Growing Concerns About Bambu Lab’s Direction, An Open Letter

Other topics
45

In addition to what Sharky said about open source, the remeshing question is tied up with the license models are published under. Open source is intended to be shared and the licenses used tend to include attribution and restrictions on making source available and not closing the source.

Trying to compare to remeshing gets tricky because it would only apply if the model to be remeshed is published under a similar share and share alike / open source kind of license.

The license is key and defines what is allowable and what is not.

46

Microsoft stole from for-profit companies.

I tend to find the open-source argument to be a bit silly, offensive even. The insistence that one constantly acknowledge where their ideas came from. At times, seemingly to do nothing more than sooth fragile egos. I have been playing with electricity since I was 10 years old. After I retired I was reading something and the author kept going on about Kirchhoff’s laws. I started thinking, what the heck are Kirchhoff’s laws? Turns out that they are the laws that say everything in a circuit adds up. Current in, current out, and all that. I went back to the oldest text books that I have, and sure enough, Kirchhoff’s laws are there, front and center. Yet anyone who plays with electricity knows that everything always adds up. Right in line with the law of the conservation of energy. Seems like it was something I never had to think about, it was just natural to expect that everything should add up around a circuit. The idea that someone’s name had been slapped on what was obvious seemed rather odd to me, which is why I ignored it all these years.

If you are going to insist acknowledging where the technology came from, why not hero worship Alois Senefelder, the inventor of lithography? Or what about the consultant group that 3D Systems hired to invent the STL file format? After all, isn’t that what slicers are, they convert STL to g-code? What about paying homage to H. Joseph Gerber who invented g-code?

47

@MZip

Thank you. I agree it’s tricky. I wish the law was better defined.

48

…sorry, I had to do it - I had exactly this image in my mind.

buffet
buffet1403×1121 246 KB

49

uhm… Pythagorean theorem. Sorry. probably not relevant, but what you said brought it to mind.

@3dEd

And then there is this (You and I are of similar age so i suspect you’ll remember this as I do):

Margaret Keane (1927–2022) created the famous “Big Eyes” paintings, featuring children and animals with large, soulful eyes. While she painted them, her then-husband, Walter Keane, falsely claimed authorship for years, leading to a famous 1986 court “paint-off” where Margaret proved she was the true creator. Walter Keane took credit for the paintings, selling them as his own and gaining fame, while Margaret was pressured to work in isolation.

So, then, where to we place the limits of creator rights vs. common use?

50

I have been in the business of designing machines that use stepper motors. The concept for stepper motors was first conceived back in the 1700’s or something. Yet it wasn’t until the advent of the microprocessor that stepper motors became a reality. Unlike regular electric motors, stepper motors require a controller. A controller that runs under software control. That is to say that the machine runs on its own software. Maybe you call it firmware but it is the same thing.

How is it that they were able to create a great machine without internal software? Slicers are file translators. They don’t control the machine. They create g-code that the machine then translates into physical motion. On my printers, once the g-code is generated, the slicer can be turned off and the printers work entirely independent of what is happening with the slicer. Since the machine is working as it does, obviously there is proprietary code in the machine, where the magic happens, not in the slicer. I can put dozens of g-code files on an SD card and print for days without opening a slicer.

When I first got a Bambu Lab printer, I had been using an Anycubic Vyper. Using the same slicer for both machines, the difference was stark. With nearly identical g-code the Bambu printer put the Vyper to shame. Obviously, it is not the slicer software that makes Bambu Lab printers great printers.

51

It’s a mix of both good hardware and good gcode generation, that only a good slicer can provide. With the advancements of the community, the generation of movements sent through the g-code, in the machine, are what make the prints come cleaner and better.

Hardware wise, Bambu is in the right place, but without a good implementation of g-code generation through software, the hardware would be irrelevant.

I can show a few prints, mention a few printers used, and people can try and guess which printer did it. That would be fun while also deceiving :slightly_smiling_face:

52

You have to look at it as a whole system: The slicer is a converter to g-code - at the end of the day. Slicers and G-code have always gone hand in hand.

So if you’d rather talk about the G-code now, yeah, that’s been around forever. Many CNC machines use it. G-code is just a set of instructions telling the machine how to move and when to perform certain actions. The slicer does the heavy thinking, determining the optimal paths, infill, and wall thickness. It’s a team effort.

Back then, I used to sit there with a calculator and do the math right in front of the machine, but I don’t think anyone wants to do that anymore.

As for your comment: Of course, hardware and software play a role in achieving perfect results - as a team. Otherwise, you could just buy a $100 3D printer on AliExpress and turn it into an H2D by changing the software or firmware … but that’s obviously not possible. It can’t work the other way around either.

By the way: G-code itself is open-source and isn’t owned by any company.

53

Not just in USA! It was the same in Europe.

54

Product segmentation isn’t new and is widely used across all industries to capture a wide range of consumers. It is never going to change. For all products, the burden of due diligence lies with the consumer when purchasing the right tool or product to do the job you want it to do—caveat emptor. Most people don’t care about the pitch of a belt or VFA’s because the change is so minute or unnoticeable for most users, depending on what you’re printing. etc.. Name a product that isn’t segmented, where desirable features are implemented in its higher-end models. People worry about cloud-based products. That’s where eventing is going; just like product segmentation, it isn’t new, as long as there is a choice, it works offline. Just my 2 cents.

55

Back in the 1920’s or so American electronic manufacturers formed alliances for the purpose of creating standards so that one manufacturer’s products would be compatible with other manufacturer’s products. That is how we ended up with standard plug types and why electronic circuits tend to be built on 2.54 mm spacings, which is 1/10 of an inch. When one specifies an electrical motor to a specific NEMA (National Electrical Manufacturers Association) standard, a motor from any manufacturer that meets the standard should be a drop in replacement by one made by another manufacturer.

In the early 1960’s the Electronic Industries Alliance published RS-274, a standard that specified g-code based largely on the the format used by Joseph Gerber and others. Gerber had freely published and promoted his code in the hopes that potential customers would learn the code and buy one of his machines. Which incidentally is also how lithography came into the equation. The format has been used to standardize the way CNC machines are operated. Sometime in the 1980s RS-274 was copied into ISO 6983. That is to say that g-code is a standard in the same way that screw thread sizes are standards. Standards are not open source as they are tightly controlled by a regulating body and no license or ego stroking is required to use them.

56

Probably back to the roots: an artisan, open-source TEMU glue gun.

57

There are a lot of fools and a lot of weapons already. I don’t see banning weapons on 3d printers as having any real effect on either. It’s not even possible to really prevent printing weapons. To me it just appears to be theater by politicians.

58

True. People are going to do what they are going to do, legal or not, advised or not.

59

I’m wondering if anyone here, or in those so fearful governments, have any realistic idea, or at least some basic level of understanding, as to what really entails building a ghost gun, that could be used as a real firearm, only from 3d printed parts (no matter how strong is the filament used), on how durable and, especially how reliable under repeated fire such a weapon is. The Liberator was indeed the very first ghost gun entirely 3d printed, but it exploded after one shot. Improved versions of the Liberator, using more technical advanced filaments were printed and fire tested on the fire range, being proven time and again that these guns do not last long under the powerful stress generated by repeated exploding ammunition. The most successful ghost guns were proven to be those combining 3d printed parts with dedicated metal parts (barrel, hammer/firing pin, fire chamber…). There’s plenty of information available on the internet, there are plenty of gunsmithing forums and gunsmith exchange platforms, especially in the US, where highly knowledgeable people discuss, model, print and test various types of 3d gun print files, share their test results and even gun testing videos, and although some progress seems to have been made with specific filament types, including increased gun resilience to successive repeated fire (and not exploding in one’s hands after the 2nd or 3rd shot), the overall experts opinion on 3d printed ghost guns is that there’s still a long way to go before such “guns” might become a real threat. And I do wonder, once more, why one would risk losing his hands, if not his life, by relying on a 3d printed gun, when real, untraceable, metal guns aren’t that difficult to come by, for a reasonable price, no matter where in the world one is.

60

TLDR.

I just wish they stocked their spare parts properly. Surely they have enough data to predict demand. It takes soooo long to get a spare part :pensive_face:

When I get more printers I will have to investigate other options solely because of this issue.

62

Seems obvious it’s a technical challenge that’s not worth the effort. Real weapons are readily available faster and cheaper.

It’s all about the optics.

Politicians need to be seen to be “doing something.” They pass stupid laws that are impossible to implement.

Manufacturers have to respond so they make something that looks like it works but really doesn’t.

Everybody has now “done something” and it looks good.

It’s all optics and we pay for it.

64

As a Master Gunsmith and a super Novice 3d printer, I can tell you that a semi-auto pistol lower can be successfully made with PLA Tough + (Bambu) or PLA + (Other brands), ASA, and stronger level filaments without issue, and last for many rounds. PLA and PETG for some reason don’t last “in the field” (In the field meaning range/in use/etc.). The uppers typically need to be metal, to include the barrel, firing pin, extractor, etc., and the lower parts, Fire Control System (trigger, trigger bar, hammer, sear, etc.) should also be metal.

I have made a 1911 from scratch using a self programmed CNC with metal (lower aluminum and upper steal) to include the screws, springs, trigger, trigger bar, hammer, etc. to test things out and have over 2000 rounds through it. (I did not make the barrel (due to not have the right lathe), wood hand grips (due to being crappy working with wood) or sights (wasn’t liking what I was making, and it was cheaper to buy a good set, next time I may perfect that though and make them better than I was.)

I have made a Glock 17 lower from scratch and have over 1500 rounds through it testing it out. No problems what so ever. Granted, I can’t and won’t make them for others, but as a Gunsmith wanted to see what would work and not work… (And I wouldn’t share the files for it either before anyone asks…)

Hybrid guns can and are made successfully and with long term use. Full 100% 3d plastic (in any version of it, PA, PLA, ASA-GF, ASA, etc., I have seen or talked to anyone that has made one that withstands moderate to heavy use yet. It will happen though, I’m sure.

65

So, as a master gunsmith, and from your own experience, you confirm that only hybrid guns would deliver the expected results on medium to long term use, not those made entirely from 3d printed filament. So the government’s arguments are mot and unsupported technically at this time. As you pointed out, with tech development (more resilient filaments, improved 3d printers…) someone with an inquisitive and creative mind, might eventually come up with a solution allowing for a relatively reliable gun to be 3d printed. Eventually … but not now, not yet.

66

I guess no one here likes Apple products too.