With Core One, Prusa's Open Source Hardware Dream Dies | Hacker News Reader

With Core One, Prusa's Open Source Hardware Dream Dies

hackaday.com

125 pointsiancmceachern76 comments

There are too many cheap clones. Too much stealing of the open source work. This isn't remotely shocking, just look at Redis, Elastic and many many others... Open Source works until it doesn't.

I don't buy Prusa because they are OSH, I buy them because they are great printers. They are an open platform, if not open source. Which is good enough for my needs. If these changes they are making will allow Prusa to keep producing world class devices at reasonable prices, then more power to them.

And yes, I know some people hate Prusa or have had major issues. But they do a lot to move 3D printing forward, rising tide lifts all boats and all that jazz. We want all respectable and reputable 3D printer companies to succeed - because then everyone wins.

> keep producing world class devices at reasonable prices

At the current price points can you really recommend a Core One over an X1 to someone with a tight budget? Without resorting to arguments about open platforms and the big picture?

I took a look at the price. They're almost comparable if X1/AMS combo wasn't (always?) on sale.

The better comparison is a Core One vs a P1S/P1P. You can almost buy two P1 printers for the price of a Core One.

Disagree. The better comparison is Core One vs X1E. As frankly the main selling point of the X1E is Active Chamber Heating.

With your logic you can also say you can just get 2 P1S printers instead of an X1C, but an X1C is still sells just fine.

The Core One doesn’t have chamber heating, it just has chamber exhaust. Not the same thing.

Thanks for the correction, it does not have a dedicated chamber heater. But I don't think simplifying it to "chamber exhaust" is correct.

It sounds like the real claim is the device can actively keep the chamber at 55C. Other than semantics, I don't understand how this is different from having a dedicated chamber heater. I can close my X1C, but it won't maintain any stable temperature. The bed and nozzle are heating the chamber. This is all assuming the time to get to 55C is reasonable, and that 55C is enough. I personally have a need for 55C chamber often, but never 100C (X1E can only get to 60C).

Again, feel free to correct if I'm wrong, but directly from Prusa: "The automatic ventilation system and active temperature control"

The way they describe the active chamber management in their blog post is basically to say “the enclosure allows it to reach temps as high as 55C, but you can keep it enclosed and drop the temperature so you can print PLA and PETG”. There might be some insulation perhaps, but getting to 55C is not particularly hard on almost any printer. Uninsulated printers with enclosures that are reasonably tight will hit 55C no problem. Add some insulation and you can hit 70C pretty reasonably all without any chamber heating.

Firstly, neither of these are "budget". I think if you need a budget, you a probably best sticking to a Prusa Mini, Bambu P1s or A1 Mini.

Without a doubt. An X1 is 1k USD. This is 1,199 USD.

Truly this is a competitor to the X1E though which costs 2.5k (!!!) with basically the only notable addition being the heated chamber (which the Core 1 comes with for free).

I have multiple Prusa Mark 3s, a Prusa XL and an X1 carbon, and frankly I only use the Prusa XL nowadays (and sometimes the Mark 3s).

Bambu makes a good printer, but it has lots of annoying issues and proprietary annoyances. I also don't like them as a company, but that wouldn't prevent me from buying another if I needed and used it.

In my experience Prusa printers "just work" more often than Bambu printers do.

It's worth noting that the X1 is on sale today. Normally it's $1199 as well.

Frankly, the only advantage I see from a spec list for the Core One is a chamber exhaust (not heater, just exhaust).

Compared to a P1P it’s missing a camera.

Compared to the X1C it’s missing a camera, the LiDAR, and carbon rods.

Also, the AMS solution on Bambu printers is much better than the MMU by Prusa.

> Compared to a P1P it’s missing a camera.

Built in yes, and that's disappointing to see. Unsure why they aren't just including that at this point. Assuming its like the XL, you can use any camera you can find (phone, rpi, esp) and link it to the printer to get a "first party"

> Compared to the X1C it’s missing a camera, the LiDAR, and carbon rods.

I never use the LiDAR on my X1C. In my opinion is produces worse results then calibrating manually. Agree on the camera.

Regarding carbon rods, I would be surprised if it made any tangible difference to 99.99% of people.

> Also, the AMS solution on Bambu printers is much better than the MMU by Prusa.

Agreed. I never use AMS or MMU. For some its a dealbreaker to not have something as good as AMS. I never print multi material, other than using multiple heads on the XL.

I rarely print multi material in the same print just because of the filament wastage, but damned if the AMS isn’t worth it for not having to mess around rethreading the filament for filament changes. I know that’s a luxury “nice to have”, but after years of fighting with an Ender 3 Titan extruder and the filament curling just enough to not make it through the PTFE coupling, the AMS is such a time saver. Pop in a roll, push it in an inch, then forget about it and pick it from a list when I print. Done. Love it.

I got my P1P pretty hesitant about the closed ecosystem, but having to never really think about my printer and whether a print is worth the time it’s going to take, which prevented me from printing a bunch of times on the Ender 3, I’m sold at this point.

I wish Prusa the best of luck with their new printer, and I’m sure it’s a solid piece of kit. I think they’ll do alright. Just like I feel as though Bambu have really changed the market in the last couple of years, they’re building in the footsteps of the paths carved by Prusa.

> Compared to a P1P it’s missing a camera.

The video showed a camera with an "optional" legend.

The core one doesn't have a heated chamber. It has a fan at the top that regulates air leaving the chamberm hence "active chamber heating" rather than "heated chamber" in their marketing materials. The heating is done by the heatbed, making it comparable to a P1S rather than an X1 or X1E.

The P1S has the same heater (the heat bed) and the same concept of a variable speed fan that regulates how much air is drawn from the chamber. The only real difference here is that the core has a vent cut out, whereas the P1/X1 tell you to open the door for PLA to let it very slowly pull in cooler air (the fan still runs, just at a lower speed to prevent warping).

The Core One is technically not even comparable to the P1. It's not got a camera, nor an AMS system (the MMU is well known for being incredibly unreliable and finnicky to get working well vs a box that you plop on top and plug in).

The only real compelling thing about it is the upgrade path from the MK4, and the nice design cues like the integrated spool holders with potential for a dryer.

So it has a chamber, a heater, a way to control chamber temperature. But its not a heated chamber because it does not have a single purpose heater vs using the bed for it? Sounds like semantics until testing is done.

Fair enough. Then we must declare that every single printer with an enclosure and heatbed has a heated chamber.

Why would I recommend anyone buy a printer that cannot be repaired? That’s just throwing money away and creating e-waste. Even a Prusa Mk4 makes more sense than the X1 when you consider repairability.

Bambu sells replacement parts for their printers. Very little of the printer is not repairable.

Not only does Bambu sell parts, but there’s a pretty healthy market of third party parts as well. Saying these printers are irreparable is plain false.

At some point the value proposition makes sense. People buy non repairable 2D printers all the time.

Also, as the other commenter noted, they actually are quite repairable. Bambu offers pretty much every part you could imagine and at prices that are extremely reasonable. Any wear component you’d expect is easy to replace.

> At some point the value proposition makes sense. People buy non repairable 2D printers all the time.

Where can I buy a repairable 2D printer? I would prefer this if I could make a choice.

Anything commercial that is also the size of a small fridge. You don't actually want one.

Have you looked at the parts shop? You absolutely can repair a Bambu printer, and as someone with a farm of the things can attest that it's no more complex than working on Prusa's. You still need to buy the parts from somewhere. Bambu's own pricing on parts is pretty reasonable in my opinion.

> Too much stealing of the open source work

How do you steal Open Source? Can Pruse no longer use it themselves or something? Sounds wrong calling "companies creating products from other projects" stealing when the intention from the beginning is that others can freely use the created project for whatever.

> This isn't remotely shocking, just look at Redis, Elastic and many many others... Open Source works until it doesn't

Isn't those examples that Open Source builds great software? Companies trying to wrestle control of projects after making them Open Source doesn't mean what's already there didn't have a great impact.

> How do you steal Open Source? Can Pruse no longer use it themselves or something? Sounds wrong calling "companies creating products from other projects" stealing when the intention from the beginning is that others can freely use the created project for whatever.

Thing is, the fundamentals of Open Source have changed over the last decades - and the assumptions people made Back Then no longer hold. Let me expand a bit:

Back in the late 80s and 90s, up until the early '00s a lot of popular open source software was developed by academic institutions or with scientific grants. For them, it didn't matter - the money way paid for anyway and sharing source code fits with the ideals of science. In some projects it's very clear that they have an academic history - my to-go example is OpenStack, the myriads of knobs it has absorbed over the years all come from universities wishing to integrate whatever leftover hardware they had.

But ever since academic funding all but dried up, life has gotten difficult. We got a few rockstar projects that manage to survive independently (cURL), godknowshow (OpenSSL), with consulting services (sqlite with their commercial comprehensive test suite, mysql, mariadb, psql), on corporate contributions (Linux kernel, ReactJS/Facebook), on donations (everything in the FOSS graveyard better known as Apache) or, like Prusa, on hardware they sell. The general idea behind many projects is the implicit assumption: if you use a project commercially and the developer has a commercial support platform, be so kind and pay the original developers a bit so they can improve upon the project.

The problem is when juggernauts with deep money pits, be it companies with net market values in the trillions of dollar range or companies being under influence of the CCP, come on the field and take the hard work of others to make money without contributing back either financially or with code. Legally, they are absolutely in the clear, if the project isn't under AGPL, CC-NC or other such terms. ElasticSearch got ripped off that way by AWS for example.

It's not stealing in a traditional sense, but it is breaking the ethos and expectations.

Proprietary companies always have a license to print money.

People who do open source don't usually do it for the money or have the expectation of just making a living from it, never mind making a lot of money. They don't even charge a nominal price for their software. So you have a mismatch between funding and enthusiasm.

Yes. Too many people in this community seem to be believe that Open Source is a marketing tool and somehow even more bizarrely, a business model. And then pretend to be disappointed when they find out that it is a poor fit for both and that people and businesses aren't tripping over each other to throw money at them.

Open source is a vehicle for giving the world something neat and useful, with no other obligations implied. (Other than perhaps the continuation of said freedom for downstream users, a la GPL.)

Re "or with code", none of the "open source companies" these days actually care about that, they are all about paying back their VC investors and making money.

PS: A blog post related to this situation:

https://drewdevault.com/2021/01/20/FOSS-is-to-surrender-your...

There can be many ways open sources comes into being. The way any open source software I've written is I've needed it myself and made it available to others. There has never been an expectation of getting paid for it, it doesen't even matter if anyone ever uses them, because the software's primary purpose is to solve _my_ problem

>How do you steal Open Source?

Stealing isn't just a legal concept, it also applies (among others) in a social context where you "steal someone's joke" or "steal someone's girlfriend".

With open source the social contract is that you're going to contribute to the project if it's a substantial part of your business.

> This isn't remotely shocking, just look at Redis, Elastic and many many others... Open Source works until it doesn't.

I would argue that redis and elastic are signs that open source does work, albeit not well as a for-profit business. Open source hardware has a completely different set of problems.

If it doesn't work well for such high profile names, why would we expect it to work at all for an unheard of nobody? Doesn't that mean that kpen source doesn't actually work?

> If it doesn't work well for such high profile names

What, exactly, doesn't work well? Almost anybody knows of Redis and ElasticSearch, many of the ideas they implement spread in the ecosystem and everyone can still use the old versions as the FOSS they were made as.

If you're talking about that they were unable to build a for-profit on top of giving software away for free without any concrete plans on how to actually make money, yeah, that might not have been sustainable. But that makes it sound like their business plans were what didn't work, the Open Source part seems to have worked out just fine for what the purpose is, to give away software for free.

I think it depends a lot of what one see as a success and "make it work". If we think that success is only high earning, growing forever, vc fuelled companies, then no. Open source will never work for that.

Those clones that you speaks of are often of questionable quality. Unless we're talking about creality printers, which were open source(at least with the Ender 3), and are also low quality.

But my question is "what's the point?" If you have an open source project and yet the commmunity is largely uninvolved in its development, why do you even care to be open source?

Yes, freedom is important, but hardly anybody but developers take advantage of it. The most important aspect of FOSS is that it's a marker of a project/product that won't take advantages of its users with shady business practices, and that's probably the most important thing about it.

The main issue with this move is that it's not going to cut down on clones very much. Chinese 3D printer companies already clone all kinds of parts from other companies that don't provide design files, including stuff very similar to the now-proprietary extruder. They won't need to spend much effort replicating it. The people who lose out the most are open-source hardware hobbyists.

Prusa gave everyone permission to make copies of their i3 series machines, make modifications, and distribute modified versions. It’s not theft if you have been given explicit permission to make copies.

“You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty.”

“You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above.”

https://github.com/prusa3d/Original-Prusa-i3/blob/MK3S/LICEN...

> Prusa gave everyone permission to make copies of their i3 series machines, make modifications, and distribute modified versions. It’s not theft if you have been given explicit permission to make copies.

Indeed. But I guess Prusa also expected that people will buy the original such that this support this open-source mission can be sustained. This is where I personally see Prusa's fallacy of thinking.

Sure, their choice to give it away may not have gone as they hoped, I’m just pushing back against the idea that making copies of a thing the creator explicitly gave permission to copy is “stealing”. The core of open source is that making and distributing modified or unmodified copies is good. Calling it stealing undermines the very important social work we can do with open source.

> The core of open source is that making and distributing modified or unmodified copies is good.

The core is not that it is "good", only that it is allowed (i.e. we won't sue you if you do).

The Chinese are good at cloning, with or without access to design files. You'll be a fool to not have a plan.

Besides Apple, which other <$2000 electronics/hardware purchase isn’t “open”? As in you can buy parts for repair/upgradability right?

this is such a bad take. there is a huge subset of the community that made contributions to 3d printing both randoms and other companies.

ie bambu pushing the slicers and printers to actually not be dead slow

I don't think the cheap clones were ever the problem. Prusa changed gears more heavily when Bambu came, took the code, ideas and learnings, modified, added, closed it and held for themselves.

For people in the 3D printing space, the most important points so far are not the fact printers are designed on open source hardware, but:

1. That they are easy to fix. This is still the case with Prusa, and that's a good thing, together with their great support.

2. That replacement parts are relatively cheap. This has been an issue with Prusa: open hardware helps very little if you need to pay an unreasonable amount of money to get a nozzle and heatbreaker or so. Bambulab parts are much cheaper, even if the printer is completely closed.

3. The OSS nature & hackability of software: that, yes, mattered a lot, and Bambulab, Prusa itself, and many other companies benefitted from reliable and powerful open source software to drive 3D printers (slicers, firmware). This had the effect of accelerating the field.

A bigger danger than closed hardware is patents. Also in the field of 3D printing the feeling is that the small incentive to innovate (Prusa was really stagnating before Bambulab) was also a result of providing the same value instantaneously to all the competitors.

I believe in open source as an accelerator of society. I also like open hardware. However both open source and open hardware can fail in certain setups, and in this case it is better to move away.

I don't understand why Prusa thinks keeping their designs proprietary addresses the "unfair competition" problem they seem to be concerned about. Anyone wanting to release a printer can use freely available designs, like those from Voron. The openness of the Prusa Core ONE is not what would allow a competitor to enter the market "unfairly" with a competitive product. Maybe it would make sense if they were bringing some new innovations to the market, but for a catch-up product like the Core ONE restricting access feels like slighting your customers for no gain.

They're on a bit of a streak of very poor business decisions really. They did the same with the XL, they've been caught lying in marketing materials (the Mini for example was advertised as having power loss recovery at launch, which changed to coming soon, and eventually when they realised it wasn't possible was dropped entirely, about 2 years later).

On top of that there's the incredibly slow response to Bambu and the other Corexy options overshadowing them, and the stream of lies from Josef Prusa regarding Bambu labs (e.g his tweets claiming they stole code and violated the MIT license, which he's since removed from Twitter but thankfully was backed up in several reddit discussions as well as archive.org).

I've got a lot of respect for Prusa and what they've achived but they really do seem to be fumbling pretty hard. The Core One will certainly get them back in the right direction but things like cheaping out and not including a camera when its already a worse product than the one they're trying to compete with feels like an incredibly stupid decision.

It's such a cheap part to include, for some sort of comparison a Raspberry Pi Zero camera is £14 on Pimoroni and thats a consumer price. Even if it was costing Prusa £10 per camera, thats absolutely nothing.

I tried to stick to Prusa stuff through the release of Bambu products in order to support the notion of a group that can give-take within the OSH concept -- now they offer zero value comparably.

The Bambu products are better if you're willing to buy into proprietary stuff and you're not willing to put the leg-work into building something proper-open like a railcore.

Really sucks, but the writing has been on the walls for some time -- it has been harder and harder to find source/designs/models/etc regarding Prusa machines since the MK3 period.

Remember that Mr. Prusa himself has an open source logo tattooed on his arm. This must hurt him as much as it hurts us. I see myself as a huge proponent of open source ideals, but the company needs to make money, so I understand this move. It feels less bait-and-switch-y than the relicensing of other prominent foss products.

I am skeptical of Mr. Prusa's reasoning for not being open source. I am not doubting the sincerity of his belief, but I think he's wrong.

The Chinese are very good at cloning. Releasing the design doesn't change this, as they don't need the PCB layout because they already have their own PCBs and have lot of people who can design electronics. In any case, there are hardly any secret sauces on a Prusa. Rather, Bambu doesn't even need to copy and are seen as surpassing Prusa in some respects(true or not).

He mentioned about keeping things secret to prevent supply chain competition. I wish I knew more about this issue, so I can't confirm or refute it, but it seems dubious to me as well.

Anyway, open source is only as good as how you use it or develop it. Prusa seems to be for the most part a closed shop, so they don't benefit from community development and seem to have Not Invented Here syndrome and allergic to using community developed solutions, except for software.

For the majority of people, the open source label signifies that you are much less likely to get scammed or get taken advantages of companies or individuals. That's probably the most important thing.

Anyway, I don't run an open hardware company, so take it with a grain of salt.

This is a terrible mistake. People see Prusa as a premium brand mainly because of its long history in open source. You can buy a similar Creality printer for like $200 from Amazon vs Prusa's $800 or whatever shipped from Europe already.

Barring any overwhelming breakthrough, other brands have them beat on price and performance. Hopefully they at least keep middle of the road models as open-source for those who really value that. I anticipate that the open-source printers will outsell the closed-source ones, unless they conceal the fact that they are now closed source.

By the way Prusa, if you don't want to help the competition, maybe don't let Prusa Slicer slice for other printers. But that cat is out of the bag, and it's also probably a marketing tool at this point.

Prusaslicer and forks all benefit from the sharing of codes. Prusaslicer even copied code from Cura.

Doesn't the Core One share a lot of common parts and accessories with the Mk4? An upgrade kit is promised.

It's mainly things like rods and motors from the looks of it. The electronic kits used to be the most expensive parts when building your own printer so makes sense to reuse these and just pay for the frame. It's a smart move on their part thats for sure.

Bambu obviously killed it.

I follow 3d printing pretty close but can't claim to be an expert. With that said, I truly thought they served different consumer segments with the only overlap being those who bought a Prusa pre-assembled beleving it to be a one stop shop machine. Bambu is a black box from China for an end user with little knowledge or care of maintaining a machine themselves (down to printing replacement parts)

Bambu Labs' quality and feature set is much, much higher and larger than Prusa's, and the price is right. Prusa bet on people wanting to continually fiddle with their 3D printer, but that segment is already niche and likely dying off.

Most fiddling these days have to do with the printing surface being unclean. I also experienced issues with my X1C too.

But the most common problem is the surface is unclean(on both printers), and my soap to water formulation not being quite dialed in.

What printing surface are you using? I use a PEI sheet that I clean with straight isopropyl alcohol, and I almost never have issues.

PEI smooth and textured. Isopropyl alcohol works until it doesn't. That is why it is recommended that you use warm water with soap. I suspect the ratio of water to soap isn't dialed in quite right in my case, but I haven't bother to fix it just yet.

Either that, or don't touch the surface with your bare hands.

I can only speak anecdotally, but I've been using this sheet for ~4 years while only cleaning it with 90+% IPA, and I haven't seen any loss of adhesion. I expect to replace it due to scratches before I have any problems with the cleaning method.

I always use IPA to clean the bed too.

I have once used glue for a very thin print with lots of intricate holes in it.

I think the problems arise from using glue on it.

If I only print PLA without glue, one wipe with IPA and it's clean.

Glue on the other hand, it really sticks and you need water and soap.

The experience I have on MK4 and X1C are similar, as far as reliability, etc.

There's different annoyances for each; if you calibrate each time X1C is slower to get going. X1C is faster overall on bigger jobs. X1C has weird wifi error-out issues more often. MK4 gets a bit more gunk on the nozzle. X1C wastes more filament. X1C had some issues with retracting filament at first that I printed someone else's bracket design to fix, while MK4 just worked. X1C quality seems slightly better with PLA; MK4 does a slightly better job with PETG.

When wear makes major maintenance necessary, it's going to be easier on MK4.

> the only overlap being those who bought a Prusa pre-assembled beleving it to be a one stop shop machine

Thats a surprisingly large segment of the market, though.

Yes I agree, I suppose my point was as soon as Bambu went mainstream that entire Prusa appeal was killed

Prior to Bambu, prusa was as close as you could get to “put it together and it’s ready to print” including printer profiles and such. Bambu did this cheaper and better, and much faster, so basically took that entire market from Prusa.

For anyone that wants a printer that “just works”, there’s little reason to choose Prusa over Bambu at this point.

Prusa grew up with the market. Their printers sold very well, that I had to wait for quite awhile for my (MK4) kit to get delivered.

I bought my Prusa before Bambu became popular, and honestly I always see Prusa’s in school's and libraries and feel their main market in the United States is in that higher end role of something that is just fairly reliable and used where organizations want to provide 3d printing where a lot of different users are going to use them.

But I regularly see Bambu winning the reviews and awards these days, and I’m not sure if I would have been aware of Prusa if I were in the market today.

I really would love a multi-tool change core x y, but it’s soooo expensive.

Bambu didn't killed its open source dream. Prusa did.

It’s just another Bambu clone anyway

Getting a bambu soon. Happy I didn't go for Prusa, as the open source aspect was the only advantage

Going for a Bambu if you valued the open-source aspect? There are a few comparisons here I won't make about how silly that sounds to me.

They clearly valued some other things higher, even if they also valued the open source aspect. There's no dissonance or contradiction in that. I went the same route they did, despite also putting some value in the open source aspect.

As I clearly stated, the open source aspect was the only advantage they had over Bambu. And now that advantage is gone, which means Bambu holds all of the advantages instead of all advantages except one

Reply on news.ycombinator.com