> I use Linux daily and don't want to switch to Windows just to connect to the printer's FTP service
I wonder if the author tried using their file manager to connect? I haven't needed any kind of external file management system since switching to Linux, Dolphin just handles everything (sftp, ftp, samba, etc) for me natively in the same window.
My favorite way to connect to FTP servers on Linux was with lftp from the command line. I say was because I don’t really use FTP anymore. I do use Linux still though.
lftp is available in every package manager I know.
I'm guessing bambu implementation returns the server socket's listen address, and they bind to 0.0.0.0. (Typography pet peeve, how do I disambiguate that dot?)
What's surprising is how this got shipped. Do the devs use some other client that has the WinSCP setting on by default, or is that feature only used by their slicer and their SDK does it by default?
One theory I have is they bound the server to the printer's address originally and it behaved properly, but then changed to 0.0.0.0 later.
I wish there was a common convention for logical grouping like we have in math when disambiguating operator precedence with parenthesis, but those are already taken for asides in regular prose. Maybe curly braces?
Instead of recompiling the source and installing it again, is there a way to monkey patch the already existing package? It seems like a few lines of code.
I'd rather more like to imagine a software packaging/distribution regimen whereby dropping a patch(1)-compatible patch file into "an appropriate folder" meant that it would instantly take effect the very next time you run the program.
Bambu and Prusa are the two default picks. They're both very common choices for first 3D printers, they're both about as high quality as you'll find without spending substantially more money, and they're both about as idiot-proof as 3D printers gets. I personally prefer Prusa as a bit more open and good for hacking on, but Bambus target end users a bit better and have their own advantages. You can't really go wrong either way.
I will say that the answer may change a bit depending on what you're hoping to print. If your goal is, for example, high detail miniatures for tabletop gaming, you may want to be looking into something like SLA printers. Or if you need a specific exotic material, or if you anticipate needing multiple filaments, the answer similarly changes.
This was true a few years back, but in recent times Creality has fallen victim to quality control issues, a huge number of printer variants and generally having a not so great out of the box experience. You'll find this sentiment echoed in comments on any of the 3d printing subreddits as well.
I upgraded from my buggy, annoying Ender 3 Pro to a Bambu A1, and it's been pretty wonderful so far. I haven't had any need to "babysit" it, and I can trust it to just start a print, and finish it when I get back. It self-levels the bed, etc.
I got the most basic model - a single feed for filament, etc. I recommend it.
People are right that you shouldn't spend too much money, but don't spend too little, either. If you think to yourself, "Well, $300 is a lot for a 3D printer, I'll just get an Ender 3 for $200, or a used Ender 3 for $100", you'll end up getting significantly more frustrated if all you want to do is 3D print things.
I second this. Occasionally the head will jam, but it is easy to clean them out. But, the A1 mini is the first device that really just works. It's so much fun.
Do not be an idiot like I was and try to print in an outdoor atrium to avoid fumes. That's really not an issue these days and humidity will kill your filament. For many reasons related to humidity control, it is useful to invest in a humidity monitoring filament holder.
Go for it, don't spend a lot of money though on the first one. If you enjoy it then figure out the next one to spend the money on.
The big issue for me right now is that a lot of the smaller bed printers can't really do some of the larger projects I want to do like wall hanging systems or drawer organization systems.
Also Bambu the company mostly is fine, but there's some worry that they'll eventually lock people into using only their filament, but doesn't seem to have happened yet. So buyer beware.
> Also Bambu the company mostly is fine, but there's some worry that they'll eventually lock people into using only their filament, but doesn't seem to have happened yet. So buyer beware.
I'm not sure how Bambu could actually do that. They use RFID tags to identify their filament type/color. I taped a tag from a used roll to some prusa filament and the printer couldn't tell the difference.
Just in case, my Bambus are LAN only and don't get updated. I use Orca Slicer instead of the Bambu slicer.
Bambu can't even keep their filament in stock, plus they ship the printers with multiple preloaded profiles for other filament vendors. I don't foresee them making that change any time soon.
Get an old Prusa MKIII and stick a Revo in there, then learn everything there is to know about 3D printing without spending a fortune or getting locked in. Once you have processed a couple of rolls of filament you'll be much wiser about your needs and that would be the moment to pull the trigger on a 'proper' printer.
Bambu AI is a very good printer (we have 10's of them, and 10's of Prusas as well), but the Bambu eco-system is not ideal and they push really hard to get you to use their cloud connect, the printers have cameras and send footage to servers in China if you get them connected to the point that they are usable. In contrast, there are many open source solutions that will connect a Prusa to your LAN and allow various degrees of remote management (Octoprint, for instance).
Prusa's are extremely hackable, I've adapted them to do all kinds of stuff they were never meant for (1x1x.25 meter for instance, or standard width and height but 65 cm tall). Bambu's are quite closed, though in theory you could hack on their slicer but it's infuriatingly bad compared to the alternatives.
I think I'd largely disagree with your recommendation, unless they specifically wanted to get into 3D printing (the hobby) rather than 3D Printing (the tool). I got my printer wanting to make things, and didn't enjoy the tinkering with my mk3 at all. It was a great printer for the time! But I swapped to a P2S and never looked back. I hear Prusa is competitive these days, though not perhaps in price at the low end.
I print a couple of tons every year and I would not be able to do that without the knowledge required to operate a farm reliably and productively. Yes, they're tools, but like all tools it helps to know what you're doing. If I hand you a machining center you won't be able to learn much without a lot of breakage and expense. If you get a lathe to learn and play with, to build up an intuition for feeds & speeds and how materials handle and chip then you will be able to use that machining center to the maximum of its abilities.
Tools require knowledge. 3D printers are no different in that respect and to toss $100 on a printer just to learn is money very well spent. And those old Prusa's excel at precision work, we can do stuff on those that we can not touch with any of the others.
Most people don't intend to print tons every year nor desire to manage a print farm. Most people interested in creating object rather than managing printers will have an infinitely better experience getting a Bambu that is ready to crank out amazing prints right out of the box.
Other than basic troubleshooting (which they have documentation on) there isn't really a need to take a deep dive into how exactly each piece works.
I say this as someone who started printing many years ago with an i3 clone and has replaced nearly every piece of multiple printers (control board, bearings, hotend, extruder, etc!) over the years for better performance. I moved away from wanting to tinker with the printers and haven't touched them since getting a P1S years ago.
I mean, it's kind of like comparing a tuner car with a new EV. Both will take you from A to B but one requires a lot more work to take you from A to B in a very specific manner while the other just turns on and goes. The tuner car has a tremendous amount of power and is really good at driving fast and accurately but it also requires a lot of work and custom parts to get it to perform like that. The most amount of work in maintaining the EV is that it may need new tires eventually, otherwise it just works.
Most people just want something that just works out of the box using models they downloaded from the internet. It's great that you want to have a 3D printer that performs at the absolute limit of the hardware but that requires work.
Not the MK3. I love that printer to death, but asking somebody new to 3D printing to do Z leveling manually is a tall order unless they're in it for the hobby.
The MK4, with its load cell, eliminates this requirement and is therefore a way better choice for someone new to 3D printing.
The MK3 does have auto Z leveling, but it does it with a PINDA sensor, and you're still going to need to do a Z calibration plus maybe some fine tuning, and that's not the hardest thing in the world, but it's not the easiest.
The loadcell on the MK4 completely eliminates that whole mess.
I've put 10's of thousands of hours on MK3s and the PINDA works fine as long as it is mounted well (which isn't always the case) and set to the right height relative to the nozzle. The automatic calibration routine - on a well built printer, which also isn't always the case, runs exactly once after you've built the printer. Fine tuning is possible, and can be quite useful if you want to play around with the degree to which your prints stick to the bed, for instance for automatic print release tricks (and bed temp as well).
The fact that you have all this control comes at a price compared to say a Bambu A1: you need to know a bit more about how a printer works. But that knowledge will come in very handy when sooner or later you want to do something that isn't in the manual.
The MK4 loadcell is a solution for some problems and a the cause of some others, it is not the best implementation of a loadcell either but it is smarter than the way for instance the Creality K1 and variations are set up (they have four, underneath the bed).
I would recommend against the Creality series printers by the way, they're dirt cheap and wicked fast but they break all the time and the stock extruder is complete junk.
Finally, the best investment you can probably make alongside any printer is a multi-roll filament dryer with a heating element and a fan. That more than anything else will affect your print quality.
We do the bulk of our work on Bambu's now, but for really high precision stuff we will use the MK3s and it's amazing what those old printers can churn out. Nozzle wear is a bit higher on the MK3s than on the Bambu's. The oldest MK3 we have has done > 100 Km of filament...
bambus are great in the "just print stuff" market (no modding needed, everything comes properly configured, the calibration is done automatically, profiles are tested, etc.)
There are some controversies about them locking their printers to their own software and some other issues though.
Good old FTP. I am a bit sad that it kind of died. We should have some modern FTP with super-simple everywhere implementations.
Edit: Actually, some things should be fixed too if any future FTP-like protocol would be added. I never liked the difference between active mode and passive mode in FTP. A user really should never have to care about that. Things should "just work". It's only data transfer in both cases anyway.
IMHO, there's a better answer to any thing you like about FTP. For simple stuff, TFTP seems... fine? I've never looked under the hood, but I'm pretty sure it fixes the really bizarre choices in FTP. For everything else, either SFTP or HTTP seems to be the rule. For moving files around between machines, I've become quite fond of https://github.com/9001/copyparty over http
> I use Linux daily and don't want to switch to Windows just to connect to the printer's FTP service
I wonder if the author tried using their file manager to connect? I haven't needed any kind of external file management system since switching to Linux, Dolphin just handles everything (sftp, ftp, samba, etc) for me natively in the same window.
My favorite way to connect to FTP servers on Linux was with lftp from the command line. I say was because I don’t really use FTP anymore. I do use Linux still though.
lftp is available in every package manager I know.
Man page: https://linux.die.net/man/1/lftp
Actually lftp supports a lot of different protocols but I only ever used it for FTP, FTPS and SFTP.
Ironically, Microsoft recently (few years ago) removed the FTP client from Explorer.
That was my first thought as well, there is special client needed on all major desktops.
Or use mc (midnight commander).
I'm guessing bambu implementation returns the server socket's listen address, and they bind to 0.0.0.0. (Typography pet peeve, how do I disambiguate that dot?)
What's surprising is how this got shipped. Do the devs use some other client that has the WinSCP setting on by default, or is that feature only used by their slicer and their SDK does it by default?
One theory I have is they bound the server to the printer's address originally and it behaved properly, but then changed to 0.0.0.0 later.
> Typography pet peeve, how do I disambiguate that dot?
I have resorted to "0.0.0.0".
Reasonable.
I wish there was a common convention for logical grouping like we have in math when disambiguating operator precedence with parenthesis, but those are already taken for asides in regular prose. Maybe curly braces?
`0.0.0.0`.
I always put an extra space after such things. Especially URLs. Makes it easier to copy/paste, too. Debatable grammatical correctness
Plus there's also ::/0 to bind to which will break/fix things as well.
Instead of recompiling the source and installing it again, is there a way to monkey patch the already existing package? It seems like a few lines of code.
That would be incredibly complicated and crash-prone to do.
With nice distros like gentoo, you can just drop the .patch file into an apropriate folder and it'll be applied with every re/install/upgrade
I'd rather more like to imagine a software packaging/distribution regimen whereby dropping a patch(1)-compatible patch file into "an appropriate folder" meant that it would instantly take effect the very next time you run the program.
you can recompile and not install it anywhere, just run the binary you compiled yourself.
Too bad FileZilla can’t be forced to use EPSV which might just work with this buggy server.
The A1 mini has FTP O_o that's actually great to hear
Somewhat related I am thinking of picking up a Bambu A1 as my first foray into 3d printing ... seems to be a really solid move can anyone comment?
Bambu and Prusa are the two default picks. They're both very common choices for first 3D printers, they're both about as high quality as you'll find without spending substantially more money, and they're both about as idiot-proof as 3D printers gets. I personally prefer Prusa as a bit more open and good for hacking on, but Bambus target end users a bit better and have their own advantages. You can't really go wrong either way.
I will say that the answer may change a bit depending on what you're hoping to print. If your goal is, for example, high detail miniatures for tabletop gaming, you may want to be looking into something like SLA printers. Or if you need a specific exotic material, or if you anticipate needing multiple filaments, the answer similarly changes.
The Creality line of printers is also another line of 3D printers to look at. They've got some entry models are quite good (and cheap) to start with
This was true a few years back, but in recent times Creality has fallen victim to quality control issues, a huge number of printer variants and generally having a not so great out of the box experience. You'll find this sentiment echoed in comments on any of the 3d printing subreddits as well.
I upgraded from my buggy, annoying Ender 3 Pro to a Bambu A1, and it's been pretty wonderful so far. I haven't had any need to "babysit" it, and I can trust it to just start a print, and finish it when I get back. It self-levels the bed, etc.
I got the most basic model - a single feed for filament, etc. I recommend it.
People are right that you shouldn't spend too much money, but don't spend too little, either. If you think to yourself, "Well, $300 is a lot for a 3D printer, I'll just get an Ender 3 for $200, or a used Ender 3 for $100", you'll end up getting significantly more frustrated if all you want to do is 3D print things.
For $300 you can get an A1 mini and it's a pretty solidly engineered printer. We're running them until they break. But they don't break...
I second this. Occasionally the head will jam, but it is easy to clean them out. But, the A1 mini is the first device that really just works. It's so much fun.
Do not be an idiot like I was and try to print in an outdoor atrium to avoid fumes. That's really not an issue these days and humidity will kill your filament. For many reasons related to humidity control, it is useful to invest in a humidity monitoring filament holder.
I got the A1 - I knew I would want to print bigger things eventually, and spending an extra ~$100 seemed like a no-brainer.
I also always get the bigger disk on phones, etc.
Go for it, don't spend a lot of money though on the first one. If you enjoy it then figure out the next one to spend the money on.
The big issue for me right now is that a lot of the smaller bed printers can't really do some of the larger projects I want to do like wall hanging systems or drawer organization systems.
Also Bambu the company mostly is fine, but there's some worry that they'll eventually lock people into using only their filament, but doesn't seem to have happened yet. So buyer beware.
> Also Bambu the company mostly is fine, but there's some worry that they'll eventually lock people into using only their filament, but doesn't seem to have happened yet. So buyer beware.
I'm not sure how Bambu could actually do that. They use RFID tags to identify their filament type/color. I taped a tag from a used roll to some prusa filament and the printer couldn't tell the difference.
Just in case, my Bambus are LAN only and don't get updated. I use Orca Slicer instead of the Bambu slicer.
Go the HP way. May be throw annoying errors based on the expected amount of printing a filament with a particular tag is supposed to last.
If corporations can put a chip in a tooth brush head...
https://www.reddit.com/r/hacking/comments/145fsx6/just_finis...
They could write to the RFID how much filament "it has left".
Bambu can't even keep their filament in stock, plus they ship the printers with multiple preloaded profiles for other filament vendors. I don't foresee them making that change any time soon.
Good choice, keep in mind that you'll probably spend more on filament rolls if you get hooked
The new X2c just came out. Consider buying a used X1c someone is selling to upgrade.
Get an old Prusa MKIII and stick a Revo in there, then learn everything there is to know about 3D printing without spending a fortune or getting locked in. Once you have processed a couple of rolls of filament you'll be much wiser about your needs and that would be the moment to pull the trigger on a 'proper' printer.
Bambu AI is a very good printer (we have 10's of them, and 10's of Prusas as well), but the Bambu eco-system is not ideal and they push really hard to get you to use their cloud connect, the printers have cameras and send footage to servers in China if you get them connected to the point that they are usable. In contrast, there are many open source solutions that will connect a Prusa to your LAN and allow various degrees of remote management (Octoprint, for instance).
Prusa's are extremely hackable, I've adapted them to do all kinds of stuff they were never meant for (1x1x.25 meter for instance, or standard width and height but 65 cm tall). Bambu's are quite closed, though in theory you could hack on their slicer but it's infuriatingly bad compared to the alternatives.
I think I'd largely disagree with your recommendation, unless they specifically wanted to get into 3D printing (the hobby) rather than 3D Printing (the tool). I got my printer wanting to make things, and didn't enjoy the tinkering with my mk3 at all. It was a great printer for the time! But I swapped to a P2S and never looked back. I hear Prusa is competitive these days, though not perhaps in price at the low end.
I print a couple of tons every year and I would not be able to do that without the knowledge required to operate a farm reliably and productively. Yes, they're tools, but like all tools it helps to know what you're doing. If I hand you a machining center you won't be able to learn much without a lot of breakage and expense. If you get a lathe to learn and play with, to build up an intuition for feeds & speeds and how materials handle and chip then you will be able to use that machining center to the maximum of its abilities.
Tools require knowledge. 3D printers are no different in that respect and to toss $100 on a printer just to learn is money very well spent. And those old Prusa's excel at precision work, we can do stuff on those that we can not touch with any of the others.
Most people don't intend to print tons every year nor desire to manage a print farm. Most people interested in creating object rather than managing printers will have an infinitely better experience getting a Bambu that is ready to crank out amazing prints right out of the box.
Other than basic troubleshooting (which they have documentation on) there isn't really a need to take a deep dive into how exactly each piece works.
I say this as someone who started printing many years ago with an i3 clone and has replaced nearly every piece of multiple printers (control board, bearings, hotend, extruder, etc!) over the years for better performance. I moved away from wanting to tinker with the printers and haven't touched them since getting a P1S years ago.
I believe I gave two options, one for more in-depth learning and one for appliance style use.
As for the P1S: it is not nearly as well engineered as the A1s, and the firmware absolutely sucks. It also takes forever to get started on a print.
I mean, it's kind of like comparing a tuner car with a new EV. Both will take you from A to B but one requires a lot more work to take you from A to B in a very specific manner while the other just turns on and goes. The tuner car has a tremendous amount of power and is really good at driving fast and accurately but it also requires a lot of work and custom parts to get it to perform like that. The most amount of work in maintaining the EV is that it may need new tires eventually, otherwise it just works.
Most people just want something that just works out of the box using models they downloaded from the internet. It's great that you want to have a 3D printer that performs at the absolute limit of the hardware but that requires work.
Gah and here I was thinking I'm on hackernews rather than 'appliances are us'. Sorry, I got the audience mixed up ;)
Not the MK3. I love that printer to death, but asking somebody new to 3D printing to do Z leveling manually is a tall order unless they're in it for the hobby.
The MK4, with its load cell, eliminates this requirement and is therefore a way better choice for someone new to 3D printing.
The MK3 does automatic Z leveling.
The MK4 is nice but a lot more expensive even today.
The MK3 does have auto Z leveling, but it does it with a PINDA sensor, and you're still going to need to do a Z calibration plus maybe some fine tuning, and that's not the hardest thing in the world, but it's not the easiest.
The loadcell on the MK4 completely eliminates that whole mess.
I've put 10's of thousands of hours on MK3s and the PINDA works fine as long as it is mounted well (which isn't always the case) and set to the right height relative to the nozzle. The automatic calibration routine - on a well built printer, which also isn't always the case, runs exactly once after you've built the printer. Fine tuning is possible, and can be quite useful if you want to play around with the degree to which your prints stick to the bed, for instance for automatic print release tricks (and bed temp as well).
The fact that you have all this control comes at a price compared to say a Bambu A1: you need to know a bit more about how a printer works. But that knowledge will come in very handy when sooner or later you want to do something that isn't in the manual.
The MK4 loadcell is a solution for some problems and a the cause of some others, it is not the best implementation of a loadcell either but it is smarter than the way for instance the Creality K1 and variations are set up (they have four, underneath the bed).
I would recommend against the Creality series printers by the way, they're dirt cheap and wicked fast but they break all the time and the stock extruder is complete junk.
Finally, the best investment you can probably make alongside any printer is a multi-roll filament dryer with a heating element and a fan. That more than anything else will affect your print quality.
We do the bulk of our work on Bambu's now, but for really high precision stuff we will use the MK3s and it's amazing what those old printers can churn out. Nozzle wear is a bit higher on the MK3s than on the Bambu's. The oldest MK3 we have has done > 100 Km of filament...
bambus are great in the "just print stuff" market (no modding needed, everything comes properly configured, the calibration is done automatically, profiles are tested, etc.)
There are some controversies about them locking their printers to their own software and some other issues though.
Good old FTP. I am a bit sad that it kind of died. We should have some modern FTP with super-simple everywhere implementations.
Edit: Actually, some things should be fixed too if any future FTP-like protocol would be added. I never liked the difference between active mode and passive mode in FTP. A user really should never have to care about that. Things should "just work". It's only data transfer in both cases anyway.
IMHO, there's a better answer to any thing you like about FTP. For simple stuff, TFTP seems... fine? I've never looked under the hood, but I'm pretty sure it fixes the really bizarre choices in FTP. For everything else, either SFTP or HTTP seems to be the rule. For moving files around between machines, I've become quite fond of https://github.com/9001/copyparty over http
Isn't the what sftp is? sftp multiplexes commands and data, and is single port without all the old fashioned two port negotiation of FTP.