Ancestors

Written by requiem 🏴 on 2025-01-21 at 19:30

@djsundog if this is a nuisance let me know, but I have questions about the orangestorm and you are the only person I know who has actually used one 😁

First off, if you were buying one today, would you get any of the add-ons they offer as a package?

https://us.elegoo.com/products/orangestorm-giga

=> More informations about this toot | More toots from requiem@hackers.town

Written by DJ Sundog from the new toot-lab on 2025-01-21 at 19:33

@requiem no worries, happy to give you any answers I have!

I wouldn’t, with the possible exception of a second print head/extruder, not to use in tandem with the first, but to make swapping between the 0.6mm nozzle and the 1.0mm nozzle we got easier (just swap the whole head rather than the nozzle). (we did not get the second print head yet so I cannot guarantee it’s easier to swap than just swapping the nozzle, but that’s my bet.)

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by requiem 🏴 on 2025-01-21 at 23:01

@djsundog cool, that makes sense.

What size (kg) spools do you use?

=> More informations about this toot | More toots from requiem@hackers.town

Written by DJ Sundog from the new toot-lab on 2025-01-21 at 23:04

@requiem we got some 5kg spools that have been working out okay, running from the provided large spool holder, running a meter of filament up to a pulley and through the filament sensor, and then down a meter to the extruder and hotend. I’m not thrilled with the filament path being so dang long and triggering a runout pause with a meter of filament left stuck into the top of the extruder, but I haven’t come up with a preferred mode of operation that’s not that yet haha

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by requiem 🏴 on 2025-01-23 at 17:59

@djsundog did you try their filament? If so, how is it?

Not sure where else I’ll get 5Kg spools but I don’t want to buy a bunch of crap either 🤣

=> More informations about this toot | More toots from requiem@hackers.town

Written by requiem 🏴 on 2025-01-23 at 18:56

@djsundog oh one more: what slicer do you use (I don't see any options for linux)?

=> More informations about this toot | More toots from requiem@hackers.town

Written by DJ Sundog from the new toot-lab on 2025-01-23 at 20:55

@requiem we’re using OrcaSlicer as elegoo made a profile for it and I don’t enjoy cura haha

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by requiem 🏴 on 2025-01-23 at 20:55

@djsundog ah cool, I’ve never used Orca, is there a Linux build?

=> More informations about this toot | More toots from requiem@hackers.town

Written by DJ Sundog from the new toot-lab on 2025-01-23 at 21:24

@requiem yeah, it’s a fork of prusaslicer merged with a fork of something else lol - https://github.com/SoftFever/OrcaSlicer

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by amd on 2025-01-23 at 21:51

@djsundog @requiem

Orca Slicer has Bambu slicer’s layout and some of its functionality along with a bunch more.

Softfever has done an amazing job with it and it runs great on Linux. It’s my daily slicer for both my machines.

=> More informations about this toot | More toots from amd@gts.amd.im

Written by requiem 🏴 on 2025-01-28 at 18:37

@amd @djsundog my first run with Orca and I crashed it 😂

I wonder if it needs more ram or something...

=> More informations about this toot | More toots from requiem@hackers.town

Written by amd on 2025-01-28 at 23:48

@requiem

I don’t know how much ram you have but it doesn’t use a crazy amount. Probably 1 gb if it’s slicing a pretty complicated plate.

If you want any troubleshooting help:

Did it throw an error or just crash silently?

What distro are you running?

I do run software rendering on Wayland with nvidia-drivers

@djsundog

=> More informations about this toot | More toots from amd@gts.amd.im

Written by requiem 🏴 on 2025-01-29 at 02:51

@amd I ran a few more tests, it’s running out of memory 😁

This machine has 32gb, I’ll try my other laptop that had 64 and see what happens 😁

@djsundog

=> More informations about this toot | More toots from requiem@hackers.town

Written by amd on 2025-01-29 at 05:59

@djsundog @requiem

Wow! That’s a surprise to me. I see no issues on my 16gb machine… maybe you’re slicing way cooler models than I am.

=> More informations about this toot | More toots from amd@gts.amd.im

Written by requiem 🏴 on 2025-01-29 at 15:02

@amd just a 760mm tall beaver... 😇 @djsundog

=> More informations about this toot | More toots from requiem@hackers.town

Written by amd on 2025-01-29 at 16:56

@requiem

I guess it’s just the number of layers? My biggest printer is 300mm^3 and can’t print half that height.

I’m way from my machines but I’m curious to add a giga profile to my orca and see what happens.

@djsundog

=> More informations about this toot | More toots from amd@gts.amd.im

Toot

Written by requiem 🏴 on 2025-01-29 at 17:24

@amd yeah im sure thats it, but its something. I’ll have to solve before the giga arrives 😅

I thought my other laptop had more ram, but it doesn’t. If I can find some time I’ll take a look at the code and see if it’s leaking or something that I can do something about.

@djsundog

=> More informations about this toot | More toots from requiem@hackers.town

Descendants

Written by amd on 2025-01-31 at 01:15

@requiem

I just got back to my laptop and added the default elegoo giga profile with the 0.6mm nozzle.

There is definitely a lot more memory usage when you're slicing bigger stuff though not so bad as you saw.

Slicing the biggest benchy I could fit on the plate was about 9gb ram and didn't crash on my X230 with 16gb memory.

Let me know if you'd like me to slice some sample file for any comparison.

@djsundog

=> View attached media

=> More informations about this toot | More toots from amd@gts.amd.im

Written by requiem 🏴 on 2025-01-31 at 01:58

@amd cool, ill try and find a link to the file I can share.

@djsundog

=> More informations about this toot | More toots from requiem@hackers.town

Written by requiem 🏴 on 2025-01-31 at 02:04

@amd Here we go:

This is the original STL:

https://github.com/jjg/models/blob/main/unsorted/beaver.stl

Here's the Orca project file:

https://github.com/jjg/models/blob/main/unsorted/big_beaver_orca.3mf

In particular I simplify the original model (I forget the exact setting but one click below highest quality, which is still a lot fewer polygons than the original) and scale it up to about 670mm in Z.

It does OK slicing until it starts computing the layers. About half-way through I see RAM consumption climb until finally it grabs too much and gets oomkilled.

@djsundog

=> More informations about this toot | More toots from requiem@hackers.town

Written by amd on 2025-01-31 at 02:37

@requiem

Yup. That’s a whole lot of triangles.

I simplified it at the medium level and got to layer 2233 before my machine slowed to a crawl, started swapping, then crashed.

I don’t generally print highly cosmetic models like that and clearly never at that size. TIL that’s a recipe for disaster.

@djsundog

=> More informations about this toot | More toots from amd@gts.amd.im

Written by requiem 🏴 on 2025-01-31 at 02:46

@amd we do a lot of that 😁

I actually think it’s a fixable thing. Based on what I’m seeing, I think it’s buffering all the gcode to ram (and possibly creating multiple allocations, I don’t think there’s 10GB of gcode being produced). If instead that buffer were spooled to disk when it starts getting big I think the oom could be avoided.

That said I haven’t had a chance to look at the code yet so I could be wrong.

@djsundog

=> More informations about this toot | More toots from requiem@hackers.town

Written by DJ Sundog from the new toot-lab on 2025-01-31 at 02:54

@requiem @amd gotta say, that’s a fantastic beaver you’ve got there

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by DJ Sundog from the new toot-lab on 2025-01-31 at 02:56

@requiem @amd I’ll toss my 128GB RAM workstation a shot at slicing it tomorrow once I get Orca Slicer installed on it (I don’t use it as a desktop - in fact, a tiny slice of it is sending this toot lol)

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by requiem 🏴 on 2025-01-31 at 03:02

@djsundog thanks, I’m excited to see what happens!

Out of curiosity, what’s the biggest thing you’ve printed on the giga so far?

Also, have you had to add filament mid-print?

@amd

=> More informations about this toot | More toots from requiem@hackers.town

Written by DJ Sundog from the new toot-lab on 2025-01-31 at 03:04

@requiem

we did a tall Millennium Falcon scaled to the width of the bed and it came in at just over 750mm for height, and we’ve done prints with three filament swaps mid-print - the filament sensor pauses it and starts an unload/load cycle for you at the front corner of the bed and waits for you to tell it to resume and does so no problems so far.

@amd

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by Andrew (Television Executive) on 2025-01-31 at 13:28

@djsundog @amd @requiem we might print a 2.5 ft tall astronaut skeleton today.

=> More informations about this toot | More toots from ajroach42@retro.social

Written by Andrew (Television Executive) on 2025-01-31 at 13:28

@djsundog @amd @requiem i mean "today"

It'll take a week.

=> More informations about this toot | More toots from ajroach42@retro.social

Written by requiem 🏴 on 2025-01-31 at 14:18

@ajroach42 I'm very excited to see this one!

Do you all have any photos of your prints from this thing?

@djsundog @amd

=> More informations about this toot | More toots from requiem@hackers.town

Written by Andrew (Television Executive) on 2025-01-31 at 14:45

@requiem @djsundog @amd

C3P0 and trashcan for scale. Standard pla, printed nose up with no supports.

=> View attached media | View attached media | View attached media | View attached media

=> More informations about this toot | More toots from ajroach42@retro.social

Written by requiem 🏴 on 2025-01-31 at 16:22

@ajroach42 that is kind of amazing… I can’t believe it stood on-end like that 😲

@djsundog @amd

=> More informations about this toot | More toots from requiem@hackers.town

Written by Andrew (Television Executive) on 2025-01-31 at 16:24

@requiem @djsundog @amd

=> View attached media

=> More informations about this toot | More toots from ajroach42@retro.social

Written by DJ Sundog from the new toot-lab on 2025-01-31 at 16:25

@ajroach42 @requiem @amd

it was so scary to watch it print as it got taller haha

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by amd on 2025-01-31 at 16:31

@ajroach42 @requiem @djsundog how’s the strength of the print that size when it only has lightning support?

=> More informations about this toot | More toots from amd@gts.amd.im

Written by DJ Sundog from the new toot-lab on 2025-01-31 at 16:34

@amd @ajroach42 @requiem the Malcon is more structurally solid than I expected tbqh - some of the air ship prototypes felt a little flimsy in the big voids, but the Malcon is thin enough that the voids aren’t nearly as big heh

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by Andrew (Television Executive) on 2025-01-31 at 16:35

@djsundog @amd @requiem and its not just lighting infill, it's lightening infill at, what, 15%? 10?

=> More informations about this toot | More toots from ajroach42@retro.social

Written by DJ Sundog from the new toot-lab on 2025-01-31 at 20:29

@ajroach42 @amd @requiem yeah, think we ran the Malcon at 15%, we’ve gone as low as 10% for sure but that was what got us the flimsy air ships I am pretty sure. I need to start a diary.

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by requiem 🏴 on 2025-01-31 at 20:57

@djsundog most of my repos have a “journal.md” file for exactly this reason 🤣

@amd @ajroach42

=> More informations about this toot | More toots from requiem@hackers.town

Written by requiem 🏴 on 2025-01-31 at 16:37

@djsundog I haven’t used lightning support yet, but I’ve designed a few things that are very strong with no infill and just beefed-up the walls (I have some peony cages like this that have survived multiple seasons outdoors).

@amd @ajroach42

=> More informations about this toot | More toots from requiem@hackers.town

Written by Andrew (Television Executive) on 2025-01-31 at 16:39

@requiem @djsundog @amd we're doing 3 shell layers at .36.

Thick shells, minimal supports.

=> More informations about this toot | More toots from ajroach42@retro.social

Written by requiem 🏴 on 2025-01-31 at 18:34

@ajroach42 sounds like a Sir Mix-a-lot lyric

@djsundog @amd

=> More informations about this toot | More toots from requiem@hackers.town

Written by amd on 2025-01-31 at 16:53

@requiem

I have used lightning super successfully for things that aren't really intended to be handled. It does a really good job of supporting top surfaces with very minimal amounts of material.

I've just never thought about using it for a durable toy. It's cool to see it working for @djsundog and @ajroach42

=> More informations about this toot | More toots from amd@gts.amd.im

Written by requiem 🏴 on 2025-01-31 at 14:13

@djsundog this is good news since that beaver is going to need more that 5kg and I've been pretty disappointed with swapping filament on some other printers.

:flan_excite:

=> More informations about this toot | More toots from requiem@hackers.town

Written by DJ Sundog from the new toot-lab on 2025-01-31 at 03:13

@requiem @amd

this is what I get slicing it at 0.36mm layers with lightning infill after bringing it down to 43k triangles (on my lil laptop with 32 gigs of ram)

=> View attached media

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by DJ Sundog from the new toot-lab on 2025-01-31 at 03:14

@requiem @amd but it is giving me hella side eye about not having supports turned on as well

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology

Written by DJ Sundog from the new toot-lab on 2025-01-31 at 03:39

@amd @requiem

okay, here’s your 3mf file sliced on my laptop with the infill switched to lightning and the supports Orca wanted turned on – resident memory usage peaked at 15GB according to htop

=> View attached media

=> More informations about this toot | More toots from djsundog@fedi.reclaim.technology