Logitech G500 sensor optimization help

[hfcobra]

So is there a mouse sensor guru out there that can help me out with this?  I have installed the latest setpoint and the settings are set to:
Level 1 CPI 800
Level 2 CPI 1800
Level 3 CPI 3500
The sensitivity is in the middle and acceleration is set to zero (all the way to the left) even though I told the program to use the settings from windows.  The windows settings are 6/11 and I have disabled "Enhance Pointer Precision" to stop the OS acceleration as much as I know how to do.  The liftoff distance is amazing.  I can set it on 2 separate towers (for stability) of pennies that are one penny high and it still works, but when I add a second penny to each tower the sensor is too high off the pad and no longer tracks.

The computer I use it on is using Windows 7 Ultimate 64-bit, Intel i7 920 @4.2GHz, 6GB 1600MHz 6-8-6-24 RAM, Gigabyte GA-X58A-UD3R rev. 2.0 motherboard, EVGA GTX 580, Crucial C300 128GB for OS and games with 1TB Samsung HDD for movies, music etc.  I have a Corsair HX580W power supply if that matters at all, although I don't think it does.  Lastly, I have a sound card if you need to know that.  If you need to know what it is just say so, I don't want to list everything that does not directly affect the speed of the computer or overclock in some way.  

Is there anything else that I need to do to make the sensor "perfect?"  I am not going to get a new mouse just for a better sensor so please don't suggest that.  I tried the Deathadder Black and while the sensor was nice, its liftoff was higher than the G500 I am using and the sensor placement being in the center of the mouse was a little weird feeling to me.  Overall I really liked the censor, but I hated the mouse body.  It was not comfortable for me and made my hand cramp up.  The censor placement was a little hard for me to get used to since the only mouse I have ever really used for gaming is the G500.  Since it was in the middle instead of the top-middle of the mouse it felt weird and I would have to retrain muscle memory to get used to it.  I would have done this if I had to but the G500 was so much more comfortable that I just could not use the Deathadder at all.

So all in all, I guess that I am asking how do I make my sensor as "perfect" as possible?  I am not saying I want it like the Deathadder sensor, I was just using that as an example to show that I did like it overall (except for a few small things that only apply to me or a very small consumer base) because I have heard that its censor is as near-perfect as they come.  If I am wrong in this then please let me know.  I don't want to spread misinformation about the Deathadder or anything really.

Sorry for the wall of text but I am really enjoying typing today!  :biggrin:

[fingers of death]

Perhaps, this will be of some assistance: http://forums.guru3d.com/showthread.php?t=335510

[hfcobra]

Well that just told me about the mouse.  Didn't say much about optimizing it.  Thanks for the article though, I learned a little about some laser sensors while reading it!  

[fingers of death]

I'm pretty sure that you're a member at 'overclock.net', but perhaps you missed this thread.
http://www.overclock.net/mice/173255-cs-s-mouse-optimization-guide.html

[hfcobra]

I'm pretty sure that you're a member at 'overclock.net', but perhaps you missed this thread.
http://www.overclock.net/mice/173255-cs-s-mouse-optimization-guide.html

I actually did read all of that completely.  There was just so much stuff to do that I was not sure if I had to do it or not for a laser sensor of particularly the one on the G500 (Avago ADS-9500 or something like that).  So I decided to use that as a last resort after asking for some help.  I figured someone who already knows all about what that article is saying could help me out by only telling me what I have to do and could probably describe it to me much more simply.

Thanks for your help though, I am just hoping someone hops in here who basically has that article memorized inside and out and knows a lot about the G500 mouse and laser.  

[fingers of death]

If you take nothing out of it, but this formula: [(Current dpi) x (In-game sensitivity)] / (Maximum dpi) = (New Sensitivity for max dpi), you're way ahead of the game. Try it and see for yourself.

P. S. Of course, it goes without saying that you have to un check 'Enhance Pointer Precision' in mouse settings and set acceleration to 0 in set point.

[hfcobra]

If you take nothing out of it, but this formula: [(Current dpi) x (In-game sensitivity)] / (Maximum dpi) = (New Sensitivity for max dpi), you're way ahead of the game. Try it and see for yourself.

P. S. Of course, it goes without saying that you have to un check 'Enhance Pointer Precision' in mouse settings and set acceleration to 0 in set point.

yep, I did all of that already.  There is still some acceleration built into windows from what I have heard though....

I don't really play enough games to have a set sensitivity yet so I don't really need to use the formula right now.  I just use whatever is comfortable for me which changes from time to time     Haven't found the perfect sensitivity just yet.  

Thanks for the help though!  

[fingers of death]

You're welcome. We're both looking for the same thing. The 'MarkC Windows 7 Mouse Acceleration Fix' is a strange bird. It works in Combat Arms, but doesn't do **** otherwise.

http://www.sevenforums.com/hardware-devices/9035-windows-7-mouse-acceleration-help-2.html

[fingers of death]

Just a head's up on the 'Mouse Acceleration Fix': I uninstalled it.The mouse acceleration is the same normally and in game with it or without it. I think the best advice is lowering mouse  sensitivity on site and maxing the sensitivity of the actual physical mouse, as per [(Current dpi) x (In-game sensitivity)] / (Maximum dpi) = (New Sensitivity for max dpi).

[suntorytime]

there isn't much you can do to 'optimize' a mouse.

my steps in short (with the most important steps in bold):

1. plug mouse in and not install any crap drivers unless you really need to. (to change polling rates etc, generally the new 'gaming' mice use 500Hz and I find this more than adequate)
2. make sure sense is 6/10 in windows
3. turn acceleration off in windows
4. If you are beyond OCD you can either install CPL mousefix if you are using XP or MarkC mouse fix if you're on Win7.
5. change DPI of mouse to whatever you're normally accustomed with, i use 800 DPI only.

There are a few sensors that inherently have negative or positive acceleration, depending on DPI. Also some sensors react badly to certain mouse mats (high lift off, cursor move when lifting etc). I'm not sure what sensor the g500 uses so you might want to do some googling there.

If you play newer games you will have more issues trying to work out how to turn off funky console like acceleration and using sensitivity sliders, but these issues are game dependent. Some older games which use direct input such as counter strike and quake should not have these problems.

ps. your computer specs won't affect the performance of your mouse at all unless you're on some 486 trying to run 1000hz polling. And generally a high lift off distance is bad.

[Dr.Pepper]

I am personally using a G500, and I really like it, the G5 I used to have got to jumpy at 2000dpi, and 9 sensitivity in windows, now I'm comfortable at 5700, and 6 out of 10 in windows sensitivity, smooth, fast and no acceleration.

[manfaux]

there isn't much you can do to 'optimize' a mouse.

my steps in short (with the most important steps in bold):

1. plug mouse in and not install any crap drivers unless you really need to. (to change polling rates etc, generally the new 'gaming' mice use 500Hz and I find this more than adequate)
2. make sure sense is 6/10 in windows
3. turn acceleration off in windows
4. If you are beyond OCD you can either install CPL mousefix if you are using XP or MarkC mouse fix if you're on Win7.
5. change DPI of mouse to whatever you're normally accustomed with, i use 800 DPI only.

There are a few sensors that inherently have negative or positive acceleration, depending on DPI. Also some sensors react badly to certain mouse mats (high lift off, cursor move when lifting etc). I'm not sure what sensor the g500 uses so you might want to do some googling there.

If you play newer games you will have more issues trying to work out how to turn off funky console like acceleration and using sensitivity sliders, but these issues are game dependent. Some older games which use direct input such as counter strike and quake should not have these problems.

ps. your computer specs won't affect the performance of your mouse at all unless you're on some 486 trying to run 1000hz polling. And generally a high lift off distance is bad.

Is looking at "6/10" not bothering you?

6/11.

[suntorytime]

Is looking at "6/10" not bothering you?

6/11.

Not really, but ^ that's correct.

[NKRO]

Just a head's up on the 'Mouse Acceleration Fix': I uninstalled it.The mouse acceleration is the same normally and in game with it or without it. I think the best advice is lowering mouse  sensitivity on site and maxing the sensitivity of the actual physical mouse, as per [(Current dpi) x (In-game sensitivity)] / (Maximum dpi) = (New Sensitivity for max dpi).

When "Enhance Pointer Precision" is disabled, mouse acceleration in Windows (and hopefully your game) is also disabled.

Previously, some older games would call a Windows function that would disable mouse acceleration automatically. However in newer versions of Windows, this function now enables the EPP setting.

What the MarkC mousefix does is change the acceleration curve that Windows uses for the EPP setting so that it does not add any acceleration whether the option is on or off. Most modern games should be using direct input methods for the mouse and should be unaffected.

There is nothing that can really be done about fixing a mouse that has acceleration built into the hardware though, unless the manufacturer releases a firmware update that disables it.



One other thing you should do is check that your mouse polling rate is stable. Mouse Rate Checker

Move your mouse around quickly inside the test area. Your mouse should be averaging whatever you have set your mouse polling rate to, without any big drops. (low speeds will report lower polling rates)

On most systems, 1000Hz is not going to be stable and 500Hz should be used. On my system 500Hz reports 500±5% consistently and always averages 500Hz. 1000Hz is anything between 500-1000Hz and while it often does average 1000Hz many times it does not.

[Dr.Pepper]

1000HZ is stable on my computer, and I never have a problem with it, and 500 isn't... and not to mention that it feels wired, go figure.

[Quest-]

The G500 uses a good sensor, the same one as the G9x and the steelseries Xai, and many more.

First off, Steelseries only rates this sensor to 5000DPI, so i sugest you dont use more then that unless you really need to.

It has one weak spot though, Black cloth mouse pads. I dont know which pad you are using, but if its black cloth and you use the mouse at a relatively low sensitivity (actual movement, DPI doesnt matter for it) it shows up to 10% hardware acc.

You might want to look at a glass / aluminum pad, but personally i never had a problem using this mouse with a Steelseries QcK (black cloth)

Also make sure you have applied the MarC mouse fix as someone posted earlier.

The sensor is sensitive to dirt also, so subtly clean it every month at least.


In Setpoint, make sure angle snapping / path correction is off

Ingame sensitivity should be a logical number either or below 1.0 (such as 1.0 - 0.5 -  0.25), if you want your sensitivity higher or lower, use the DPI settings.
 
All ingame stuff like mouse smoothing etc, all off

Disabling or lowering the frame buffer can help reduce input lag in games, for ATI cards install ATI Tray tools and change the Flip Queue Size to 0 or 1

Ok im tired now, let me hear it if you want more pointers .

[fingers of death]

I like angle snapping! :argue:

As I said previously, the 'Mouse Acceleration Fix' is completely ineffectual (on my com, at least) and what does make a dramatic difference is using the g500 at 5700 dpi and lowering 'in game' mouse sensitivity. I only wish that a button could be assigned to toggle between two dpi settings, instead of requiring two buttons for switching.

[Enzyme]

that sensor is rated at 5001dpi, setpoint uses emulation of sorts to make it 5700max

for ultimate accuracy dont install setpoint, as any dpi settings you apply there will go thru the emulation system, which is known to work badly on 64bit systems.
but you will be limited to the default drivers 5 buttons.

if you have to, turn off acceleration, apply the mark-C mousefix, set the windows pointer control to the center notch,
keep the speed at 1, and the dpi at 5000.
or speed at 2 and dpi at about 3000-3600 (more noticable acceleration effects but smoother)

and dont let people confuse you with tales of enhance pointer precision and acceleration.
the enhance pointer precision affects deceleration not acceleration.
which basicly brings your mouse pointer to a stop faster, this can be an aid without affecting accuracy of movement.

also the mark c fix removes the windows acceleration curve and replaces it with a linear sensitivity.
this may or may not make a difference depending on the mouse software applied and how well it deals with windows being *****y about its settings.

[Enzyme]

yup theres a lot of "my mouse is the best evr, yours sucks" about.

you can trust me on this issue, as an aging tech head ive worked closley for the last two years with both logitech tech support, avid gamers, users and experienced tech review personel from well known websites in attempt to find a fix for this issue.

tests have been performed extensivley and on various configurations.

logitech is just greedy and has no consideration for those its ripping off.

personally i wont ever put my money theyre way again.

[NOMiS]

Try and set your dpi to a multiple of 90, the sensor in the G500 was designed to operate that way. iirc, the G500 can only be set to multiples of 100, so the only dpi settings I recommend are 900, 1800, and 2700. You can go higher, but at that point it's useless. Imo having 2 buttons on your mouse just to change dpi is dumb. Just choose one that you're happy using in windows, and stick to it. Use those buttons for something else, like quick save/quick load for games, weapon fire mode/grenade toggle, etc. They are much more fun to use that way.

For the hertz, 500hz is sufficient, as 1000hz gives you 1ms, 500hz will bring that to 2ms, so a very small difference.

I'd keep angle snapping, and accel off and use driver to set accel not windows. Make sure your LEDs are set to orange, that's the best color for the sensor.

Apart from that, just get used to the mouse, don't think of it as having a ****ty sensor, cause it doesn't. The first ESEA match I played with mine (and first time using it overall) I top fragged both sides with a Qck.

Everything you read on OCN is dumb, no one on there knows anything (except for a select few).

[fingers of death]

NOMIS,

Thank you. Amazingly, the multiples of 900 works. I'm now using the mouse at 5400 dpi with an in game sensitivity of 2 instead of the default 10. That's based on the "[(Current dpi) x (In-game sensitivity)] / (Maximum dpi) = (New Sensitivity for max dpi)" formula, but modified so the highest dpi is a multiple of 900.

I'm truly blown away by how well it works.

I've upped the polling rate to 500 (see 'G500 left click sticking' thread) to see if a previous left click sticking problem still manifests.

Thanks again.

[Skylit]

Try and set your dpi to a multiple of 90, the sensor in the G500 was designed to operate that way. iirc, the G500 can only be set to multiples of 100, so the only dpi settings I recommend are 900, 1800, and 2700. You can go higher, but at that point it's useless. Imo having 2 buttons on your mouse just to change dpi is dumb. Just choose one that you're happy using in windows, and stick to it. Use those buttons for something else, like quick save/quick load for games, weapon fire mode/grenade toggle, etc. They are much more fun to use that way.

For the hertz, 500hz is sufficient, as 1000hz gives you 1ms, 500hz will bring that to 2ms, so a very small difference.

I'd keep angle snapping, and accel off and use driver to set accel not windows. Make sure your LEDs are set to orange, that's the best color for the sensor.

Apart from that, just get used to the mouse, don't think of it as having a ****ty sensor, cause it doesn't. The first ESEA match I played with mine (and first time using it overall) I top fragged both sides with a Qck.

Everything you read on OCN is dumb, no one on there knows anything (except for a select few).

I believe Logitech's SROM is designed to work in intervals of 100 without interpolation.

First off, Steelseries only rates this sensor to 5000DPI, so i sugest you dont use more then that unless you really need to.

Steelseries may only rate the sensor at 5001 CPI, but they also use Avago's default SROM which is set in intervals of 90 meaning any setting other than a multiple of 90 is either interpolated or downscaled to the next lowest setting. Avago's native rating on the sensor is 5040 CPI.

 I'm sure Logitechs 5700 CPI setting is fine if it was properly tested by engineers using Avago's proprietary software (Which they wont give me *sadface*). Roccat and Gigabyte have this sensor running at 6000 DPI, although I'm unaware of any interpolation is involved.

[Quest-]

Nice, getting some advanced info in this topic , hard to find nowadays.
Some questions while were at it:

Is it really beneficial to keep the D/CPI dividable by the rate set in the SROM?

Can there be a negative impact if sampling rate is set to its max - 1000hz? (besides the claims of harming the usb bus)

Is it true that sampling rate should be kept at something dividable by monitor hz? (steelseries claim)

Can/should the firmware of the G500 be changed?

[Bullveyr]

I believe Logitech's SROM is designed to work in intervals of 100 without interpolation.

Jep

Steelseries may only rate the sensor at 5001 CPI, but they also use Avago's default SROM which is set in intervals of 90 meaning any setting other than a multiple of 90 is either interpolated or downscaled to the next lowest setting. Avago's native rating on the sensor is 5040 CPI.

Allthough they use the default CPI settings their SROM is far from default, but you shoud know that.

Is it really beneficial to keep the D/CPI dividable by the rate set in the SROM?

Yes but there is a good chance you will not feel a difference.
Still, there is no reason to use a setting that at least can be bad unlike you need it for your "perfect desktop setting".

Can there be a negative impact if sampling rate is set to its max - 1000hz? (besides the claims of harming the usb bus)

Higher CPU load and less stable pollingrate compared to 500 Hz.

Is it true that sampling rate should be kept at something dividable by monitor hz? (steelseries claim)

Only in theory, the polling rate isn't stable enough anyway.[/QUOTE]

[fingers of death]

Multiples of 100 means any goddamn setting is good. It's not! Multiples of 900 works beautifully. Smooooooth.

[Shuki]

Isn't a mouse always polled at a set frequency but the hz testing programs just can't determine this if you move the mouse too slowly?

[Arc'xer]

Isn't a mouse always polled at a set frequency but the hz testing programs just can't determine this if you move the mouse too slowly?

Polls based on the need of the OS, when moving slowly. If you move your mouse at a consistent speed, it polls at or near the set maximum. But when you move it slower, it polls itself to the needs of the moment so instead of seeing the maximum set; you see 100hz, 40hz, 2hz, 18hz, 63hz etc.etc.