Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Graph takes two times to display #4148

Closed
wassimj opened this issue Jun 6, 2021 · 14 comments · Fixed by #4157
Closed

Graph takes two times to display #4148

wassimj opened this issue Jun 6, 2021 · 14 comments · Fixed by #4157

Comments

@wassimj
Copy link

wassimj commented Jun 6, 2021

Problem statement

When I open a sverchok graph (using Topologic nodes), I could see from the terminal window that the topologic nodes are executing correctly. However, when the Blender interface finally displays on screen, the visual results are not visible and all nodes report empty outputs. I then have to press "Update all" to re-run the graph and the output is then correct. This effectively means that graphs are taking twice the time to run. Why is that happening? Am I doing something wrong?

Steps to reproduce

  1. Install TopologicSverchok
  2. Run Blender from a Terminal window
  3. Open a sample topologic graph
  4. Watch the results in the terminal window and then in the Blender GUI

Expected result

The graph runs once and the visual results displayed in the Blender view

Actual result

Empty screen and the nodes have empty output

Sverchok version

0.6.0.0

@zeffii
Copy link
Collaborator

zeffii commented Jun 6, 2021

is this upon the first load of a .blend ?

it's helpful to also establish if this is behaviour unique to your suite of nodes, or does this happen if you make a blend with just a few sverchok-core nodes outputting to a viewer node.

@zeffii
Copy link
Collaborator

zeffii commented Jun 6, 2021

but i have seen reports of users needing to press update all for sverchok to fully "get a grip",
#4078

@wassimj
Copy link
Author

wassimj commented Jun 6, 2021

is this upon the first load of a .blend ?

it's helpful to also establish if this is behaviour unique to your suite of nodes, or does this happen if you make a blend with just a few sverchok-core nodes outputting to a viewer node.

Yes exactly. It is on first load of sverchok:

  1. I launch blender.exe from Terminal
  2. I choose open recent and choose my .blend file
  3. It executes (as evidenced in the Terminal window (Topologic nodes are outputting debug information correctly)
  4. Nothing displays on screen and the output of the topologic nodes is NULL/Empty
  5. I choose Open Recent again and choose the same file
  6. It asks me if I want to Save the current file, I choose Don't Save
  7. It opens the file again and this time, everything is correct

@wassimj
Copy link
Author

wassimj commented Jun 6, 2021

It is not specific to Topologic nodes. I just tested it with a simple:
Sphere node connected to a Temporal Viewer node.
It does not display on first load until I either press "Update all" or I re-load the file one more time without quitting Blender.

@zeffii
Copy link
Collaborator

zeffii commented Jun 6, 2021

testing_temporal_viewer_2021_06_06_08_48.zip

this should load:
image

i don't have to press "update all". which makes debugging kind of hard..

@zeffii
Copy link
Collaborator

zeffii commented Jun 6, 2021

there must be something that unites all of the users who have to do this.

@zeffii
Copy link
Collaborator

zeffii commented Jun 6, 2021

@wassimj when you load blender, do you

  • always have to create a new sverchok node tree?
  • or have you stored a startup.blend with an empty nodetree (which is what I do..)
    • Files -> Defaults -> Save Startup File

@zeffii
Copy link
Collaborator

zeffii commented Jun 6, 2021

@wassimj just out of interest, how many cores does your machine have? :)

@Durman
Copy link
Collaborator

Durman commented Jun 6, 2021

Most likely it is relevant to, as I consider this, Blender bug - https://developer.blender.org/T88837
There is no such bug in 2.92

@zeffii
Copy link
Collaborator

zeffii commented Jun 6, 2021

yeah, the order can be this one sometimes:

Read blend: c:\see_life1.blend
do update general was called!
handler: pre load: start
loading:
cleaning Sverchok cache
handler: sv_clean: start
handler: sv_clean: end
handler: pre load: end
handler: post load: start
handler: post load: end
handler: sv_main: start
handler: sv_main: end

@wassimj
Copy link
Author

wassimj commented Jun 6, 2021

Most likely it is relevant to, as I consider this, Blender bug - https://developer.blender.org/T88837
There is no such bug in 2.92

I can confirm that this bug does not exist in 2.92. It opens and loads correctly in 2.92.

@wassimj
Copy link
Author

wassimj commented Jun 6, 2021

@wassimj just out of interest, how many cores does your machine have? :)

Oh I am on a relatively weak machine: An Intel NUC i5 Quad core. I use it for software development but not for any advanced rendering or animation.

@wassimj
Copy link
Author

wassimj commented Jun 6, 2021

@wassimj when you load blender, do you

  • always have to create a new sverchok node tree?

  • or have you stored a startup.blend with an empty nodetree (which is what I do..)

    • Files -> Defaults -> Save Startup File

I have not done the save startup thing. I either start blender, press scripting and then switch to sverchok view or I already have a blender file so I start from terminal (to see errors) then File>Open a blender file. I have also experimented with saving and loading JSON graphs. Works well.

@zeffii
Copy link
Collaborator

zeffii commented Jun 6, 2021

a NUC ? cool :) it can be quite deceiving to dev on a high spec machine.

There's not a whole lot we can do about Blender behaving differently between 2.92 and 2.93, except go into discussion with Blender devs. Thanks for the report, it's really weird that my machine doesn't suffer from this.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants