WTF? VIA KT400 > NF2?

[SecretIan]

What the hell? I just went and upgraded my a7v8x-x to an a7n8x (non x series) and my fps in ut2k3 actually went down by 5...

Neways, I am too tired to go into specifics right now, I'll finish this post tmrw, I just wanted u guys to know that I am pissed

-"Shoot coward, you will only kill a man"-

 

[SecretIan]

Ok so... I figure it is most likely some BIOS setting which has affected my vid card performance since my score in 1024x768 w/ noAA or AF didnt really change. Unfortunately I am very unfamiliar with this mobos BIOS so I would really appreciate it if someone either gave me a link or told me about some BIOS settings to take a look at. The manual isnt really very helpful with this unfortunately...

I would really appreciate some help with this, thanks a lot in advance.

-"Shoot coward, you will only kill a man"-

 

[ChipDeath]

Are your Memory timings the same as they were on the old mobo? Probably under 'Advanced Chipset Settings' I reckon.

Also, Are you running Dual channel or not? Athlon Systems don't really benefit from the extra Bandwidth, but [apparently] there's lower latency when the load is split between the controllers....

---
The end is nigh.. (For this post at least)

 

[SecretIan]

Nah, I am still running in single channel, havent gotten my new RAM yet.

Yes my timings are the same though.

-"Shoot coward, you will only kill a man"-

 

[ChipDeath]

You OverClocking? It's possible that your old mobo was 'cheating' by having a slightly higher FSB or something (e.g. 134.3Mhz instead of the 'official' 133.3333etc). And cos VIA (for some odd reason) don't include any sort of AGP/PCI locking in their mobos, it would also effectively OC your PCI and AGP buses too, giving slight performance gains.

So if you're running at stock speeds, the Nforce2 mobo is probably running at the speed it's supposed to, whereas the VIA one was not.

[This is just a theory, I don't know for sure ]

---
The end is nigh.. (For this post at least)

 

[Crashman]

I don't know if it will help, but you should at least try setting your AGP Aperture to 128MB.

Also, when you reinstalled the video card, it may have enabled AA or AF or both by default. Were these previously disabled?

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[SecretIan]

No I am not overclocking, and I always make sure whether Anisotrophic filtering and Anti Aliasing are on or off (Whatever I want). Anyways, the wierdest thing is, every other benchmark I ran gave me higher or equal scores w/ this mobo, exept 1280x960 w/ AA and AF on... I swear to god. I have no [-peep-] clue why, I have ran the 1280x960AA+AF benchmark several times, trying all different kinds of settings but nothing helps. ANyways here is a comparison(the first number will be VIA chipset and second will be NF2):

PCmark2002:
CPU-5500 vs 5706
MEM-4704 vs 5317
HDD-716 vs 666 (ok well NF2 lost this one, but thats cuz I didnt defrag b4 I ran test with NF2 but I did with VIA)

3dMark2003:
3DMark Score 5399 3DMarks vs 5530
GT1 - Wings of Fury 148.0 fps vs 166.2 fps
GT2 - Battle of Proxycon 38.3 fps vs 38.2
GT3 - Troll's Lair 33.4 fps vs 33.4
GT4 - Mother Nature 34.4 fps vs 34.4
(All tests done with driver sliders all set to "high performance")

3DMark2001:
3DMark Score 14445 vs 15463
Game1 Low Detail - 189.3 fps vs 206.3
Game1 High Detail - 60.2 fps vs 67.0
Game2 Low Detail - 285.5 fps vs 302.8
Game2 High Detail - 145.3 fps vs 156.9
Game3 Low Detail - 157.2 fps vs 175.6
Game3 High Detail - 68.6 fps vs 76.1
Game4 - 132.2 vs 130.8
(All tests done with driver sliders all set to "high performance")

Unreal Tournament 2k3:
-1024x768-
4xAA and 8xAF = 83.7 fps flyby, 49.0 fps botmatch vs 135.3,62.0 fps
noAA and NoAF = 167.0 fps flyby,54.5 fps botmatch vs 189.7,63.3 fps
-1280x960-
4xAA and 8xAF = 95.2 fps flyby,49.4fps botmatch vs 95.1,45.7 fps
noAA and noAF = 156.2 fps flyby,54.2 fps botmatch vs 168.8,63.2 fps
-1600x1200-
4xAA and 8xAF = 50.2 fps flyby,24.8 fps botmatch vs 60.5,30.1 fps
noAA and noAF = 120.7 fps flyby,53.6 fps botmatch vs 121.8,60.8 fps
(All tests done with driver sliders all set to "high quality")

I swear to god I havent altered any of these scores, and I ran the 1280x960 several times after rebooting, letting comp cool down etc. but it stays the same... Isnt that like the wierdest thing you've ever seen? LOL this scared the [-peep-] outta me though cuz 1280x960AA+AF was beginning to become the standard for me since thats what I was playing in usually...



-"Shoot coward, you will only kill a man"-

 

[Crashman]

Strange behavior indeed! At least you got a gain in MOST benchmarks!

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[endyen]

What I find most strange is .4 more fps in 1280 than 1024. Must be a counter error. But huury and get the other stick of ram. I got about a 3% increase in most benchies. Not great, but a nice extra just the same.

 

[SecretIan]

u only got 3% increase? What kind of system u got? Similar to mine?

-"Shoot coward, you will only kill a man"-