A quick test using fraps to see what sort of FPS hit night racing and headlights have on my system:

Each test consisted of 2 full laps of Montreal, starting from the grid, (test 1 pit lane).
During tests 2 & 3, the AI were kept in front of human driver at all times, with lead AI never being more than 200yds away.

Test 1: Open practice 8:30pm only me on track
Test 2: Race weekend 8:30pm 4 AI, All lights OFF
Test 3: Race weekend 10:00pm 4 AI, All lights ON

Using Jerry's Tip for tests 4 & 5
Test 4: Race weekend 10:00pm 6 AI, All lights ON
Test 5: Race weekend 10:00pm 12 AI, All lights ON
Test 6: As Test 5 but without the Jerry fix

Frames.....Time (ms)....Min....Max....Avg
23003.......195375.........96.....144....117.738.... Test 1
19605.......214007.........73.....112.....91.609..... Test 2
14373.......214030.........39.....103.....67.154..... Test 3
17567.......218148.........55.....109.....80.528..... Test 4
13157.......215106.........42.......78.....61.165..... Test 5
9009.........214756.........20.......76.....41.95....... Test 6

Nearly a 50% performance hit on min frame rate just because the lights are on! I can't imagine what it would be like being in the middle of say a 6 or 7 cars, fighting for position, or being on a grid of 20+ cars with their lights on.

Edit:
As you can see Jerry's tip does make a huge difference. And for what it's worth I think it looks just as good "eye-candy" wise. Test 5 are about the min figures I'd feel happy racing with. These are pretty good as there were 3 times as many cars on the grid compared to test 3!

Oh and I forgot to add, that on both tests 4 & 5, the AI had a mass pile-up at the exit of the final chicane which dropped the FPS even further. And in test 5 I was actually racing the AI, so cars were very close and a lot of the time I was in the middle of the pack.

System Specs:
AMD 4400+ dual core
2Gb ram
single 7900GTX
24" LCD 1900x1200
19" LCD 1024x1280

Average of 67 fps seems pretty good to me, wanna swap pc's? :-]

Nothing wrong with the average in any of the tests. It's the min FPS that is the concern.

Remember thats only 4 cars, I'm not really racing that tightly just keeping them in view. Also there's no smoke from locking up of brakes etc or dust from anybody running wide and putting a couple of wheels on the grass. In fact none of the "nastys" that can drop FPS in a "normal" race.

you can edit the plr file to help the situation. i now onlt see a 20% frame drop on my PC

Max Headlights="8" // Max headlights visible relative to your car.
Headlights On Cars="0" // Headlights illuminate other cars.

In your .plr file. Works wonders.

Edited first post to include tests 4, 5 & 6 to show the improvement with Jerry's TopTip.

Starting at the back of the grid, 23 cars ahead of me, at 22:00, on Montreal with Panoz series I get 56 FPS. Mmmmmkay?

System specs:
AMD 64 3000+, sck.754
1GB DDR400
Radeon 9800PRO

Having hard time fending of those antique dealers lately.

Would I be correct in assuming that you not running with everything turned on maxed out and/or not a screen rez of 1900x1200 Enis? (Or 2924x2480 if you include 2nd monitor)

Also is that from the cockpit or from one of those "other" girlie views :p

You assumptions would be amazingly correct, though point was that if your PC can't handle max details atm you should adapt.

I use cockpit view exclusively.

I realy had no problem with last night race in night (FPS problem that is..., my skill was problematic:) Man you have quite some monitor there and thats why the FPS loss, quite a high resolution... with probably some AA and AF?

I myself with 7900GTO and quite a high OCed E6300 didnt notice any "nondrivable" fps slowdown, with 1280x1024 and 4AA 16AF and ingame its all on max...

You should be fine to run the series with a PC like that, even with night transition, unless u r using 8aa 16af, since that combined with the resolution you have is quite a performance hit...