[Bug] UVTools produces corrupted pm5s files for the Anycubic Photon Mono 5s #742

cdheiser · 2023-07-19T00:25:02Z

System

UVtools v3.15.1 X64
Operative system: Darwin 22.5.0 Darwin Kernel Version 22.5.0: Thu Jun  8 22:22:22 PDT 2023; root:xnu-8796.121.3~7/RELEASE_X86_64 X64
Processor: Intel(R) Core(TM) i9-9980HK CPU @ 2.40GHz
Processor cores: 16
Memory RAM: 15.84 / 28.12 GB
Runtime: osx.13-x64
Framework: .NET 6.0.18
AvaloniaUI: 0.10.21
OpenCV: 4.7.0

Sreens, resolution, working area, usable area:
1: 1792 x 1120 @ 200% (Primary)
    WA: 1792 x 1095    UA: 896 x 548
2: 3200 x 1800 @ 200%
    WA: 3200 x 1775    UA: 1600 x 888

Path:       /Applications/UVtools.app/Contents/MacOS/
Executable: /Applications/UVtools.app/Contents/MacOS/UVtools
Loaded file: Diablo One.pm5s [Version: 518] [Class: PhotonWorkshopFile]

Printer and Slicer

Printer: Anycubic Photon Mono M5s
Slicer: Prusa Slicer 2.6.0

Description of the bug

Initial generated pm5s files are corrupted resulting in a print that does not proceed past the first layer. Print times displayed are several orders of magnitude greater than what's expected (hundreds of hours), and never proceeds past the first layer.

I've made both the bad and working pm5s files available here: https://file.io/JoWgslx6cLfQ as well as the sl1 file these were converted from. I can also share the original prusa project.

How to reproduce

Slice a file in Prusa Slicer
Export in SL1 Format
Open in UVTools
Copy converted pm5s file to the printer
Print the file
Observe the print time looks correct until the first layer goes to print and then jumps by a factor of approximately 100x

If you reopen the pm5s file and make a no-op change and re-save it, the resulting file prints successfully.

Files

No response

The text was updated successfully, but these errors were encountered:

github-actions · 2023-07-19T00:25:40Z

This is your first time submitting an issue with UVtools 🥳Please review your issue and ensure that the submit template was followed, the information is complete, and not related to any other open issue. It will be reviewed shortly. Debugging is very important and make the program better. Thanks for contributing and making the software better! 🙌

sn4k3 · 2023-07-19T02:33:56Z

If I understand correctly, the conversion out of SL1 generates a bad file, but if you reopen the file and force a resave without any other change it makes files good?

cdheiser · 2023-07-19T02:44:57Z

That's correct. I've spent the last few days figuring out how to consistently reproduce the problem. My method of forcing a re-save is to edit the print settings to increase the base layers by 1 and then set it back.

…

On Tue, Jul 18, 2023, 7:34 PM Tiago Conceição ***@***.***> wrote: If I understand correctly, the conversion out of SL1 generates a bad file, but if you reopen the file and force a resave without any other change it makes files good? — Reply to this email directly, view it on GitHub <#742 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACQARWXHD2BEXRUIP6L2SRLXQ5BR5ANCNFSM6AAAAAA2PD2BWA> . You are receiving this because you authored the thread.Message ID: ***@***.***>

cdheiser · 2023-07-19T02:58:13Z

I forgot to mention that this happens on both Windows as well as MacOs

…

On Tue, Jul 18, 2023 at 7:44 PM Chris Heiser ***@***.***> wrote: That's correct. I've spent the last few days figuring out how to consistently reproduce the problem. My method of forcing a re-save is to edit the print settings to increase the base layers by 1 and then set it back. On Tue, Jul 18, 2023, 7:34 PM Tiago Conceição ***@***.***> wrote: > If I understand correctly, the conversion out of SL1 generates a bad > file, but if you reopen the file and force a resave without any other > change it makes files good? > > — > Reply to this email directly, view it on GitHub > <#742 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/ACQARWXHD2BEXRUIP6L2SRLXQ5BR5ANCNFSM6AAAAAA2PD2BWA> > . > You are receiving this because you authored the thread.Message ID: > ***@***.***> >

sn4k3 · 2023-07-19T03:00:02Z

There is a better way to force a full save: Pixel editor and erase a black pixel on a black pixel (Does nothing but turns the full encode flag). Also, it's possible to change the save flag via Terminal with two lines of code.
Your method rebuilds all layer's properties which is "aggressive"

What I find between both:
Model XY info is missing on bad file, but I think it's not required, anyway I provide a file with only that fixed to you to try and report back.
Then the per layer settings is enabled which will follow layer table instead of the global settings, while your good file have that set to false and I confirm that force a full encode without any other setting turn that to false

cdheiser · 2023-07-19T05:21:35Z

The file you provided had the same behavior. I've attached a photo of the status screen on the printer.

I ran this test with the build plate and vat off so I could see what's going on. The first layer does what you'd expect, 10 second exposure, but then it just sits there, screen off, like there's some extra wait time it needs to do.

sn4k3 · 2023-07-19T15:14:50Z

Can you check if theres an firmware upgrade for your printer?
Because that is not normal, it is triping by follow layer table and those are correct set as I have checked.
Please report back with your firmware version

cdheiser · 2023-07-19T16:56:38Z

The printer is on 1.0.2.1 which is the latest version on Anycubic's site.

…

On Wed, Jul 19, 2023 at 8:15 AM Tiago Conceição ***@***.***> wrote: Can you check if theres an firmware upgrade for your printer? Because that is not normal, it is trip by follow layer table and those are correct set as I have checked. Please report back with your firmware version — Reply to this email directly, view it on GitHub <#742 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACQARWXML4QWBAWAMMWB6CLXQ72XLANCNFSM6AAAAAA2PD2BWA> . You are receiving this because you authored the thread.Message ID: ***@***.***>

sn4k3 · 2023-07-19T17:36:16Z

Ok, so let's do another test...
Slice a file on Anycubic slicer, test if its printing and attach here. (Without UVtools processing)

cdheiser · 2023-07-19T19:14:23Z

I sliced this file in Anycubic Photon Workshop and it prints without error: https://file.io/PEquEQfzs8y3

…

On Wed, Jul 19, 2023 at 10:36 AM Tiago Conceição ***@***.***> wrote: Ok, so let's do another test... Slice a file on Anycubic slicer, test if its printing and attach here. (Without UVtools processing) — Reply to this email directly, view it on GitHub <#742 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACQARWT4HFLKEFRWTNKEXFDXRALJXANCNFSM6AAAAAA2PD2BWA> . You are receiving this because you authored the thread.Message ID: ***@***.***>

sn4k3 · 2023-07-19T20:04:11Z

Now try the same file but with per layer flag on (Manually edited): Diablo WingsPerLayer.zip

cdheiser · 2023-07-19T23:27:56Z

Looks like it's the per layer flag. Your file exhibits the identical behavior to all the other problematic files.

…

On Wed, Jul 19, 2023 at 1:04 PM Tiago Conceição ***@***.***> wrote: Now try the same file but with per layer flag on (Manually edited): Diablo WingsPerLayer.zip <https://github.com/sn4k3/UVtools/files/12099704/Diablo.WingsPerLayer.zip> However, that produced file uses version 518 and UVtools was using 517. — Reply to this email directly, view it on GitHub <#742 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACQARWQJEFFLOQNMKNCU6IDXRA4ULANCNFSM6AAAAAA2PD2BWA> . You are receiving this because you authored the thread.Message ID: ***@***.***>

sn4k3 · 2023-07-19T23:52:28Z

Then is a firmware bug, and a bad one that prevent you from use that feature.
UVtools have a bug in that conversion stage that set flag ON when it shouldn't, will be fixed on next release.

For now use this workaround after conversion and save again (Reset layers properties):

DanJaramillo · 2023-08-23T20:19:50Z

I have the same issue when running Lychee Slicer 5.2.201 to slice followed by UV Tools 3.15.1 to post process. The Lychee generated file displays correct times and layers, however the UV Tools file does not print beyond the first layer, show a correct preview, or show correct times. This is evident on both the AnyCubic M5S interface as well as the AnyCubic Photon Workshop.

- **UI:** - **Settings:** - (Add) Themes: Fluent default, which set light/dark mode depending on system settings - (Add) Theme density which only applies to Fluent theme - (Change) Themes: Rename "Default" to "Simple" - (Remove) Simple theme - **Exposure time finder:** - (Fix) "Multiple layer height" panel visibility was wrongly check for `ExposureTime` instead of `PositionZ` - (Fix) "Multiple exposures" panel was always visible even if not able - (Improvement) Disable "Multiple exposure" and "Multiple heights" for the AnyCubic format - **Tool windows:** - (Fix) Align the description to the left to fix the misalignment when resizing the window - (Fix) Make layer selection shortcuts to work - (Fix) Some invalid bindings - (Fix) Save properties to disk/clipboard was disabled when file has no thumbnails - (Fix) File - Reset layer properties was no unlocking the File - Save - (Improvement) When sending a command under "File - Terminal", it will now unlock the "File - Save" option and refresh the UI with the new information if changed - (Improvement) When moving model by an operation or suggestion it will redetect the issues to update their positions (#752) - **Layer actions - Import layer(s):** - (Add) Button to set the current layer - (Improvement) If any file isn't imported it will show an message informing why - (Fix) Layer properties were getting rebuilt (#739) - **File formats:** - (Change) Thumbnails logic, it's now a `List` instead of a fixed array, this allow to dynamic add or remove thumbnails without having a fixed number - (Fix) The flag for using per layer settings (`IsUsingGlobalParameters`) now take into account if layer is able to use each property in check before compare (#742) - (Fix) Accessing `BoundingRectangleMillimeters` was not calculating the `BoundingRectangle` if necessary - (Fix) SL1: Files with thumbnails size different from original SL1 would not display under the UI - (Fix) Anycubic: Model information is not set after an auto-conversion from SL1 - (Fix) CTB: Error when open file in partial model with malformed file due invalid checksum - (Fix) Image files: Could not find a writer for the specified extension (#755) - **UVtoolsCmd:** - (Add) "compare" command to compare two files and output the differences - (Add) "print-formats" command to print the available formats - (Improvement) Improve "extract" command to allow extract specific thumbnails or layers in a range or indexes (#754) - **Project:** - (Change) Rename UVtools.WPF to UVtools.UI - (Change) Make UVtools.UI nullable enabled - (Remove) UVtools.GUI - (Improvement) Use CompiledBinding instead of ReflectionBinding - (Improvement) Sign the libraries and project with a strong key - (Upgrade) .NET from 6.0.18 to 6.0.22 - (Upgrade) AvaloniaUI from 0.10.21 to 11.0.4 - (Upgrade) OpenCV from 4.7.0 to 4.8.0 - (Add) PrusaSlicer printer: Elegoo Saturn S

sn4k3 · 2023-09-17T12:13:18Z

Fixed

Rubiniuz · 2023-09-28T08:18:34Z

So the above comment

Then is a firmware bug, and a bad one that prevent you from use that feature. UVtools have a bug in that conversion stage that set flag ON when it shouldn't, will be fixed on next release.

For now use this workaround after conversion and save again (Reset layers properties):

this is the fix? cant this be fixed in a uv tools update? or is this a anycubic issue?

just clicking that button fixes the issue? or are there more steps required?

I really want to use uv tools but i also get that first layer issue. a step by step guide also posted on reddit could save a lot of people a lot of time.

edit: I am using version 4.0.1 and have the same issue

sn4k3 · 2023-09-28T14:09:40Z

It's fixed on v4, however the bug persist on the firmware, if you use the suggestions on UVtools it will enable the flag that trigger the problem. If you only use to fix issues it won't be enabled.

If and only after you save the file the following highlight "PerLayerOverride" is True printer will produce that problem

In your case if you see True you must turn that False by the trick of "reset layer properties".

Skorinn · 2023-10-23T17:31:49Z

I am able to reproduce this bug with 4.0.3:

After slicing the file with Lychee, and openning in UVTools the PerLayerOverride property is False:

I used the Pixel Editor to delete a blank pixel and saved the file (Save As). After which the PerLayerOverride is True:

Using the Reset Layer Properties does reset it back to False. Please let me know if you need any additional information.

Thanks

Rubiniuz · 2023-10-23T17:46:01Z

I am able to reproduce this bug with 4.0.3:

After slicing the file with Lychee, and openning in UVTools the PerLayerOverride property is False:

I used the Pixel Editor to delete a blank pixel and saved the file (Save As). After which the PerLayerOverride is True:

Using the Reset Layer Properties does reset it back to False. Please let me know if you need any additional information.

Thanks

Yeah exactly. thats why we closed it. thanks for the heads up again

Skorinn · 2023-10-24T01:37:45Z

Yeah exactly. thats why we closed it. thanks for the heads up again

I'm not sure I follow. Based on the comments above, I thought that 4.0.0 included a fix for UVTools to not always set PerLayerOverride to True when using it to detect and repair issues (i.e., islands, resin traps, etc.). Is that not the case?

If not, I'm curious if there are plans to make that change or if this is the intended behavior? The workaround provided (thank you) is easy enough to make this a low priority, but it is easy to forget to do,

If the change has been made, is there something I am doing incorrectly that is resulting in that property getting set to True? What triggers that change?

Thanks

sn4k3 · 2023-10-24T01:48:48Z

You need to provide the file, for me replicate and find the problem.

I thought that 4.0.0 included a fix for UVTools to not always set

Yes, until you apply suggestions or have them to auto apply the PerLayerOverride will turn true when using wait times.
Also, if slicer sets an individual value on layer different from global it will cause UVtools to set that to true on save time.

Skorinn · 2023-10-24T02:07:18Z

You need to provide the file, for me replicate and find the problem.

Here is the file created by Lychee before post-processing (PerLayerOverride=False): M5_100um_3DSLA_TestModel.pm5s

This is the file saved by UVTools after solely using the Pixel Editor to delete a black pixel (PerLayerOverride=True): M5_100um_3DSLA_TestModel_00h32m_1ml_copy.pm5s

Thanks

sn4k3 · 2023-10-25T18:44:10Z

The problem is on slicer or how you have configured it:

All layers are set with 1200 mm/min lift speed, however the global properties are set with 180mm/min. UVtools find that and turn flag on to respect layer data instead.
Only the global lift speed and retract speed for normal layers shows 1200 mm/min, but all others does not match layer data. Something is odd, that value ins't normal either, please inspect your slicer print settings (profile). Also make sure you have your slicer up to date.
Note that reset print properties will not sane that insane speed. As so I recommend to reset values, go to edit print parameters and set good values there.

Skorinn · 2023-10-25T20:24:08Z

The problem is on slicer or how you have configured it:

That makes sense. I'll take a look at fixing the profile in the slicer. Thank you for looking into this for me.

cdheiser assigned sn4k3 Jul 19, 2023

sn4k3 closed this as completed Sep 17, 2023

sn4k3 mentioned this issue Oct 27, 2023

[Bug] Breaks m5s .pm5s files #775

Closed

sn4k3 mentioned this issue Apr 3, 2024

[Bug] M5s Pro Files saved from UVTools sometimes never stop exposing the first layer. #859

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Bug] UVTools produces corrupted pm5s files for the Anycubic Photon Mono 5s #742

[Bug] UVTools produces corrupted pm5s files for the Anycubic Photon Mono 5s #742

cdheiser commented Jul 19, 2023

github-actions bot commented Jul 19, 2023

sn4k3 commented Jul 19, 2023

cdheiser commented Jul 19, 2023 via email

cdheiser commented Jul 19, 2023 via email

sn4k3 commented Jul 19, 2023 •

edited

Loading

cdheiser commented Jul 19, 2023

sn4k3 commented Jul 19, 2023 •

edited

Loading

cdheiser commented Jul 19, 2023 via email

sn4k3 commented Jul 19, 2023

cdheiser commented Jul 19, 2023 via email

sn4k3 commented Jul 19, 2023 •

edited

Loading

cdheiser commented Jul 19, 2023 via email

sn4k3 commented Jul 19, 2023 •

edited

Loading

DanJaramillo commented Aug 23, 2023

sn4k3 commented Sep 17, 2023

Rubiniuz commented Sep 28, 2023 •

edited

Loading

sn4k3 commented Sep 28, 2023

Skorinn commented Oct 23, 2023

Rubiniuz commented Oct 23, 2023

Skorinn commented Oct 24, 2023

sn4k3 commented Oct 24, 2023 •

edited

Loading

Skorinn commented Oct 24, 2023

sn4k3 commented Oct 25, 2023

Skorinn commented Oct 25, 2023

[Bug] UVTools produces corrupted pm5s files for the Anycubic Photon Mono 5s #742

[Bug] UVTools produces corrupted pm5s files for the Anycubic Photon Mono 5s #742

Comments

cdheiser commented Jul 19, 2023

System

Printer and Slicer

Description of the bug

How to reproduce

Files

github-actions bot commented Jul 19, 2023

sn4k3 commented Jul 19, 2023

cdheiser commented Jul 19, 2023 via email

cdheiser commented Jul 19, 2023 via email

sn4k3 commented Jul 19, 2023 • edited Loading

cdheiser commented Jul 19, 2023

sn4k3 commented Jul 19, 2023 • edited Loading

cdheiser commented Jul 19, 2023 via email

sn4k3 commented Jul 19, 2023

cdheiser commented Jul 19, 2023 via email

sn4k3 commented Jul 19, 2023 • edited Loading

cdheiser commented Jul 19, 2023 via email

sn4k3 commented Jul 19, 2023 • edited Loading

DanJaramillo commented Aug 23, 2023

sn4k3 commented Sep 17, 2023

Rubiniuz commented Sep 28, 2023 • edited Loading

sn4k3 commented Sep 28, 2023

Skorinn commented Oct 23, 2023

Rubiniuz commented Oct 23, 2023

Skorinn commented Oct 24, 2023

sn4k3 commented Oct 24, 2023 • edited Loading

Skorinn commented Oct 24, 2023

sn4k3 commented Oct 25, 2023

Skorinn commented Oct 25, 2023

sn4k3 commented Jul 19, 2023 •

edited

Loading

sn4k3 commented Jul 19, 2023 •

edited

Loading

sn4k3 commented Jul 19, 2023 •

edited

Loading

sn4k3 commented Jul 19, 2023 •

edited

Loading

Rubiniuz commented Sep 28, 2023 •

edited

Loading

sn4k3 commented Oct 24, 2023 •

edited

Loading