Print time estimate is bonkers
 
Notifications
Clear all

Print time estimate is bonkers  

Page 1 / 2
  RSS
mchahn
(@mchahn)
Trusted Member
Print time estimate is bonkers

I have a sliced octopus (yum) that prusaslicer estimates a print time of 1 hr 54 mins.  Octoprint says 5 hrs 19 mins which is quite accurate.  I can't think of anything I could be doing wrong.  Has anyone else seen this problem?  I've enclosed my 3mf file.

 

Edit: Second try to add file.

Edit2: I cannot add the file.  I click the Choose File button, choose my file, and then when I save the post I don't see the file.  Is there something else I have to do after choosing the file?

Best Answer by mchahn:

My printer was set to a 35% feed rate with the M220 command.  I had set it slow when doing some tests.  After the tests I did an M502, which restores the hardwired settings from the marlin build.  Now I know that doesn't reset the feed rate (wtf?).  After setting the feed rate to 100% octoprint and prusaslicer agree within 18 seconds.  Apparently octoprint reads the feed rate before calculating print time.

The moral of the story is check printer settings when times don't match.  Maybe this will help some lurkers.  Sorry to waste everyone's time.

Posted : 10/10/2021 10:03 pm
MysDawg
(@mysdawg)
Estimable Member
Slicer

What version of the slicer are you using?

Posted by: @mchahn

I have a sliced octopus (yum) that prusaslicer estimates a print time of 1 hr 54 mins.  Octoprint says 5 hrs 19 mins which is quite accurate.  I can't think of anything I could be doing wrong.  Has anyone else seen this problem?  I've enclosed my 3mf file.

 

Edit: Second try to add file.

Edit2: I cannot add the file.  I click the Choose File button, choose my file, and then when I save the post I don't see the file.  Is there something else I have to do after choosing the file?

 

Posted : 10/10/2021 10:16 pm
sylviatrilling
(@sylviatrilling)
Honorable Member
zip

You need to zip your 3mf file in order to attach it to a post. 

Mk3S+,SL1S

Posted : 10/10/2021 10:22 pm
mchahn
(@mchahn)
Trusted Member
Topic starter answered:
added zip file

Thanks.  I'm not sure how I was supposed to know that 🙂   Zipped file is enclosed.

Posted : 10/10/2021 10:51 pm
mchahn
(@mchahn)
Trusted Member
Topic starter answered:
slicer version

My slicer version is 2.3.1 + win64.

Posted : 10/10/2021 10:54 pm
Swiss_Cheese
(@swiss_cheese)
Noble Member

@mchahn

 

My guess would be that Octoprint is slower then printing from an SD card, Ive printed similar products with derometer 60A using SD card and the estimation was within a couple of minutes.

 

BTW, way don't be a dick about posting photo's when people here help they are giving their time to help you, this is a community not the prusa business get off your ass and take some freakin pictures if it makes the people that are helping you feel better.

 

Regards

 

Swiss_Cheese

The Filament Whisperer

Posted : 11/10/2021 2:19 am
mchahn
(@mchahn)
Trusted Member
Topic starter answered:
sorry

Didn't mean to be a dick.   I am curious though what you mean by posting photos and to take some pictures.  Nobody said anything about pictures.

I don't believe octoprint would be 2.5 times slower.  The print is only 15 mm/sec and the gcode is the same either way.  I've done a zillion prints where prusaslicer and octoprint agreed on the time.  Something is wrong with this particular part.  Did you look at my 3mf?  Not saying you should, just asking politely.

Posted : 11/10/2021 4:18 am
bobstro
(@bobstro)
Illustrious Member
Slicer & OctoPrint can use different methods of time estimation
Posted by: @mchahn

I have a sliced octopus (yum) that prusaslicer estimates a print time of 1 hr 54 mins.  Octoprint says 5 hrs 19 mins which is quite accurate.  I can't think of anything I could be doing wrong. 

The PrusaSlicer estimates should be pretty close if you're using a Prusa printer. If you're using another printer, the machine limits do need to be set correctly to get good estimates. PrusaSlicer, depending on how you've configured it, will use a number of different methods of calculating print times, and some plugins will update print times during the print, so there's no reason they should match exactly.

My notes and disclaimers on 3D printing

and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan

Posted : 11/10/2021 5:06 am
mchahn
(@mchahn)
Trusted Member
Topic starter answered:
I will try to find the configuration that triggers the prusaslicer problem

> if you're using a Prusa printer.

I'm not printing this slice on a prusa.  This slice is for an old creality ender 3.  It doesn't have an sd card so I can't compare sd with octoprint.  But octoprint has nothing to do with this problem.  Octoprint is only slower when the printer eats the gcode faster than the raspberry pi can't send it.  At 15 mm/sec that's not going to happen.  I should not have mentioned octoprint in the original post.

> the machine limits do need to be set correctly to get good estimates

Getting a good estimate is not the problem.  The problem is that prusaslicer is giving an absurd estimate.  Two hours versus five hours is not a difference that can be explained with printer settings.  And the print really does take five hours.

I also have the latest prusa printer and on previous slices the two printers have printed the (almost) same gcode with the same print time, within minutes.  The gcode determines the print time, not the printer.  

As I said, I will try to find the configuration that triggers the prusaslicer problem (bug).  I posted here hoping someone else had seen the problem, or if anyone has any ideas what I could be doing wrong.

Posted : 11/10/2021 4:51 pm
geekandi
(@geekandi)
Member
Ender 3 V2 vs your set up

Ender 3 V2 on left, your settings on the right

Your 3MF file is printing in PLA though at these speeds and set up I gotta assume you're doing TPU.

Hope this helps you!

(ignore output filename format, I changed mine across the board for all my printers for better categorization)

Posted : 11/10/2021 6:03 pm
mchahn
(@mchahn)
Trusted Member
Topic starter answered:
ok, maybe I *am* a dick after all

My printer was set to a 35% feed rate with the M220 command.  I had set it slow when doing some tests.  After the tests I did an M502, which restores the hardwired settings from the marlin build.  Now I know that doesn't reset the feed rate (wtf?).  After setting the feed rate to 100% octoprint and prusaslicer agree within 18 seconds.  Apparently octoprint reads the feed rate before calculating print time.

The moral of the story is check printer settings when times don't match.  Maybe this will help some lurkers.  Sorry to waste everyone's time.

Posted : 11/10/2021 6:10 pm
geekandi liked
geekandi
(@geekandi)
Member
Timewasting..

is what the internet is for, and pr0n.

Glad you figured it out - and that may have been on your display as well 🙂

Posted : 11/10/2021 6:15 pm
mchahn
(@mchahn)
Trusted Member
Topic starter answered:
My settings were wrong, but in the printer, not prusaslicer

> Ender 3 V2 on left, your settings on the right

Thanks, I will fix that.  Apparently though that mistake didn't affect anything.  See my answer above.

Posted : 11/10/2021 6:15 pm
bobstro
(@bobstro)
Illustrious Member
You're not the first!
Posted by: @mchahn

My printer was set to a 35% feed rate with the M220 command.  I had set it slow when doing some tests.  After the tests I did an M502, which restores the hardwired settings from the marlin build.  Now I know that doesn't reset the feed rate (wtf?). 

If you change those settings mid-print via the menu or knob, they are persistent until reset or a power cycle. I include gcode commands to reset M220 and M221 at the start of every print in my start gcode along with a few other settings. If you change those settings mid-print via the menu or knob, they are persistent until reset or a power cycle. 

My notes and disclaimers on 3D printing

and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan

Posted : 12/10/2021 1:16 am
mchahn
(@mchahn)
Trusted Member
Topic starter answered:
RE: settings

I include gcode commands to reset M220 and M221 at the start of every print in my start gcode along with a few other settings.

Good plan.  I will do so also.

Edit: Although that will possibly make the octoprint time estimate wrong.

Posted : 12/10/2021 1:51 am
Guido666
(@guido666)
Active Member
RE: Print time estimate is bonkers

 

Posted by: @bobstro
Posted by: @mchahn

I have a sliced octopus (yum) that prusaslicer estimates a print time of 1 hr 54 mins.  Octoprint says 5 hrs 19 mins which is quite accurate.  I can't think of anything I could be doing wrong. 

The PrusaSlicer estimates should be pretty close if you're using a Prusa printer. If you're using another printer, the machine limits do need to be set correctly to get good estimates. PrusaSlicer, depending on how you've configured it, will use a number of different methods of calculating print times, and some plugins will update print times during the print, so there's no reason they should match exactly.

@bobstro Is there a way to adjust Machine Limits (or other necessary print time calculation settings) for "G-Code Flavors" that disable the Machine Limits settings page?  My prints are consistently 25-35% off on time estimation.

Posted : 06/02/2023 5:20 pm
Neophyl
(@neophyl)
Illustrious Member
RE: Print time estimate is bonkers

Change to marlin and set them to match your machine, change the drop down at the top of the limits to use for time estimate, then change to the flavour that doesnt use that tab.  Even though the tab is gone the settings are still 'there' and will be used for the time estimate.

Posted : 06/02/2023 9:01 pm
Guido666
(@guido666)
Active Member
RE:

@Neophyl

 

Perfect, thank you, that's what I was wondering.

This post was modified 2 years ago by Guido666
Posted : 06/02/2023 9:24 pm
Guido666
(@guido666)
Active Member
RE: Print time estimate is bonkers

 

Posted by: @neophyl

Change to marlin and set them to match your machine, change the drop down at the top of the limits to use for time estimate, then change to the flavour that doesnt use that tab.  Even though the tab is gone the settings are still 'there' and will be used for the time estimate.

I just tried it, and PrusaSlicer does remember the Machine Limits settings after you switch the G-Code Flavor back (and the tab disappears), but it does not continue to use the Machine Limits to calculate print time unless a G-Code Flavor is selected that enables the Machine Limits tab.

I asked on the Discord channel, and Tojik (Prusa Mod) responded...

well it looks like you should be able to switch to RepRapFirmware and it being fine, it supports machine limits and so on. I've read something about a wipe tower complicating things (edited)

some sources:

https://forum.duet3d.com/topic/19389/prusa-slicer-2-3alpha2-adds-new-g-code-flavor-for-rrf

https://forum.duet3d.com/topic/5181/slic3r-prusa-g-code-flavor (M900 M907 and wipe tower stuff)

Posted : 06/02/2023 10:51 pm
Neophyl
(@neophyl)
Illustrious Member
RE: Print time estimate is bonkers

I tried it with 2.5 and a flavour that didn’t have the tab and the slicing time estimate was same between marlin Legacy and the one I changed it to .  Think there was about a minutes difference. 

Posted : 06/02/2023 11:03 pm
Page 1 / 2
Share: