do you think a single 4870/gddr5 will outperform 3870 CF?

hmmmmmmm???

 

No.

why are you looking at this it's white so it's hard to see

 

To be honest it shouldnt be that big difference.

A single card gets around 8-9k in 3dmark so we can assume that the 4870 will get around 10-11k.

Depending in the cpu usually the 3870's in crossfire get around 11-13k (not counting quads). So they seem to be pretty close.

 

little off topic but i was using matts custom black theme for NBR and i could not figure out what you ment because it showed up the same color....so i switched to opera man i felt stupid

 

The Crossfire will win the pointless 3Dmark battle, the 4870 will compete very well in the real benchmarks.

 

same cpu and same everything else, do you guys think 4870 with t9400 will get 11k 06 3dmark? cuz 3870cf can get that score

 

3dmark 06 is a snythetic bench mark and in terms of real gaming it doesn't matter tweak it for the game not 3dmark06. When they come out i will be able to tell you eaxctly how good they are please be pateint and just know the 4870's will be better ?!?!?!?!

obviously if a card with a slower clock speed and memory can score that then why in the world would a faster one score lower?

EDIT thats what i ment to put

 

Come on Mr. Moo. Settle down.

4870s have already been extensively benchmarked using the W90vp-X1, and has shown some great overclocking ability. It has broken 20K (albeit with a massive CPU OC), 17K and 14K stock with a T9400, with CF 4870s.

How do you know that it won't get 11K? A 9800M GTX coupled with a C2D can easily reach that with a slight OC, and the 4870s are supposedly better.

And what do you mean by "3DMark06 is a benchmark"? Were you trying to be smart? Did you mean "3DMark06 is a synthetic benchmark that should not be used as a full guide on a GPU's performance"?

 

yes sorry plasma i got mad and wasn't thinking... its a synthetic bench mark to test the *saposed capabilitys of a card in gaming"

lol unno what i ment. the 4870 will obviously get higher then the single 3870 and also maybe the xfire 3870 sinces they are easy to OC(hypothetically here) and over all a much better card

plasma ur right i was so mad at the time i just put every thing backwards lol

 

^^^^

My 4870 desktop outperforms or is at a dead heat with 9800M GTX's in SLI. A mobility 4870 with GDDR5 will likely be very, very close to Crossfire 3870's.

 

its also a 1gb version of the 4870 and and there mobile 9800m gtx...the 4870m will be underclocked as well, i am not sure but i think it will be like 80% of the desktop card. dont hold me to it because i cannot remeber where i read it ...

but the 4870's will be a step up thats what im trying to say... i just want true bech marks against other mobile cards.(not the ddr3 versions)

 

Hey Jl, have you tried testing your 4870 clocked down to 680mhz core and 700mhz memory? Those are the specs of the Mobility gddr5 reference model from AMD/ATi

 

SO basicaly the ATI Mobility HD4870 will be about equivallent to a desktop HD4850 then ?? If so not bad God Bless

 

doubtful, but not impossible.

 

I think the mobility 4870 will bench a little lower (in 3DMark06) than 3870CF...but I think gaming performance will be pretty close. And personally I always prefer a single-card solution if possible.

 

would you not also need to take into consideration shader models, stream proc units, and other differences?
i mean to be fair, just to downclock does not constitute an exact match.

 

actually, if you compare the specs of the two cards, they are exactly the same except for the clock speeds.

 

i tried, and kept getting net errors....so i thought i would post anyway....