View Single Post
  #26  
Old 02-15-2011, 08:58 PM
MasterTec MasterTec is offline
Registered Member
 
Join Date: Feb 2011
Posts: 8
Default

Quote:
Originally Posted by robtek View Post
This "reversal-bug", imho, just adds a bit more realism as the mechanical links
to the rudder, elevators and ailerons in the real planes had for shure more play then 2 or 3%.
OK GUYS! I'm the one who created that youtube video and am very active in the Logitech forums about this bug. I registered here just so I could respond to this post.

If the "hysteresis" as Logitech calls it (or reversal-bug as we call it) worked like you described, most of us would have little to no problem with it. The problem is it doesn't work like that.

With no play in the controls at all in the rudder control and if you moved the rudders from center, to left, and then back to the right, it would look like this (with 0 being full left, 100 being full right):
Starting at center, then going left: 50,49,48,47,46
Now reversing and going back right: 46,47,48,49,50
That's pretty cut and dry.

If you pretend you were in a real plane and there was a little bit of slack in the rudder control and you moved the rudders from center, to left, and then back to the right, it would look like this:
Starting at center, then going left: 50,50,50,49,48,47,46
Now reversing and going back right: 46,46,46,47,48,49,50
The fact that it does not immediately change shows the play in the linkage that must be overcome.. this would be expected in a real aircraft.

Here's what the G940 does:
Starting at center, then going left: 50,49,48,47,46
Now reversing and going back right: 46,46,46,52
See how it jumps from 46 to 52? If it was simply mimicking play in the controls, that jump would not be present... *THIS* is what makes it so annoying.

I don't fee like going back and researching my figures, but IIRC the reversal 'jump' in the throttles was over 5%, and in the rudders and trims was over 3%. This is *HUGE* when precision flying is involved.

I can understand why Logitech decided to put some Hysteresis in there... to compensate for 'pot flicker' where very small adjustments upon reversal tends to produce some variance, but what they should have done is programmed it to 'pick up where it left off' rather than to 'compensate for the amount of hysteresis'
Reply With Quote