3D printing. I bought an X1 Carbon in early 2023, about six months after the Kickstarter, after watching a ton of positive reviews about it on YouTube. Yes there are a ton of other 3D printing companies with their own CoreXY machines, but with the Kickstarter release of the X1 Carbon, nearly every other company shifted their focus to copying Bambu Lab. I believe that there are a few key reasons for this.

1. When Creality started dominating the 3D printer market, it was by making the worst possible 3D printer that could still function at the cheapest possible price. This was the original Ender 3, and at around $200 it was by far the cheapest entrypoint to 3D printing. 

2. As other companies tried to steal market share from Creality, nearly all of them decided the way to do this was to sell a printer with more features but the same overall form factor for slightly more money. Feature creep was the name of the game here. Charging a bit more money for more features. This created a race to the bottom for all companies, including Creality, whereby they would try to cram as much functionality into their machines for the lowest possible price.

3. Bambu Lab took a VERY different approach, and went after a VERY different target market. Rather than targeting hobbyist 3D printer tinkerers who wanted to modify and upgrade their machines but pay as little money as possible upfront, Bambu Lab went way upmarket. They designed a sleek, attractive, fast CoreXY machine with a full enclosure, massively more sensors and preprint calibrations, capable of reliably handling engineering materials which were next to impossible to work with on the standard open frame bedslingers. 

4. Most importantly, they launched it at a $1200+ price point, and their focus was on the customer experience and ease of use above all else. They made almost no compromise for the sake of cost cutting, instead opting to get their cost savings by investing in economies of scale. The moment the X1C launched, Bambu had tooled up to produce and sell tens or hundreds of thousands of machines. They made a huge financial gamble by doing this but it paid off for them, as they did not end up with 3+ month delivery times after the kickstarter ended. Prusa on the other hand routinely suffered from extremely long queues to receive a printer after theirs launched, with the assembled machines being more months delayed. 

5. Bambu innovated in many ways with their X1C by developing the AMS which was, while not entirely new to the world of 3D printing, still vastly superior to pretty much every multimaterial solution that existed beforehand. This was another way they focused heavily on user experience, while opening up the possibility to print multicolor models for everyday consumers and making loading and unloading filament much less of a chore. 

6. By launching the X1C at a price more than double that of most other 3D printers on the market, all the companies who were scrounging for market share with shrinking margins at a $300 price point suddenly could offer a machine for $1000 and still undercut the market leader. This opened up a huge amount of potential profit for the competition. 

7. Bambu Lab also proved definitively that there was a new market of “prosumer” and small business 3D printer buyers out there, and larger companies who were previously using $20000 Stratasys and other industrial machines realized they would be able to get much more productivity from ten X1 Carbons at a fraction the cost. 

8. Among serious 3D printing hobbyists, many have been concerned about the closed source walled garden approach Bambu Lab has taken. Whether it is concern about data collection by a Chinese company, the possibility of more substantial vendor lock in the future, Bambu Lab’s hostile actions towards some open source projects/communities, or simply the lack of ability to tinker with Bambu Lab machines, there are many people who would never buy from them. Other companies moved to capitalize with similar printers that weren't as locked down. 

I ended up selling my Artillery Sidewinder. I kept the Mk3S and X1C at the same time for a while, but the Mk3S barely got touched ever since. The X1C also brought a new baseline of 3D printing speed to the market, able to print sometimes 4-8 times faster than the older bedslingers with no loss in quality. So eventually I sold that to a friend as well and got a Bambu Lab P1S in November 2023 as a second printer. At the time, the P1S was around $700 and offered 90% of the functionality of the X1C for half the price. This let Bambu Lab be competitive once again with all those clone machines that had appeared at a sub $1000 price point. Bambu had almost obsoleted the X1C at this point, which was a bit too expensive for print farms to buy tons of. 

The short lived P1P, and better value P1S remained strong value for another year or so, but Bambu quickly followed up with their first bedslinger printers in late 2023 with the A1 and A1 Mini. While these open frame printers harkened back to the era of 2019 in a way, they had the same bulletproof reliability due to advanced preprint calibration that Bambu was known for, a much more user friendly touchscreen than the P1-series machines, and brought standalone printer prices back into the $200 and $300 range while keeping the amazing user experience and ease of use for those new to 3D printing. The A series machines are pretty much perfect for anyone looking to open up an Etsy 3D printing store with a print farm, and to this day offer incredible value for the money if you don't need to print anything more advanced than PLA, PETG, or TPU except on rare occasions. Personally my A1 Mini has been at least as reliable for TPU printing as the bigger machines. Once again, they also inspired waves of clones and lookalikes from Creality, and a resurgence from other companies like Sovol, Anycubic and Elegoo looking to snap up customers who cared more about price and openness than quality assurance or customer support.

Meanwhile, Prusa has been in a perpetual game of catchup. With a disappointing and buggy Prusa XL launch at a $3000+ price point after a year of delays, poor Mk4 launch timing, and all their modern machines being far more expensive than rivals with similar or better feature sets. Their line of Core One and Core One L machines appear to be solid printers, but they make some serious compromises on features to maintain a semi-reasonable price point with a Made in EU sticker. That said, with nearly all their competition coming from China, they have positioned themselves well for the more cybersecurity conscious buyers and companies, and those looking to support a western brand name.  

Part 4: 2025-Present, Multimaterial, Toolchangers and More

The modern day and future 3D printing landscape are looking bright. We are starting to see Bambu Lab’s once market leading innovation fall behind some of their competitors. Bambu Lab releases only a few new printers each year while competitors collectively launch dozens. 

Bambu has been focussed on the engineering market for their more upscale printers, and has split their attention to processes beyond just 3D printing, a move many people are confused by or critical of. Bambu introduced the consumer to easily accessible multicolor 3D printing, but their filament switching AMS wastes a ton of time and material. In the last 2 years we have seen a handful of new systems from competitors like Sovol, Elegoo, Creality, Snapmaker, Bondtech/Prusa, and brand new companies like Lightmake, all of whom are offering their own multicolor systems that are faster, produce less waste, and often allow even more color or multi material options to be utilized in the same print.

 

Bambu Lab’s AMS system has always had some major limitations - any filament that is too brittle and stiff or too flexible would crack or jam and the original AMS wasn’t airtight enough to be useful for Nylons. Multicolor TPU was not possible, nor was printing with dissolvable PVA support. Bambu attempted to solve this at first with their dual extruder H2D which dedicates a moving right nozzle to some materials and a stationary left nozzle to others. However they do not recommend printing TPU in the moving nozzle to avoid jams, and in general the long feed tubes make it difficult to print with at times. The AMS is still needed to swap between materials when more than 2 are utilized in a print, with super long wait times to retract, feed, and purge materials during each swap, and a purge tower needed to prevent color bleed. Contrast this with a true toolchanger like the Snapmaker U1 at $849 which can print with 4 colors of TPU with no problems, or the Bondtech INDX system which produces almost 0 waste and can in some cases eliminate the purge tower entirely. It takes Bambu Lab’s AMS over 45 seconds to switch colors and wastes almost a gram of material each time. Toolchangers cut the waste to a tiny fraction of this and save considerable time as well.

Bambu has tried to address the waste issue with their Vortek inductive nozzle swapping system, but at launch the H2C fitted with it was around $2399 and currently retails for $2199 with AMS. This system eliminates most of the waste material but relies on the AMS so it's still a lengthy swap process.

I recently sold my P1S and replaced it with the new Bambu Lab X2D. The X2D has a heated chamber reaching 65°C, a main left extruder and auxiliary bowden right extruder which is designed specifically for support materials and accent colors, and it has the same print volume as the X1C it replaces in Bambu's lineup for just $649 USD without an AMS or $899 with the AMS 2 Pro which is also a heated filament dryer. Altogether it's a fantastic machine for engineering filaments, though the change to a servo-driven extruder and the bowden aux system does make it annoying to adjust print profiles from their other machines and 3rd party filaments. One of the best use cases for this is to print TPU with PLA support, or PLA with PETG support, both of which break away with little force leaving excellent overhang finishes.

Still, Bambu remains the king when you need a fully enclosed, active heated chamber, along with multimaterial capabilities... for now. 

 
Sovol announced an IDEX toolchanger still only on preorder, but can print up to 7 colors at a time and can use 2 toolheads at the same time to print duplicate or mirrored objects twice as fast. A newcomer, Lightmake, has a kickstarter right now for an extremely innovative design utilizing linear motors and beltless motion, and an independent Z axis on every toolhead, for 4 complete toolheads to be printing at the same time or collaborate on one multimaterial object. 

While all the other printer companies are making efficient material switching machines, open source developers have expanded the abilities of 4 color printing to replicate almost any color using Full Spektrum or similar processes. Bambu Lab meanwhile as of today just launched a standalone CO2 laser, and it has been clear since the H2D they have ambitions outside of just 3D printing as that machine supported both vinyl cutter and diode laser attachments. Their newest 3D printer, the A2L, to me seems like a reactionary push to prevent wholesale losing the battle for the big print volumes that cosplayers are after. This is the first time they have offered a printer with a 300mm x 300mm or larger build plate for under $1000, something that the CR10 offered way back in 2016 and has had a million clones since. Being able to one-shot a cosplay helmet has been the often mentioned target of these larger format machines for a decade. Personally as an engineer printing functional parts, it's rare a 3D print should be used for a part this large where a traditional machined, lasercut, or other subtractive part wouldn’t be better. But I can understand the appeal of wanting to just hit print and come back a day later to a full build plate of a dozen small parts. 

The MakerBot Replicator 2 hit the market in September 2012, almost exactly 14 years ago today. It took around 10 of those years to get to the point where a <$1000 3D printer could be considered a tool rather than a project. In just the last 2 years, there have been pretty massive changes as well. 

Part 5: Where do we go from here?

The future of the consumer and prosumer 3D printing world looks to be uncertain at the moment. I fully expect that multi toolhead printers will pop onto the scene more and more in the coming years. I’d love to see more innovation in allowing multiple nozzle sizes in the same print to get fine details while the bulk prints super fast. True 4th or 5th axis 3D printing to eliminate supports and make stronger parts is another direction that has been almost entirely ignored by commercial printers, but chased by hobbyists for years. And IDEX or Lightmake’s 4 toolhead simultaneous printing would allow huge productivity gains without taking up more space.

A few years ago when Bambu Lab first hit the scene, running a 3D print farm was a massively laborious affair. You were constantly dealing with printer maintenance and swapping filament spools and emptying the print beds. With a machine like an A1 Mini coupled with an AMS Lite however, a new spool can be fed automatically when one runs out mid print up to 3 times, so you can replace spools 4 times less frequently. And these printers are carefully calibrated before every job eliminating most of the pain of maintenance and print failures. This means that for under $600 you can essentially have a machine printing plastic objects 24/7 with almost no human intervention, something which was inconceivable a decade ago. However you still need to remove each print from the bed. 

There are a few commercial or open source projects that allow automatically ejecting prints or swapping entire print beds. In my opinion, the next horizon that needs to be tackled by Bambu Lab and their competitors is full print farm automation, making automated production work more approachable for a layman. This is something Prusa has teased for close to a decade, with a robot that removes prints from the bed automatically. IMO there is a really easy opportunity for a way cheaper and simpler solution to just crap parts out in front of the printer into a bin using the 3D printer’s own toolhead to push them off like existing hobby solutions already do. Having the ability to automatically queue up the next print from a queue with good software made by the printer manufacturer that's aware of all the other objects that need printing and every other printer in your fleet seems to be an obvious next step to me. One of the well known 3D printing youtubers already made an open source version of this called Watchtower in fact that's brand agnostic. 

Another direction Bambu Lab seems to be heading is making your printer into a “manufacturing hub”. Snapmaker tried and, in my opinion, massively failed to do this years ago with their 3in1 laser/printer/CNC machines. The core issue is that you need a lightweight toolhead and fast accelerations with low force to make a good laser or 3D printer, but you need an extremely robust rigid frame with ultra precise movements at consistent speeds to make a good CNC router or mill. I do not anticipate this compromise will ever really be worth it to have a single machine do it all. Also, the dust and fumes from a laser caking up all your 3D printer motion mechanics is a massive problem. But I digress...

Part 6: Winding Down my 3D Printing Service 

The original reason I wanted to start this article was to explain this decision. As you can see from all the above, 3D printing has been a major part of my life for more than a decade now. I’ve watched hundreds of 3D printer reviews for machines I would never buy to keep up on the industry as a whole. There is something amazing about the “just hit print” nature of some of these modern machines which makes it more and more a tool that anyone can use. I will keep using my printers to make parts for all my own robots and projects.

If you want to cheaply and reliably 3D print your own parts, now is the best time to get a printer for yourself. For better or for worse, that statement is likely to be true every day from now on. The tech is advancing fast, so it's tempting to wait for what the "next big thing" might be, but there will always be a "next big thing". Any recommendation I make will probably make me come across as a shill or be outdated in 6 months or less. That being said, if you want to print TPU and PLA combat robot parts for 3lb, ant/plastic ant, and smaller weight classes, it's hard to beat the Bambu Lab A1 Mini today. If you want to print more advanced materials, the Bambu Lab X2D is an incredible value. I own and use both of these machines personally. If you prefer to avoid Bambu's ecosystem, I also have a Prusa Core One and they are solid machines, though I would recommend hunting around on FB Marketplace for a used machine as they are fairly expensive at full retail. 

However, keep in mind there are still limitations for what a 3D printer can print. 

For one thing, materials are not free, and expensive engineering filaments can easily pass $50-100+ per kg. Another issue is supports - needed to print steep overhangs, removing them sucks, often it can ruin a delicate detailed part, and reorienting to eliminate or minimize them sometimes has other costs like making the part weakest in the direction it will be loaded. Drying filaments constantly uses a lot of energy and in my climate it’s a necessity for the majority of non PLA material I use. On their own, these are not huge problems, but consumer 3D printers have gotten so good and so affordable that most of my past customers just bought their own. The 3D print requests I still receive today are largely for printing the most difficult or obnoxious part geometries, or with the most expensive materials, or both. And it isn't sustainable to keep a huge stock of materials on hand when customers want like 20-50 grams at a time, once or twice a month, of a 1000g spool I paid for in whole. 

I usually try to bill for only about 15-30 minutes of my time, but when I need to actually print a part with a material that’s been on the shelf for a week, usually it needs to be dried. That means loading the dryer, setting it to dry for 6-12+ hours, then tediously feeding filament through these long bowden tubes to my printer, loading the filament, hitting print, and hoping the annoying geometry the customer needs does not cause the job to fail. Whenever a job fails, the time and material needed to retry and fix the issue usually vaporizes any profit I would have made. But I never feel great about billing the customer for my own failures, so I accept a paltry $5-10 of profit on a job where the customer paid $40 for the material, shipping, and operating costs (electricity, maintenance, wear and tear, etc). Quite simply, when you then factor in the time of slicing, quoting, emailing back and forth, the “oops that part doesn't slice please send the correct one” or “This is impossible to print please redesign it” or “That part is way too big to fit on my largest printer build area” etc, I really am making no money doing this anymore, and my time has been increasingly eaten up by far more important things. Like developing and testing amazing custom robotic components that you all need for your robots!

For those reasons and more, I’ve decided that this is the end of the line for offering 3D printing as a service. There are many other people who offer this service such as Team Malice, a combat robotics team I’m happy to support and who are probably accepting enough frequent orders to keep the service profitable. Just please, for Bunny’s sake, make sure your parts are actually printable before sending them over.

-Seth Schaffer