Job status by filament usage

gnydick

@fma hallelujah. I've been saying this for months.

dc42

@Phaedrux said in Job status by filament usage:

Let's try to keep the discussion limited to the relevant reported bug.

I have completely lost track of what bug this is. The first post in this thread gives only a vague sense of what it might be. There may just be time to get it fixed in 2.04RC4, but only if I am given a clear description of the symptom by tomorrow morning.

Phaedrux

@dc42 your guess is as good as mine. I wasn't able to find reference to anything else from searching other than this: https://forum.duet3d.com/topic/10581/fix-to-show-correct-filament-usage-in-dwc-for-slic3r-gcode/12

But I think the prusa slicer filament usage was fixed.

Perhaps @gnydick can restate the problem?

dc42

@Phaedrux said in Job status by filament usage:

@dc42 your guess is as good as mine. I wasn't able to find reference to anything else from searching other than this: https://forum.duet3d.com/topic/10581/fix-to-show-correct-filament-usage-in-dwc-for-slic3r-gcode/12

But I think the prusa slicer filament usage was fixed.

Perhaps @gnydick can restate the problem?

I've looked through the list of posts by @gnydick at https://forum.duet3d.com/user/gnydick and apart from this thread and one post in the thread you linked to above, I don't see any posts where the title appears relevant to his original post in this thread. So I hope he can clarify what he means.

robm

gnydick believes the time to finish estimation based on filament usage became less accurate after a recent firmware update.

msbailly thinks that the filament usage counter employed for that estimate is off because it counts retractions but not the return from retraction.

I think prints take a long time anyway and I just check back later.

Thanks David for all your work.

dc42

@robm said in Job status by filament usage:

gnydick believes the time to finish estimation based on filament usage became less accurate after a recent firmware update.

Can you link to the post where he says that?

msbailly thinks that the filament usage counter employed for that estimate is off because it counts retractions but not the return from retraction.

If that were the case, the filament used so far that is reported just before the print finishes would be much less than the filament usage reported by the slicer. I haven't noticed that happen, but I'll watch it carefully in my next print.

It's also possible that the slicer has the opposite bug, i.e. in reporting the filament needed it counts re-prime moves but not retractions. If a slicer has this bug, I'm not sure than anyone not running RRF would notice it.

One thing that might throw off the calculation is if firmware retraction is used and in M207 the re-prime amount is set differently from the reaction amount. This will obviously cause the filament consumption to be different from the amount reported by the slicer, and I don't think RRF allows for that. Again, the opposite bug could affect the slicer if it doesn't take account of the "extra restart distance".

dc42

I watched the figures given by DWC for my latest print as it approached the end. The filament-based estimated end time remained fairly accurate and converged to zero time remaining at the end of the print. The filament usage count at the end tallied exactly with the expected usage count as far as I could tell, certainly within 1mm. This was on a print sliced by S3D version 4.0.0 and not using firmware retraction.

If there really is a problem in RRF in this area, I need more information so that I can reproduce it; for example a sample GCode file along with config.g settings that will enable me to replicate it.

dc42

PS - one thing I noticed during testing is that if you turn the printer off in the middle of a print and then resurrect it, the print times are inaccurate. This is because the resurrection data doesn't include the amount of filament consumed, so it counts from zero again. I'll make a note to fix this.

gnydick

@dc42 sorry I've been away, @dc42 . The prusa slicer bug is what I'm referring to. The DWC shows nothing for filament usage.

When was that fixed?

dragonn

@gnydick said in Job status by filament usage:

@dc42 sorry I've been away, @dc42 . The prusa slicer bug is what I'm referring to. The DWC shows nothing for filament usage.

When was that fixed?

Did you read what dc42 has written? Please post a sample g-code file with this problem and config.g. And maybe M122 response because I use Prusa Slicer too and don't have any problems with filament usage.

dc42

@gnydick said in Job status by filament usage:

@dc42 sorry I've been away, @dc42 . The prusa slicer bug is what I'm referring to. The DWC shows nothing for filament usage.

When was that fixed?

There is an entry for Prusa slicer in the version 2.03 section of the whatsnew file. Is that the one you mean?

chas2706

Prusa slicer just starting to print skirt of model...……. Start of Print.jpg

chas2706

In Cura, all jobs have max layer of 25...…….

dc42

Firmware version?

chas2706

Its a Duet Ethernet on version 2.02RTOS. Luckily I have a camera so I can guestimate myself.

dc42

@chas2706 said in Job status by filament usage:

Its a Duet Ethernet on version 2.02RTOS. Luckily I have a camera so I can guestimate myself.

Please try 2.04RC4.

chas2706

Ok thanks I will try it out later and let you know.

droftarts

"Recognise filament usage comment in Prusa slicer" was added in RRF 2.03.

If you're going to update, don't forget to check through the What's New document to see how changes to the firmware may effect your configuration. And back up your existing, working config, so you can revert to it if necessary.

Ian

chas2706

Ok will do, thanks for that.

chas2706

@dc42

Just Updated to 2.04RC4 and everything looks good.
Many thanks.