Overclock.net banner
Cancel
Create thread New Forums

Consensus on 500 Hz vs 1000 Hz polling

Jump to Latest
1 - 15 of 15 Posts

shapul

· Registered
Joined
·
7 Posts
Discussion starter · #1 ·
Hi,

I was wondering if there is any consensus on 1000 Hz vs 500 Hz polling rate? Is there any study or objective way that demonstrate there is any benefit in using 1000 Hz polling rate? Is it just marketing hype or is there any solid evidence behind it?

Technically, I have problem figuring out how 1000 Hz polling rate would be any different at all from 500 Hz, even at the age of 144 Hz monitors or G-Sync. In other words, if you believe that there is a difference between 500 and 1000 Hz, what is the explanation behind it?

The reason that I ask the question is that a while ago I was working on a software library to capture USB events. I noticed that some computers really struggle to receive 1000 Hz packets on USB without dropping one or two every few seconds. Dropped packets are not noticeable from user's point of view since the mouse hardware compensate that by including the missing counts in the next packet. However, this observation was telling me that on these computers, using 1000 Hz polling rate is not really optimal.

I for one cannot notice the difference between 500 and 1000 Hz but well, I am not a pro and possibly I don't push the games that far.

Cheers,
Shapul
 
#2 ·
well at the end of the day 125 Hz is 8 ms between the updates and 500 Hz is 2ms while 1000 Hz is 1 ms.

I doubt you can really tell as humans have a reaction time between 100-300 ms anyway

But on many mice 500 Hz has been much more stable in the past so due to that fact many have avoided 1000 Hz
 
Discussion starter · #4 ·
Quote:
Originally Posted by maximdymok View Post

The only real advantage of 1000hz over 500hz is less microstutters, http://www.blurbusters.com/mouse-125hz-vs-500hz-vs-1000hz/
Click to expand...
Very interesting. I have recently got a G-Sync capable monitor. I guess I'll need to repeat these test to see it for myself. If observing micro-stuttering at 500 Hz turns to be a repeatable process, that could be an objective argument for the 1000 Hz for high end machines.
 
#5 ·
You can use a program like MouseMovementRecorder to monitor the polling rate consistency at high speed to make sure your computer can consistently poll at 1000Hz, and also to monitor how well your computer and mouse handle fast deceleration, where it should go down from 1000Hz towards 0 in a fairly even manner (i.e., it shouldn't bounce around between say 1000Hz and 250Hz several times during a rapid deceleration). If your computer can handle it, there's really not a good reason to not use 1000Hz. Though it may not really matter unless you are operating on a 100+Hz refresh rate monitor.
 
#6 ·
I doubt humans can feel the difference between 4GHz and 4.1GHz cpu, but still they (we) OC till there is room left
wink.gif

1ms updates are technically better and benefit high frame and game simulation rates. Everything in a computer runs at a frequency, the higher the closer we get to real time.

Said that, the nature of polling rate is dynamic, the faster you move the mouse the higher it goes. For that reason 500Hz will have a more consistent latency across more variance in mouse movement speed.

Hardware limitations aside that is the only reason for using 500Hz, being used to (or just liking) the feel, as it is worse from a technical standpoint. It is to say that a pc built for gaming these days (or even some years ago) should not remotely have problems with 1000Hz mouses.
 
#7 ·
^Polling rate doesn't change based on the speed of the mouse AFAIK. You get lower polling rate measurements while moving the mouse slowly because the programs measure it by checking for new data every second - if you're not moving your mouse fast you aren't producing enough data to "fill" every poll, not that the mouse isn't being polled.
 
#8 ·
Quote:
Originally Posted by Hackshot View Post

^Polling rate doesn't change based on the speed of the mouse AFAIK. You get lower polling rate measurements while moving the mouse slowly because the programs measure it by checking for new data every second - if you're not moving your mouse fast you aren't producing enough data to "fill" every poll, not that the mouse isn't being polled.
Click to expand...
Yes polling it self is constant, data from the mouse is not sent every millisecond (or 2) when moving slowly as you don't cover enough distance to register a position change every polling period.
Mouse frame rate is dynamic and based on movement speed.

I believe measuring software has higher temporal resolution than 1sec..
 
#10 ·
Quote:
Originally Posted by ghostlacuna View Post

well at the end of the day 125 Hz is 8 ms between the updates and 500 Hz is 2ms while 1000 Hz is 1 ms.

I doubt you can really tell as humans have a reaction time between 100-300 ms anyway

But on many mice 500 Hz has been much more stable in the past so due to that fact many have avoided 1000 Hz
Click to expand...
This is like 'the eye can't see more than 30fps' argument.

Reaction time ≠ input lag.

1ms of input lag is noticeable. And also 1000Hz avoids micro-stutters, like was mentioned above. The problem's always been stability.
 
#11 ·
1000hz all the way!!



I play Warsow.

My settings.

144hz monitor, with FPS locked in at 144hz in game V-Sync OFF.

6 of 11 in windows mouse settings with mouse acceleration off ( also know as enhance pointer precession ) .

Razer Orochi, with 1000dpi at 1000hz.

I dont know if this affects it, but my cpu is at 5.1ghz.

Super smooth game play! Yes it makes a huge difference, I can easily tell the difference between 500hz and 1000hz while moving the mouse in game. I have multiple profiles for starcraft and warsow, and when I forget to switch over, i can tell right away in game.
 
#12 ·
Quote:
Originally Posted by Trull View Post

This is like 'the eye can't see more than 30fps' argument.

Reaction time ≠ input lag.

1ms of input lag is noticeable. And also 1000Hz avoids micro-stutters, like was mentioned above. The problem's always been stability.
Click to expand...
well not really since fighter pilots have shown that humans can identify things at at least 200 fps but i get what you mean trull.

My bad i was going to make a reference to how the change from 125 Hz to 500 Hz is clearly noticeable, while not everyone notice going from 500 Hz to 1000 Hz.

So yeah my post was quite bad.

I game at 1000 Hz but i have done tests with my friends who mostly play RTS and RPG games in the past and they have a hard time noticing the input lag. They do however notice stuttering and hate it.

But as you mention and i did as well stability was the main reason why people choose 500 Hz over 1000 Hz:)
 
#13 ·
I currently use Rapier V1 on 2 ms, but it's individual for reach mouse. Rapier felt better with 2ms and 2400 DPI, other mouses might work great at 1600 DPI and 1 ms. It depends on sensor and internals.

8 ms is very, very slow and it's for people with sensomothoric problems, and for these who lacked foresight and damaged theirs coordination by menial work.
 
Discussion starter · #14 ·
Quote:
Originally Posted by Trull View Post

Reaction time ≠ input lag..
Click to expand...
I have to agree with that. Reaction time is much more longer than the "noticeable delay". While 100~200 ms reaction time is well known and well studied, I don't know if there is a universally agreed upon "noticeable delay" time since it varies a lot depending the situation. On an unrelated and totally different subject, one could argue that the brain can perceive sub millisecond events: pops in recorded music can be as short as a single sample (at 44.1 kHz) and are still audible.

Here, the context is complicated since there are many variables involved. Screen refresh rate, gray to gray response time, USB polling rate, cursor movement speed, etc.

Originally I didn't expect that anyone could see the difference between 500 and 1000 Hz but the microstuttering argument is hard to dismiss.

I just wonder if there are any other ways to demonstrate the difference.
 
#15 ·
Quote:
Originally Posted by shapul View Post

I just wonder if there are any other ways to demonstrate the difference.
Click to expand...
Sure, get a copy of Quakeworld or any other FPS game that you can run a super high framerate on. Here's what I just tested.

144Hz with 1008fps cap (7x refresh rate), with vsync OFF
Find an in-game location where tearing is highly noticeable (eg: most of the wall lamps on DM3 have pure black on their sides, and extremely bright yellow/white on the front), and wiggle your mouse back and forth while looking at it.
You should see 7 tear lines across your screen. The tearing will be more noticeable at lower polling rates. Compared to 1000Hz, 500Hz will have twice the view angle offset under each tearline. 125Hz has 8 times more noticeable tearing.
 
1 - 15 of 15 Posts
About this Discussion
14 Replies
10 Participants
Last post:
povohat
Overclock.net
posts
28M
members
562K
Since
2004
The premier forum for overclocking experts and enthusiasts. Discuss hardware optimization, custom builds, benchmarking, cooling solutions, and pushing the boundaries of computing performance. From beginner guides to extreme overclocking, join our technical community to master system tuning.
Full Forum Listing
Explore Our Forums
Intel CPUs Hardware News NVIDIA PC Gaming Water Cooling