Hd5850 vs. hd5870

[DTV DRAGON]

Do you want me to throw in my 5850 test results that I've got so far? If I could find a program that would allow me to change the voltage on it (I have the toxic and someone said it might be just different enough that programs can't access the voltage) I'd be able to get well over 950/1350 which is where I'm sitting on 1.088 volts. It could make it even more interesting to see if the 5850 toxic could outclock a 5870 for 60 bucks cheaper

sure that would be great! post anything you like, the more numbers we have the better, I didnt get around to anymore benching today as it turned out to be a tuff day,

 

[jasper1605]

K, I'll have my replacement RAM in tomorrow so I should be able to get some numbers out tomorrow afternoon/evening.

 

[DTV DRAGON]

ok here are the gpu scores for vantage on performance
hd5850
gpu-16829
hd5870
gpu-19110

Ok Ive done a few comparisons between my old 5850 sapphfire and my new 5870 sapphire, and I must say I realy wanted to see the 5870 kill the 5850, you can see that it is clearly faster in pretty well everything but it did not dominate it, I did notice that it jumped ahead alot when there was intensive graphics going on, It as well was alot cooler than my 5850 was when they were both o/ced quite abit, was it worth that extra $$ ?for me I will say yes, as Iam always looking for something faster everytime, but I realy wanted to see the 5870 pull ahead by a longshot, but it didnt. I cant speak for the 5850 toxic's or vapors, but only about my sapphire 5850 and my sapphire 5870 They are both killer cards in my books, and either one would be a great choice!
hmm I wonder how much I can sell this 5870 for so I can get a 5970??? I would love to test this all over again lol

 

[PaulieG]

I have seen no discernible differences b/t my old Asus 5870 and my new XFX 5850 Black. Clock for clock, I see maybe a 500 pt difference in Vantage, and while gaming I found MAYBE a 2-3 fps difference at comparable clock speeds. For me, the $100 difference in price is not worth it.

 

[wolf]

I have seen no discernible differences b/t my old Asus 5870 and my new XFX 5850 Black. Clock for clock, I see maybe a 500 pt difference in Vantage, and while gaming I found MAYBE a 2-3 fps difference at comparable clock speeds. For me, the $100 difference in price is not worth it.

Owing both now I can clearly say that a 5850 is easily better value IF you intend to overclock at all.

 

[DTV DRAGON]

yes I agree the 5850 is by far a better value, but i still think the 5870 has a slight edge on performance, however small it may be

 

[wolf]

yes I agree the 5850 is by far a better value, but i still think the 5870 has a slight edge on performance, however small it may be

no doubt is has an egde, mine together at 950/1250, when the heat is on, the 5850 is @ 99% gpu usage and the 5870 is @ 95-97%, so there is extra power there.

 

[DTV DRAGON]

no doubt is has an egde, mine together at 950/1250, when the heat is on, the 5850 is @ 99% gpu usage and the 5870 is @ 95-97%, so there is extra power there.

exactly what I was trying to say!!

 

[Sasqui]

Like to see each card @900/1000 run Furmark, clock for clock comparison. Do the extra shaders help with OpenGL?

 

[DTV DRAGON]

I think the extra shaders do help, I kinda compare these two cards to a smallblock and bigblock chevy, The small block will keep up to the bigblock for a good while until you open the 4 barrles up on the bigblock and thats when you see it pull away, It just has that little extra under the hood lol
I found this at another site and copied it but cant remember where one other point*** my WEI went froma 7.7 with 5850 to a 7.8 with 5870, not that it realy means anything significant

 

[jasper1605]

Here's what my CPU will be at for my tests. I also can't seem to get furmark to run the test with msaa at 0. It's always setting itself to 4x, which defeats comparison purposes again. Anyone have any ideas where I need to change things at to let it drop to 0?

Edit: Nevermind, it has fixed itself

 

[jasper1605]

factory OC vs my OC on sapphire 5850 toxic

Factory OC: 4547 at 1920X1080 no MSAA for 60 seconds
MY OC at 965 core 1370 mem on 1.088v: 5574 at 1920X1080 no MSAA for 60 seconds.

I wish I had a true 1920X1080 monitor though Also note these were run in fullscreen mode.

I'm curious as to why they're so low compared to DTV's?

 

[jasper1605]

full screen DX 11 at 938 core 1335 mem scored 1004 on heaven

 

[DTV DRAGON]

yeah something isnt right , mine were run in fullscreen mode as well, any thoughts?

 

[DTV DRAGON]

maybe you have it o/ced to high as with these cards if you go to high you actually start to lose performance.. just a thought

 

[erocker]

Factory OC: 4547 at 1920X1080 no MSAA for 60 seconds
MY OC at 965 core 1370 mem on 1.088v: 5574 at 1920X1080 no MSAA for 60 seconds.

I wish I had a true 1920X1080 monitor though Also note these were run in fullscreen mode.

I'm curious as to why they're so low compared to DTV's?

Unstable RAM, that's why it slowed down. With these cards when you OC the RAM it's not going to shut down or artifact on you if you push it too far, it will instead throttle performance. Lower your RAM clocks.

 

[DTV DRAGON]

I think we hit the nail on the head erocker!

 

[jlewis02]

ok here are the gpu scores for vantage on performance
hd5850
gpu-16829
hd5870
gpu-19110

Ok Ive done a few comparisons between my old 5850 sapphfire and my new 5870 sapphire, and I must say I realy wanted to see the 5870 kill the 5850, you can see that it is clearly faster in pretty well everything but it did not dominate it, I did notice that it jumped ahead alot when there was intensive graphics going on, It as well was alot cooler than my 5850 was when they were both o/ced quite abit, was it worth that extra $$ ?for me I will say yes, as Iam always looking for something faster everytime, but I realy wanted to see the 5870 pull ahead by a longshot, but it didnt. I cant speak for the 5850 toxic's or vapors, but only about my sapphire 5850 and my sapphire 5870 They are both killer cards in my books, and either one would be a great choice!
hmm I wonder how much I can sell this 5870 for so I can get a 5970??? I would love to test this all over again lol

Could you post a compare link for the 5850 and the clocks it was at?

 

[jasper1605]

looks like I'll be tinkering with the ram speeds then to find the best performance on it. Thanks a bundle erocker

edit: why would my stock OC's still be insanely low? That doesn't make any sense either, aren't the factory OC's supposed to be an increase in performance not a decrease?

 

[jlewis02]

http://service.futuremark.com/compare?3dmv=1891584 ok this is 5850 on E
http://service.futuremark.com/compare?3dmv=1847535 and 5850 on P

both were done at clocks of 898/1195
hope it helps and this is what you were asking for

Thanx
I hope to see scores like that when I get my 5850.
My GTX280 only gets 13.4k without physx and 15.9k with physx Overclocked to its max.

On a side note I got ya by 900 points on the CPU score.
http://service.futuremark.com/compare?3dmv=1912453

 

[DTV DRAGON]

yeah Ive been meaning to up my cpu to 4.0, but never got around to it

 

[jlewis02]

I can't wait till my card gets here now.
Thanx for posting this you made me get it lol.

 

[DTV DRAGON]

I can't wait till my card gets here now.
Thanx for posting this you made me get it lol.

iam sure you are very upset with me right now lol

 

[jlewis02]

Not really I have been looking aa the new Ati cards for a while now.Was waiting for more benchmarks and comparesons before I spent the cash.
I ordered an XFX 5850 Black Edition for $319 shipped.
I hope it performs well.

 

[DTV DRAGON]

Iam sure it will be great I came from a gtx285 oc2 that I thought was good until i went ati and got the 5850 and it blew it away bigtime, and I was a nvidia guy for a long time