POLL: Which is the faster system for gaming? RESULTS IN!!

[ScrapSilicon]

Originally posted by: fkloster

Originally posted by: Luagsch

Originally posted by: fkloster

OMFG TAHT SI AMOST 1 % DIFFARNS

Ummm, english translation please?

lol, i won't give away that one😉 either you get it or you don't, sorry :beer:

lol got it. So not much difference but clear none the less that a lil' better latency beats a moderate amount of bandwidth non the less...

uhm...were there background processes running and were they the same on both runthrus ..😛😉😀

 

[sharkeeper]

This should bring back memories...

*********************************
* Overall Performance Index*
*********************************
|
This Computer+****************************************** 277.4
|
Compaq |
386/33MHz +****** 25.9
|
IBM AT |
286/8MHz +** 3.7
|
IBM XT |
88/4.77MHz +* 1.0
|
+----+----+----+----+----+----+----+----+----+----+--- 20 40 80 100 140 180 220 260 300 340 Overall Performance Index



******************
* Disk Speed *
******************
|
This Computer+******************************************** 301
|
Compaq |
386/33MHz +** 8.4
|
IBM AT |
286/8MHz +* 2.1
|
IBM XT |
88/4.77MHz + 1.0
|
+-----+-----+-----+-----+-----+-----+-----+-----+-----
40 80 120 160 200 240 280 320
Disk Index
Average Seek: 0.20 ms Track-to-Track Seek: 0.0 ms Data Transfer Rate: 32,165.8 Kilobytes/Second

Found this on one of our servers. Dated in 1992. It was a 486 EISA system with a caching SCSI controller card that had an incredible 12MB of cache and two hitachi 155GB Fast SCSI disks.

Cheers!

 

[Tetsuo]

Originally posted by: Acanthus

Originally posted by: charlie21
Probably system B, but I bet you wouldn't be able to tell the difference.

 

[Duvie]

I voted for A cause oldfart tested this long time ago and with real world benches....The better cas timings plus sometimes (and 3dmark usually is not one of them) overall processor speed matters....

 

[hurrikaane]

Well the nice thing about the 5:4 ratio is that the PC3500 memory is MUCH cheaper 🙂

 

[oldfart]

Originally posted by: Duvie
I voted for A cause oldfart tested this long time ago and with real world benches....The better cas timings plus sometimes (and 3dmark usually is not one of them) overall processor speed matters....

A lot of people have tested since and have proven that latency matters more than DDR speed. Once people got off the stupid SiSoft mem bench kick and tested with real apps, it became very apparent. The "high speed" PC4000 3-4-4-8 stuff is a joke.

fkloster is running 2.5-7-3-3 DDR425 on the 5:4 setup. If he had some better (and cheaper) ram like the Mushkin L2 PC3500, he's be able to run much better timings and improve performance even more.

Oh yeah, I picked system A.