mastodon.ie is one of the many independent Mastodon servers you can use to participate in the fediverse.
Irish Mastodon - run from Ireland, we welcome all who respect the community rules and members.

Administered by:

Server stats:

1.7K
active users

#marlin

0 posts0 participants0 posts today

Ocean showdown: How #marlin and #sardines outsmart each other phys.org/news/2024-12-ocean-sh

Collective anti-predator escape manoeuvres through optimal attack and avoidance strategies: Palina Bartashevich et al. nature.com/articles/s42003-024

"individual prey in groups follows simple decision-making rules, which lead to complex, collective self-organized maneuvers—and this response is something predators can capitalize on"

Replied in thread

@stooovie Frankly; I've never had Klipper behave fractionally as reliable as Marlin is.

Every time I build or radically change a print system, I go through the "I should try klipper" phase. I always spend 2-3 days on it and if I'm lucky I get a workable bed mesh and in one rare case, it extrudes.

In three different builds, I've never had klipper once running reliably. It always seems to have an entropy setting that is just RNG for 'when is it going off the rails'.

I have made forum posts in the past and from others telling me "that should work. Why doesn't it work?"

I didn't post because it wasn't worth doing so. I already burnt the weekend on it and I had the head literally moving in a square pattern and probing and it couldn't reliably hit the same points time and time again.

Sooo, I toss back in the Marlin firmware card and the Octoprint Pi SDCard and I'm back to reliable operation again.

I want to like Klipper but the damn thing never works, so it's literally just a time sink at this point. Worse yet; 1) It's not like I'm a tech/printing newb, 2) I KEEP TRYING and it somehow always lets me down.

🤷‍♂️

#rambling
#3dprinting
#Klipper
#Marlin
#Tacos
#OkayNoTacosButIWishThereWereTacos

Alrighty.

After dicking with #Klipper for two days, there's something wrong with the kinematics.

Being a Hybrid #CoreXY, my kinematics have a lot less people using them than a run of the mill CoreXY.

Hybrid CoreXY means that two steppers have to work together to make what might be a simple, lateral Y- toolhead move.

I am not an expert at #Klipper by any means, but I've noticed a couple instances when this seems to be true.

There's a small bump at homing (which could also be Klipper backing off, in case it was already flush with the sensor) but there is also an example near 0,0 when clicking the button to move Y, only moves the X stepper when it should move both.

This shouldn't happen. It needs to operate BOTH the X and Y steppers together to make that move. Even if it cannot move one (e.g. it's at the limit), it should then abort the move on the OTHER stepper. It should never move ONLY one when both are required.

So; After two days, it's back to #Marlin.

While the printer is disassembled, I have time to think about all sorts of stupid things. In particular, I came to the conclusion that this Arachne thing in #PrusaSlicer is not compatible with Linear Advance in #Marlin - when Arachne makes a decision to reduce the flow, it does not take into account that it is possible that Linear Advance will work in this particular place in the geometry of the product and reduce the flow even more , which will lead to catastrophic underextrusion.
#3dprinting

It's fascinating how well my printer can suddenly print even very silky filament. Objects printed with Tinmorry's "Ultra Silk+" filaments, except the white cable organizers.

The only difference to before is the #Klipper firmware instead of default #Marlin that came with the printer. To be fair, it was rather old firmware.

#3DPrinting will never cease to fascinate me. Likewise, people / influencers printing nonsense without proper recycling plan will never cease to piss me off. :thisisfine:

Found a super weird bug in #Marlin.

During print moves the nozzle would move then stall, move then stall. No matter the print speed it would make a stuttering print movement. Non-print moves ran fine even at exceptional speeds.

I poked through everything. Flashed an old firmware, changed print speeds down to 20, acceleration to 50 and it would still stutter between each point. A slow acceleration then a pause then it'd do it again.

Eventually I sat down and started changing all of the motion settings as it was printing.

I don't use linear advance as I have direct drive, so I had zeroed out the value. Upon changing it to 0.001 everything returned to as smooth as glass.

I think the printer is performing a division during a print move and the divide by zero caused a stall. It survived flashes and sore didn't matter so I think it's an actual #bug in Marlin.

Predatory #fish use rapid color changes to coordinate attacks phys.org/news/2024-02-predator

Rapid color change in a group-hunting #pelagic predator attacking schooling prey: Alicia Burns et al. cell.com/current-biology/abstr

"the attacking #marlin 'lit up' and became much brighter than its groupmates as it made its attack before rapidly returning to its 'non-bright' coloration after its attack ended"