GSCE v1.20 Bug Report thread

Status
Not open for further replies.

Renato Simioni

Reiza Studios
Please concentrate your bug reports on this thread.

If you may please elect to adapt the following report form below:

Track & Layout Used:
Car Used:
Applicable Settings:
Report:
Steps to reproduce:

Ex:

Track & Layout Used: Montreal 88
Car Used: Formula V12
Applicable Settings: n/a
Report: After losing front wing the car gets airborn above around 100mph.
Steps to reproduce: Crash and lose front wing, drive fast ;)

For any difficulties related to installing, running or setting up the game which are not necessarily bugs in the software, please use the v1.20 release thread.

EDITS - IMPORTANT!

EDIT1: Please allow time for us to test, verify and rectify the issues. We might not reply to you right away but rest assured if you posted here, we will have read it.

Avoid repeating reports by trying some recommended solutions below to known problems:

Measures to try regarding the various CTDs / crashdumps:

1-Run GSCSync again or install from the RAR update packages, make sure you have the lastest GSC.exe;

2- If you had mods installed, try deleting your Gamedata\Shared folder and running GSC Sync again to restore the v1.20 shaders (refer to the note regarding issues with 3rd party mods & tracks in the opening post GSCE V1.20 Released).

3- Add a DEP exception for GSC.exe. (System Properties > Advanced > Performance > Data Execution Prevention) - This is a temporary solution, we´ll fix the issue proper in an upcoming update

Regarding stuttering:

-Try deleting your USERDATA\LOG file, and creating a fresh player profile (PLR);

Regarding shadow flickering issue:

- Try creating a shortcut to GSC.exe and add -fullproc to the command line, performance will be worse but it might prevent the issue from happening.

Please direct further bug reports to this thread.
 
Last edited:
Driving on one of the latest converted tracks like Sebring and then you go back to Burnos Aires the darker part of the tracks replicating rubber build up gets affected somehow and is changed to look like the converted tracks. the below images don't quite show the dramatic change and it is a lot more apparent when your driving in game... I just used Sebring as an example i believe they all will have the same effect. Does something need flushing from memory when changing tracks?
I have only installed tracks that had been updated in the forum here since the v1.20 release as i wanted to avoid "issues" like this. None of the tracks I installed since update place any files outside of the track specific folder whatsover

easy to replicate. Install Sebring available here. load the track, drive a lap. leave track & load Burnos Aires and see what it has done to the track look.

Burnos Aires BEFORE going to Sebring for example:
GSC2014-08-2121-22-52-263_zpsa26e4b69.png


Burnos Aires AFTER driving on Sebring for example:
GSC2014-08-2121-25-22-152_zpse29962d6.png

Are you sure the two screenshots are taken at the same time of day? If the mod put any loose files into the Locations folder or ReizaCommon folder then yes it may cause issues...
 
Yes, positive same time, same session, same car. Just changed track back to a standard GSCE track from Sebring in this case and the racing groove darker buildup gets changed to what was being used at Sebring
EDIT: I'm using a test session with private on to experiment after I first noticed something amiss happening. And yes definite just one track folder was present in the Downloaded tracks installed thus far, that I placed in locations folder only.
 
I just got out of bed and went and double checked all the installed converts to make sure, these are all i have installed so far, and yep positively none of them install anything into shared folder etc, all of the specific files are with there own track folder directories:
Imola_GSCEx_V2.1
Jerez_GSCEx_V2.1
Monza_GSCEx_V1.4_DRS
Portimao_GSCEx_V1.1
RoadAtlanta_GSCEx_V1.1
Sebring_GSCEx_V2.2
Spa_GSCEx_V2.1_DRS
Vanport_GSCEx_V2.1
 
Another interesting thing i've found is this from the Sysinfo file in the support folder from GSCE

DX: DirectX 9.0c (4.09.00.0904)
CPU: Intel(R) Core(TM) i7-3770K CPU @ 3.50GHz @ 3,50 GHz
QPC: 3410244 tps
SRAM: 0 MB
OS: Windows Service Pack 1 Build 7601
GPU: NVIDIA GeForce GTX 770
VRAM: 4029 MB
DRIVER: 9.18.0013.2723
GUID: D7B71E3E-52C4-11CF-C970-58121CC2C435

When infact it should be this (from FTruck)

DX: DirectX 9.0c (4.09.00.0904)
CPU: AMD FX(tm)-8320 Eight-Core Processor @ 4.00 GHz
QPC: 3919443 tps
SRAM: 4092 MB
OS: Windows Service Pack 1 Build 7601
GPU: NVIDIA GeForce GTX 670
VRAM: 4037 MB
DRIVER: 9.18.0013.4052
GUID: D7B71E3E-52C9-11CF-277C-17A41CC2C435

Why is the sysinfo file reporting that my cpu is an intel i7? and my gpu a GTX 770 when it's clearly not.
I'm not to sure if this could be causing some problems, perhaps someone from Reiza can confirm?
 
Hmm you're right, that is interesting. There must be something wrong, as I have just checked the file you mention, and the information in mine is EXACTLY the same as yours, yet I have a 4770K, and my GPU is a Sapphire HD7950 3GB
 
Here's a strange one - If I use GSCSync right now, it wants to copy a new veh file into the Reiza Studios folder of my V12s.

The new file is called reiza15.veh and it is a completely blank file.

(I already have reiza_15.veh, which appears to be correct).
 
Not a bug, this is how multiple qualifying work with Stock Car V8 (only half the field in q1 / q2 each, top 10 from the combined sessions go into q3).

We have however verified the multiple qualifying rules have been broken along the way towards v1.20 release. This will not be fixed for v1.21 but will be fixed in v1.25. For the moment our recommendation is to to NOT run more than 1 quali sessions at all or you will likely run into trouble.
Ahh ok thanks for the reply.
 
Tried some mod cars today and some of them exhibit the same issues as the Vee, really straining the GPU for no obvious reason at all.

Also, still 100% gpu usage goes away and becomes 40% without anything changing except advancing from practice to qualifying in the same session.

Right now, running AA forced in nvidia comtrol panel and not in the game seems to reduce the issue somewhat.

On 1.10 i could run anything without problems, the later versions have been like this for me with certain car models.

There is definitely a bug in the renderer or some incompatibility with something NVidia has done since some time back...

And i am obsessing a bit i know... But as a professional software engineer these things really get under my skin and i have to investigate them (workaholic of grand proportions) :)

I even took it one step further and monitored the render pipeline and vertex buffer in real time vy hooking sandy suite to it. Found that something is definitely hammering the pipeline with calls, when certain cars are rendered, especially if AA is used. Like if some part of the renderer is being fed craploads of mesh data, whilst the mesh itself does not look that complex.

Maybe the shader caching in nvidias newer drivers can be a culprit? Dunno when it was introduced to the driverset...

Or something in the graphics plugins changed.

On the other hand, i have complete confidence in @Reiza Studios and i know it will probably be fixed sooner or later. The sim is still my number one choice, but i wish i could tweak less and go back to the more joyous all play of 1.10, that i know the hardware and software is capable of. Just a bit of polish still and the content will be the best package ever. My nvidia-friends suffering from the same gpu issues as me said the same thing earlier today when we played a bit for fun...

I also opened a bug report for NVidia to look at, on the driver side. Supplied some low level pipe and buffer data for them to investigate.
 
Status
Not open for further replies.

Latest News

How long have you been simracing

  • < 1 year

    Votes: 333 15.5%
  • < 2 years

    Votes: 229 10.6%
  • < 3 years

    Votes: 224 10.4%
  • < 4 years

    Votes: 169 7.8%
  • < 5 years

    Votes: 289 13.4%
  • < 10 years

    Votes: 251 11.7%
  • < 15 years

    Votes: 161 7.5%
  • < 20 years

    Votes: 122 5.7%
  • < 25 years

    Votes: 96 4.5%
  • Ok, I am a dinosaur

    Votes: 279 13.0%
Back
Top