• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

What's the best benchmark?

P

pigpen

2[H]4U
Joined
Aug 17, 2004
Messages
2,408
Just curious what general sentiment on this board is. 3Dmark? 01? 03? Aquamark? PCMark? What's the best?
 
For me, the best benchmark is running actual games and getting the average fps for a timedemo. This is more representative of real work performance.

Synthetic benchmarks like 3dMark family is stressing parts which is not necessarily the ones who is used in games.
 
I'm not really concerned with eye candy. I just want something that will give me a good overall representation. Meaning, i'd like a program that can run a nice quick baseline, and then give a good view of how much effect an OC has had.
 
meglamaniac said:
3DMark05 looks pretty stunning
Click to expand...
I actually think it looks horrible. I mean, it chugs along at slow-ass frame-rates in 1024x768 with no AA - yuk! I can't appreciate the eye-candy when the jaggies are sawing my eyeballs in half. I'm sure it'll look great on a decent SLi setup or next-gen GPU but for now it's pretty shocking.
 
coz said:
I actually think it looks horrible. I mean, it chugs along at slow-ass frame-rates in 1024x768 with no AA - yuk! I can't appreciate the eye-candy when the jaggies are sawing my eyeballs in half. I'm sure it'll look great on a decent SLi setup or next-gen GPU but for now it's pretty shocking.
Click to expand...

Agreed.. It looks pretty bad, even when run smoothly.
 
Any benchmark that will give you some numbers that you can compare with other similar systems...

Benchmarks are good for comparing to your own system, with regards to personal overclocks, etc...

As always, if you want a good idea of your machine's performance, use a lot of different benchmarks, both synthetic and game-based.

Keep in mind what kind of games you like to buy when deciding on a benchmark to base your purchase decision on. nVidia is good if you like Doom 3, but other types of games may favor ATI.

Personally, I think 3dMark2005 is fun. Its free, and the online compare utility is a quick way of seeing what other people are getting. If you are comparing against others with your exact same setup, it can provide a valuable tool in diagnosing where your personal setup may be lacking or incorrect.
 
3Dmark will benchmark your system vs 3Dmark, no relationship to testing your system vs Half-Life/Doom3-- completely different rendering engines/systems.

Play the games you want to play-- observe/or use in-game benchmarking to determine system performance vs your subjective workload.

Even saying that 3Dmark could be used to test driver-level performance, might be an over-statement.
 
coz said:
I actually think it looks horrible. I mean, it chugs along at slow-ass frame-rates in 1024x768 with no AA - yuk! I can't appreciate the eye-candy when the jaggies are sawing my eyeballs in half. I'm sure it'll look great on a decent SLi setup or next-gen GPU but for now it's pretty shocking.
Click to expand...

just becuase it runs like shit on your computer dosnt mean it looks like shit....the benchmark is uber stunning.....the graphics are AMAZING..and who gives a shit..its the closest we have in the best graphics, games now a days dont have that kind of graphics!...we are nowhere near that level of stunning graphics..and to say they look horrible because your computer cant play it...im guessing your a liberal
 
Brent_Justice said:
DOOM 3 and FarCry
Click to expand...

Yeah, Doom3 for Open GL since many future games will probably be based on that engine.

And FarCry for an intensive DX game. Though HL2 / CS:S may end up being a better one once its out.

And I like 3dMark2001 & 2003, also. 2005 kinda sucks... not enough tests.

I kinda like running Quake3 too, just b/c I've benched all of my systems with it, so it gives me a good historical perspective of where my performance is now compared to 3-4 years ago.

Like my old Quake 3 results:
400fps @ 2400Mhz vs 600fps @ 3600Mhz exactly 50% scaling with 50% cpu speed (plus other improvements - ram, vc, etc)
 
Warrior said:
just becuase it runs like shit on your computer dosnt mean it looks like shit....the benchmark is uber stunning.....the graphics are AMAZING..and who gives a shit..its the closest we have in the best graphics, games now a days dont have that kind of graphics!...we are nowhere near that level of stunning graphics..and to say they look horrible because your computer cant play it...im guessing your a liberal
Click to expand...

I'm going have to disagree. While it does look good, it doesn't look that good. That just might be me though. I've seen it in decent framerate too and i'll have to pass. I uninstalled that lousy benchmark seeing as they don't even give you all the scenes anymore. I did notice they lowered the price of that thing though. Probably because no one is buying it. I will agree with Brent though, Farcy and Doom 3 are the benchmarks of choice for most around here. I use D3 as mine.
 
Warrior said:
just becuase it runs like shit on your computer dosnt mean it looks like shit....the benchmark is uber stunning.....the graphics are AMAZING..and who gives a shit..its the closest we have in the best graphics, games now a days dont have that kind of graphics!...we are nowhere near that level of stunning graphics..and to say they look horrible because your computer cant play it...im guessing your a liberal
Click to expand...
games don't have graphics like that because there based off of REAL gaming engines. futuremark just creates these benchmarks with way to many lights sources, over reflective objects, etc. this doesn't look good or behave realisticly. It may look good (edit: to some :p ), but it just doesn't mean crap when it comes to real performance. just because a 6800U may chug at times doesn't mean that its a $500 piece of crap.
 
bboynitrous said:
. I uninstalled that lousy benchmark seeing as they don't even give you all the scenes anymore. .
Click to expand...
I can't run it at all, I have a GeForce 4 ti4400...

How many and what type of scenes do they cut out? In 3dmark2001 you could run every scene, but just not play the "video games" themselves...i thought...

So the benchmark isn't complete unless you actually pay for it?
 
Oops, seem's I've started an argument...!
Let me clarify my position a bit.

3DMark in my opinion, as I mentioned in the first post, is good for showing off your GFX capabilities for the "wow, pretty" factor, and that's about it. It's occasionally useful to compare scores with systems similar to yours in order to see where you could get a bit of extra performance, but that's the limit. It's no good saying "this game will run fine if you have <x> 3DMarks" because <x> could be arrived at on wildly different setups (eg. great video card with crap CPU, vs crap video card with great CPU).

For actual meaningful performance tests stick to Doom3 playdemos and FarCry, as they're the best real world stress tests available for OGL and DX respectively.

To sum up:
3DMark to impress your mates with the snazzy graphics your machine can do.
Games to give you a meaningful result.
 
Warrior said:
im guessing your a liberal
Click to expand...
I'm guessing you forgot to take your medication. :)

Yeah, it does run like shit on my computer but it runs like shit on everyone's computer! For a 30 fps average frame-rate in 3DM05 you'd need to be scoring 7500, 60 fps would be a 15,000 score. So how many folks here get playback that's as smooth as buttah?? Are you really that impressed watching those amazing graphics fly past your screen at 5fps with jaggies you could impale yourself upon? Personally I prefer fast frame-rates with high-resolution, anti-aliasing and all that stuff, but that's just me.

But the thing I really hate about 3DM05 is that no matter how much I use the strafe left & right keys the guy with the big gun never budges an inch. :)
 
You must log in or register to reply here.
Back
Top