[SOLVED] [RVGL 19.0430] Benchmark mode is FPS-dependant

Report issues or bugs in RVGL or ask for help regarding any Re-Volt topic.
User avatar
EvgenKo423
Posts: 27

Re-Volt [SOLVED] [RVGL 19.0430] Benchmark mode is FPS-dependant

Unread post by EvgenKo423 » Sun May 05, 2019 8:02 pm

I don't know if it's by design or not and it's been like this for a long time, but that's not how benchmarks are supposed to work IMO... :roll:

1. Turn off V-Sync (like people usually do when measuring performance);
2. Turn off anti-aliasing;
3. Run a game with -gamegauge command line.

The whole replay is played almost instantly, in about 1 sec.
I think it should play at usual speed, giving a game some time to show off and actually measure FPS.
User avatar
Huki
Developer
Posts: 397

Re: [RVGL 19.0430] Benchmark mode is FPS-dependant

Unread post by Huki » Tue May 07, 2019 7:53 am

It's true that the replay runs instantly, but the game should save out an 'fps.txt' file to your profiles folder containing the measured frame rates.
User avatar
EvgenKo423
Posts: 27

Re: [RVGL 19.0430] Benchmark mode is FPS-dependant

Unread post by EvgenKo423 » Tue May 07, 2019 5:36 pm

Well yes it does, but there is only a single measure, not a series of them.
I understand that the benchmarking is more or less pointless in such a game, but it can be used to compare impact of different settings on performance.
User avatar
Huki
Developer
Posts: 397

Re: [RVGL 19.0430] Benchmark mode is FPS-dependant

Unread post by Huki » Wed May 08, 2019 9:05 am

There is an FPS entry for every second during the run. Here's the output on my computer for example:

Code: Select all

264.22 RVGL Game Gauge File
225 Min
332 Max
256 Second 1
250 Second 2
264 Second 3
287 Second 4
264 Second 5
225 Second 6
332 Second 7
285 Second 8
272 Second 9
I guess on faster computers the entire run could complete within a second, which would mean there is only a single measure. In that case, it would be better indeed to disable this speedup.
Locked