Overclock.net › Forums › Components › Mice › Logitech G102 Prodigy
New Posts  All Forums:Forum Nav:

Logitech G102 Prodigy - Page 33

post #321 of 1351
Quote:
Originally Posted by wareya View Post

I'm having z-axis bugs with 3320-like sensors, not 3090

If you mean the cursor moving downwards a bit, that's really not a bug and it isn't really any detrimental. I would call it a bug if it was like the old PTEs where the cursor didn't negate the effect on reposition.
post #322 of 1351
I don't think 20 pixels is a "bit", and it really is a bug.
post #323 of 1351
Well I guess we have a different definition for the word "bug". Personally I don't think a lack of a compensation logic for a normal phenomenon is a bug, rather just a bad (and cheap) design tongue.gif A3090 does the same thing unless it is compensated with a LoD-detection.
post #324 of 1351
Quote:
Originally Posted by wareya View Post

I don't think 20 pixels is a "bit", and it really is a bug.

Then 3360 doesn't meet your requirements either. It may not be 20 counts at low DPI due to lower liftoff, but the 3366 for me with standard liftoff moves at least 5 pixels even on what seems to be a straight lift. I don't really have something to make sure it is 100% straight though I guess. It moves back very close to where it was before when I place it back down though. You are the only person I've ever heard refer to anything that doesn't cause systematic movement in a certain direction on multiple lifts and placements a z-axis bug.
Edited by chr1spe - 10/26/16 at 2:50am
post #325 of 1351
The cursor moves because the mouse itself moves, z-axis bugs exist for a completely different reason: the optically "active" part of the lense is, itself, literally off-center compared to the sensor
post #326 of 1351
Quote:
Originally Posted by wareya View Post

The cursor moves because the mouse itself moves, z-axis bugs exist for a completely different reason: the optically "active" part of the lense is, itself, literally off-center compared to the sensor

Idk it seems to me no matter what it move down on lift and up when placing it back down.
post #327 of 1351
Try kicking the lense out of where it's stuck.
post #328 of 1351

I would assume it is because of this and if by "optical active part" you refer to the light coming from an angle, then yes, I agree.

(edit, yes my timings are a tad off for the reading at times which results into a blank/max saturated pixel)
Edited by trism - 10/26/16 at 3:21am
post #329 of 1351
No I'm referring to the actual lense piece in front of the sensor

You can see it in this picture someone else took: http://i.imgur.com/7tjjEsA.jpg

In that picture, the inlet overall is perfectly centered on the "nipple" on the bottom of the sensor's IC. Said "nipple" has a pinhole directly in the center so that only the tracking surface significantly effects the sensor's image acquisition area. However, in the lense, the smooth clear "actual lense piece" is totally not in the center at all.

On other mice this is centered or nearly centered, so the only thing that affects the skew of the cursor's sensor as you lift it is the way the lense may shift around during operation, thus offsetting the "actual lense piece" but the lense is still able to get put into a position where it doesn't do this.. On the 332X (including the abyssus V2) this is totally not in the center at all, which give the sensor an inherent z-axis bug, in direct correspondence with the design, because the central pixel of the tracking surface literally must move on the sensor's acquisition area because optics 101.

I'm pretty sure they did this to reduce the amount of vertical space the sensor takes up without affecting the depth of field but it's still really bad. (my 3310 lense lets my 3320 kind of track, but it doesn't fit. and modifying the 3320 lense so that it fits with the "actual lense piece" centered on the pinhole causes it to stop tracking, probably defocusing due to being closer.)
Edited by wareya - 10/26/16 at 3:47am
post #330 of 1351
That's not the regular lens though. It's the one that drops CPI. I'll try to find the Xornet V2 default lens so I can do a comparison video and see if it is skewed even when placed on the surface. Personally I am using the Rival 100 lens with my PMW3320 thingmajigy.

If someone is interested, here is the lift-off behavior for the 3366. Please don't mind my coffee jitters. biggrin.gif



(The right frame is just showing the previous frame)
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Mice
Overclock.net › Forums › Components › Mice › Logitech G102 Prodigy