|
NOTE:
As this thread was mentioned on the day9 daily, I'm going to add this disclaimer:
When I set out to write this post I thought that it would be fairly straight forward to communicate what I was trying to say, however I was wrong and as a result this thread can be very confusing for non-mathematically minded people.
The general gist of it all is basically sc2 mouse settings that are multiples of 5 are inconsistent, and so you should avoid them. The other thing is that there are not as many different sensitivity settings as you think (there are not 100, as would be implied by a slider that works with percentages). You'll also find a lot of discussion on pixel precision and 1:1 ratios and stuff, and if you're interested you can read on, however I don't give a crap about pixel precision or 1:1 ratios or any of that stuff because as far as I'm concerned all of that is meaningless in the face of having a 'comfortable mouse speed'. However the findings are still relevant to myself, because although I don't care about pixel precision, I do care about having the same mouse speed today and tomorrow and the next day. If you're the sort of person that doesn't keep count of how many sugars they put in their coffee, then that's fine too.
Ideally you want your sensitivity as 6/11 (1:1 ratio of mouse movement to cursor movement):
http://www.teamliquid.net/forum/viewmessage.php?topic_id=142026, and the thread tells us that 50% is the same as windows' setting of 6/11.
I had to re-write this post a few times because I kept finding out new stuff. At first I wanted to test if 50% or 55% is 6/11 because:
windows sensitivity is n/11
starcraft2 sensitivity is n/100
100/11 = 9.0909090909090909...
If you want you sc2 sensitivity to be the same as your windows sensitivity..
windows sens x 9.09 = sc2 sens
6 x 9.09 = 54.54 (55% when rounded)
However I did some tests with MouseMovementRecorder (you can find it somewhere in here) and I kept getting weird results.
When set to 50% there were sometimes lost pixels and other times there weren't. (green means pixels mean lost pixels--movement that was picked up by the sensor but not performed with the cursor--this happens when you go below 1:1 ratio of mouse:cursor movement, a.k.a "6/11")
at 55% there were, again, sometimes lost pixels, sometimes no lost pixels. At this point I'm thinking maybe it rounds down decimal places?
at 51%--54% there were never lost pixels. (black=1:1 [6/11])
But this implies it also rounds up decimals :/
I kept playing around to see what was going on and I found that there are in fact fractions of percentages on this setting (i.e. there are more than 100 notches). I dragged the slider very carefully, and above ~50.5% you get a perfect 1:1 ratio, but below this you get lost pixels.
at 55% you get the same; on the lower end of the 55% setting you get perfect 1:1, and on the right-hand side you get red in the tester program=above 1:1=extra pixels thrown in.
This is pretty strange. They have a sensitivity slider with over 100 notches [probably as many notches as your horizontal resolution--in my case 1280 positions on the slider] but windows only recognises 20 different settings, so really the setting only functions at intervals of approx. 5%.
Another thing: if you're like me and have a 400dpi mouse (logitech mini wheel optical, etc.), I've found 8/11 to be a good setting (8/11 is not as bad as 7/11 or 9/11, and definitely 11/11, because 8/11 is an exact 1:2 ratio [skips every 2nd pixel in order to move twice as fast], and nothing in sc2 is 1 pixel =p)
8 x 9.09 = 72.72
CONCLUSION:
There are as many notches as there are pixels drawn on your screen, yet windows only recognises 20 different mouse settings--that is, 5% intervals.
It's kind of hard to figure out the exact border lines of the intervals, so I suggest avoiding multiples of 5% (50%, 55%, 60%, 65%, etc.) because the borders will be fractions of these numbers. Set your sensitivity between multiples of 5 (theres no difference between 51%--54%; 71--74; 96--99 [and 100]; etc. There is no 100.0000000000001% or 0.9999999999% so it works out no matter what your screen resolution is).
EDIT:
The reason why multiples of 5% are inconsistent:
- There are fractions of percentages on the slider, however the displayed whole number percentage only changes when you hit a new whole number, whereas the computer rounds the numbers---so, you are at 49%.. you slide it up to 50%... then you slide it left a few pixels. It displays "50%", however you've actually set it to 49.1%. The computer rounds it down to 49 and you see "50".
CONFUSED PEOPLE:
People seem to be confused. I'll try to summarise some stuff.
- Your windows sensitivity doesn't matter at all. SC ignores it.
- There are only 20 mouse settings in starcraft2 (1--5, 5--10, 10--15, etc.), even though there more than 100 notches on the slider.
- for 1:1 ratio of mouse movement to cursor movement, set your in-game sensitivity to anything between 51%--54% [with "enhance pointer precision" turned off in windows] -- not 50% as people previously thought (although it sort of is 50%; it's just that you can't see what the decimal point on the number is so it's safer to use 51--54) (that is, 51%, 52%, 53%, and 54% are all exactly the same mouse speed).
If you just want 1:1 pixel ratio of mouse movement to cursor movement stop here.
If however 51--54 is too slow, you should sacrifice the negligible 1 or 2 pixels of accuracy for a comfortable mouse speed. However, be careful that you avoid multiples of 5 because there are invisible decimal places that can make a big difference. (Theres no such thing as "exactly 80%". You're either playing at about 76 or about 84).
|
Good info, thank you.
I always thought adjusting Windows settings actually affected SC2 ingame mouse movement... turns out they really aren't even related in regards to speed.
|
So if you're below 1:1 you lose pixels (and therefore precision) because your cursor moves less than your physical mouse, and if you're above 6/11 (1:1) you also lose precision due to artificial generation of movement? This is the reasoning behind the theoretical 1:1 perfection point?
|
On November 02 2010 14:22 rolled64 wrote:
So if you're below 1:1 you lose pixels (and therefore precision) because your cursor moves less than your physical mouse, and if you're above 6/11 (1:1) you also lose precision due to artificial generation of movement? This is the reasoning behind the theoretical 1:1 perfection point?
Basically.
You can read more about it here:
http://www.overclock.net/mice/173255-cs-s-mouse-optimization-guide.html
|
|
|
Interesting article, I guess the ideal way to modulate mouse sensitivity is then to set windows (and SC2) at the theoretical 6/11 point and then change your DPI accordingly, as long as you have sufficient DPI. Do you know how much mouse movement is required to make a 180-degree turn in CS:S? I would guess that it would be the equivalent of moving your mouse across 1 horizontal screen length. I ask because I'm trying to create equivalent starcraft 2 formulae for the necessary dpi and such.
|
This confuses me too much what if i just have my ingame at 100% windowed fullscreen @ desktop resolution and my windows at 8/11 it willjust be 8/11 yeah
|
From what I gathered, SC2 ignores the windows mouse setting and instead uses its own psuedo-version of the same system, scaled to 100 notches rather than 11. As the OP said though, some wackiness arises with rounding. 8/11 in windows would be the equivalent of 72.72 in starcraft, to quote the OP.
|
On November 02 2010 14:36 T0fuuu wrote:
This confuses me too much what if i just have my ingame at 100% windowed fullscreen @ desktop resolution and my windows at 8/11 it willjust be 8/11 yeah
In-game setting overrides windows setting.
Try setting windows to 1/11 and sc to 100%, etc.
If you have some other mouse driver (eg. razer), it complicates things.
|
On November 02 2010 14:42 rolled64 wrote:
From what I gathered, SC2 ignores the windows mouse setting and instead uses its own psuedo-version of the same system, scaled to 100 notches rather than 11. As the OP said though, some wackiness arises with rounding. 8/11 in windows would be the equivalent of 72.72 in starcraft, to quote the OP.
The whacky rounding occurs because there are more than 100 notches (so sometimes when you set it to 50% it rounds down, sometimes it rounds up, depending on which notch within 50% you landed on).
|
edit: was only able to successful port 1 equation from CS:S. This the least confusing version of the information I've gathered so far:
If you have a mouse that can change DPI:
1) Set your windows sensitivity to 6/11
2) Set your SC2 sensitivity from 51-54 (all these are actually the same and equivalent to 6/11 in windows)
3) Change your DPI until you get the sensitivity you like for your mouse.
Equation to change your windows sensitivity to 6/11 and change your DPI without modifying the true sensitivity (how it feels)
New DPI = Current DPI * (Current Windows Sens / 6)
For example, if right now you have 9/11 windows sensitivity and 500 DPI, you would do:
New DPI = 500 * (9/6) = 750, so ideally you'd set your windows sensitivity to 6/11 and your mouse DPI to 750 (or as close as you can set it).
|
On November 02 2010 14:08 hide.X wrote:
It's kind of hard to figure out the exact border lines of the intervals, so I suggest avoiding multiples of 5% (50%, 55%, 60%, 65%, etc.) because the borders will be fractions of these numbers. Set your sensitivity between multiples of 5 (theres no difference between 51%--54%; 71--74; 96--99 [and 100]; etc. There is no 100.0000000000001% or 0.9999999999% so it works out no matter what your screen resolution is).
So you can set the sensitivity to any 51%--54%; 71--74 and still maintain 1:1? As long as you keep it in that 1-4 range your good?
Also since your knowledgeable on this information do you know how Razer's sensitively factors into the pixel ratio?
|
wow this thread is very interesting but very confusing for my tiny mouse like brain so apologies ahead of time. I always have problems at lan and always just wing my mouse sensitivity because sometimes it works sometimes it doesn't; and it seems to vary per computers.
What your saying is my in-game sensitivity should NOT be x5 ?
or
is it my computer mouse sensitivity shouldn't be x5?
or mystery option 3?
edit: my settings of right now -
computer control panel: 7/11
in game: 67%
and my mouse is an basic logitech optical mouse
|
gcubed:
A mouse maker doesn't change anything. Sensitivity should always be set to 1:1. If your DPI is too high or too low for your personal taste, lower the DPI.
|
Huk:
Windows should be 6, the middle one.
SC2 should be 51% to 54%.
|
Edited my above post to include the one equation I managed to successfully use from the CS:S article. Seems like if you don't have a mouse that can modify its DPI and your sensitivity at 6/11 is not pleasant, you're shafted.
|
On November 02 2010 15:17 sk` wrote:
gcubed:
A mouse maker doesn't change anything. Sensitivity should always be set to 1:1. If your DPI is too high or too low for your personal taste, lower the DPI.
Sorry, i mean the Razer sensitivity controller that you download on your computer.
I have windows 6/11 startcraft2 54% and the Razer sensitivity at 9/10. I'm asking how the Razer sensitivity factors into the whole 1:1 pixel ratio.
also I'm curious at how he said x1-4 are all the same (1:1) Which would mean you can set SC2 sens to anything from 11-14 all the way to 91-94 and still maintain 1:1
|
This thread confuses me. I use a Logitech G1, which is a 400dpi mouse. I've always used it with 8/11 sensitivity in windows. As for SC2 though, I've been play with it set to 80% as this has felt close to what my mouse moves like on the desktop. When I move it down to 55% or whatever like you suggest, it's way too slow, or am I missing something? What's the optimal percentage to use for a 400dpi mouse?
|
On November 02 2010 15:42 ShivaN wrote:
This thread confuses me. I use a Logitech G1, which is a 400dpi mouse. I've always used it with 8/11 sensitivity in windows. As for SC2 though, I've been play with it set to 80% as this has felt close to what my mouse moves like on the desktop. When I move it down to 55% or whatever like you suggest, it's way too slow, or am I missing something? What's the optimal percentage to use for a 400dpi mouse?
You misread this he jsut means if you like a super fast mouse speed then dont' set it to 80% set it to 81%
|
Canada1620 Posts
Can anyone confirm? This is all really confusing but basically you want 6/11 in windows, 51-54% in game, and then adjust the DPI on your mouse to decide exactly on sensitivity?
|
|
|
|
|
|
WCS
