• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

My 4870 Benchmarks o.k?

Born_2_Kill_83

Born_2_Kill_83

Born_2_Kill_83

Soldato
Joined
10 Nov 2006
Posts
8,578
Location
Lincolnshire
Q6600 @ 3.6 4870 (stock)

3D Mark 06 - 15493

Q6600 @ 3.6 4870 (overclocked) 790/1100 - Is it possible to get these any higher? This is as high as the control panel will go!

3D Mark 06 - 16205

Just off to do Crysis next, but do these look o.k for one 4870?? They seem a little slow to me??
My 8800GTX overclocked used to pull 14200!!

Will be putting my 2nd 4870 in tomorrow, how much extra can I except off that?
 
Crysis @ 1680X1050:

All High No AA - 43FPS

All High 4XAA - 35FPS

All Very High No AA - 27.8 FPS

All Very High 4XAA - 23FPS

Hoping a 2nd card will make some difference to the above, question is how much?
 
3d mark 06 does not scale well with the new gen of graphics cards and its very cpu dependent. try vantage for more accurate results.
 
As said 3dmark06 not a good comparison of the new card's capabilities.

My GTS scores 16583 with my quad and I don't expect my new 4870 will score as high.

However, comparing all my game framerates with reviews on the 4870 shows the 4870 giving faster framerates and pulling way ahead once you put 4x and 8x aa on

At the end of the day which would you rather have? A higher 3dmark06 and worse performance in games?
 
Greebo said:
At the end of the day which would you rather have? A higher 3dmark06 and worse performance in games?
Click to expand...

Thats fair enough I agree, but after all it is a benchmark to compare with other cards if nothing else. Ive seen 4850 OC pull 17k and CF 4870 22k.

I'm just concern that my new card/s aren't up to speed, if they are then I'm happy, going to put another 4870 in tonight.

My overclocked 8800GTX used to pull 14200 ([email protected]) so I guess its not all bad. Plus my Crysis results look o.k?
 
I only get 24.7fps on crysis (16580x1050 - Very high 16AF / 0AA ) with xfire 4850's and a Q6600 @ 3.8Ghz which is lower than your single 4870. This makes me believe xfire doesnt work on it.
 
run the fill rate test in 3dmark 01se and if its the same with crossfire enabled as it is disabled then your having problems.

I was having that problem when i started running crossfire, took a while to work out what was happening too. It was'nt enabling the default card (looping throught it like a gateway) its got to do with the way you set up you default monitor settings, i found some posts by kinc which back this up.

recon 24k is doable with a 24.7 crossfire setup (no lod, tweaks ect).

As for which is better benchmarks vrs games, we use benchmark tests to make sure that we have are chosen setups working to the best of there ability so when we want to run a game we know the graphics cards are giving it everything they have got so we get the best gaming experience.
 
13.4k in mark06 with my e6400 at 3.2ghz 2gb ram win xp. Got a 2k bump over my old hd 2900xt, good enough, cpu holding me back me thinks. I will begin pondering a cpu/mobo/ram upgrade.
 
hmmm i get like 18k on 3dmark 06 on vistax64, [email protected] and 4gb ram, which seems wrong I believe, but then run 3dmark vantage and score a tad over 13k which seems good:/ the drivers are iffy anyway especially on x64 systems.
Running dual 4870's btw...
 
Last edited:
just to show 20k
3dmark06.jpg
 
Sorry I may have came accross wrong, but I think you misunderstand.

In your pic the resolution says 1024x768 and not 1280x1024 the default resolution... so I'm asking what did you run it at?

Also for a more accurate result(if you want to call 3dmark accurate^^) enable all the tests, as that is how must people run it... you running at 1024x768 and 6/19 tests is not a fair measure against me running at 1280x1024 and 19/19 tests... and I believe does not help the OP gear his mark to see if there is a problem.

or maybe I'm wrong I dunno :)
 
Last edited:
Well the number of tests shouldnt matter since only those 6 contribute towards the score. But the resolution being lowered has a fairly big effect and is an unfair comparison.

NB You havent actually broken 20k with those 2900XTs since everyone refers to 1280*1024 unless they state otherwise.
 
Last edited:
3dmark06 is not good at all for showing graphics cards now.

its totally MHZ driven even a dual core at 4.5ghz will score higher than a quad at slightly slower speeds.

Get 3dmark vantage that will show you the real difference between them.
 
You must log in or register to reply here.
Back
Top Bottom