forums.naturalpoint.com

I was wondering have there been made any studies to determine the precision of the TrackIR.

With a mouse you get pixel precision, and although that�s clearly not the case with TrackIR.

I must admit I am quite impressed with the precision, robustness and performance of the product.

But still: are there any available information on the exact precision of the system?

Regards

Kristian

I believe it runs at 60hz. Not ideal for precision games like FPS but mostly fine for flight sims imho.

Even in IL2 I find that when using TIR it seems to "skip" so many pixels when turning your head which is ok for general viewing but distracting when determining the distance speed of another plane without either pausing TIR or switching to precision mode which helps a bit.

Dead zones in FB should help considerably.

But back to the topic, the loss of precision comes from the extrapolation of a small movement of your head to a larger FOV. It helps to have the dot as close as possible to the sensor unit (1 to 2 feet I find) by constructing a "boom extension" on a baseball cap and lowering sensitivity to minimize the magnification necessary.

I for one am greatly looking forward to the next-gen TIR with a more precise sensor (CCD?)but even the one used in the current TIR is pretty advanced as it is and is also the main proponent of cost for the TIR from what I gather.

I'd be interested in seeing more specs (and speculation ) myself.

Thanks for the fast repply.

The thing is im looking for actual "proof" of the precision.

Im using the system for an advanced usability test of an alternative onscreen keyboard for paralysed people (neck down).

So I really would very like more accurate information on the presicion, for papers and the like.

Something like: "Under theese conditions we were able to observe a 20-30 pixel precision." including a short description of the conditions would be like christmas =)

quote:Originally posted by Kristian Toerning:
Thanks for the fast repply.

The thing is im looking for actual "proof" of the precision.

Im using the system for an advanced usability test of an alternative onscreen keyboard for paralysed people (neck down).

So I really would very like more accurate information on the presicion, for papers and the like.

Something like: "Under theese conditions we were able to observe a 20-30 pixel precision." including a short description of the conditions would be like christmas =)In that case I think you would be better served by dropping an e-mail to Halstead or Jim personally but they may come by and read the thread, they're pretty good that way.

These guys are class acts when it comes to helping developers (and the people who post here) although I can't really comment on how open they are with the TIR's specs.

Good luck!

Thx for the tip!

Ill point them to this thread.

Regards

Kristian

Hello:

The precision of the trackIR is a hard thing to nail down. It all depends on how you look at it and what you want to measure. The trackIR can be "pixel" accurate if you have the dot very close to the unit, but the further you move away the lower the resolution. It is a big arc with the resolution decreasing as you move futher away from the unit.

quote:Originally posted by Jim Richardson:
Hello:

The precision of the trackIR is a hard thing to nail down. It all depends on how you look at it and what you want to measure. The trackIR can be "pixel" accurate if you have the dot very close to the unit, but the further you move away the lower the resolution. It is a big arc with the resolution decreasing as you move futher away from the unit.So will sub-pixel precision be possible when TIR2 arrives? I will be more than happy to thoroughly test it for you when it does.

I disagree precision is quite easy to determine. It is however true that it depends on circumstance.

Usually one would solve the problem by conduction tests, descirbe the test conditions and showing the results, but I know thats not how the industry works.

Mayby ill conduct a small experiment myself.

Thing is we need an ISO standard on this sort of stuff too.

Anyway nice product - and huge potential for disabled people.

Rock on.

Regards

Kristian