VOGONS


First post, by havli

User metadata
Rank Oldbie
Rank
Oldbie

Here is a part of my CPU benchmark project. I made these charts to support my claim that K8 is not fast enough for XP gaming rig (even for old games like NFSU2). Just to be clear - for me fast enough = avg fps > 100 and minimum fps > 60. 😀

First batch of results was made using GeForce 7800 GS (16PP) overclocked to 500/1400 MHz. Drivers 275.33, winXP SP3. All game details at maximum, AA/AF disabled, 1024x768. For Athlon Thunderbird GF 6800 Ultra + FW 71.89 was used - 275.33 doesn't work properly on CPU without SSE.
nfs_172uij.png

Second part - this time I used GTX 285 with drivers 341.44 and windows 7 x64. All game details at maximum, AA disabled, 16xAF, 1280x1024.
Clearly with same CPU fps are worse. Most likely this is due to different driver core for GF7 and GF8+ architectures. GF7 driver overhead seems to be less significant, however the driver doesn't benefit from dual-core CPU. On the other hand GF8 and later are slower on single-core but scales very well with second CPU core.
nfs_2ruu2q.png

HW museum.cz - my collection of PC hardware

Reply 1 of 4, by PhilsComputerLab

User metadata
Rank l33t++
Rank
l33t++

Very nice job!

You must really like this game, that must have been so much work to benchmark all these systems.

YouTube, Facebook, Website

Reply 3 of 4, by clueless1

User metadata
Rank l33t
Rank
l33t

Thank you for sharing those results. I second RoyBatty's question. 😀

The more I learn, the more I realize how much I don't know.
OPL3 FM vs. Roland MT-32 vs. General MIDI DOS Game Comparison
Let's benchmark our systems with cache disabled
DOS PCI Graphics Card Benchmarks

Reply 4 of 4, by havli

User metadata
Rank Oldbie
Rank
Oldbie

NFS U2 is only small fraction of my CPU testing methodology. It wouldn't be very time effective to put all these systems together just for benchmarking one game. 😀

There is no built-in benchmark. I used fraps to measure one lap at "Resort loop" quick race.

HW museum.cz - my collection of PC hardware