Overclock.net › Forums › Components › Mice › MouseTester Software Reloaded
New Posts  All Forums:Forum Nav:

MouseTester Software Reloaded - Page 4

post #31 of 72
Hey everyone! I'm new to OCN and I'm trying to optimize my Razer Imperator 4G a bit. I'll probably end up just buying a different mouse soon, but testing never hurts. Below I did a few basic tests, but ran into some issues.

If anyone could point me in the right direction on how to fix/interpret these issues I'd really appreciate it. For reference, my Imperator is set to 800 DPI. Can also post system specs if anyone would like me to.

xCounts:
I did an xCounts test and included the yCounts just in case anyone was interested. I did this twice just to be on the safe side.

First test


Second test


From what I understand, everything looks decent here.

xVelocity:
Next up, I did a velocity test to see how it handled quick speeds. I swiped pretty quickly, and ended up with some silly results. Again, I tested this twice.

First test

Points spiked to 100 m/s. No idea why this happened.

Second test

This time it spiked to 120 m/s. If my hand was really that fast, I'd probably become an MMA fighter. thumb.gif

It's hard to get useful data when these spikes happen, so that's a bit frustrating to see.

I've considered disabling my C-states in BIOS, but I read a lot on my computer when not gaming and don't want to raise my power bill too much from staring at text for 6 hours with my CPU on full blast. Will disabling C-states cause any meaningful changes in things like my PC hardware lifetime/power bills? I'd rather not tax my wallet or my computer if I can avoid it.

Anyways, we've still got one test left...

Intervals:
Finally, I made some small circles for a few seconds to check my polling rate. Mouse is currently set to 1000Hz.

First test, moving mouse quickly


Second test, moving mouse at a normal rate

For this one, are the spikes/scattered points at the beginning just junk data? If so, what would be a good ms to cut off the data at?

Not 100% sure, but my polling rate seems fine. I'm not sure what constitutes an unacceptable deviation — but I'd love to hear from everyone else.

So to cap this post, my questions are...

1) Is my xCounts tracking test done correctly? Are the results good?
2) Why am I getting huge spikes on my xVelocity tests?
3) Are the spikes on polling rate test #2 just junk?
4) Is my polling rate consistent? Are these results good as well?
5) How will C-states impact my hardware's lifetime and/or power bills?


I've been into this for some time, but mostly as a lurker. I'd like to learn a lot more, so you might see me around more regularly.

Thanks in advance for any help I get!
post #32 of 72
1. yes/yes
2. xvelocity = xcounts/interval. spikes in xvelocity correspond to intervals that come close to 0. see http://www.overclock.net/t/1550666/usb-polling-precision/0_100.
sometimes, it's for a much more innocuous reason: you have two mice plugged in, and as you swipe one, you accidentally move the other one by a pixel or something. this leads to two data points very close together, but from separate mice, but mousetester only recognizes this as a short interval.
3. hard to tell. sometimes, bad mouse firmware skips usb reports. it could also be your pc/os acting weird.
4. see haggard's thread linked previously
5. disabling some c states isn't the same as running your pc at full load all the time. i don't think the idle power from keeping a pc in c1 is something to worry about.
main
(15 items)
 
old
(14 items)
 
 
CPUMotherboardGraphicsRAM
4770k maximus vii impact nvidia gtx 970 crucial ballistix tactical 16gb 
Hard DriveCoolingOSMonitor
crucial mx100 noctua nh-c14 windows 7 ultimate sony cpd-g520 
KeyboardPowerCaseMouse
kbp v80 matias quiet silverstone sx500-lg ncase m1 v3 logitech g100s with mcu replaced by teensy2.0 
Mouse PadAudioAudio
allsop raindrop xl chord mojo hifiman re-600 
CPUMotherboardGraphicsRAM
i7 920 evga x58 sli le galaxy gtx 460 crucial something 3x1gb 
Hard DriveCoolingOSMonitor
intel 330 180gb scythe kotetsu windows 8.1 pro sony cpd-g520 
KeyboardPowerCaseMouse
logitech k120 silverstone st75f-gs nxzt h440 evga torq x5 
Mouse PadAudio
allsop raindrop mobo 
  hide details  
Reply
main
(15 items)
 
old
(14 items)
 
 
CPUMotherboardGraphicsRAM
4770k maximus vii impact nvidia gtx 970 crucial ballistix tactical 16gb 
Hard DriveCoolingOSMonitor
crucial mx100 noctua nh-c14 windows 7 ultimate sony cpd-g520 
KeyboardPowerCaseMouse
kbp v80 matias quiet silverstone sx500-lg ncase m1 v3 logitech g100s with mcu replaced by teensy2.0 
Mouse PadAudioAudio
allsop raindrop xl chord mojo hifiman re-600 
CPUMotherboardGraphicsRAM
i7 920 evga x58 sli le galaxy gtx 460 crucial something 3x1gb 
Hard DriveCoolingOSMonitor
intel 330 180gb scythe kotetsu windows 8.1 pro sony cpd-g520 
KeyboardPowerCaseMouse
logitech k120 silverstone st75f-gs nxzt h440 evga torq x5 
Mouse PadAudio
allsop raindrop mobo 
  hide details  
Reply
post #33 of 72
How to compare two mice in terms of input lag? It's been a while since I've done this.
Dual device, set CPI for both, hold them together and move both at once while collecting? Then check X/Y?
I'm finding that the angle and precise CPI they are at makes the curves for both a little different.
Edited by JackCY - 8/3/16 at 9:54am
post #34 of 72
Thread Starter 
Exactly.

Usually you need some DPI correction, just increase or decrease the DPI of a mouse until the Velocity graph shows same speed for both mice. I usually move them left and right only, so the angle difference does not disturb me. Then change input lag value until the graphs cover each other.
post #35 of 72
Thread Starter 
Quote:
Originally Posted by qsxcv View Post

3. hard to tell. sometimes, bad mouse firmware skips usb reports. it could also be your pc/os acting weird.

Just realized that graph might show that the cable is about to die. It loses data connection for a few seconds but still gets voltage, so counts the measurement. When data connection recovers, it sends all the data in one poll.
post #36 of 72
I did some Alcors test, 800dpi 1000hz. Is the tracking ok?
post #37 of 72
Quote:
Originally Posted by dobragab View Post

Exactly.

Usually you need some DPI correction, just increase or decrease the DPI of a mouse until the Velocity graph shows same speed for both mice. I usually move them left and right only, so the angle difference does not disturb me. Then change input lag value until the graphs cover each other.
xSum vs Time graph? That seems to work best and I get 8ms delay.

Please add legend with which mouse has what colors in graph.
And also note where is the delay added, is it mouse 2 is slower by Xms than mouse 1 when X is positive?

I'm not even sure I got the delay right how do you really measure it from these graphs? I look at first count but then last is also delayed in opposing way which is weird. Can you make a picture as an example?
Edited by JackCY - 8/4/16 at 5:12pm
post #38 of 72
Why is it that whenever I need this file most urgently, the download link always refuse to work at precisely this time? Can't you upload the binary to GitHub instead?
post #39 of 72
Quote:
Originally Posted by daniel0731ex View Post

Why is it that whenever I need this file most urgently, the download link always refuse to work at precisely this time? Can't you upload the binary to GitHub instead?

MouseTester.zip 158k .zip file

I'm gonna assume I'm allowed to upload it... Here's the version I have, I believe it is the latest. Obviously not responsible if you computer blows up or your mouse flies away.
post #40 of 72
Still the same:

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Mice
Overclock.net › Forums › Components › Mice › MouseTester Software Reloaded