Thread: Graphics Error
View Single Post
Cryptic Studios Team
Join Date: Sep 2012
Posts: 55
# 44
12-03-2012, 01:38 PM
Originally Posted by shadow10052012 View Post
Here you have a set of graphic options wich will not crash your video driver when in space, or if it does, it will be very rare.
Options marked with * can be changed at will.

Monitor vertical sync OFF
Half resolution OFF
Antialiasing NONE
Dynamic lighting OFF
Lighting quality HIGH*
Shadows LOW
Max lights per object 2
Texture anistropic filtering 2x*
World texture quality 150%*
Character texture 100%*
World detail distance 75%*
Terrain detail distance 150%*
Character detail distance 80%*
Show high detail objects OFF
Max physics debris objects 30*
Raise settings in character creator OFF
Postprocessing OFF
Visual FX quality LOW
Soft particles OFF
Bloom quality LOW*
Bloom intesity 100%
Screen space ambinet occlusion ON (this is important, must be ENABLED)
Cinematics focus ON*
Underwater view OFF*
Reflection quality LOW
Lens flare quality OFF

Now for "Troubleshooting" i will list only those that are ON, those not listed are OFF including "Limit frame rate".

Limit CPU usage when inactive ON
Mulitcore rendering ON
Minimum shadow buffer bias ON
Reduce CPU/GPU usage ON

Try them as listed first time, then you can modify those with * if necessary. I do not guarantee that these settings work, but they did for me and other players with same video issue.
This definitely helps us narrow it down a bit. Unfortunately, our error reporting tools can't handle the non-user-mode part of the driver exploding, so information on our side for this has been a bit scarce.

If these settings are actually helping, I'd really like to know so we can start from there to narrow it down. The settings I'm mostly curious about are...

Dynamic lighting,
Postprocessing, and

If a few want to run with one (and only one) of these settings flipped from the "non-crashy" settings, and then report back if the crashes do happen, that'd help us narrow down exactly which one of these features (which it hopefully is just one) is causing the issue.

This is one that doesn't seem to be affecting any of our internal Intel testing machines (stacks of Intel GPU laptops) and right now suspicions are pointing at it being an Intel driver bug because it's unlikely we'd be able to make the entire display driver crash on our own otherwise.

PS: Try to keep the vitriol to a minimum, pretty please.