Overclock.net › Forums › Components › Mice › Twin eye sensor, optical limitations, native dpi, smoothing : interesting discussion I had
New Posts  All Forums:Forum Nav:

Twin eye sensor, optical limitations, native dpi, smoothing : interesting discussion I had

post #1 of 4
Thread Starter 
I recently found myself discussing these topics with a guy on youtube (https://www.youtube.com/watch?v=gBjLIYFo9Ow&list=UUwhyF4Lrh7EJ_NqiWBBzuPw look for the above8/circacheese comments), and some interesting points came up in the dialogue.

He says PTE sensors don't have any angle error, thus needing no correction and so giving lowest input lag.
On top of that he says PTE is not limited by native SROM cpi, and z-axis issue seems not to be a problem with an aluminum pad.
Can anybody provide more details, specially about the cpi and how it is scaled?

He also says "But I know only 3 mouses correctly using this sensor: Razer Lachesis 3G, Razer Spectre and MadCatz R. A. T. TE."
Since native cpi and smoothing shouldn't be a problem I wonder what he means with "correctly".
Do all twin eye sensors suffer from z-axis issue in the same manner? Why are these better?

other quotes:

"Knowing that optical sensor use arrays up to 30x30 pixels, is easy to calculate their native (physical) DPI which for optical LED sensors is about 400-450 (same as for ball mouses) and for optical laser sensors about 800-900, any CPI beyond this is not reliable. Movements for about 1 or 2 native pixels aren't reliable too."

"400 dots per 1 inch = 400 dots dots per 254 mm = 30 dots per 2 mm. Lasers are like 2 times smaller. XL's A6010 has less smoothing than 3090 because A6010 only scales 880 DPI to 1760 CPI (3520 CPI is achived by skipping half pixels without any smothing) and 3090 scales 410 DPI to 3690 CPI (and 3690 CPI its smooth)."

Can anyone confirm this? Why should optical LEDs scale from ~400dpi if they have a 30*30 pixel array?
I obviously asked him to explain but didn't get any answer (actually the second quote was his explanation..) in two days so I'm posting it here.
Edited by the1freeMan - 4/12/14 at 1:37pm
post #2 of 4
Very interesting, I am actually an owner of the Lachesis 3G, but I'm not sure how it's "correct." It is an older PTE unit with a lower ~120 IPS max PCS, and the shape is too claw-y for me. I would have thought he would have mentioned the original Mamba there since it doesn't have any dynamic CPI scaling on older firmwares. I don't have an answer for your question though, this is Skylit-level stuff.
L1m1t
(22 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5 GHz ASUS P8Z68-V LX ASUS DirectCU II GTX 770 OC Patriot G2 
Hard DriveHard DriveCoolingOS
Western Digital Caviar Black Kingston SSDNow Cooler Master Hyper 212+ Windows 8.1 x64 
MonitorKeyboardPowerCase
ASUS VG248QE SteelSeries 6Gv2 MX Red Corsair TX650 NZXT Source 210 
MouseMouse PadAudioAudio
Zowie FK1 Puretrak Talent ASUS Xonar DG Sennheiser PC350 
AudioAudioAudioAudio
Creative Fatal1ty HS-800 Shure SRH440 Audio Technica ATH-M40x AKG K701 
AudioAudio
beyerdynamic DT770 AKG K550 
  hide details  
Reply
L1m1t
(22 items)
 
  
CPUMotherboardGraphicsRAM
i5-2500K @ 4.5 GHz ASUS P8Z68-V LX ASUS DirectCU II GTX 770 OC Patriot G2 
Hard DriveHard DriveCoolingOS
Western Digital Caviar Black Kingston SSDNow Cooler Master Hyper 212+ Windows 8.1 x64 
MonitorKeyboardPowerCase
ASUS VG248QE SteelSeries 6Gv2 MX Red Corsair TX650 NZXT Source 210 
MouseMouse PadAudioAudio
Zowie FK1 Puretrak Talent ASUS Xonar DG Sennheiser PC350 
AudioAudioAudioAudio
Creative Fatal1ty HS-800 Shure SRH440 Audio Technica ATH-M40x AKG K701 
AudioAudio
beyerdynamic DT770 AKG K550 
  hide details  
Reply
post #3 of 4
Quote:
Originally Posted by the1freeMan View Post

He says PTE sensors don't have any angle error, thus needing no correction and so giving lowest input lag.
On top of that he says PTE is not limited by native SROM cpi, and z-axis issue seems not to be a problem with an aluminum pad.
Can anybody provide more details, specially about the cpi and how it is scaled?

The way I understand it, the PTEs use the Doppler effect to measure how far they've travelled over the surface, meaning they don't scan the surface like "optical" and "laser" sensors do to compare before and after snapshots. Rather, they merely measure the wavelength of the reflected light. It's thus plausible that the PTEs could do with a smaller delay-causing overhead. Personally, though, I don't feel the difference. The PLN2033 feels just like an "optical" sensor to me. It's just the "laser" sensors that are inferior.
Quote:
Originally Posted by the1freeMan View Post

He also says "But I know only 3 mouses correctly using this sensor: Razer Lachesis 3G, Razer Spectre and MadCatz R. A. T. TE."
Since native cpi and smoothing shouldn't be a problem I wonder what he means with "correctly".
Do all twin eye sensors suffer from z-axis issue in the same manner? Why are these better?

From PLN2030 through PLN2033, they all have a z-axis issue in one form or another because the Doppler shift of a single beam reflected off a mousing surface doesn't allow you to distinguish between z-axis movement and x- or y-axis movement. As of the PLN2033, the issue is fairly tolerable as the cursor returns to its former position when the mouse is set down again. The R.A.T. TE (presumably using the PLN2034) promises further relief in the form of an adjustable lift-off distance.
post #4 of 4
The mentioned generations of laser and led based cmos designs don't really differ in regards to possible resolution.

Possible CPI is related to image matrix size but isn't a true representation of actual counts (per inch) as measured on desktop.

Far as the optical twin eye goes, there is no specific value or increment. Philips seems to like CPI multiples of 800.
Edited by jsx3 - 4/12/14 at 10:11pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Mice
  • Twin eye sensor, optical limitations, native dpi, smoothing : interesting discussion I had
Overclock.net › Forums › Components › Mice › Twin eye sensor, optical limitations, native dpi, smoothing : interesting discussion I had