Valve Index Stuttering/Frame Drops

[Varangia1122]

I recently built a brand new PC on next gen hardware. And for some reason, I am getting sever frame drops every few seconds to the point that it is nauseating.

New Build is as follows:
Windows 10
NVIDIA 3070 ti
AMD Ryzen 7 7700K
32gb of DDR5 5600 Ram


In comparison, my old build that had no problem running my Index:
Windows 10
NVIDIA 1660 ti
I5-9600K
16gb of DDR4 3000 Ram


I have gone over quite a few possible solutions as is already. I have turned off Motion Smoothing and adjusted the resolution of my headseat. My GPU drivers are all up to date. But neither of those solutions helped with the stuttering. I am fresh out of ideas as of right now.

[--ranXerox--]

does your motherboard have a IGPU? turn it off in bios.

[Varangia1122]

You are the first person who has suggested that. And while it did not fix my problem, it is the only solution I have done that has improved it noticeably. It happens significantly less in the SteamVr home now, and noticeably less in games.

[--ranXerox--]

you probably have the drivers lingering for some odd reason, go into control panel, device manager, display adapters, disable all in there minus nvidia. you also might need to adjust settings in steamVR like supersampling keep it at 150%. make sure you have your hdmi from both headset and pc monitor on the same nvidia card and not the motherboard igpu.

[Varangia1122]

The only driver is my 3070 ti. And the super sampling didn't particularly help either. Everything is plugged into my GPU as is. Two DP's and one HDMI. I'll try a few solutions people mentioned with straight up disconnecting other ports while in VR. But aside from that i've got nothing else I know to try.

Thank you though for your help so far. It has made a sizable difference <3

[--ranXerox--]

one last thing you can do is enable direct mode. in steamVR settings.

[Varangia1122]

Still nothing.
I may just contact Steam Support directly.

[JF0603]

do you have this problem across the board (i have the same thing- only hardware changes were a change of case, to one with increased airflow thru vs what i had prior.
running
i9-10900k @stable static OC all cores 5.1GHz,
Z490 Auros master with a set of b-dies OC'd (gskkillas well stable [normally would be ddr4-3200 CL14 (or 12 I am not on it to recall since i last touched that years ago to get it "stable"...)
EVGA RTX3090 FTW3 ULTRA [no LHR revision] . never had an issue until a few mo back, when i wanted to give HL:VR a try- noticed the issue there, figured it was the game, so went to something i am used to running and know the perf of- "beat saber" only to see in that (especially easy to visualize with the more extensive telemetry overlay from fpsVR enabled (though it happened between multiple times a second if i was changing relative rotational postition with my head, or "looking not at a fixed position while remaining stationary"- which even so produced it at a very noticable to my eyes- 2-3 times per 5 seconds on avg in just a menu being entirely stationary. SO again, i thought maybe my 3090 is just not up to the task like it used to be and purposely lowered all settings, but to no avail, then figured it has to be "beat saber" and jumped on HL:A since that uses a different backend and i recalled speicifcally optimized for the index in terms of the process of development so it might be a representitve sample for the hardware i have- and at the settings formerly used in the game which involved 100% res @90Hz vs 120 recalling that alyx had its own dynamic quality/LOD system that would maintain a fixed fps if the frametimes were not cooperating, yet it in the menu alone was nauseating enough i just all together stoppd- (went through everyones suggested steps, rolled back drivers and windows 10 to an earlier edition i recalled having 0 issues with in VR, yet still there was something that was inducing the stutter... after a week of this and muiltiple inquires i have resigned and packed it up figuring it must be device or backend issue (though i had kept the index completely in immaculate condition and not changed anything in the playspace since the first proper setup at roomscale when i was finally able to get one some years ago. packed it back up for now for storage as i see many people with the issue and yet the same suggestions that have not worked thus far and have been ones i had spent suffifciently long in terms of time on. I of course would love to use the device again i just have no way to as i would be ill if i tried- since the HL:A leviathan fully original community campaign is somethign i had been really excited to try for over a year since its announcement and i really enjoyed the way the index was fully utilized for Alyx vs other titles which didnt fully utilize the unique featuresets the index offered- and since have also declined in order to make accessible titles to the quest which has caused a degredation of quality but i can underrstand financially why studios are pandering to selling by volume now to those less capable units in terms of tracked accuracy and immersive featuresets for your "mid level" VR game user, (w/ overmatched hardware, where the core has no hw and uses the quest like all in one devices to solve that issue... hopefully psvr ports one day come out for something like the deckard if that comes out if the issues here are dealt with... maybe by then in valve time this issue will be resolved and i will also have a 4090 or 4090 ti with the need for likely DP iteration >2,x thruput as i am seeing a cutoff potential for the type of content if tethered that you could roll through the 40 series continued use of dp 1.4 (at this time)
have reverted drivers and done all these recommendations but across the board,

let me know if you make progress

[Varangia1122]

Hi, I do have an update.
My issue was fixed recently. It is technically an issue with drivers, but there was no need for me to revert them on my end.

As this is a new PC that I built, once I connected it to the internet to finish all the installations I need to, my MOBO (Gigabyte) installed some software to manage the bloody RBG settings among other things. A friend recommended I uninstall the settings completely and then reinstall my GPU drivers. After a quick reset, everything worked just fine.

What I can recommend from my own experience, is do what RanXerox recommended with the IGPU. And then in your control panel, find whatever programs affect RGB and other systems for your MOBO and remove them. For me, it was the Gigabyte Control Center for my MOBO, and I forget the name of the RGB program for my GPU, which is a Gigabyte Aorus Master 3070 ti. That is the best I can say on the matter. Do tell if that helps in any capacity.

[JF0603]

Originally posted by Varangia1122:

Hi, I do have an update.
My issue was fixed recently. It is technically an issue with drivers, but there was no need for me to revert them on my end.

As this is a new PC that I built, once I connected it to the internet to finish all the installations I need to, my MOBO (Gigabyte) installed some software to manage the bloody RBG settings among other things. A friend recommended I uninstall the settings completely and then reinstall my GPU drivers. After a quick reset, everything worked just fine.

What I can recommend from my own experience, is do what RanXerox recommended with the IGPU. And then in your control panel, find whatever programs affect RGB and other systems for your MOBO and remove them. For me, it was the Gigabyte Control Center for my MOBO, and I forget the name of the RGB program for my GPU, which is a Gigabyte Aorus Master 3070 ti. That is the best I can say on the matter. Do tell if that helps in any capacity.

we have very similar setups i am glad you got yours squared away, the one thing i hadnt consiidered though i had unistalled all gigabyte's control center related (RGB fusion etc) is the physical 3pin 5v RGB headers i have connected which is something that hadnt been prior connected since the last case and fans i ran had no mobo RGB sync connections. tbh i prefer the older noctua's i had but the cabling and spatial constraints were a mess in the much smaller than anticipated Lian-li O11d-E. also i figured i'd have a 4090 now, which i have not exactly been able to get- which would have forced me to isolate the cause with the ability to rule in/out the gpu, while needing to de-case again... it might also be the usb headers is what i am fearing on the mobo which the AIO i have is using for control and also seems to have a weird affinity for periodic gpu usage...