Overclock.net › Forums › Components › Mice › I FINALLY discovered the right combination of settings & drivers for perfect mouse response in Win 7
New Posts  All Forums:Forum Nav:

I FINALLY discovered the right combination of settings & drivers for perfect mouse response in Win 7

post #1 of 54
Thread Starter 
If you've seen any of my recent posts, you might have seen where I talked about different Nvidia drivers varying in mouse input lag, both on desktop, and in-game. I have a background of doing 3d and CAD, clicking thousands of tiny verts and dragging them around all day. If the mouse doesn't handle perfectly, you'll go insane pretty fast doing it.

Through my subjective testing with every Fermi driver released from the last year, 295.73 WHQL comes out on top in terms of mouse response. I consider 296.10 to be the last Fermi era driver and while it's "OK", it has slightly higher input lag (although not as bad as ones like 275.33). 296.10 also doesn't down clock properly either and will sit at full 3d clocks while browsing web pages a lot of the time, heating your house up to Venus level temperatures in the process.

As for 300 series drivers, each one I have tested (up to 301.40), is SIGNIFICANTLY worse than 295.73 WHQL in input lag. Even though they're "unified" drivers, I would avoid anything higher than 295.73 right now until this situation changes, if ever. You also need to combine that driver with the settings below.


1) Install 295.73 driver

2) Change the control panel settings to the following settings and reboot: (reboot is mandatory). I've done extensive testing with all of these settings and things like anti-aliasing DO add input lag regardless of what some mongoloid with quad SLI on Guru3d will tell you. All single GPUs should be set to prerender of 1, never 0. I believe SLI people are supposed to use a minimum of 2 to avoid issues, but I'm not an SLI expert.

412

3) Change Nvidia scaling to "Display - No Scaling"
This setting makes a large difference in input lag for me. Nvidia, in their infinite wisdom, makes it so even while running your monitor in native resolution, makes it so scaling lag STILL APPLIES unless you have this setting completely disabled in control panel. Maybe some LCD aren't affected by this, but mine sure as hell is.
392

4) Change to Windows 7 Basic UI:
The most accurate in terms of what mouse response you see on desktop versus in-game.

5) Go to Control Panel --> Performance Information & Tools ---> Advanced Tools ---> Clear all Windows Experience Index Scores and re-rate the System: I've noticed not having run this after a new driver install can affect desktop response.

6) Disable HPET (High Precision Event Timer) in BIOS - mostly used to sync video frames better but increases overall system latency

7) Control Panel - Power Plan: The last one may or may not make a difference, but if you're running a desktop and not a laptop, you might as well change it to be safe

443
Edited by r0ach - 6/1/12 at 1:42am
post #2 of 54
i have heard some claim that 0 pre-rendered frames is better when you have more fps than your refresh rate wackosmiley.gif
post #3 of 54
Good that progamers are no where near this picky about stuff or tournaments would be delayed for hours biggrin.gif

I stopped doing crap like mousefixes and junk years ago, no point imo.
I go by the "Can progamers do without so can I"
post #4 of 54
Thread Starter 
Quote:
Originally Posted by deewe View Post

i have heard some claim that 0 pre-rendered frames is better when you have more fps than your refresh rate wackosmiley.gif

0 isn't a good setting. It will work fine in some games, others will stutter, and some games will interpret 0 as invalid and use a default of 3 instead !
post #5 of 54
Quote:
Originally Posted by r0ach View Post

0 isn't a good setting. It will work fine in some games, others will stutter, and some games will interpret 0 as invalid and use a default of 3 instead !

Which games? I thought using 0 with NVIDIA works still.
Edited by Yahar - 12/27/12 at 8:44pm
post #6 of 54
Quote:
Originally Posted by r0ach View Post

0 isn't a good setting. It will work fine in some games, others will stutter, and some games will interpret 0 as invalid and use a default of 3 instead !

I would like source on this one, since ive been on max prerendered 0 for the past 8 years or more, and havent had a single issue with it.
post #7 of 54
Tests done for prerendered frame values:
http://forums.laptopvideo2go.com/topic/29333-max-pre-rendered-frames-benchmarks/

I've run similar tests with my AMD 6950. You don't need to wait for an entire benchmark to finish to see the results. Simply look at the gpu usage. With no buffering/faffing about behind the scenes, your card should never run at full load for extended periods. This is why ATI and nVidia introduced the feature - to fool benchmarks.

Heaven @ 1680x1050 4xaa 4xaf
Flip Queue Size = GPU Usage
0 = 95-99%
1 = 91-93%
2 = 95-99%
3 = 95-99%
4 = 95-99%
5 = 95-99%

Heaven @ 640x480 0xaa 0xaf
Flip Queue Size = GPU Usage
0 = 78-87%
1 = 53-60%
2 = 78-87%
3 = 78-87%
4 = 78-87%
5 = 78-87%

As my results seem to show, there are only 2 values of Flip Queue Size. A value of 1 is off and everything else is on.
My System
(17 items)
 
  
CPUMotherboardGraphicsRAM
3570k @ 4.4ghz Asus Z77 Deluxe GTX970 SLI 24GB 
MonitorKeyboardPowerCase
Acer XB270HU CM QuickFire Red - PBT white keycaps Seasonic X-850 Raven RV02 
MouseMouse PadAudioOther
Roccat Savu Zowie P-TF Onkyo SE-90 Marantz PM7003 amplifier 
OtherOther
Wharfedale 10.1 speakers Sennheiser HD598 
  hide details  
Reply
My System
(17 items)
 
  
CPUMotherboardGraphicsRAM
3570k @ 4.4ghz Asus Z77 Deluxe GTX970 SLI 24GB 
MonitorKeyboardPowerCase
Acer XB270HU CM QuickFire Red - PBT white keycaps Seasonic X-850 Raven RV02 
MouseMouse PadAudioOther
Roccat Savu Zowie P-TF Onkyo SE-90 Marantz PM7003 amplifier 
OtherOther
Wharfedale 10.1 speakers Sennheiser HD598 
  hide details  
Reply
post #8 of 54
Thread Starter 
Quote:
Originally Posted by whybother View Post


As my results seem to show, there are only 2 values of Flip Queue Size. A value of 1 is off and everything else is on.

I don't understand what you mean by this. Your results don't show that there's only two settings for prerender, it shows that 0 defaults to 3 and that the lowest setting "1", results in GPU utilization hit on your ATI card.
post #9 of 54
My results shown above are for AMD. The link detailing prerender settings for NVIDIA is not mine. Flip Queue Size = AMD.

As for GPU utilization "hit", that's the whole point of this exercise. Less input lag means running without buffering behind the scenes to artificially increase fps (which in turn increases gpu utilization).
Edited by whybother - 4/19/12 at 6:06am
My System
(17 items)
 
  
CPUMotherboardGraphicsRAM
3570k @ 4.4ghz Asus Z77 Deluxe GTX970 SLI 24GB 
MonitorKeyboardPowerCase
Acer XB270HU CM QuickFire Red - PBT white keycaps Seasonic X-850 Raven RV02 
MouseMouse PadAudioOther
Roccat Savu Zowie P-TF Onkyo SE-90 Marantz PM7003 amplifier 
OtherOther
Wharfedale 10.1 speakers Sennheiser HD598 
  hide details  
Reply
My System
(17 items)
 
  
CPUMotherboardGraphicsRAM
3570k @ 4.4ghz Asus Z77 Deluxe GTX970 SLI 24GB 
MonitorKeyboardPowerCase
Acer XB270HU CM QuickFire Red - PBT white keycaps Seasonic X-850 Raven RV02 
MouseMouse PadAudioOther
Roccat Savu Zowie P-TF Onkyo SE-90 Marantz PM7003 amplifier 
OtherOther
Wharfedale 10.1 speakers Sennheiser HD598 
  hide details  
Reply
post #10 of 54
Quote:
Originally Posted by Nivity View Post

Good that progamers are no where near this picky about stuff or tournaments would be delayed for hours biggrin.gif

Alot of them just know their settings by heart so setting up at a stage pc goes rather quickly. I remember some belgian LANS as well, I just had a USB stick with some basic stuff like registry tweaks and what not. Double click all the stuff, auto apply my tweaks and reboot and I was set to play.

Even my autoexec.cfg from games had everything non default in, including binds and gfx settings so just dropping it in the cfg folder was enough.

BTW, roach, disable Core Parking as well, it can also cause some minor input lag on quad or more cores:
http://forum.cakewalk.com/tm.aspx?m=1861804

Also experiment with USB 2.0 and 3.0 it's kind of a subjective thing but there's a small difference, up to you to decide which you prefer.
Edited by CorruptBE - 4/19/12 at 7:00am
Desktop (Main)
(21 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 5820K ASUS X99-A/USB 3.1  ASUS ROG Strix GeForce GTX 1080 Kingston HyperX 16GB 3Ghz  
Hard DriveHard DriveHard DriveHard Drive
Samsung 840 Pro Seagate Barracuda 1.5 TB Seagate Barracuda 2.0 TB Seagate Barracuda 4TB 
Hard DriveOptical DriveCoolingOS
Seagate Barracuda 4TB Le random dvd writer Corsair Hydro Series™ H100i Windows 10 Professional 64 bit 
MonitorKeyboardPowerCase
iiyama ProLite G2773HS Corsair K70 Super Flower 1200W (Platinum) Corsair 600T Black 
MouseMouse PadAudioAudio
Nixeus Revel Allsop Raindrop XL ASUS Xonar Essence STXII Sennheiser HD650 
Audio
Musical Fidelity X-Can v3 
  hide details  
Reply
Desktop (Main)
(21 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 5820K ASUS X99-A/USB 3.1  ASUS ROG Strix GeForce GTX 1080 Kingston HyperX 16GB 3Ghz  
Hard DriveHard DriveHard DriveHard Drive
Samsung 840 Pro Seagate Barracuda 1.5 TB Seagate Barracuda 2.0 TB Seagate Barracuda 4TB 
Hard DriveOptical DriveCoolingOS
Seagate Barracuda 4TB Le random dvd writer Corsair Hydro Series™ H100i Windows 10 Professional 64 bit 
MonitorKeyboardPowerCase
iiyama ProLite G2773HS Corsair K70 Super Flower 1200W (Platinum) Corsair 600T Black 
MouseMouse PadAudioAudio
Nixeus Revel Allsop Raindrop XL ASUS Xonar Essence STXII Sennheiser HD650 
Audio
Musical Fidelity X-Can v3 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Mice
  • I FINALLY discovered the right combination of settings & drivers for perfect mouse response in Win 7
Overclock.net › Forums › Components › Mice › I FINALLY discovered the right combination of settings & drivers for perfect mouse response in Win 7