Search Issue Tracker

Fixed in 2018.3

Fixed in 2019.1

Votes

224

Found in

2018.3.0b8

2018.3.0b9

2018.3.0b10

2019.1.0a9

Issue ID

1099125

Regression

Yes

[VR] Particle Systems cause "task.rasterData.indexBuffer == NULL" error and lead to graphical glitches

VR

-

Steps to reproduce:
1. Create a new project
2. Enable VR
3. Create a particle system
4. Enter playmode and put it in your HMDs view
5. If error doesn't show, repeat Stop / Play of playmode again. (when doing this i can repro it with the MockHMD)

Result: "task.rasterData.indexBuffer == NULL" error in the console and potential graphical glitches

Reproduced on 2018.3.0b8, 2018.3.0b10, 2019.1.0a9
Not reproduced on 2018.2.16f1, 2018.3.0a1, 2018.3.0b7
Regression introduced in 2018.3.0b8

notes:
- Is known to also cause memory leaks

  1. Response avatar

    Resolution Note (fix version 2018.3):

    The fix has now landed in our various release versions. 2019.2.0a4, 2019.1.0b2 (available approx 1st February), 2018.3.5f1 (available approx 6th February). Thank you for your patience.

Comments (142)

  1. E567259fc57a48aaffae92dd2ae60eed?d=mm

    wagaga

    Jan 22, 2019 09:20

    Does anyone have this situation, use unity 2018.3 build PCVR game,when close the game will lead windows crash!

  2. 82d4744b969643be9df4e146dcc80788?d=mm

    theguywiththelemon

    Jan 22, 2019 01:44

    Still facing this issue on 2018.3.2f1 in a VR project with the standard render pipeline. I notice it happens as soon as my character shoots (instantiating the first particle systems in the scene)

  3. 503834d37736f47cd539abd411df3a5e?d=mm

    julian-moschuering

    Jan 21, 2019 18:21

    We have multiple different 0x0 access violation errors in particle jobs that started somewhere during 2018.3 _NOT_ VR related. Eg in ParticelGeomWrite->WriteVBOData

    As this is a null buffer two and it started to occured at the same time I hope they are related.

  4. 0c0c0d7be806131e8f24eab69f6dcf74?d=mm

    boco

    Jan 21, 2019 17:05

    What's the ETA to having this resolved, this is a real show stopper and we have a deadline. Hopefully it gets resolved quickly.

  5. 22204384958ada0286352f83fd2dbb97?d=mm

    myxolobe

    Jan 19, 2019 20:36

    Confirmed that this issue still exists in 2018.3.2f1

  6. 95cfe7fd4e884b0551712b76f8de72be?d=mm

    SG_roboryantron

    Jan 18, 2019 22:22

    We actually hooked into the PlayerLoopSystem and disabled all particle related loops and saw the issue go away.... but we don't have particles. It is a useful high level switch to turn particles off so non-art devs can keep moving.

  7. 4d77ef840b5f005369801b7c2da69234?d=mm

    RWeijermars

    Jan 18, 2019 17:11

    Thx for thx for the update Richard. Any news on if a possible solution has been found?

  8. 4d77ef840b5f005369801b7c2da69234?d=mm

    RWeijermars

    Jan 18, 2019 13:44

    Thx for thx for the update Richard. Any news on if a possible solution has been found?

  9. 85f86a16c477dca3280a348adaf8d591?d=mm

    dimitris_baud

    Jan 18, 2019 11:26

    Yes, you are not alone. You can tell by the number of votes.

    My current workflow is:
    - Disable all particle systems and work on the rest of the project fine
    - If I need the particle systems, restart Unity after 2 times I hit Play, when RAM usage is around 30GB (!)

  10. 7f13c2609ade1034b0eab69f8e485e5a?d=mm

    PatrickVRNerds

    Jan 18, 2019 09:09

    Pew.. finally found a thread about this. It's really dramatically - if we start playmode once, everything seems to be fine. If we start playmode a second time our memory usage increases until it's full. We can't do anything right now, computers are crashing when we are closing a build after like 10 minutes runtime. High prio on this please!

Add comment

Log in to post comment

All about bugs

View bugs we have successfully reproduced, and vote for the bugs you want to see fixed most urgently.