*** The 100% Complete 3DMark05 Forceware Comparison ***

God damnit....I'm so ridiculously bored and not a single person I know is willing to go out.....I guess I could get started on the Aquamark3 scaling analysis. ;)
 
oozish said:
I'm bored too, can ya tell? ;)
Click to expand...

No, not at all....I mean, we're only essentially having an MSN convo on the forums.

K guys, I'm out.....I might even just stay up and get all the 6800GT tests done tonight, and do the ones with teh card at ultra clocks tomorrow. Might not put up a thread until it's done this time though. :cool:

cya in a bit.
 
oozish said:
Two things I would like to say:

1. Have you tried bumping up the resolution in 3dmark? I notice that the framerates hardly change from 1024x768 to 16x12 (and in between). This leads me to believe that this bench is already cpu limited with the GT for the most part. If you have the full ver. try bumping it up.



[/URL]
Click to expand...


I only gain 100 pts or less by going from 2ghz stock, to 2.5ghz, on my A64. It seems 3DMark2005 is GPU limited, not cpu limited...

I got like 4919 stock speeds, and only 5001 overclocked. I see a bigger delta in actual games than that crappy bench...
 
TheRapture said:
I got like 4919 stock speeds, and only 5001 overclocked. I see a bigger delta in actual games than that crappy bench...
Click to expand...

That's interesting... I don't wanna turn this into a discussion of 3DMark's worth but a lot of people that claim this is a useful benchie point toward it's future outlook on technology, the fact that it's meant to stress today's hardware like a game that's a year or two years away from now would...

Now, that's actually been the trend so far, games these days are more and more GPU dependant than ever, look at D3... But you'd think at some points dev would realize thare's gotta be something they can do with all these idle cycles no? :eek: <Impulse stops talking to himself>
 
TheRapture said:
I only gain 100 pts or less by going from 2ghz stock, to 2.5ghz, on my A64.
Click to expand...

Funny, you gained 100 points or less from a 500Mhz overclock.....whereas a 700Mhz overclock on my P4 (bumping it up to 3.5GHz) nets me close to 500 points, and the gain is well over 500 points when I'm up to 3.7GHz. I know that 500 points is still a relatively small gain for a 900MHz overclock, but 3DMark05 IS very gpu intensive, so it's no wonder that the system overclock of 900MHz doesn't play more of a role.

Man, it's almost getting to the point where it's like overclocking your video card to try and get better SuperPi 1m times.....sad. ;) :p

But yeah, I'd appreciate the thread not being turned into a general 3DMark05 discussion, I just had to say that. :)
 
Well, the P4 is much more clock-dependant than A64s... That's a big discrepany in gain from OC tho, interesting.
 
I suppose I could do the XT-PE cause mines stable in 3dmark05. But maybe somebody with a more modern cpu and without non generic ram running at crappy timings should do it :p
 
9mmx19 said:
Wow man crazy - you seem to have quite a bit of free time haha
Click to expand...

A little bit to much sometimes.....but other times I'm quite busy and I really dont have time to sit and think about what I'm doing in this time off. :p
 
Running the system in my sig ([email protected]) I get ~5400 with my GT at 430/1.15. The bump in score over what you're showing must be entirely due to the GPU.

If this is truly indicitive of games to come, then I don't see a point in upgrading my CPU anymore (a long time anyway) since it's plenty fast now and the margin of increase per dollar spent on mb/CPU is really small.
 
Well I'm getting an FX-53 system on Wednesday with a VIVO -> XT that can do 520 / 560 easily. Maybe if I get bored I'll do the ATI version :p
 
Wow, I was sure surprised to see this make the front page again......thx to all those who have supported me.

Note: I'm almost finished the Aquamark3 6800 scaling tests already....so I should have the thread for that posted up some time today.

EDIT: Guys, I just got an email back from clockgen's guys and the CG-ICS952607 version actually works just fine, and I've got it up and running now.....works great, and I'm gonna be off to finish the Aquamark3 testing pretty soon, because I'm done the GT clock testing, and half done the Ultra clock tests. :)
 
Why doesn't the Vcard manufacturers do this for us?
Hey I got a great idea, If The [H]ardCorn had all the diferent versions of the 6800Gt's we could see which manufacturer had the best card!

Probably never see him again for a long time :D

You have served us well. !
Thankyou from the bottom of my [H]arddrive!
 
OPUS1 said:
Why doesn't the Vcard manufacturers do this for us?
Hey I got a great idea, If The [H]ardCorn had all the diferent versions of the 6800Gt's we could see which manufacturer had the best card!

Probably never see him again for a long time :D

You have served us well. !
Thankyou from the bottom of my [H]arddrive!
Click to expand...

You're right, I might not be around for a good while. I would really like to be able to get onboard with a review site or start my own to write articles and hardware reviews, I mean.....what am I already spending most of my time doing? :p

[H]ardCorn.....I like it. :cool:
 
OPUS1 said:
Hey I got a great idea, If The [H]ardCorn had all the diferent versions of the 6800Gt's we could see which manufacturer had the best card!
Click to expand...

Or you would see the reality that all the cards performed within 1% of each other as long as they have the same clocks and drivers.

Granted pre-overclocked cards like a BFG or GoldenSample will perform a little better but you can OC yourself to the same clocks.

Why dont people get it through their heads all (highend) videocards now-a-days are built using the same exact reference PCB, and the quality has to be very high to run 1Ghz memory clocks, etc. The only difference in overclocking comes from the GPU chips and memory chips, neither of which are produced by the card manufactures. They all buy the same chips from Nvidia, and they pretty much all buy the same ram. They all use the same BIOS, that whys you can flash any card with other manufacturer's bioses, etc.

THEY'RE ALL THE SAME for the 111th freakin time!!!!

Granted bundles and warranty's are different, but peformance is all the same, and overclocking is luck of the draw and nothing to do with the manufacturer.

Now motherboards, those are different, everyone comes up with their own PCB's, own bioses, etc... and manufacturers really matter. But videocards nope.... doesnt matter. Its like asking which Ford Dealership sells the best Ford cards.

Sorry OPUS, but it wasnt a great idea.... its actually kinda dumb. :D lol
 
Be nice chris....and I'm not entirely joking around either.....that is something that could have/should have been done via PM.
 
Hey cornelious, thanks for all the time well spent. Kinda feel bad now that I've been runnin the 65.73's since my last LAN party :p

I do have a request tho, could you label the WHQL drivers on your graph? Kinda like you listed the beta drivers? Thanks again.
 
Gargoyle_Hunter said:
I do have a request tho, could you label the WHQL drivers on your graph? Kinda like you listed the beta drivers? Thanks again.
Click to expand...

For sure bro.....thx for the idea. :)

EDIT: K, done. :)
 
Gargoyle_Hunter said:
I do have a request tho, could you label the WHQL drivers on your graph? Kinda like you listed the beta drivers? Thanks again.
Click to expand...

Now, that is a good idea! :)

yeah, I might have been a little too harsh, but my post is like a public service anouncement.... helping him out. b/c its the facts.

BTW, can someone tell me which Ford Dealer ship sells the fastest Mustangs. B/c I want to buy one and I dont know who sells the fastest ones. :p
 
chrisf6969 said:
THEY'RE ALL THE SAME for the 111th freakin time!!!!

Sorry OPUS, but it wasnt a great idea.... its actually kinda dumb. :D lol
Click to expand...

Darn You just took away The [H]ardCorn's next assignment
I guess I should a put :rolleyes: behind it so you could see a joke (albeit a bad joke)
just an aside:
But while we are on the topic, You go out and buy x amount of dollars worth of anything, ie cameras,sterieos,computers and they will all pretty much have the same stuff inside!
Just different bells and whistles

Hey [H]ardCorn we will continue to search for your next assignment, dumb or smart :D
 
OPUS1 said:
Hey [H]ardCorn we will continue to search for your next assignment, dumb or smart :D
Click to expand...

Good stuff guys, let me know if there's something that you wanna see done, and as long as it'd within reason I'll give it a go. :) :cool:
 
Just wanted to say one more thing, I think I've decided on my next project.....

I have already tested out the scaling of my Aquamark3 scores at these two different video card clocks, so I figure what's next is for me to run 3DMark2001, 3DMark03, and 3DMark05 in the same manner.....to see how each benchmark "deals" with the two video card setups and scaling CPU.

What do you guys think? Interested in the lengthy results? :p
 
How 'bout something 'sides 3DMark... :p I don't much see the point of testing with an older version of 3DMark, sorry.
 
Impulse said:
How 'bout something 'sides 3DMark... :p I don't much see the point of testing with an older version of 3DMark, sorry.
Click to expand...

The reason i was going to use 'em is because I wanted to do all the testing in my benchmark partition so that some tests aren't with a whole lot extra running in teh background. I guess I could just re-do everything in my main partition and include Doom3 and Unreal2k4 eh?.......you might NOT see me for awhile, but woudl the full 3DMark2001, 3DMark03, 3DMark05, Aquamark3, Unreal2k4, Doom3 scaling analysis thread be worth it? :p
 
Or just restore the testing partition then create a "super testing partition" with a few more programs! Heh, I'm just kidding, anything goes... What are you using to restore/create your partitions by the way? I was planning to get Ghost when I build my new rig (Dantz is starting to aggravate me even tho it's very versatile, but it doesn't image, just does full backups).

Only hitch is I'm not sure if Ghost will create images thru the network (for the other computer at home) or if it'll even run scheduled backups to my external 250gb, so I'd love to hear of other options based on experience.

Far Cry might be more valuable a test than the slightly less stressing UT too... 'till HL2 anyway. :D
 
I basically just used PartitionMagic to create a seperate 5GB partiiton to install XP on.....so that i can tweak the hell out of the OS to the point where "regualr" windows wouldnt function here and now....but so that i can still benhcmark just fine and get a nice little score boost.

I'm not doing any ghosting between tests/drivers/etc....just uninstalling the drivers and using driver cleaner in safe mode to make sure it's all cleared up.....just how I've got it outlined here.

I might give FarCry a shot....except I dont own the game, and I dont think I've got it burnt either.......Unreal might be a bit more dated, but It'll still be somewhat representative of current D3D performance. I figure that Unreal2k4 is something that ppl with GeForce3's and stuff can still play very well (i know, my bro's comp has one in it....runs just fine) and by using Unreal2k4 instead of FarCry I'll be possibly helping out a wider range of people, because there are WAY more ppl with systems that are more then capable of playing Unreal2k4 that might be interested in an upgrade then there are ppl who are able to crank FarCry. ;)
 
CPU scaling isnt as interesting as a driver comparison for me.

I'd rather see you compare the drivers on 3dMark 2001 and 3dMark2003 and Doom3 (and maybe FarCry or UT or maybe even something really old like Q3)

Then see if the tweaks that they're working on in the drivers help other games or just 3dMark and Doom3 b/c thats what so many people seem concerned with. Or if they're sacrificing gaming performance to tweak their benchmark #'s.
 
chrisf6969 said:
CPU scaling isnt as interesting as a driver comparison for me.

I'd rather see you compare the drivers on 3dMark 2001 and 3dMark2003 and Doom3 (and maybe FarCry or UT or maybe even something really old like Q3)

Then see if the tweaks that they're working on in the drivers help other games or just 3dMark and Doom3 b/c thats what so many people seem concerned with. Or if they're sacrificing gaming performance to tweak their benchmark #'s.
Click to expand...

That is actually a very good point. While the driver testing would take a fair bit longer, would you guys prefer to see another full driver comparison like this, but with 3DMark2001, 3DMark03, 3DMark05, Aquamark3, Doom3, Unreal2k4, and possibly Quake3?

I'd have to do this comparison using my main windows partition to do the game testing, but that's alright, I'm just not sure what kind of pace I would try to complete it at.....probly one driver a day on the 5 days a week I'm working, and 2 drivers a day on the 2 days a week I have off.

What do you think?
 
You must log in or register to reply here.
Back
Top