ffhein @ fhein @lemmy.world

Posts

12

Comments

311

Joined

2 yr. ago

Do you mean that you want to build the docker image on one computer, export it to a different computer where it's going to run, and there shouldn't be any traces of the build process on the first computer? Perhaps it's possible with the --output option.. Otherwise you could write a small script which combines the commands for docker build, export to file, delete local image, and clean up the system.

ASA seems like an easier-to-print alternative to ABS, afaik only downsides are higher cost and fewer colours to choose from. Supposedly possible to print ASA without enclosure.

snap/flatpak >500mb

And to make it worse, snap keeps copies of previous versions of all programs. Which can be good if you need to roll something back, but at least last time I used Ubuntu it didn't provide any easy way to configure retention or clean up old snaps.

Yea, definitely keep a fire extinguisher nearby if you decide to do this experiment :) But on the other hand, one should probably already have one of those close to any 3d printer.

One could try "printing" 1.3mm EL wire using a nozzle of similar diameter, low enough temperature to only melt its outer PVC layer, and thick layer height so that its core isn't squished. And it would of course have to be vase mode.. I'm not saying it would work, but it might be an interesting experiment.

Have you compiled your own firmware? It can be a little bit tricky figuring out what you need to change, but it's definitely doable if you have a little bit of experience with computers.

Btw. I think Creality made 2 versions of Ender 5 with different Z leadscrews, so regardless of if you're trying to use prebuilt firmware or compiling your own you might need to verify that the z-steps is set correctly for your hardware.

I did some quick testing yesterday and my initial impressions were that Mythalion and Pyg2 (13B q5_K_M versions btw) were a bit more eloquent and verbose in some situations, but they would often take this too far and start writing novels instead of a dialogue. It also felt like they were more prone to take a sentence and repeat it verbatim as part of all their turns. It's possible that these issues could be toned down by adjusting generation parameters, but MythoMax has been very easy to get good results out of.

It's interesting that you can specify which "mode" pyg2 should operate in as part of system prompts but I didn't test how much difference it actually makes on generation. I told it to be in "instruction following mode" and it seemed good enough at general tasks as well.

If I understand pyg2's model card you're supposed to prefix all turns with <|user|> or <|model|> which I didn't manage to get text-generation-webui to do in chat-instruct mode, so I just used the notepad tab instead.

Cyberpunk feels like it so much missed potential it almost made me sad playing it.. The game is gorgeous and in many ways it really nails the cyberpunk feeling, which I've been very fond of since I was a kid so I would just love to be able to immerse myself in a game like this.

However it keeps slapping me in the face with stupid things that break the immersion.. Primarily the low effort CRPG item system, where each weapon and piece of clothing has random stats. So you find 10 identical looking guns but they all do different amount of damage and add some random elemental damage, which would've made more sense if they were magical weapons in a fantasy game.. When I last played it I found an oversized dildo that does 4 times as much damage as my katana.. And of course a tiny bikini can have better armour value than actual armour..

I would guess it ultimately depends on how much you're interested in modding, fine-tuning settings and calibrating the printer to perfection. If you only want to load a model and hit "print" then I guess you wouldn't notice much difference from just a mainboard swap. But since you're about to install Klipper, you probably have some interest in spending time optimising your printer :)

I've upgraded a lot of things on my printer, but not got to the mainboard yet so it still has TMC2208 driver hardwired in stealthchop mode. I never got much out of pressure advance while the printer was still in bowden mode, even with a BMG extruder which is much stronger than the stock extruder.. Can't say if it would've turned out better with TMC2209 drivers, but after modding the printer to direct drive I have no issue with pressure advance. Now the main reason I want to upgrade mainboard some day is that I would like a 5:th driver so I can do proper dual Z.

And a general Klipper recommendation, in case this is the first time you try it out; Resonance compensation is like magic, and it works perfectly fine calibrating it manually (i.e. without buying an ADXL). After calibration, I could increase acceleration to 4000 without visible ringing. For reference the default acceleration for Ender-sized bed slingers is usually 500, which IIRC can be upped to about 700-900 before you start to get ringing.. Not only does high acceleration make it print faster, but at least for me it had a bigger impact on corner quality than pressure advance ever had.

With a previously installed client I was able to buy and download Diablo II: Ressurected, but after starting it it immediately stated that my setup was to outdated to run this game. I highly doubt that, since I have more than the minimum requirements.

Every time I launch Diablo 4 it says my GPU isn't supported, but I just click "ok" and the game starts anyway.

From what I can find on search engines, the Aquila probably has a 32 bit board, so from a "processing" perspective I think it would be equivalent to the E3 Turbo.

It seems a bit unclear what kind of stepper drivers the Aquila has, but some posts I found claim it has clone TMC2208. I believe the E3 Turbo has TMC2209 which should be more powerful, and supposedly is better for pressure advance. You can search for "tmc2208 vs tmc2209" if you want more opinions on if it's worth upgrading just for that.

I also believe the E3 Turbo will give you UART access to the stepper drivers. IIRC this allows you to configure each stepper motor for either stealthchop (less noise) or spreadcycle (more power) mode, so you could for example set the extruder stepper to spreadcycle.

The E3 Turbo has 5 stepper drivers while the Voxelab board only has 4, right? If you upgrade/mod your printer to dual Z it's nice to have each stepper on their own driver, or if you rebuild the printer to IDEX, but otherwise there's not much use for it.

Thank's a lot mate! Assumed a bot would instantly grab all the keys, as usually happens on reddit, but it was still there after I checked my Steam account so I didn't already have it (Disco Elysium)

edit: sorry @CAPSLOOKFTW@feddit.de ,your comment wasn't there when I first loaded this page, it only showed up now when I reloaded it. Better luck next time.

I believe the standard amount of blocking in Firefox is:

• Social media trackers
• Cross-site cookies in all windows
• Tracking content in Private Windows
• Cryptominers
• Fingerprinters

Since the line between "ads" and "tracking everything people do on the internet" has been pretty blurred, perhaps the anti-adblock checker triggers on any of those.

Which browser are you using? Perhaps it has some built in blocking

Right, I forgot that they make other extruders than the BMG :) Sounds like the motor gear is press fit mounted on the shaft, which is great (as long as you don't need to remove the gear).

If you have the printer in parts, I suppose you could run some tests directly on the motor.

First of all, I would recommend resetting the eeprom to the firmware's default settings by issuing the following commands:

    
M502   ; Reset to factory defaults
M500   ; Save firmware to EEPROM

  

Then, with the stepper motor not installed in the extruder, set esteps to 200 and instruct the printer to extrude 1 unit filament, to verify that the motor gear turns exactly 1 full turn in 1 second.

    
M83   ; Set E stepper to "relative" coordinates
M92 E200   ; Set E steps/unit to 200
G1 E1 F60   ; Extrude 1 unit, with a feed rate of 60 units per minute = 1 unit per second.

  

The reason I suggest setting e-steps to 200 for testing is because this stepper motor should require exactly 200 steps per revolution, according to this page. This should make the printer send 200 pulses to the stepper motor for each "unit" (very unlikely that your firmware is configured with something else than unit=millimeter) you tell it to extrude.

If that works as expected, you can try setting esteps to 400, which should result in the motor gear making 2 full turns per unit you tell it to extrude.

    
M83   ; Set E stepper to "relative" coordinates
M92 E400   ; Set E steps/unit to 400
G1 E1 F30   ; Extrude 1 unit, with a feed rate of 30 units per minute = 0.5 unit per second.

  

I'm lowering the feed rate to make the rotation speed the same, i.e. it should make 2 full turns in 2 seconds now.

If all of this is working, then it seems like your firmware/esteps behaves as expected, and you can set the correct e-steps for this extruder, which should be close to 415.

    
M502   ; Reset firmware to factory default again, to make sure you aren't accidentally saving something else.
M92 E415   ; Set E steps/unit to 415
M500   ; Save firmware

  

Also check your slicer start/end gcode so that it doesn't contain any M50x commands that mess up the EEPROM.

If the extruder is still acting up, at least you know that the firmware is probably ok :)

Did you upgrade the extruder yourself btw? Did you adjust the stepper motor current to the correct value? Otherwise here's a guide how to do that

It could be something completely different, but if you want to rule out a mechanical issue you could check these things:

• The orientation of the drive gear, and its distance from the motor housing. Step 12 in this guide It's quite common that people install it the other way because they think it looks wrong. I was pretty close to doing it myself, and had to double check during installation :)
• There ought to be 2 grub screws, circled in this image. The one on the silvery hobbed gear should be aligned with the flat side on the drive shaft (pointed to by arrow). The one on the brass motor gear should be aligned with the flat side on the stepper motor's shaft. If your stepper motor doesn't have a flat side you can create one using a metal file.
• IIRC when the BMG is installed you can remove the thumb screw completely to verify that the hobbed gears are aligned with the filament path. If you need to adjust it you can turn the extruder until the grub screw is facing the opening, so you move it without removing the whole extruder.
• Ensure that the thumb screw has the correct tightness. Step 21 in the official Bondtech guide linked above.

What printer do you have? If you link a .gcode file where you set esteps in the start code, but it feels like nothing happen, we could check if there's something weird going on. If your extruder has grub screws, have you verified that they're tightened and aligned with the flat side of the shaft?

As others have said, it's not a great extruder, but I've never heard of any "crushing" the filament. Perhaps you could post some photos of the filament and the extruder? Despite its flaws, it should still be usable (unless the plastic tension arm has started to break).

Does it always happen at the same height? Could be Z axis binding. What kind of printer do you have?

Perhaps we'll give that a try since we're going to have to wipe his computer anyway.. They live quite far away so the top prio is something that won't require me to go there and fix the computer every now and then :)