Doom 3 Bfg Edition Texture Lod Bias
Claude
I decided to replay Doom 3 BFG Edition but I'm not being able to "maximize the settings". I remember I did a comparison back then and the original Doom 3 had much better textures, after changing the lod bias in the settings at the time I said I resolved the problem, but this time it's not working (the textures are still blurry) and my search for a Doom 3 BFG tweak guide is having very frustating results.
Any help is very appreciated (what files I should edit, what parameters, etc).
Most important:
- Achieve highest resolutions on default game textures (no modding needed)
- restore imp fireball casting shadows (I remember they included it again but I don't remember where to reactivate)
The game was pretty blurred, like it was running on a lower resolution than the actual set one, maybe this is affecting you, either seta rs_enable 0 or seta r_skipPostProcessing 1 fixed it, with no noticeable drop in visual quality.
Recently I wanted to replay the original DooM 3 and everything in fact looked much better and less blurry than BFG, so I decided to take some screenshots for comparison:
I knew something was wrong when I played through the BFG edition. The game not only wasn't "remastered" but actually looked worse than the original!
The only thing good in it were the hell levels (aesthetically speaking) from the Lost Mission and even then the level design itself was mediocre.
I have no idea why the would want to make thnigs look worse (I guess "they don't give a f*ck" is an appropriate answer, must be related to the console ports), but that's just plain lazy.
/rant.
???
If you actually read the post you'll see the BFG EDITION is blurry while the original game (from 2005) is NOT. "
Also, as the posts above indicate, it was a texture LOD slider that was set to low for whatever reason.
Yup, it's just an expansion pack and a pretty mediocre one save for the last two levels. Also, since the Oculus Rift is reportedly not working on D3 BFG anymore, this whole endeavor seems like it was a waste of time for Id.
This can be demonstrated in 3DCenter Filter Tester when comparing with and without DXVK api wrapper.
With native D3D10, there is hardly any shimmering with a LOD bias of -1.5 for chosen ground2 texture (shipped with Filter Tester in its directory):
With Vulkan (copy 32 bit DXVK DLLs into Filter Tester directory), there is much more texture shimmering, despite HQ filtering (which makes quality of negative LOD bias much better with D3D10-12) configured in driver:
This should also affect all Vulkan games with DLSS support (such as Doom Eternal), as it is recommended to developers to utilize negative LOD bias with DLSS. Also in Red Dead Redemption 2 with Vulkan, DLSS seems to make texture more shimmering than with D3D12 accordingly.
Note: The videos should be watched in native resolution (e.g. 1440p on a 1440p screen). When you try to reproduce with 3D Filter Tester, make sure Windows high DPI scaling is disabled (either globally in Windows display settings or set high DPI scaling to override app controlled in Windows compatibility options for Filter Tester.exe). If Filter Testser UI gets scaled, it also affects its texture rendering, making it appear very blurry and impossible to really judge the result.
I just wanted to give you an update that this issue is still under triage internally. And to ask a question regarding your enhancement request. Is that something which is already specified in the Vulkan standard? If not, than adding this feature might be more difficult, regardless of implementation complexity.
Because screen percentage causes geometry to render at a lower pixel density, temporal upsample requires more texture information from the Surface and Deferred Decal Material Domains to maintain the same output sharpness. For this purpose, the Texture Sample expression can use, by default, Automatic View Mip Bias .
It can't stick to 60fps. At the very beginning of the game, looking toward the stairs immediately drops to 30fps. Looking the other way it's 60fps.
I suspect it's an old-engine compatibility or setting issue because even the menus, before starting a new game, are at 30fps.
Not sure if it is a driver issue but I had strange performance issues with using in game AA on a system with a GTX 770. My only solution for running the game at a smooth 60 FPS with AA is to use SweetFX 2.0 and use the SMAA option. It is not a perfect AA solution but it is better than nothing.
When I mean strange, I mean when combat occurred on my end the game would slow down, not lag or stutter, literally slow down like bullet time in Max Payne. It only happened when I used AA in the game or forced it in my GPU control panel.
I tried this with GTX 550 SLI at 1920x1080 with all settings at max, including x16 AA. I get 45-50 fps when looking at the stairs before moving. It doesn't drop below that while walking towards the door, but when I turn around it drops to about 35 when I look at the guy on the bench.
The fps you get is with vsync on? I should check with Fraps what's going on in more detail rather than rely on the game's counter. The 50fps I get (with x4 AA) is actually an odd figure for standing still in the game with vsync on. If correct, it would imply a rate that changes between 30 and 60fps in a regular pattern.
By default I had regular stuttering/hiccups, quite an odd bug that apparently many experience. Setting com_fixedTic to 1 fixes or at least improves it, but causes motion slowdowns when the frame rate is lower than 60. The best solution I found, instead, is setting com_preciseTic to 0.
There are a few spots where x4 drops down to consistent 30fps, but it's not common. In one spot I disabled vsync to see the effect, and the result was about 50 fps. I guess frames there draw consistently a few ms too long for 60Hz.
I didn't find x8/x16 blurry, maybe something renders differently on your setup? You can try changing the game's texture mipmap bias by setting image_lodbias to small negative values. Maybe -0.5 or -1. You can also try forcing it in the driver using Nvidia Inspector or similar.
@ bakcom - Which version of D3 are you running? On modern rigs v1 can confer a 10-15% (or more) frame boost over the 1.2 and 1.3+ patches. Thread on the ongoing D3 benchmark extravaganza here:
Doom 3 timedemo shootout with period correct hardware.
Turn on FXAA and see if that fixes the issue. For some bizarre reason, antialiasing causes major frame drops unless you have FXAA enabled, at which point it will apply all of those settings while still maintaining a smooth 60fps. At least for me, anyways. Dunno about the others but it's worth a shot.
I played through Doom III again a few years ago on a GTX760 with a decent overclock and an IvyBridge i5. It held 60fps without a problem, averaging around 116fps in timedemo, but that was without AA. Factor in 30% fps reduction for the AA, and another 30% for the 750Ti being slower than the 760 by quite a margin, and it seems you're getting about the right framerate. Play it without AA! ?
It held 60fps without a problem, averaging around 116fps in timedemo, but that was without AA. Factor in 30% fps reduction for the AA, and another 30% for the 750Ti being slower than the 760 by quite a margin, and it seems you're getting about the right framerate.
Redux is a large-scale modification that attempts to modernize Doom 3 with a massive overhaul of the graphical, sound and UI aspects of the game along with bug fixes, restored content and new quality of life features while keeping the gameplay and visual style intact. Here's a temporay list of the mod's features as it was highly requested: (currently applies to version 2.0RC)
For the most part, the new release seems to be pretty much the same as the original digital version on Steam. However, I did spot a few things that are different: the Resurrection of Evil expansion pack is now automatically installed with the base game and doesn't require an additional download, which should be more practical, but I seemed to have some issues registering the included CD keys at first and had to fill "doomkey" and "xpkey" manually. The same could happen with the Steam release.
Being a GOG release this is a DRM Free version with the executable seemingly being lifted from patch 1.3.1, which could already be used to make the Steam version DRM Free. But not having to go through any steps for that to happen is pretty great, and it will help with setting up mods that require their own executable while still running the game through Galaxy.
Another change in this release comes in the form of an autoexec file that GOG has included in both the "base" and "d3xp" folders. It's mostly positive, with the included tweaks preventing the texture quality issues of the Steam release due to modern hardware not being properly recognized by the game.
c80f0f1006