• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

STICKY: ATi 5xxx pre-release thread

Page 22 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
At 1680x1050 the GTX285 and the 5870 would probably be about equal. But at 1920x1200 I think the 5870 would cream the GTX285....Wait how long is the 5870 supposed to be? Is it longer than a GTX285 or the same? Because the GTX285 in my case (Antec 180B) barely fits..
 
I think it is probably the same length as 285 which is 10.5". Or maybe 10". Either way it sticks out of a typical ATX motherboard.
 
well, if nvidia is already making a 10.5" board then it would make sense for 5870 to also use it. most high end users are conditioned to it by now.
 
Originally posted by: StinkyPinky
The only thing that worries me is the lack of physx. Nvidia are pushing that pretty hard.
Click to expand...

yes, and as we all know, Nvidia's $129 card outperforms the 5870 because it can do real world physix calculations to take advantage of today's most demanding games 😉

I find it telling that somebody who actually HAS one of these cards and is testing it right now seems to think that these slides are BS.

@dflynchimp: gtx 280 eventually surpassed 9800gx2 because nvidia focused most of their driver development on improving the new card, not because it was legitimately a FASTER card. Ati does it, too, they are both completely focused on benchmarks at release (and benchmarks when the competitor has a release). Once a card is eol, if you own it you're sol.
 
Luckicly I have a 9800 GX2... I'm good to go for an upgrade. I like PhysX and would rather have a NV card, but I also hate NV for pimping it.

I sure hope the tops of those cards are open (not encased all around) to help with heat. And I'm not too pleased with their design. Stuffing all those ports on there and giving it a half vent for exhaust...
 
I know. I am nitpicking here. It doesn't cost me. 😛

P.S. Holy Poodle! @the encyclopedic post @XS. Good job.
 
Originally posted by: happy medium
Originally posted by: Shaq
I don't like whoever is doing this. They cherry pick where the 5870 beats the 295 and highlight and ignore the other settings where it loses. It makes me question their veracity/accuracy. For instance Stalker:CS 1920 with 4AA is on the bottom and not 2560 becuase the 295 actually ties at 2560 so you can't highlight that. They only highlight the most favorable 5870 settings-- Definitely a biased chart. Wolfenstein is even better. The 295 beats it at every resolution but 2560 with 8AA so that is what is highlighted. lol

The Xfire vs. 295 quad-SLI results don't surprise me as 4 cards is a lot less efficient than 2. The 295 competed very well against 4870 quad-fire in certain games and in bang vs. buck.

How long is the 5870x2? It has to be 11" at least.
Click to expand...

5870 is 10.46 inches and the 5870x2 is 11.66 inches
Click to expand...


Okey My HD4870X2 is about 10.5 and your saying its a 1 inch longer?

When i first got my HD4870X2 i thought DAYAM... thats a long card, and when i threw in the second one, i had to move my water stuff.

But if u guys want a gross approx, my HD4870X2 is longer then my P6T mother board and then some when mounted.

Infact its longer then my motherboard backplate:

http://i125.photobucket.com/al...aigomorla/IMG_1270.jpg

So yes your screwed on a standard PC case.
 
The current rumor is that ATI is able to launch a big-ish chip on the problematic 40nm process because a large part of it is currently disabled. There's a supposed die shot circulating around the net that people are counting 2560SPs on. Looks like it originally came from chiphell, which has given us a lot of solid RV870 info but has also posted a fake die shot in the past. If true, ATI might be able to take on G300 with a single die solution in a future HD5890 refresh once yield improves.

Edit: Nevermind, fake
 

Originally posted by: zagood

Even so, take a closer look at the 5870 v 295 image, ignoring the gray highlights, you'll see much more balanced results with 5870 in general pulling ahead at max res. but that's about it.

Hypothetically, if GT300 has as much of a performance increase over GT200 as 580xx has over 48xx, then ATi is in trouble if they don't price it right and/or launch far enough ahead of nV.
Click to expand...

One single GPU that performs just as fast or faster then the fastest last generation dual GPU card! What, did you expected? 50% more performance on a single GPU 5870 over a dual GPU GTX 295? I think it's plenty fast, just as everyone expected to be and just as it needs to be. And we all know that ATi will probably have a crappy Catalyst at the beginning. Things will further improve with newer drivers.

5850 is faster and probably cheaper then gtx 285, so, from the day ATi releases these cards, Nvidia won't have anything to sell. GT300 needs to appear as soon as possible.

 
Wow. If these benches are true I think I will bite the bullet. Im on a GTX 260 right now so this might be a nice upgrade. :beer:
 
Originally posted by: Bateluer
Oh, Qbah, your statement about the 2900XT's performance is inaccurate. I had both cards and the 2900XT handily bested the 1950XTX in DX9, more so once its drivers were refined.
Click to expand...

True, but it wasn't anything "generation class" - especially when one ran games with AA or compared it to recent performance jumps (not to mention the 8-series completely slaughtered the 7-series and there were huge expectations from the HD2900 series). And the fact that the card improved with newer drivers was mostly due to the fact, that they stopped improving on the R580 core card drivers. It is kinda normal to try and squeeze the most from your newest cards though. Same can be said about the 9800GX2 🙂

But that doesn't change the fact that the HD48x0 did that in hardware alone and so will most likely the HD58x0 (as we're still in unofficial status and you never know 😉). Can't wait for Anand and Xbitlabs to do their testing and give us proper numbers 😀
 
Graphs with the line starting at 0.8 should be agaisnt the law :| although they arent actually lying, its just so cheap

"ZOMG the card is 10x faster!!"
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top