USB polling precision - Page 48 - Overclock.net - An Overclocking Community
Forum Jump: 

USB polling precision

Reply
 
Thread Tools
post #471 of 1083 (permalink) Old 10-11-2016, 03:46 AM
OCN is bad civilization
 
Melan's Avatar
 
Join Date: Feb 2015
Location: 0,0
Posts: 3,293
Rep: 112 (Unique: 86)
Leave it at default. On in BIOS, off in OS.
Melan is offline  
Sponsored Links
Advertisement
 
post #472 of 1083 (permalink) Old 10-11-2016, 04:14 AM
New to Overclock.net
 
SweetLow's Avatar
 
Join Date: Jan 2016
Posts: 487
Rep: 55 (Unique: 31)
As i read this thread (and other same threads) .
People, why do you think, that enabling or disabling HPET, changing "UsePlatformClock" or "/usepmtimer" (actually in this case you simply select different real hardware sources for abstract OS timer) make POLLING more precise?
SweetLow is offline  
post #473 of 1083 (permalink) Old 10-11-2016, 10:52 AM
New to Overclock.net
 
PLUSPUNKT's Avatar
 
Join Date: Oct 2016
Posts: 2
Rep: 0
Some drops, but looks good so far
Logitech G303 + W10

PLUSPUNKT is offline  
Sponsored Links
Advertisement
 
post #474 of 1083 (permalink) Old 10-11-2016, 03:54 PM
OCN is bad civilization
 
Melan's Avatar
 
Join Date: Feb 2015
Location: 0,0
Posts: 3,293
Rep: 112 (Unique: 86)
Those are xCounts not Update time.
Melan is offline  
post #475 of 1083 (permalink) Old 10-11-2016, 10:24 PM
New to Overclock.net
 
VolsAndJezuz's Avatar
 
Join Date: Nov 2014
Posts: 779
Rep: 95 (Unique: 53)
Quote:
Originally Posted by SweetLow View Post

As i read this thread (and other same threads) .
People, why do you think, that enabling or disabling HPET, changing "UsePlatformClock" or "/usepmtimer" (actually in this case you simply select different real hardware sources for abstract OS timer) make POLLING more precise?

https://www.overclock.net/t/1609933/terminology-pet-peeves-thread-misnomers-that-annoy-you/0_20#post_25472818

Referring to this stuff as 'USB polling' is just much easier than saying something like 'Operating System induced jitter relative to USB report rate'. Just think of it as colloquial use of the term.

sourceGL software
CS:GO and TF2 sensitivity - 1 @ 850DPI, 48.9cm per 360°
baby boy
(22 items)
CPU
Intel Core i7-4790K @ 4.8GHz - de-lidded (w/ ROCKIT COOL Copper IHS)
Motherboard
Gigabyte GA-Z97X-UD5H-BK
GPU
EVGA GTX 980 Ti Classified @ 1532MHz core, 4192MHz memory
RAM
OCZ Reaper HPC 2GB [x4] @ 2400MHz, 8-12-8-28 1T
Hard Drive
Samsung 850 PRO 256GB [x2]
Optical Drive
Lite-On DL-8A4SH
Power Supply
EVGA Supernova PS 1000W
Cooling
Cooler Master Nepton 140XL [x2] (w/ EK-Vardar F3-140ER fans)
Cooling
Thermal Grizzly Conductonaut liquid metal thermal compound
Cooling
Fujipoly ModRight Extreme thermal pads
Case
SilverStone Fortress FT05B
Operating System
Windows 7 Ultimate 64-bit (w/ NVIDIA Driver 399.24)
Monitor
BenQ XL2411Z 24" @ 144Hz
Keyboard
Das Keyboard Model S Ultimate (w/ MX Green modifiers and PBT keycaps)
Mouse
Logitech G900 - de-braided and de-battery'd (w/ Hotline Games Master feet)
Mousepad
Logitech G640 3mm
Audio
ASUS Xonar Essence STX II (w/ UNi modded driver)
Audio
beyerdynamic DT 880 PRO headphones (w/ AntLion ModMic 4.0)
Audio
Blue Sky eXo 160W 2.1 studio monitors
Other
Solarflare SFN6122F network adapter
Other
MikroTik RouterBoard RB751G-2HnD
Other
CyberPower 1350AVR UPS
▲ hide details ▲
VolsAndJezuz is offline  
post #476 of 1083 (permalink) Old 10-12-2016, 02:03 AM
New to Overclock.net
 
SweetLow's Avatar
 
Join Date: Jan 2016
Posts: 487
Rep: 55 (Unique: 31)
Quote:
Originally Posted by VolsAndJezuz View Post

https://www.overclock.net/t/1609933/terminology-pet-peeves-thread-misnomers-that-annoy-you/0_20#post_25472818

Referring to this stuff as 'USB polling' is just much easier than saying something like 'Operating System induced jitter relative to USB report rate'. Just think of it as colloquial use of the term.
It's all true, but i told about some other stuff. In your terms i say "playing with some parameters don't change (real) OS introduced jitter at all" wink.gif
SweetLow is offline  
post #477 of 1083 (permalink) Old 10-12-2016, 02:09 PM - Thread Starter
New to Overclock.net
 
HAGGARD's Avatar
 
Join Date: Jun 2013
Posts: 657
Rep: 56 (Unique: 44)
Quote:
Originally Posted by SweetLow View Post

It's all true, but i told about some other stuff. In your terms i say "playing with some parameters don't change (real) OS introduced jitter at all" wink.gif
Well, most people assume those parameters make a difference because the measurements change (see for example most recently post #469).

I'm not really sure myself whether, and how "enabling or disabling HPET, changing "UsePlatformClock" or "/usepmtimer"" significantly affects polling precision - differences in measurements between those are within negligible range on my setup. If you however have that knowledge, please do share.
HAGGARD is offline  
post #478 of 1083 (permalink) Old 10-12-2016, 02:58 PM
New to Overclock.net
 
SweetLow's Avatar
 
Join Date: Jan 2016
Posts: 487
Rep: 55 (Unique: 31)
Quote:
Originally Posted by HAGGARD View Post

Well, most people assume those parameters make a difference because the measurements change (see for example most recently post #469).

I'm not really sure myself whether, and how "enabling or disabling HPET, changing "UsePlatformClock" or "/usepmtimer"" significantly affects polling precision - differences in measurements between those are within negligible range on my setup. If you however have that knowledge, please do share.
It's obvious wink.gif You use keyword "measurement" two times but ignore it.
When we profile "jitter" we don't have only one process - getting messages from hardware to software. Second process (running on the same PC, not standalone hardware) - it's measurement itself. And it have its own precision, "jitter" and its own affecting settings (which i mentioned in my first post).
P.S. AFAIK, this thread may be precisely named "Optimization of reactivity of soft real time task running on non real time OS" biggrin.gif
SweetLow is offline  
post #479 of 1083 (permalink) Old 10-12-2016, 03:12 PM - Thread Starter
New to Overclock.net
 
HAGGARD's Avatar
 
Join Date: Jun 2013
Posts: 657
Rep: 56 (Unique: 44)
Quote:
Originally Posted by SweetLow View Post

It's obvious wink.gif You use keyword "measurement" two times but ignore it.
When we profile "jitter" we don't have only one process - getting messages from hardware to software. Second process (running on the same PC, not standalone hardware) - it's measurement itself. And it have its own precision, "jitter" and its own affecting settings (which i mentioned in my first post).
Very true. Better results are still better results, are they not? Less jitter is still less jitter. How are you to know where that jitter has its root - OS handling of USB reports or handling of the software used to make measurements? Again, if you do have that knowledge, I'd be glad to hear all about it. And besides, it would still be an optimization either way, even if not strictly related to input handling.
Quote:
P.S. AFAIK, this thread may be precisely named "Optimization of reactivity of soft real time task running on non real time OS" biggrin.gif
True. Not quite as catchy though! ;D
HAGGARD is offline  
post #480 of 1083 (permalink) Old 10-12-2016, 10:54 PM
New to Overclock.net
 
VolsAndJezuz's Avatar
 
Join Date: Nov 2014
Posts: 779
Rep: 95 (Unique: 53)
I think that optimizations that improve USB polling graphs (or whatever you want to call it) in MouseTester are a good way to measure very small changes to how the OS (or user mode programs in the OS, if you prefer) can handle very rapid and particularly timed function calls. A single optimization that shows a small but statistically significant change in MouseTester might not really be reflected by a difference in a game's FPS. But when you compound the effect of many good optimizations, there is a staggering difference in performance. With my overclocking and optimizations, I get higher FPS in CS:GO FPS benchmark and fps_test3 runs than I've even seen claimed by anyone else, including friends with GTX 1080s and much more expensive CPUs than me who I asked to run those tests. And it is smooth as butter, like my brain is directly controlling my aim in game and my arm and the mouse aren't even being used.
Quote:
Originally Posted by SweetLow View Post

It's obvious wink.gif You use keyword "measurement" two times but ignore it.
When we profile "jitter" we don't have only one process - getting messages from hardware to software. Second process (running on the same PC, not standalone hardware) - it's measurement itself. And it have its own precision, "jitter" and its own affecting settings (which i mentioned in my first post).
P.S. AFAIK, this thread may be precisely named "Optimization of reactivity of soft real time task running on non real time OS" biggrin.gif

To answer your concern about what exactly we are measuring with these polling precision graphs, I would ask the question: what is a game? A 'second process' that has its own precision, jitter, and settings, and collects USB mouse data much the same way a program like MouseTester would go about collecting mouse data. So the assumption here is that observing changes in MouseTester plots represents changes in in-game performance.

I really don't see how you can argue that point. The only real (and perfectly reasonable) argument is on what scale we as humans can detect the changes we are observing. Can we detect a consistent 500us jitter, like you see with 1000Hz polling on horribly optimized and outdated systems? Almost certainly. Can we detect repeating 1us jitter, like the difference in 0.999ms and 1.000ms update interval? Almost certainly not. Maybe the most sensitive of us can detect something like repeating 10us jitters. Maybe for most people it is something like 100us. I don't know but the point is that we can detect polling precision on some scale, so optimizing it is probably a worthwhile endeavor.

sourceGL software
CS:GO and TF2 sensitivity - 1 @ 850DPI, 48.9cm per 360°
baby boy
(22 items)
CPU
Intel Core i7-4790K @ 4.8GHz - de-lidded (w/ ROCKIT COOL Copper IHS)
Motherboard
Gigabyte GA-Z97X-UD5H-BK
GPU
EVGA GTX 980 Ti Classified @ 1532MHz core, 4192MHz memory
RAM
OCZ Reaper HPC 2GB [x4] @ 2400MHz, 8-12-8-28 1T
Hard Drive
Samsung 850 PRO 256GB [x2]
Optical Drive
Lite-On DL-8A4SH
Power Supply
EVGA Supernova PS 1000W
Cooling
Cooler Master Nepton 140XL [x2] (w/ EK-Vardar F3-140ER fans)
Cooling
Thermal Grizzly Conductonaut liquid metal thermal compound
Cooling
Fujipoly ModRight Extreme thermal pads
Case
SilverStone Fortress FT05B
Operating System
Windows 7 Ultimate 64-bit (w/ NVIDIA Driver 399.24)
Monitor
BenQ XL2411Z 24" @ 144Hz
Keyboard
Das Keyboard Model S Ultimate (w/ MX Green modifiers and PBT keycaps)
Mouse
Logitech G900 - de-braided and de-battery'd (w/ Hotline Games Master feet)
Mousepad
Logitech G640 3mm
Audio
ASUS Xonar Essence STX II (w/ UNi modded driver)
Audio
beyerdynamic DT 880 PRO headphones (w/ AntLion ModMic 4.0)
Audio
Blue Sky eXo 160W 2.1 studio monitors
Other
Solarflare SFN6122F network adapter
Other
MikroTik RouterBoard RB751G-2HnD
Other
CyberPower 1350AVR UPS
▲ hide details ▲
VolsAndJezuz is offline  
Reply

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off