What's a good DPI for a mouse?

[Braiboo99]

I just bought a new Corsair gaming mouse and trying to work out what is a suitable DPI, any tips?

[r.linder]

You can prove us right by simply setting your display to 60Hz, and then setting polling rate of your mouse to 125Hz (which is the MINIMUM setting for polling rate) and moving the mouse around before changing polling rate back to at least 500Hz, you'll see the freaking difference because 125Hz has a crap ton of latency, 8ms, compared to 1ms of 1000Hz.

[_I_]

the diff between 500hz and 1000hz polling rate is only 0 to 1 ms max
some cases it will be no difference, and the biggest it will be 1ms more

not a huge deal when the display is refreshing every 16ms on a 60hz display
165hz is about 6ms, it will get 2-3 updates at 500hz
240hz is about 4ms, 1-2 updates at 500hz

as long as it is getting one or more its good imho
the next frame takes longer to create and cursor will move or display will update accordingly

every edge helps with 'gamers' so having it in a different position or slightly faster could help, but not always

but marketing, they want the biggest or smallest number they can put on the buzzword tag

[PopinFRESH]

Publicado originalmente por Jamebonds1:

Publicado originalmente por PopinFRESH:

That is because you are looking at your monitor... you need to learn how to follow instructions. Get better glasses and then make sure you STARE REALLY HARD are your mouses USB CABLE. Then you'll be able to see the 1000Hz polling.

It is impossible on 60 Hz and 240 Hz monitors. Even if you have 1000 Hz polling rate mouse, you can only see what is on the display refresh rate which is either 60 Hz or 240 Hz polling rate on display.

AGAIN READ... You are looking at your monitor. STOP looking at your monitor. You should be looking at your MOUSE USB CABLE so you can see the mouses 1000Hz polling rate. You must need to go to an eye doctor because you seem like you can only see the world in 60Hz.

My eyes have a polling rate for 8000MHz so I'm able to see the mouses 1000Hz gigapoller multi-duplexed double hashed inverter speed. It makes my monitor work 4 times its 144Hz "polling" rate. 🤡

[r.linder]

Publicado originalmente por Jamebonds1:

Publicado originalmente por 尺．し工几ᗪヨ尺:

You can prove us right by simply setting your display to 60Hz, and then setting polling rate of your mouse to 125Hz (which is the MINIMUM setting for polling rate) and moving the mouse around before changing polling rate back to at least 500Hz, you'll see the freaking difference because 125Hz has a crap ton of latency.

I can still see only 60 Hz polling rate on my 60 Hz monitor from my 1000 Hz polling rate mouse.

That's not how it works, the minimum setting that polling rate can actually be is 125Hz and there is a visual difference when you switch between 125Hz and 500Hz+.

A 60Hz polling rate on mice isn't possible.

https://devicetests.com/mouse-rate-test

[_I_]

yes you only see the results of one poll every refresh, but having that one from a sooner makes a difference
it may be tiny, or almost none

[PopinFRESH]

Publicado originalmente por 尺．し工几ᗪヨ尺:

Publicado originalmente por Jamebonds1:

I can still see only 60 Hz polling rate on my 60 Hz monitor from my 1000 Hz polling rate mouse.

That's not how it works, the minimum setting that polling rate can actually be is 125Hz and there is a visual difference when you switch between 125Hz and 500Hz+.

A 60Hz polling rate on mice isn't possible.

https://devicetests.com/mouse-rate-test

Yeah, but can you hear the color blue or feel the smell of lasagna? My eye doctor said I could because I have 12THz polling rate tastebuds.

[cSg|mc-Hotsauce]

Publicado originalmente por PopinFRESH:

Yeah, but can you hear the color blue or feel the smell of lasagna?

Synesthesia

https://www.livescience.com/61930-synesthesia-hear-colors-genes.html

[r.linder]

Publicado originalmente por Jamebonds1:

Publicado originalmente por 尺．し工几ᗪヨ尺:

That's not how it works, the minimum setting that polling rate can actually be is 125Hz and there is a visual difference when you switch between 125Hz and 500Hz+.

A 60Hz polling rate on mice isn't possible.

https://devicetests.com/mouse-rate-test

It is right there. That is why you won't see 1000 Hz on the 60 Hz monitor.

https://images.app.goo.gl/Ca65fwfDMs5msgBv8

FPS and polling rate are completely different things.

[Supafly]

Shocked no one has just blocked the user everyone knows is talking rubbish. I know who as his blocked but I see who when you lot quote it.... only be talking rubbish for half the thread.

No point providing proof of any sort because he won't even look

[Zefar]

The way I do when I look for mouses is to look for one with high DPI. As I'm sure it means it has a good sensor.

Then I stick the DPI to about 800-850-950-1100-1200 that I can switch between.

The theory I have is that the better sensor is able to report the lower DPI better and making it feel smoother.

[Rod]

Publicado originalmente por Piston Smashed™:

Publicado originalmente por Rod:

Battlenonsense youtube channel. Theres gains up to 3200dpi typically 1ms over say 400dpi. After 3200dpi he said theres no gains. To prove him wrong, You need some serious equipment.

The DPI is just the speed at which you track across the screen, the higher the DPI the faster it allows you to move across the screen. Having a 3200 DPI over a 400 DPI mouse does not increase or decrease the input time of a mouse.

It is the polling rate and response times of a mouse which would make the sort of differences between mice that you are suggesting not the DPI.

My Corsair Dark Pro SE RGB Wireless
800 DPI (my setting)
2000 Hz Polling Rate
0.5 ms Response Time

Changing the DPI to 3200 or anything up to its max of 18,000 does not change its input time. Now if I go in to the mouse settings in iCue and lower the polling rate to 1000 Hz it changes the response time to 1 ms, 500 Hz changes the response time to 2 ms and so on thus making the inputs of my mouse slower.

Another random guy on the internet trying to prove a well established YT tester. He tested it and you are wrong.

[A&A]

Publicado originalmente por PopinFRESH:

Not how it works at all.

https://www.igorslab.de/wp-content/uploads/2020/09/Latency-Scheme-1.jpg
(The above diagram is from the NVIDIA LDAT documentation on measuring total system latency)

end-to-end total system latency (e.g. "click to photon", or in other words the total time it takes from when you "click" and actuate the switch on your input device to when the first "photon" change occurs, e.g. pixel changes on your display) encompasses more than just one thing; and each of those different things are independently additive to the total system latency.

Referencing the above diagram; The USB peripheral polling rate will impact both the "peripheral latency" and the first block, USB SW (Software) in the CPU (OS/Game) portion of "PC Latency".

This is, again, additive to the other things that you guys are talking about; which would be those "game latency", "render latency", and "display latency". As Piston Smashed™ noted a 500Hz polling rate will add 2ms of input latency to whatever other latency is being induced by your CPU, game, GPU renderer & compositing, and your display. None of those other things will impact or change that additional 2ms of latency from your mouse/keyboard. Whereas a 1000Hz polling rate will only add 1ms of additional latency; and a 2000Hz polling rate on newer mice/keyboards will only add 0.5ms of additional latency.

A game being "CPU bound" wouldn't change the peripheral latency at all. It would extend / add latency to the "game latency"; specifically in the "render submission". The relationship between those portions of total system latency, game latency and render latency, is what the gpubusy metric in Intel's PresentMon exposes so you can see the balance between game latency (being impacted by your CPU) and render latency (by how busy your GPU is being kept).

https://game.intel.com/story/intel-presentmon/

Everything you said is indeed true. However, it's important to note that increasing the pooling rate of the mouse will lead to more interrupt calls, which is expected. The issue arises when most of these interrupts from peripheral devices are handled by a single core. As a result, increasing the interrupt rate can cause delays for the rest, negatively impacting performance. Quad-core processors are more affected than hexa-core processors by a lot.

[Piston Smashed™]

Publicado originalmente por Rod:

Publicado originalmente por Piston Smashed™:

The DPI is just the speed at which you track across the screen, the higher the DPI the faster it allows you to move across the screen. Having a 3200 DPI over a 400 DPI mouse does not increase or decrease the input time of a mouse.

It is the polling rate and response times of a mouse which would make the sort of differences between mice that you are suggesting not the DPI.

My Corsair Dark Pro SE RGB Wireless
800 DPI (my setting)
2000 Hz Polling Rate
0.5 ms Response Time

Changing the DPI to 3200 or anything up to its max of 18,000 does not change its input time. Now if I go in to the mouse settings in iCue and lower the polling rate to 1000 Hz it changes the response time to 1 ms, 500 Hz changes the response time to 2 ms and so on thus making the inputs of my mouse slower.

Another random guy on the internet trying to prove a well established YT tester. He tested it and you are wrong.

I once read they found a double-decker bus on the moon...

[Phazar]

I personally use between 1200-2000 DPI. I just upgraded to a 32 inch 2K and am now using 2000 DPI.

[Admiral1215]

Set mine to: 800/1000/1200/1600 But 1200 seem to be the best of all