RSC was initially released for the Xbox and later on for the PC. It has been
one of my favourite racing games ever since and I’ve been playing it
so far on three different platforms up until today.
With each upgrade I have been able to both scale in game settings and resolutions
ever since, so the challenge in this case would be if the game remains playable
in it’s maximum 1600*1200 resolution with all in game settings set the
absolute maximum on the current setup.
I maximized all in game settings and used through the driver control panel,
4xAA/16xAF (quality). I measured performance from start to finish line (one
round) with FRAPS, in one of each map genre and the results are as follows:
Frames: 7402 - Time: 200282ms - Avg: 36.957 - Min: 21 - Max: 60
Frames: 3311 - Time: 93234ms - Avg: 35.512 - Min: 22 - Max: 56
Frames: 4566 - Time: 113140ms - Avg: 40.357 - Min: 22 - Max: 62
Vista Peak Base
Frames: 6069 - Time: 164703ms - Avg: 36.848 - Min: 23 - Max: 67
Frames: 4113 - Time: 136594ms - Avg: 30.111 - Min: 14 - Max: 64
Frames: 4224 - Time: 139422ms - Avg: 30.296 - Min: 17 - Max: 48
North Hill Circuit
Frames: 1854 - Time: 56265ms - Avg: 32.951 - Min: 24 - Max: 57
Frost Ridge 1
Frames: 2284 - Time: 56000ms - Avg: 40.785 - Min: 29 - Max: 79
Lake Bauer Short
Frames: 2835 - Time: 63047ms - Avg:44.966 - Min: 26 - Max: 72
Lake Point Ring
Frames: 2785 - Time: 73266ms - Avg: 38.012 - Min: 21 - Max: 63
Deer Park GP
Frames: 2653 - Time: 60531ms - Avg: 43.828 - Min: 26 - Max: 70
Frames: 1912 - Time: 61906ms - Avg: 30.885
Maps like Koenig Base (or those that basically presented a few minor stutters)
caught my attention, as there seemed to be something wrong there. I tried
to lower the resolution to 1280*1024, yet much to my surprise the resulting
performance was far from being acceptable and a lot worse than in 1600. Instead
of a few stutters, in that resolution the game is constantly dropping occasionally
to single digit framerates, in order to recover for a short while and then
it’ll start again. No difference in 1024*768 either. The fact that
it leads to only minor stutters during gameplay in 1600*1200, would lead
me to guess that there’s some sort of framelock going on , that mostly
affects the CPU (or even the GPU/CPU parallelism).
I recorded a small sequence in 1280*1024 (Koenig Base) with Lens Flares on,
via FRAPS’ per second framerate logging function and it looked like this:
RSC is up to a specific degree CPU limited too, which of course gets minimized
in a very high resolution like 1600*1200. I then switched off just the Lens
Flares in the game settings and not only was the game now playable under all
resolutions, yet the minor stutters in 16*12 were eliminated.
Note that the problem doesn’t appear in all but in some selected maps.
I tested those again, Koenig Base amongst others and one of the worst case
scenarios. Here are the results with Lens Flares switched off this time, all
other settings being identical as in the tests above:
Koenig Base - w/o Lens Flare
Frames: 5061 - Time: 146016ms - Avg: 34.660 - Min: 22 - Max: 57
That’s a rough 14% performance increase.
I did search for an official patch for the game, yet couldn’t find anything.
Frankly it’s only a minor issue in my opinion, since the Lens Flare implementation
of the game is something I can live without by far. It’s far from being
anything special or worth mentioning.
I also tested the game without FRAPS enabled, to ensure that it problem doesn’t
originate from a 3rd party application, yet the behaviour with Lens Flares
To come back to the question whether the challenge I mentioned earlier has
been met, the answer for me is yes. Especially since the Lens Flare problem
doesn’t seem to be restricted to one platform only and it’s not
relevant to the used 16*12 resolution (rather the exact opposite in the given