Hacker News new | ask | show | jobs
by commandar 906 days ago
Prusa was better positioned than anyone else in the market to do exactly what Bambu did, but actively chose not to out of a combination of stubborness and hubris.

Prusa had an established reputation for relatively low hassle, premium machines.

There legitimately isn't much in the Bambu machines that didn't already exist in some form in the high end hobbyist space. What Bambu did was take the best of what was out there and combine it into a single coherent, easy to use package. And, just as critically, they cost-engineered it for mass production.

Prusa has long resisted anything outside their own ecosystem. They invest crazy amounts of effort into reimplementing and reinventing things because NIH syndrome is deeply engrained into their company culture. They clung to using 8 bit controllers long past their expiration date. They spent years working and reworking their flagship $2000 XL which still isn't widely available and suffers fundamental design flaws that will result in it being slower and producing worse prints than its competitors when it is.

A lot of this is Prusa refusing to move past their hobbyist roots. Their printers are still made largely out of aluminum extrusion and printed parts -- this is a great way for a hobbyist to prototype one-off designs, but it's an abysmal way to mass produce a machine.

Bambu's great innovation is the fact that the machine is largely stamped sheet metal and injection molded parts. This results in a machine that is sturdier, higher quality, and cheaper than the way Prusa insists on doing things. It requires a significant investment in tooling upfront -- this was the great tell that Bambu wasn't some scrappy startup when they launched their Kickstarter; they were easily deep into six if not seven figures in tooling costs before the X1 was ever publicly announced.

Prusa was one of the few players in the industry who had the resources and could have done all of this. They actively chose not to. They were stagnant and seemed to think they could continue selling $1000 bed slingers forever. The market finally moved on.

1 comments

What are the fundamental design flaws of XL?

I haven't looked into it in detail but the design appears from a glance to be really good.

If you ignore the cost, which is really hard to do in this case. But that is also somewhat justified by them targeting businesses. It just means they are lagging (which is evident by how delayed mk4 was, if it had been on time mk5 would be imminent by now).

I do feel that the XL is a good stepping stone for their consumer line though, and up until the mk4 launch that is what I thought mk4 would be all about.

The gantry is on a cantilevered open frame.

This is something that the hobbyist world recognized as a really bad idea a long time ago because it makes the machine fundamentally less rigid than a well-supported closed frame would be.

Lack of rigidity results in higher vibration which results in lower quality parts and limits the speed that you can run the machine at. Prusa has tried hard to frame this as them prioritizing quality over speed but that's absolute marketing bullshit -- the things that allow a printer to run fast are largely the same things that allow it to produce high quality parts. A fast printer simply has a higher ceiling and while a fast printer can run slower if you need it to, a slow printer will always be slow.

Jo's said explicitly in interviews that the choice to use that frame design is because he wanted to prioritize easily being able to see and remove completed parts from the print bed.

The toolchanger on the XL has a problem where the tool heads don't like up properly/accurately all the time.

Current community solutions are slamming the toolhead into a printed block after every toolhead swap.

A lot of experienced YouTubers have also had problems with the machine:

https://www.reddit.com/r/prusa3d/comments/17samr9/the_5_tool...