| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0005223 | The Dark Mod | Design/Coding | public | 20.04.2020 18:18 | 15.12.2022 04:40 |
| Reporter | kingsal | Assigned To | nbohr1more | ||
| Priority | normal | Severity | normal | Reproducibility | always |
| Status | assigned | Resolution | fixed | ||
| Product Version | TDM 2.07 | ||||
| Summary | 0005223: 2.08 Bad internal format console error | ||||
| Description | I'm getting a R_BitsForInternalFormat : bad internalFormat : 0 error while using cvar r_showPrimitives in 2.08. This floods the console with the warning. | ||||
| Steps To Reproduce | Set "r_showPrimitives" "1" | ||||
| Tags | No tags attached. | ||||
 |
@kingsal, can you confirm if this still needs looking at? |
 |
Yes it does. |
 |
Rev 9621 |
 |
I'm not sure that adding a cvar check in global error handler is a proper way to fix it. It should be fixed properly: find why error is happening and fix the core reason, instead of masking the consequences. |
|
|
To me this looks like unintentional log spam. The r_showPrimitives function is generating a test draw and it does not have an associated image for the draw so it falls back to "default" and therefore also spams the warning. I guess the "true" fix is to assign an image to r_showPrimitives or find a way to generate the data without triggering the image association? Is it worth it to do this for the single reason of avoiding console spam? I guess I should check to see if other debug cvars do the same thing... |
|
|
Why does r_showPrimitives generate test draw? Shouldn't it simple print out numbers collected during current frame? |
|
|
Spam is gone for now. Moving the bug out of the 2.10 roadmap. We can revisit the root cause in the future |
