csgo-benchmark VS fpsbenchmark

Don't use fps workshop map for benchmarking. Its inconsistent, every time different results. Use https://github.com/samisalreadytaken/csgo-benchmark

I used samisalreadytaken's csgo benchmark script along with fREQUENCYCS' additions of ancient, anubis and inferno. As a benchmarking tool I used CapFrameX version: 1.7.0.

I think we all know the FPS BENCHMARK map on the workshop. It was created by ulletical,the same guy who created the most downloaded map in CSGO,aim_botz. Its a good map to get an estimate of what you should expect from your PC,but is it? When benchmarking something, in this case your ingame performance,the ideal way to do it,is use a realistic scenario. To begin with,the map was last updated in 2015,8 years ago. The campath takes you to some scenarios of explosions with model death animations at the start. The issue is when you drive through the smokes, your frames tank even below 200 and the performance is also similar when you drive through the many bots that mimic one animation. If i am not mistaken, this test uses old smoke particles ,but i am not sure. What i found to be more realistic and up to date, was samisalreadytaken's dust2 benchmark. Its pretty much the same idea, but on a more realistic and accurate scenario. The number difference is quite different. Today, i made one more benchmark to test out Ancients performance. You can the script and some footage of it here. I tried to make it as more realistic as i could and any feedback is much appreciated. HERE you can find the difference between these 3 tests. The FPS benchmark map has approx 200~ fps difference. I have done this tests over 30-40 times to numerous systems, and i can assure you the difference is noticable. On the second image of the post, you can find the frametime analysis of the FPS BENCHMARK map. You can see that the graph spikes quite a lot, which makes the game less smooth and pleasant. On the other hand, the third image, you can see that the difference is night and day. Frametimes are way lower and more consistent. Excuse the mini spikes, viber messenger on windows causes stutter sometimes lmao. I also made some benchmarks regarding some CSGO myths about launch options,maps and more.You can find it here. I also tried the "userdata folder" deletion myth that confirmed my early thoughts about it 6 years ago. You can find the tests here.

samisalreadytaken's csgo-benchmark script

https://github.com/samisalreadytaken/csgo-benchmark I used CapFrameX with the recording time set to 44 seconds (the length of the dust2 benchmark)

This CS GO benchmark on de_dust2 (at 64 tickrate which is the default) with CapFrameX running to record 1% and 0.1% lows.

I used Sam's benchmark script, along with afterburner/rivatuner for benchmark output.

Looks like they've used this one: https://github.com/samisalreadytaken/csgo-benchmark

I used samisalreadytaken's csgo benchmark script along with fREQUENCYCS' additions of ancient, anubis and inferno. As a benchmarking tool I used CapFrameX version: 1.7.0.