Overclock.net › Forums › Components › Mice › Death Adder sensor issues with circular motion
New Posts  All Forums:Forum Nav:

Death Adder sensor issues with circular motion

post #1 of 15
Thread Starter 
Here are some tests done on the Death Adder 3.5G at 900 DPI and 10/10 sensitivity in the Razer software and windows acceleration off and 6/11. All pictures were done with the Death Adder because of how it's widely considered to have one of the best sensors. However the same thing occurs with different DPI settings, sensitivities, and even other mice (sensei, although this is harder to test because of acceleration issues). This also occurs on all tested surfaces (func, white paper, cloth pads, etc).

There's lots of tests being done by starting the mouse on an edge of something, moving in one direction fast/slow, then back again at a different speed to test for acceleration. The test is done in a similar way, except for also using circles and see if the cursor can return to the same location.

The results are odd, because clockwise circles have the mouse cursor not return home, and counter-clockwise overshoots the original location quite consistently.

It would be great if other people could also test this out, perhaps with better positioning tools and speeds. Also if anybody knows WHY this happens, that would also be great as well.


338338338338
post #2 of 15
I get similar results. Clockwise and it's short of start. Counter-clockwise and it overshoots.
post #3 of 15
I believe it happens because you're rotating the mouse.
post #4 of 15
Thread Starter 
This still occurs when not rotating the mouse (as best as a human arm can do). Even if was caused by rotation, the mouse returns to the same angle as it started and still has a major errors. That and it'd be impossible to not rotate the mouse during normal use.

Anybody else have this issue with the Death Adder, or any other mice?
post #5 of 15
Similar results with both my DA and IMO 1.1
post #6 of 15
Quote:
Originally Posted by Grinkers View Post

This still occurs when not rotating the mouse (as best as a human arm can do). Even if was caused by rotation, the mouse returns to the same angle as it started and still has a major errors. That and it'd be impossible to not rotate the mouse during normal use.
Anybody else have this issue with the Death Adder, or any other mice?

I didnt have problems with mine until I read this lol. it def jumps around a bit going counter-clockwise
    
CPUMotherboardGraphicsRAM
Intel Core i7 2600K @ 4.7 MSI Z68A-GD80 AMD Radeon HD 7900 Series Corsair  
RAMHard DriveOptical DriveCooling
Corsair  hitachi Lg Blu-ray Corsair H100 
OSMonitorKeyboardPower
Windows 7 Asus 24'' Ducky shine green 800w 
CaseMouse
coolermaster cm2 advanced razer deathadder 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel Core i7 2600K @ 4.7 MSI Z68A-GD80 AMD Radeon HD 7900 Series Corsair  
RAMHard DriveOptical DriveCooling
Corsair  hitachi Lg Blu-ray Corsair H100 
OSMonitorKeyboardPower
Windows 7 Asus 24'' Ducky shine green 800w 
CaseMouse
coolermaster cm2 advanced razer deathadder 
  hide details  
Reply
post #7 of 15
Thread Starter 
I didn't want to post possible reasons for causing this to see if other people have different ideas, but I guess I will. Doing tests on orthogonal lines is extremely accurate, but even just straight lines at various angles also have problems with not returning to the original location on one axis, so I'm thinking at certain times the sensor's dsp switches to a less accurate algorithm because of processing power limitations (it's significantly less for 1D than 2D image processing).

Unfortunately I don't have access to any mechanical arms or a way to pull the information directly from the sensor/dsp. It could also very easily be a problem with the way the data is sent from the mouse or inside Windows. If anybody has any more information or can do any more tests it would be greatly appreciated.
post #8 of 15
Quote:
Originally Posted by Grinkers View Post

This still occurs when not rotating the mouse (as best as a human arm can do). Even if was caused by rotation, the mouse returns to the same angle as it started and still has a major errors. That and it'd be impossible to not rotate the mouse during normal use.
Anybody else have this issue with the Death Adder, or any other mice?

Let's assume that at the beginning of the movement mouse rotation angle is 0*. Move your mouse right then rotate it to 90*, continue to move it to the right, then rotate it back to 0* and move left back to the place where you started. See what you got in paint :>

I'm not saying that sensor is 100% accurate but you can't check it without special equipment.
post #9 of 15
Thread Starter 
Quote:
Originally Posted by AquaSurfer View Post

Let's assume that at the beginning of the movement mouse rotation angle is 0*. Move your mouse right then rotate it to 90*, continue to move it to the right, then rotate it back to 0* and move left back to the place where you started. See what you got in paint :>
I'm not saying that sensor is 100% accurate but you can't check it without special equipment.

I was thinking more along the lines of keeping it at the same angle while moving then returning to the original angle at the end, but good point still. I'm thinking about building a mechanical device to measure this much more accurately, but what concerns me most about the sensor is actually from straight lines at angles. But it's really so hard to check with a human arm.

I'm thinking about making my own mouse to try and compensate for these issues, but I'm trying to figure out what the issue is. If the sensor is "perfect" adding a gyroscope could potentially do a lot to fix this, but if it's the sensor/dsp that's the problem, it's significantly harder to fix (and expensive to make my own sensor/dsp in small quantities).
Edited by Grinkers - 5/4/12 at 12:58am
post #10 of 15
Same thing happens with my Logitech g1 800dpi.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Mice
Overclock.net › Forums › Components › Mice › Death Adder sensor issues with circular motion