BYF wrote:
>Anyone know why the gfx test in gp4 is so misleading? because if i set fps
>at what the test said i would have PO of 150 at start of race= slowdown.
>Ive always found that by setting the fps 10 fps lower than what the test
>says keeps the PO at 100-115 at start (bearing in mind that actually
>holding down the PO actually increases the PO by 15-25% what a joke!!)
>so its probably really at 85-100 PO.
GP4 runs the graphics test in t-cam view, while the cockpit view uses a bit more resources. Try this: as soon as the test starts, quickly hit <cursor right> to switch to cockpit and let the test run in this view, the suggested frame rate will be closer to what you actually need if you drive usually from the cockpit (particularly if you have different mirror settings for onboard cams/cockpit view).
If you play with different cams while running the test, you'll also notice that the top-front cam gives the lowest PO/highest framerate by a big difference (because there's no rendering of your car, mirrors, tyres, steering wheel etc), so it's worth giving it a try if you cannot get high enough framerates in the cockpit or t-cam views, it's a way to enjoy driving with higher detail settings for the track and other cars. The front cam gives a very good speed feeling too, I would use it more often if only it didn't require hitting 10 times the PgUp key to get to it, that's a shame
Also, I have found that a good test to make sure that that GP4 is performing well is to use a chronometer to check against the lap times or any time display in the game. So after all your graphics tweaking, do this final test in a race with a full grid and wet weather, and if the the PO peaks too high, the in-game time display gets slower. On the other hand, when the program is running smoothly, its timers are accurate to the point that can be checked with a chrono in your hand (a couple of 1/10 s). If you find that the in-game chrono is, say, 2 secs late after a 5 min period, then you know that performance is not optimal. Or in other words, you may find that your fastest 1'14.657" or whatever lap actually took a few secs longer to drive!
Bear in mind that, since the circuit length always remains the same, a longer driving time means less actual speed, no good when you want as close a simulation as possible.
* S y r i u s *