Prusa XL - Alpha Firmware Testing!
Vložit
- čas přidán 17. 05. 2024
- Prusa just announced the alpha firmware that supports input shaping for the XL, so let's test it out and see what it does! In this video I see how much you can reduce print times, how much faster files transfer and show the quality difference between the current firmware and the new alpha input shaping firmware.
Prusa XL - www.prusa3d.com/product/origi...
Thomas 'Salamander' Sanladerer (MK4 Input Shaping) - • Is Prusa's Input Shapi...
Prusa XL Update Blog Post - blog.prusa3d.com/update-origi...
Zombie Hand STL - www.printables.com/model/2935...
Prusa XL Playlist - • Prusa XL
00:00 Intro
00:40 Video Overview/Topics
02:11 Blog Post Recap
03:00 Stringing Update
03:30 Cancel Object
04:23 Shipping Updates
05:10 XL Enclosure?
05:54 Input Shaping Test Process
08:05 Speed/Network Comparison
12:43 Quality Comparison
16:00 Side by Side Comparison
16:30 Ringing Comparison
17:54 Alpha Input Shaping vs Bambu X1C?
20:56 Conclusion and Final Thoughts - Jak na to + styl
Thanks again Robert for all these XL tests and summaries of your finding 👍👍😎👍👍 Very helpful and useful.
Thanks!
This is such a useful, level-headed evaluation. Well done! Thank you! I'm glad to hear your thoughts on 2- versus 5-head as well. I just placed an order for a 5-head printer, and I was wondering whether it was excessive. Now the waiting begins.
Glad it was helpful!
Thanks so much for this video.
i am continuing to wait to finalize my XL purchase based on your videos!
Is there anything you're looking to see? I think it does 'work', there are just some things to look out for. But I've been using it pretty much non-stop since I got it and it's basically a larger MK3 with a tool changer and more speed.
Loving your XL reviews, you have been fair, consistent, and touching on the points I'd care about. I finally got my 5 tool head XL and built it last weekend. Easily the longest calibration I've dealt with out of the box. However, after watching everyone else's oopsies the first benchy and key rings were perfect. Just got my spool of HIPS filament and solvent. Looking foward to some HIPS/ABS tests. Next is using a 0.25mm nozzle for either just the outer layer, or spot printing where their is detail/text and/or the top layer printing. If I can get 0.25mm detail, 0.6mm interior/supports and dissolvable supports I'll get my ideal.
Sorry for the late reply but congrats on the printer! I'm glad to see other people are getting good results.
Good analysis and i liked the logical and physical print comparisons between prusa and bambu. Thank you.
Thanks, I'm glad you saw what I was trying to do!
Great comparison tests
I like your apples to appels comparison.
Thank god for finding your channel.
Thanks for sharing your experience with All of us 👍😀
Thank you!
I love Thomas Salamander's channel!
🤣
I'm NEVER going to use my internal nicknames for youtubers again, people did not like that. :-/
@@RobertCowanDIYFWIW, I didn't think it sounded disrespectful. I think it's an endearment.
@@ken830That was 100% my intention. I don't give people nicknames to people I don't like.
The XL has so much potential. They could do dynamic input shaping with the accelerometer, spaghetti/clog/failed object detection with the load sensor, support for a webcam, etc.
It looks like the main bottleneck is the firmware team. Every exciting feature released is coming from them, but they are the slowest to do so.
Yeah, I think that's a fair take. The hardware (aside from the networking) is good. The printer is robust and the tool changer is rock solid. I wish more people would focus on the positives. The tool changer has been FLAWLESS. It swaps heads without skipping a beat. I think once the firmware is dialed in and we get REAL input shaping, it will be quite the machine.
I suspect the team is a modern, Agile kind of deal. I never liked that system, looks good in theory, never seems to work well in practice.
Agree, tool changer is the long game on this machine. Firmware will catch up, and there’s a ton of room for growth on this platform.. @@RobertCowanDIY
@@VincentGroenewold From someone who claimed to have interned at Prusa, the problem is not the methodology as much as the lack of skilled software devs in Prague. Many companies have been very successful in doing fully remote development using talent around the world, but Prusa seems more conservative.
Once i read the blog i was waiting for yountube greatb
ha, right? I hadn't planned on making this video, but once the firmware was announced, I had to try it out.
"Cancel object" navigation on a rotary dial seems like it would be painful to do if you have 50 objects on a bed. Perhaps unlocking the (non )touch screen?
I have the 5 head. Stringing didn’t seem much more than my other printers. Thanks for the video.
Good to know. When going faster, it strings a bit more, but you have to get up to 250% speed to get that.
Nice! and thanks for your video!
Thanks for watching!
Possible improvement for comparison to other printers(specifically the bambulab). Select the 0.6mm nozzel option for the printer, and since bambu studio is based on prusa slicer, it should be easier to change settings and make a comparison.
Yeah, I did do this, but Bambu profiles are still highly tuned to print as little filament as possible while still getting a cosmetically nice part. It's tough since Bambu is focusing on speed whereas Prusa focuses on part quality and reliability.
Thanks! Its good to watch someone being scientific about it and impartial
thanks!
Thanks for your videos. Really enjoying my XL5. I could use more help with the Slicer. Best way to setup colors and materials.
I'm not sure what you need? Just select the part you want to be a different filament, select the filament for the extruder and assign it to the extruder.
One of the most objective and informative reviews for XL atm.
I really appreciate that comment. I'm not sure people realize how incentivized CZcamsrs are to say good things about certain brands. Prusa is one of the few that doesn't give a dollar to 'influencers', so it's hard to find good content on them.
@@RobertCowanDIY Indeed. Although they have huge delay problems I still support them because of their transparency, open source innovation and fact that they will always support their existing customers with hardware and software updates.
@@brezovprut4431Yeah, they just feel more like a company I'd want to work for/with, and I like how they approach things.
Great analysis. Thanks
I'm starting to wonder if just using .4mm nozzles makes more sense.
Are there currently profiles to use .4mm nozzles for the XL?
There are profiles for 0.4mm. It just ships with 0.6mm. I should probably get some 0.4mm nozzle, but I'm waiting for the hardened ones to come back into stock, since I print a lot of abrasive stuff.
Great video. As you said, getting a XL (and i have) would be about the multiple extruders. But when i finally could finish my order from 2021 i hadn't the means to upgrade to a dual one. Sadly because of the big backlog of orders the focus is more towards that, which i understand. But it will delay my upgrade options for some months now. All things considered the printer does it job nicely for me. I recognize the time difference too in a large print. My last acutal print was about 1h 30min longer then predicted by the slicer.
I'm curious if the new alpha firmware is better at predicting the actual times. I'm exclusively using that now and all times have been accurate.
Thanks for the great video
You bet
Fascinating. I’d love to see a review of multimaterial printing since that’s the only area that the XL might have an advantage over the Bambu (other than print size until Bambu come out with something to rival it).
Have you seen my other videos? I've tested this out and it seems like it works pretty flawlessly, with some caveats in the slicer.
I'm interrested too about 3D multimaterial printing ... What are th ebest alternatives in your opinion ? I'm actually hyped by the 5 arms
I want a good multi flexible material printer
@@JJ-wm9lo Currently the Prusa XL is the only printer on the market that can do true multi-material (flexible, abrasive, and dramatically different filaments on a single machine). The Bambu with the AMS can do similar materials, but it can't do flexible or abrasive filament through the AMS, so you're limited. Those are your only options really.
@@RobertCowanDIY Thanks
Thanks, great overview and like for like comparison. This printer has so much potential. No doubt Bamboo Labs will launch a multitool or idex in the not too distant future, I just hope Prusa can get the XL and Prusa slicer developed quickly enough not to lose the initiative (again).
Check my more recent videos, I haven't NOT been able to do anything I need with it. It's sometimes a bit more 'hacky' than I would like, but the slicer does allow for most situations if you know how what boxes to check/uncheck.
A comparison between a bambu and prusa would be great content and would probably finalize my decision about which one to purchase. I've seen a lot of 3D pritner "review" channels that print a few basic shapes on the bambu and say "yep it looks good and its fast" but they don't go in depth in the print quality, settings, or strength and I haven't seen anyone compare it to the Prusa.
How much do you like to tinker/modify?
Do you want a machine that you can modify or that just works?
Do you want an integrated solution or a tinker project?
Do you want to tinker with 3d printing as a hobby or print as a hobby?
That should be the questions that determine your decision, not comparisons. The quality of the bambu is excellent and it just works. If that's what you're after, go bambu.
The prusa is a tinkerers dream and very modifiable.
This is such a hard question to answer and I'm not gonna touch it with a 10 foot pole! It's a TOUCHY subject. I think you either like Prusa (slow and steady wins the race, very reliable and has a lot of flexibility) or you like Bambu (slicker and fancier, more 'flashy', and prints FAST. but less reliable, relies on cloud, less flexible). I think the best analogy is photography. Are you looking to buy a fancy camera body and use it in manual mode (Prusa) or do you want to get the fanciest tech out there and just shoot in Auto and let the camera do everything for you (Bambu)? The XL is VERY different from an X1C. If you want to print pretty colors and don't care about engineering or mechanical parts, go Bambu. If you want to try strange things and push the boundaries on what can be done with 3d printing, get the XL.
@@RobertCowanDIY It's not that hard really...
Pros for the XL
If you need the bed size and the much faster filament changer AND get the XL dialed in it can be a pricy winner. Filament changer is about 10-20 seconds on the XL vs 90 seconds on the Bambu.
Pro for the Bambu
You can always slow a fast printer but you may not be able to speed up a slow printer. THE CLOUD IS OPTIONAL for bambu printers if you want a bunch of extra features like cellphone access/control or printing without using a slicer with makerworld. After initial setup you can go completely offline or use a very functional lan only mode. For the money you can get at least 2 bambu x1C with ams and A1 mini with ams or 4 p1s with AMS for the cost of an assembled 5head XL. You can get a print farm up and running with the amount the prusa XL 5 head costs all while handling 4 rolls of filament on each machine.
Bambu with an enclosure can print more challenging filaments vs an open air XL printer.(unless you enclose it for more $$$)
Or you can print with 1 machine that is still in beta/alpha firmware. Oh and the biggest elephant in the room. If you order a Bambu printer you will have it in less than a week. As for the XL you should have ordered it years ago.....otherwise you will see it in 2024.
@@RobertCowanDIY I don't think the subject has to be touchy. People are highly opinionated about both Prusa and Bambu Labs, but comparing their printers with identical print settings and filaments (including engineering filaments) would give objective results that people can't really argue about. It would need to be rigorous though, with even slicer differences resulting in (fair or unfair) discrepancies.
@@TROPtasticHave you been on the internet before?! People will always find something to complain about, even if you control all the variables. I'm not really interested in any further comparisons, I'm just going to use it. I've seen such comparisons and it turns into "the results were faked!". I have no interest in changing people's minds, I'm just gonna do my thing and make stuff with it.
Robert, are all your prints on your stock .6 nozzle? I'm finalizing my preorder now was gonna get additional .4 nozzles but think i may wait for now. I know they include a v6 adapter so i got a nozzle x which i can use on my Mk2s if something better comes out for the xl. Edit: got to later in video you are using .6 nozzle. thanks!
Yep, everything you see is the stock 0.4mm nozzle. I haven't switched to 0.4mm yet because I'm waiting for them to have the hardened nozzles or possibly the high-flow solution.
@@RobertCowanDIY thanks! My 5 head is on its way. I’m also waiting for the obsidian nozzles to come back in stock. One thing is can you select different nozzle sizes for each tool head?
@@beeradfpvNice! I'm waiting on the obsidian nozzles as well. At this time, you cannot select different nozzle sizes per extruder, it's global for the whole print unfortunately.
That is positively surprising that a 1:1 comparison of the silcer profiles of the prusaslicer and bambooslicer end up similar print times. If you were to test this a bit more in depth that would be very interesting...
why would it be surprising? they use the same core slicing engine, bambu and orca and superslicer and galaxy slicer are all prusaslicer forks.
As others have said, it's not that surprising. As I said at the end, Bambu is just really good at tweaking the print profiles to get cosmetically/visually great parts with minimal print time.
@@RobertCowanDIY true, well from what i gathered it always sounded like the xl was not optimized for speed. i think prusa himself said something like that. but it could very well be that we just need to wait for the same optimization work from prusa, that already went into the bamboo
@@FlulessYeah, that was my take as well. When the XL was announced, it WOULD have been one of the fastest printers out there (other than Voron and assuming input shaping, etc), but now it's certainly not 'fast' when compared to fast printers. It's more of a versatile semi-industrial machine.
You can crank up the printing speed in the slicer and you will see the XL is capable to print nearly as fast as the Bambu Carbon. I printed a benchy in 25 minutes. The quality is rather poor, but I guess when Prusa is ready with the input shaping and using the values from the accelerometers instead of generic values, the quality will improve a lot.
the bambu is literally twice the speed of the prusa, because the prusa hotend has half the max volumetric flow rate. this is simple physics, and thats pretty straightforward. width x height x actual speed = flow, and the flow on the bambu before you cold core or underextrude is roughly double a 0.6 nozzle prusa xl while only using a .4 nozzle, and will push further when you up it to a .6 nozzle. people keep comparing apples to orange profile settings instead of measuring the actual machine capabilities and its a bunch of nonsense. the only way the XL can catch up is with a hotend swap to a higher flow hotend with a longer meltzone. if you do a hotend swap to something like say the new dragon ace, with a bozzle, you could outperform the bambu on an xl in terms of max limits on net part speed vs a stock bambu this would be roughly comparable to a bambu with a modded hotend running a cht nozzle. neither printer is limited by the motion system. prusa cant do magic, physics still exists, prusa needs an upgrade to their hotend, its that simple.
Justin
@@JustinBuildsThings I literally did an apples to apples slicer settings comparison and the prusa was a min faster on the parts I run from my business. I got a screenshot proving this. The XL uses the same hotend as the MK4 and the MK4 has no problems flowing at bambu speeds.
@@LilApeWhat is the max flow rate of the prusaXL hot end with PLA and 0.6mm nozzle?
@@liamventer Its the same hot end as the MK4. People with MK4's are print with over 30mm flow rates with ease
The mk4 uses a system that's pretty much just v6 with the ol v6 block. Doubt it can go over 30 unless you use the adapter + cht nozzle
impressive! if it can get that close in print speed, it will walk circles around the bambu once you test multi-color or multi-material prints
Eh, I think they will always be 'roughly comparable', because the Bambu is better designed for speed.
@@RobertCowanDIY even on a multi color print? ams on my a1 mini is certainly a decent improvement over my x1cc's, but still seems miles behind your XL's tool switching. i'm guessing you plan to test this for our viewing pleasure once you get the extra heads?
@@ame7165Eh, I'm done testing against Bambu, people really want to like the brand. Unless I have something positive to say about Bambu, I'll just skip it from now on.
@@RobertCowanDIY yeah that's for sure. the facebook group is a cesspool
@@ame7165It kinda reminds me of Tesla. A bunch of non-car guys just went and bought a car that does 0-60 really fast and now they're 'car guys'.
Great stuff. Does the machine still want to move around as much?
I'm not sure what you mean? It doesn't wobble at all.
Mine is moving the table a fair bit. Maybe I need a patio stone under it.@@RobertCowanDIY
@@iimuch3760Huh. It is a very heavy printer, so maybe you have a wobbly table? The stand I'm using is pretty heavy-duty so it takes a LOT for it to shake.
"Once you get 2 you realize you want 5" - that's me 100%
Yeah, I really wish I had just gotten 5. But I might still be waiting for it.
There is some design for a “Prusa XL Sumo” Enclosure that seems like a decent Third Party Option in the works.
Yeah, I've seen it. For right now I'm just waiting to see what Prusa offers. I have bad experiences with 3rd party enclosures.
So the "standard" one has a pretty clear difference (that isn't hidden by the filament's sparkles): the scar on the back of the hand is "sharper"
Yeah, I think there is some fine detail loss, but it's oddly hard to really see. For MY purposes it shouldn't make much of a difference.
All of the components I print for my business take the exact same time on both the X1C and MK4 with IS. The time for the MK4 was actually a minute faster. Using prusaslicer and bambu studio. Using *ALL THE SAME SETTINGS* Both using .6mm nozzle, .4mm layer height, .8mm extrusion widths, accelerations, same flow rate etc, etc. Speed and all various little settings are the same. Yet an "old tech outdated" bedslinger MK4 finishes prints in the same exact time. The XL will be the same.
I mentioned over a year ago the time differences I got between the X1C and my MK3 were minimal when using all the same settings. And the BL fanboys couldn't believe it and said I was wrong and just told me to print more benchys and crystal dragons. Your conclusions in the end matches mine. Using the same settings, they take the same amount of time.
Yeah, I KNOW I'm going to get some negative comments (well, I already have), but that's what I've found. Bambu is just really good at tweaking the profiles to save as much time as possible and print less filament. And they're pushing those print profiles to the MAX. If you increase the speeds by even 10%, the extruder starts to give up, it can't melt fast enough. I have no issue with Bambu, they're just trying to do something VERY different than Prusa.
@@RobertCowanDIY Prusa definitely cares more about quality and strength over speed, which is why their default IS profiles are pretty conservative. But yeah, ramp the settings up to match BL's default settings and you end up with the same print times.
I took a chance with BL and their P1P earlier this year and it was a bad call. I've had better results from anycubic. My parts on my MK3 only took 10 mins longer(17%) than they did on the P1P using the same settings, minus print speeds. Speeds were default on the P1P and MK3 was all at 55mm/sec
You're right about that on bambus. That is why you add a CHT nozzle and you can print at much faster speeds.@@RobertCowanDIY
@@LilApeYeah, that's what I'm finding. I'm using the alpha firmware right now, and decided to try just setting the print speed to 200%, and it's totally fine. They're VERY conservative.
@@luke8874 Then i guess go and add a CHT to an XL and then you're matched in speed again.
Do arcs and curvatures seem to suffer any with the 'compressed size?' I'm curious if it's just compression or they changed the minimum segment lengths, but maybe an arc won't show that if actual arc commands are supported. Many 8-bit boards used to choke from the density of gcode it had to run through for a given portion of time, a lot of that correlates with bloated file sizes due to VERY tiny segment commands being poorly optimized, that translates to gcode file bloat even today.
Also do the toolheads even have an accelerometer? Can't just software update to user-based input-shaping without that really. I've heard the MK4 relies on the fact they are all tensioned to specification and the tool-heads weigh the same across machines etc.
I'm not exactly sure how it's all working, they're using some sort of arc-welder and there will of course be some tolerances. To some degree, it will remove some detail I'd assume. BUT, you might not ever notice it. The toolheads on the XL DO have accelerometers! So they can actually do input shaping right, not sure why the alpha doesn't do it that way. I'm sure it has something to do with the interface, adding it to the slicer, etc, etc.
Interesting video, i hope you will do more XL testing further on, especially as software evolve. The Bambu fanatics filling every comment section and FB group is annoying though.
Ha, right? I do love the irony of the BL fanatics 'calling out' all Prusa people as 'fanatics. Strange stuff. But yeah, I will be doing more videos. I just got the 0.4mm nozzles, so I'm going to see if that makes a difference.
@@RobertCowanDIY i dont know why,, but the prusa groups on FB has been an absolute train wreck since the bambu came out, with all the fanatics going berserk. Maybe its because of the average age that is low, i dont know. But the parts of the internet that I usually hang out in is nothing like that. Anyway, you got a new subscriber at least with your calm and not attention seeking video style:)
@@garagecedric Yeah, too many YT channels are always tying to be the first to find some problem and 'expose' it to the internet. I just like making stuff. I think regarding Bambu, they did something similar to Tesla. They converted a group of non-car people into car people. People that were NEVER into car before bought Teslas and now they're bragging about how fast they are. Bambu is a lot like that. Many of these people came from really cheap entry-level printers and are showing off their print quality and how fast they are. But for long-term enthusiasts, it's nothing really new (especially the speed for the Voron crowd). It happens in many industries, you make the technology accessible for the mainstream audience and you get a lot of 'newbies' that are suddenly at the level of the die-hard enthusiasts.
@@RobertCowanDIY Good sumnary, that makes sense. Ill hope it calms down a bit. And I hope more companies starts making quality printers to make better competition in the segment, as long as they dont come with stupid fanatics :)
Input shaping makes a difference at high speeds. The zombie hand example doesn't have long straight lines to see the impact.
Fair point.
how come tha the network speed is only 200kB/sec? is it using like, an ancient CPU or what?
I suspect it's how they're implementing the USB. The processor is fine, but needs to store the file on the USB storage so there might be a hardware bottleneck between the main buddy-board and the USB interface. But 200kb/sec is awful for something in 2023.
I have a custom CoreXY running the Marlin with input shaping on it, and its complete trash when it comes to ringing. I spent countless hours if not months in actual time trying to tune it, and it just wont translate into quality prints at the high speeds I am going for. I can print just as fast as my Bambu X1C, but the quality is nowhere near the same. I guess what I'm trying to say here is that Prusa is probably wasting time trying to make this whole Marlin firmware work in 2023, and they really need to try some custom firmware or focus on Klipper. You just cant do enough manual tests that translate properly at different speeds, unless you have an accelerometer. Prusa XL is a good looking machine with high potential though, so I really hope they can make it a success.
I'm not entirely sure, but it sounds like they're not really running Marlin, it's something more custom. They discuss klipper, but for some reason they want something more in-house. We'll see if they can get it working once it goes into production and we can actually tune the extruders with the accelerometers.
@@RobertCowanDIY They're not running Marlin on the XL? It is my understanding that it is their own custom fork of Marlin, unless they are using something completely different on the XL than what they run on their MK printers. Where did you read this?
@@user-lx9jm1wo3hI could be wrong, I'm just reading from the blog post. They claim they have 'some snippets of Marlin', so I was just assuming it's not entirely Marlin.
Tom's name can be thought of as "sand loader" - maybe that helps!
That's actually a good tip, thanks!
You have took testing to get certain results to another level, you tried to slow down the bambu wich is wrong you should speed up the prusa and give it a normal nozzle size for normal comparison.
You're not wrong. If I do more testing, I should see if the Prusa can do the 2.5 hour print with a 0.4mm nozzle, the same settings as Bambu. I'm really not trying to make them go head to head though, they're quite different machines.
@@RobertCowanDIY yet you exactly tried that, this is so misleading. Your end conclusion is that the xl does the same while you slowed the bambu down wich is wrong. Testing is worthless if you make the test to fit a certain view.
@@RobertCowanDIY 2200 for the prusa(1) extruder and 1500 for the bambu ams both large format corexy but still you claim not comparable.. it's only because the bambu outperforms the xl on every level but you are a prusa fan so you change the test to make it seem the prusa is same or better while it's not.
@@OneHappyCrazyPersonYou aren't as happy as your name would lead me to believe. But jokes aside, I'll do a better comparison next time.
@@OneHappyCrazyPersonYikes, no one is trying to deceive or even convince you. Like what you want to like, I'm not forcing anything on you. But hey, the Bambu is by no stretch of imagination a large format corexy. I wouldn't really even say the XL is large-format. I'm TOTALLY fine with you preferring the Bambu if you're fine with me preferring the XL. Let people have their preferences.
So I was playing in Bambu studio and set an X1C to 0.6mm nozzle and .24 layer height and 7 walls(or 2 walls 48% infill) to get it up to 140g. and the time was about 3 hrs.
So using the same amount of filament and a crazy number of walls you will still print faster on a bambu printer.
Or just do 0.4mm nozzle 4 walls 45% infill and you are still under 4 hrs with 141grams used.
The key is to use the "bambu filament" preset 21mm/s vs 11mm/s) for generic pla. It will print the same trust me the generic profiles are very conservative and a simple profile switch means the bambu wont even break a sweat at 21mm/s
Gotcha!
Still faster because BL default print speeds are set significantly faster. take those same settings in prusa slicer for the XL/Mk4, but match the print speed settings to BL and you get the same time. I did this exact thing and got a min faster on the MK4 over the X1C while still using the same amount of filament. Only a gram in difference between the two.
@@LilApe what is the max volumetric flowrate of the XL?
I would argue that the Bambu 0.20mm Standard settings are perfectly adequate for the type of model that you demonstrated. I understand what you were trying to show by adjusting the slicing settings to get them to the same mass, but to me this shows far better optimization of the Bambu profiles. "Prusa is coming at it from a different angle" sounds like a bit of a cop out - for this type of model the Prusa profile is just wasting material. For the record I enjoy your content - just an observation!
Best would be swapping the nozzle on the prusa for a 0.4 or swapping the bamboo to 0.6 Then it would be a 1:1 comparison. Comparing by printed weight is the next best thing.
No its not wasting material it has a 0.6 nozzle by default in comparison to 0.4 bamboo. If every wall is way bigger of course it will print more filament.
Yes, a fair comparson is to select 0.6mm nozzle for the X1C
You are 100% right. I don't usually print stuff like this, but the 2.5 hour part would have been the better application here, no question. I don't agree that 'coming at it from a different angle' is a cop-out. I print a lot of mechanical or functional parts, and the Bambu profiles are fairly weak. It prints too little material and too fast (poor layer adhesion). The print profile are aimed at cosmetic prints, not functional prints. If I was just printing cosmetic prints, the Bambu is the smarter one to use. For the record, I have both and just sold my MK3 in favor of keeping the X1C. I don't dislike the Bambu, but it's been far more fiddly, but it GREAT for rapid prototypes. Also, I saw another comment from you, but it seems to have disappeared, I was going to reply but it seems like it's gone now. Thanks for the support!
Yeah, I'll try that. But I really just want to get back to actually making stuff. As much as I like comparisons, I'd rather just use the thing :-)
Crikey im very surprised the state at which this launched. I have an e3d tool changer with 4 tools and while it is a bit rough and ready it seems to print better than the XL although probably not as fast. I really would be expecting a heck of a lot more from a Prusa XL. Im puzzled why the launched this early if they still have issues. Its still on my roadmap to get one eventually, but by the time it will be available im pretty sure there will be a lot of competition from other providers.
what issues do you see? there are some shortcomings, but I've been using it non-stop since I got it and it's been reliable and been able to do everything I wanted from it, with some small tweaks.
Wow. The Input shaping one is Not Sharp. Very smooth. I dont want it
Yeah, that's why I did the comparison. I usually print engineering models and fine detail isn't all that important to me. It depends on what you're looking for. But for sure the input shaping reduces fine detail.
6 minutes to upload 47mb, wtf.
Right? This was known right when they came out and has been mentioned MANY times. There's a hardware problem there, it's the same on wifi or ethernet.
Honestly maybe just flash klipper on this thing if possible.
They did address this in the blog post. In short, I think they sell to a lot of government/EDU customers who are picky about open-source stuff hanging out on a network. From my day job, we fill out forms from large companies every week; they want to know what libraries you use, what talks to their network, if cloud is required, etc, etc. There are security risks when you just slap a raspberry pi on a network. For most hobbyists, this isn't really a concern, but for industrial customers, it's something to consider for sure. This is largely why Bambu came out with the X1E, they were probably unable to service that market.
@@RobertCowanDIY Good point. I believe if a hobbyist is using this / non governments, klipper seems like a better place to go instead of using Prusa's own firmware.
the bambu will always be more capable than the prusa. this is really simple, its the max flow of the hotend. the bambu hotend has a better meltzone and can handle more than 2x the flow the prusa can. width x height x actual speed = flow. any two printers running the same speed, accel, widths, height should be exactly the same "print speed" on outputs assuming the motors have sufficient power, not losing steps, etc. the motion system is better on the bambu's and the hotend is better, its always going to be faster, and thats never ever going to change, unless prusa releases a high flow hotend. when i asked jo about it at ERRF he blamed e3d for there not being one.
input shaping will never make a printer faster, it smooths the motion curves out to reduce ringing at the expense of smoothing out surface detail. on klipper and RRF printers this is tunable, but on prusa and bambu, this is not. the bambu will auto set the midpoint for input shaping models, but currently prusa is just setting to a fixed point. in order for a universal fixed point to be valid, because your accel performance is going to be build/printer specific iin conjunction with the exact surface and positiong placement, like a fingerprint, they are going to have to be using a very wideband and high smoothing model to account for this, which means lots of smoothing potentially. if you want to print faster, only the hotend can lift that cap. typically you can kludge a hotend to have a bit higher flow by moving to a fatter nozzle and changing the backpressure at a given flow rate, and thats an impact on the classic 3d printing paradox of speed (being able to print faster by going taller or wider but actually moving slower while at an increased flow rate, and being able to extend this with fat nozzles) but on modern printers this isnt really necessary because you can push 0.8 easily on .4 nozzles with modern geared extruders so this makes comparing things apples to apples pretty straightforward. when you use equivalent nozzles. a .6 on a bambu will push something like 32 cubic if not more if the hotend is sufficiently hot. a .6 on a prusa xl will push something like 15 on the same material. its just not comparable. you can push these printers any width/height/speed you want as long as you stay under the flow cap, but the bambu is going to be about twice as fast any way you dice it when you actualy do an apples to apples comparison. all the input shaper is doing is smoothing the surface as a way of mitigating ringing. its not making the printer faster. it is making the printer quieter. you can print at the exact same settings with and without input shaper, the only difference will be ringing performance and noise. so because of this, no, the prusa mini is not magically suddenly faster, nor is the mk4 nor xl. you can run those exact same speeds and feeds before.
how do you compare printers in an apples to apples context? you compare how fast you can move the motion system before steploss, max flow with same nozzle size, put a bowl of water under the toolhead at a fixed distance to measure the cooling, look at the spectral charts from a nozzle mounted accelerometer to understand the ringing behavior/performance and how that limits accel vs quality. nobody does this. and thus every pritner comparison on youtube is just....wrong.
you should always get the exact same time with the exact same settings......on any two printers assuming they can handle those speeds and feeds in the first place. the physics isnt changing, you're setting it to some accel, some velocity some extrusion width, height etc. where this might vary is between different slicer families where the slicers make different assumptions. this is most notable with slic3r derivatives (PS/SS/bambu/orca) vs cura engine stuff, as they make completely different assumptions about the actual shape of the extrude line. if you use the same slicer and same slicer settings on both printers you will get the same result unless something is wrong, excluding things like startup sequences. you can also use something like klipper estimator to digital twin a slice vs a fw config to actually get an accurate print time estimate as prusaslicer is notoriously and comically wrong the faster you go up in accel, something thats been known in voron, annex, and ratrig communities for years. .
Looking at all your comments on all my videos, you really really like Bambu. It's a great fast printer, no question. I won't throw in a 'but' after that because I think you have your mind made up. Just let people have preferences. These are tools, not identities.
@@RobertCowanDIY I have 3 prusas 1 bambu 3 vorons 2 annex printers, a raise3d raise2, an ender5-mercury one, a saturn2 2 kp3s's I'm pre-ordered on the magnetoX and I have a stratasys unprinted se and a bunch of other printers. I used to run a 75 printer farm for production but I run 15-20 printers for prototyping now. Some of the machines rotate in and out now. I have access to a ton of other equipment and I do consulting for farms. Over the years I've probably worked on at least 50 different kinds of printers first hand. I actualy was sponsored by prusa for a hackathon in 2020, they gave us a dozen spools to give away, and my charity cobranded with them in the past, we made and donated 260-ish thousand face shields in 2020. Back then, the mk3s was top of class, it just flat out isn't even in the conversation today. I sold my xl already.
@@JustinBuildsThingsCool! So real talk, what's your 'go-to' printer? Which one do you feel more comfortable with?
@@RobertCowanDIY at this point I'm probably most comfortable with my voron 2.4 and trident. I first built the 2.4 in early 2021 just before the kits era, and was an early adopter of moving into like custom metal parts, 2.5 amp motors, 48v, and extreme flow hotends, so they have been continuously modified over time to be purpose built for my workflows, very fast prototypes and batching small functional parts. I suspect that will shift to my annex k3's when I get to spend some more time with them, I just built them and havent put the time yet.
@@JustinBuildsThingsGotcha. I've always wanted to build a Voron, but I just have too many other projects and don't need another one! Thanks for the info.
Thanks for the update.
About Tom, it's "Tom San lader er". It's not that hard.
imho Salamander is kinda offensive. Just call him Tom if you don't know how to pronounce his last name..
Absolutely zero offense is meant. I've just never heard him say his own name, which is why I don't know how to pronounce it. He and CNC Kitchen are some of the best 3d printer channels on CZcams right now, I have lots of respect for both of them.
"Tom from Made With Layers" (=
Your humble opinion blows if you find calling someone who's last name sounds like salamander salamander. It is a joke. Maybe stop playing so much guitar outside and touch some grass while you're out there
@@algeflixYeah, that would have been better.
@@alexanderpierzchala1615My apologies, I must have struck a cord. I meant zero offense by it. I follow hundreds of channels and can't always keep them all straight. I'll keep my nicknames to myself next time.
The xl is not even close in speed to the bambu. The bambu can print high speed and accurately. There is no comparison between the two with accuracy and speed. If the xl could even reach the speeds of the bambu, I am sure it would not be as accurate. The bambus have great motion systems.
k.
I'd argue motion system wise Prusa is better, it just depends on how fast you want to push it.
The parts I print for my business take the exact same time on both the X1C and MK4 with IS. The time for the MK4 was actually a minute faster. Using prusaslicer and bambu studio. Using *ALL THE SAME SETTINGS* and an old bedslinger finishes prints in the same exact time. The XL will be the same soon enough, Its alpha FW for a reason. Got a screenshot for proof if you want it, but I already know the answer.
Yes show me the screen shot. I dont even know what you are trying to show me with the screen shot. But I would like to see it. The MK4 does not have accleration that is even close to the bambus. So even if it could travel the same speed, it could not get there as quickly as a bambu. @@LilApe
@@luke8874 Here you go. Just remove the spaces in the link. And I upped the acceleration settings on the MK4 to match what was on the BL. It only sped it up like two minutes lol so not a big difference. Using all the same settings as I've been running for years on my MK3. Even used the same amount of filament. Now you gonna tell me its fake and I'm doing it wrong?
Obviously youtube doesn't like links. But the screenshot showed both fies with their print times. Just like it did at the end of this video. Taking the same amount of time using all the same settings. It's okay, an old "outdated" bedslinger runs just as fast as the best thing out of shenzhen