------------ so what are we really looking at when we see that display of FPS dipping into the low 12's :)Mark

-----------------------------------I assume (and that may be dangerous) that "12 FPS" represents 12 complete redraws of that entire monitor screen per second. Quite unsatisfactory if that screen is 100% of the pilot's view. But if it represents only 33% of total view and the remaining 67% of the view is ALWAYS infinitely smooth-unchanging, then the brain has a problem! One third of what seems to be a single picture is "jerky" and 2/3 is absolutely smooth. M'lords- I submit to you that the brain will compute the (total pic) odds as 2:1 in favour of smoothness -UNLESS the displacement of pixels between updates is so great that it can no longer be ignored! (ie somewhere between 12- 7 FPS)I rest my case.( "Officer- I simply didn't see that the car ahead had stopped.")Alex Reid

It seems to me though that the problem with what you are saying is the same problem we have with a single screen display - the lower the framerate, the more the image changes between frames and therefore the more "jerky" the animation appears. Flying a plane at a given speed means you move a specific number of feet per frame of animation. That number of feet (or miles actually in most cases) is larger and larger per frame the lower your frame rate. Therefore, that building (or airport or mountain) approaching in the distance is going to jump towards you in steps. If you are getting 60fps, those steps are going to be much smaller than if you are only getting 20fps. I would think (and again, I don't really know since I only have one monitor) that having two monitors in a three screen display sit idle while the third gets an update would lead to a pretty unstable animation - particularly since you'd mostly be focused on the middle monitor. Actually, I find it a little hard to believe thats how the setup even works. I would have assumed that in a three display setup the image would simply be drawn as though it were a single image and then spread over the three displays. As for how the framerate is calculated, I've always understood it to be a simple count of the number of calls to display the screen image. I find the counter in FS9 too sensitive though - it seems as though the rate is re-calculated too often and it constantly jumps up and down. I use fraps frame counter as its more stable.

------- Actually, I find it a little hard to believe thats how the setup even works. I would have assumed that in a three display setup the image would simply be drawn as though it were a single image and then spread over the three displays.

--------------------------------Zevious- For a long time I assumed, like you, that the frame redraw was for all three images- not one at a time.But when I realized that the smoothness/fluidity of 16-24 FPS with triple mons was apparently identical to the smoothness with only one monitor/view active and yielding 30-45 FPS, I figured something else must be happening. It then dawned on me that with only one CPU and three completely separate views, the redraw must be occuring on a time share basis- one monitor/view at a time. It is NOT like one view being stretched over 3 monitors- these are 3 entirely separate views- but adjusted for the bezel separations so that the eye/brain combo sees it as a single scene. (Like your car where the windshield posts separate the views but the parts are seen as a single view by your brain.) I assume that the CPU can compute ONLY one view redraw at a time- it seems reasonable that that result would be outputted to a GPU before commencing the next view redraw. Otherwise you would have 3 times the amount of data cluttering up the CPU. And that would certainly create big variances in smoothness/fluidity between single & triple views configurations- which there aren't!And remember there are are only 2 video cards. It would be a neat trick for the CPU to compute all three views and only then, to squirt out 1/3 of its newly computed data to GPU #1, then call up #2 and squirt another 1/3 and then #2 GPU again with data for the third monitor.I think the evidence is pretty compelling for separate redraws of each view which time share each other for CPU access.But hey, I'm no techie- I get all sweaty just taking off the box cover to blow out dust!!Alex Reid