HTC Vive resolution and focus
- Thread starter Zac Antonaci
- Start date
Also there's been a steam update and now always on reprojection isn't the same as it was. Having better results with this switched off and both interleaved and async on. Anyone else see something similar? (Non beta)
After longer betatesting I have to admit that 90 fps is still possible with my previous stettings but some systems and locations are very bugy now ( telemetry and rendering issues), so all that fps drops may come from early beta stage. There is a hope guys that all new ED 2.3 features haven't impact VR seetings.
- - - Updated - - -
how about planets and SRV drive , have you test it? In my opinion reprojection "alway on" make a good balance at planet sites.
Are you using steamvr beta client with SS slider or old non beta?
Last edited:
PSA: SteamVR Beta antialiasing/supersampling has issues, update expected next week
(steam discussions)
1080ti just arrived. I'll test this weekend and let you know whether it's worth the upgrade from 1080fe
congrats, I'm pretty sure that 35 % fps boost is worth that money
So I tested it last night and for me it is definetely worth it. It feels like a big step up from the 1080. Previous settings 2x SteamVR sampling with 0.65 in game (leaving hmd in game at 1.0) . Async and always on etc.. if I leave on these settings with the 1080ti I had wonderfully smooth performance in all areas which is great.. clearly out performing the previous card. However it was such a performance boost that I couldn't resist pushing it further and ended the night playing at 2.0 in steam vr with 1.0 in game. These settings look incredible, especially the stars and text. Performance in some areas wobbled a bit with some skipping but I'm sure I can find a happy medium by adjusting down some other things. It looks and feels so good. I am so happy with it. You can tell how it will be in the future. If you have disposable go for it.
I thought since putting the SS into Elites settings (HMD) that it was better to do your SS'ing within Elite, and that it out performs SS in Steam. ie, SS in Steam to 1.0 and HMD in Elite at 2.0 in your situation. Is there a reason you set the SS setting in Steam instead of within Elite?
Last edited:
There is , picture is more sharp and clear with 2.0 in steamvr than 2.0 in HMD Q in game.
I did some test and also used in that way. My rig is based gtx 1080 but I'm using the same settings with async rep. kicking...
Interesting. It used to be the other way around. When the HMD/SS setting arrived in Elite, I did some testing and got a better FPS using the Elite super sampling settings Vs doing it in Steam - all other settings being equal. Earlier in this thread that is the norm, too. Perhaps, since then, Steam has patched up to overtake in performance - I'll try tonight 
Cheers
Cheers
Last edited:
Interesting. It used to be the other way around. When the HMD/SS setting arrived in Elite, I did some testing and got a better FPS using the Elite super sampling settings Vs doing it in Steam - all other settings being equal. Earlier in this thread that is the norm, too. Perhaps, since then, Steam has patched up to overtake in performance - I'll try tonight
Cheers
don't forget that steamvr is frequently updated, so try now non beta and use so called 'vanilla' steam vr . ( to do that mark an option at steam vr "start steam vr when an app starts" and than shutdown Steam and SteamVR and run ED launcher)
Last edited:
Oh and dont forget to check sometimes if Steam changed the default-SS. I found the last time my setting at 2.0 and was wondering why my 980ti had so many Drops
WOW just read your link, sounds awesome. An confirmation that someone in FD are aware and pushing for a fix. My Vive is saved. Anyone notice any changes from the latest beta?
Kevin has SteamVR 3 Mar @ 1:42am
I did some testing, It seems the Application SS settings doesn't seem to work anymore, but Compositor SS still works and seems to render games at higher res. This is bad for games that have in game SS options and adaptive SS. Boosting up the SS doesn't do anything any more. This means games using adaptive SS like the LAB no longer see any benefit with rendering on a higher end system.
We are using adaptive sampling in our game to automatically increase or decrease render quality based on the system. Before, our game looked amazing on a gtx 1080 because it was able to do Supersampling and make everything look crisp even with default SteamVR settings. Now, it doesn't improve itself and is clamped to 1. This is very frustrating as a dev as we took the time to make sure it works to improve quality.
This NEEDS to be fixed ASAP.
#23
Valve:
Alex V [developer] 3 Mar @ 3:12am
Thanks for the feedback. We are aware. Working on an update for early next week.
#24
Oh I'm quite up on SteamVR and beta. I stopped using the beta [apart from periodically upgrading to the beta to compare/test] when Elite added the HMD setting, which does the same thing. That's why I was surprised by your post about Steam being the better place to handle SuperSampling, as that was the old way of thinking and if that has changed I missed the memo (not the first time)
Keeping all other settings equal, I moved the HMD setting over to SteamVR SS, to see the difference - and it hasn't made any improvements that I could see. It just means, now, I would have to change my Steam SS setting when I enter a different game... So I've moved that setting back into Elite, as I'd rather each game manage their own settings when available and functional. I also tried lowering the in-game SS setting to .65 like you suggested, but normally I can leave that at 1.0, HMD at 1.5/1.75 and SteamVR SS at the default 1.0. With those settings It looks equally good, but I get a better FPS
I only 'tested' it by playing the game and watching the FPS counter, but over the next few days I have some time off work, so I'll try it again and log some stats to compare recorded figures... Every frame counts!
(not the first time)Keeping all other settings equal, I moved the HMD setting over to SteamVR SS, to see the difference - and it hasn't made any improvements that I could see. It just means, now, I would have to change my Steam SS setting when I enter a different game... So I've moved that setting back into Elite, as I'd rather each game manage their own settings when available and functional. I also tried lowering the in-game SS setting to .65 like you suggested, but normally I can leave that at 1.0, HMD at 1.5/1.75 and SteamVR SS at the default 1.0. With those settings It looks equally good, but I get a better FPS
I only 'tested' it by playing the game and watching the FPS counter, but over the next few days I have some time off work, so I'll try it again and log some stats to compare recorded figures... Every frame counts!
Last edited:
Oh I'm quite up on SteamVR and beta. I stopped using the beta [apart from periodically upgrading to the beta to compare/test] when Elite added the HMD setting, which does the same thing. That's why I was surprised by your post about Steam being the better place to handle SuperSampling, as that was the old way of thinking and if that has changed I missed the memo
Keeping all other settings equal, I moved the HMD setting over to SteamVR SS, to see the difference - and it hasn't made any improvements that I could see. It just means, now, I would have to change my Steam SS setting when I enter a different game... So I've moved that setting back into Elite, as I'd rather each game manage their own settings when available and functional. I also tried lowering the in-game SS setting to .65 like you suggested, but normally I can leave that at 1.0, HMD at 1.5/1.75 and SteamVR SS at the default 1.0. With those settings It looks equally good, but I get a better FPS
I only 'tested' it by playing the game and watching the FPS counter, but over the next few days I have some time off work, so I'll try it again and log some stats to compare recorded figures... Every frame counts!
Well, I never claimed to be noticable differences in FPS but my settings for a record: steamvr 2.0 Elite HMD Q 1.0 and Elite SS 1.0, AA and ambient OFF, low shadows, rest high/ultra and compositor/vanilla (get around~ 85 fps in the starting hangar) picture is less fuzzy, more legible, clearly sharp in relation to reverse settings namely: HMD Q 2.0 ,steamvr default 1.0 and so on.
I can imagine with the settings such as in game SS 0.65 makes it more difficult to see differences especially when we combine with blurry AA ( I don't) . So try out
Last edited:
Well, I never claimed to be noticable differences in FPS but my settings for a record: steamvr 2.0 Elite HMD Q 1.0 and Elite SS 1.0, AA and ambient OFF, low shadows, rest high/ultra and compositor/vanilla (get around~ 85 fps in the starting hangar) picture is less fuzzy, more legible, clearly sharp in relation to reverse settings namely: HMD Q 2.0 ,steamvr default 1.0 and so on.
I can imagine with the settings such as in game SS 0.65 makes it more difficult to see differences especially when we combine with blurry AA ( I don't) . So try out
No, I don't use AA either, of course. I can see the difference though, here, I have shadows set to high[because it looks great.] SS.65 was only tried because it was mentioned in the post here, so i duplicated settings, I would never otherwise. I've no problem with my settings, it looks and plays great, I was just supersized you still use SteramVR SS for Elite. I'll try your settings, but from what you've just said, the reason I'm using 1.5/1.75 HMD and, comparatively you're using 2.0 in SteamVR is because I use shadows set to high - otherwise I could run HMD 2.0, too. I didn't notice a difference in sharpness between HMD 1.75 and SteamVR 1.75 when I tried, but I'll give it another go, and also try 2.0 withouy the shadows
Last edited:
No, I don't use AA either, of course. I can see the difference though, here, I have shadows set to high[because it looks great.] SS.65 was only tried because it was mentioned in the post here, so i duplicated settings, I would never otherwise. I've no problem with my settings, it looks and plays great, I was just supersized you still use SteramVR SS for Elite. I'll try your settings, but from what you've just said, the reason I'm using 1.5/1.75 HMD and, comparatively you're using 2.0 in SteamVR is because I use shadows set to high - otherwise I could run HMD 2.0, too. I didn't notice a difference in sharpness between HMD 1.75 and SteamVR 1.75 when I tried, but I'll give it another go, and also try 2.0 withouy the shadows
Well based my test you should see also less noticable difference between 1.75's . Have you marked this first option in steamvr http://m.imgur.com/a/DnOcM ? ( translating : "start steamvr when an apps starts" ) and dont have other unecessary option marked. When you do this close steamvr and steam and run game from ED launcher. Latter I will try paste here my screenshots with differences.
Last edited:
Steam vr beta update today mentions supersampling improvements. Had anyone tested this yet??