[ut2004] UT2004 benchmarking: what does the "rand" value mean?

Xipher xipher at xipher.us
Wed Apr 21 19:13:08 EDT 2004


is their any correlation between the rand number, and the average FPS of 
the benchmark?

Corey Hickey wrote:

>I've been benchmarking UT2004 quite a lot recently in conjunction with
>tweaking/overclocking my system. I've noticed that often the "rand"
>value (printed on the line along with the fps values) varies.
>
>I ran a set of 50 identical 1-minute benchmarks (without altering any
>settings) and came up with this distribution of values:
>
>1425745073	2
>1919451183	28
>2051811764	9
>560406543	11
>
>
>Different values seem to correspond with the test running for slightly
>different durations; at least, with rand[560406543] the benchmark ends
>about a second later than it does with rand[1919451183], while
>rand[2051811764] ends about a half a second earlier. I don't see
>rand[1425745073] often enough to notice how long it takes.
>
>Does anybody know what this means?
>
>Thanks,
>Corey
>
>  
>



More information about the ut2004 mailing list