AMD's Radeon HD 5870: Bringing About the Next Generation Of GPUs
by Ryan Smith on September 23, 2009 9:00 AM EST- Posted in
- GPUs
Angle-Independent Anisotropic Filtering At Last
For a number of years now the quality of anisotropic filtering has been slowly improving. Early implementations from AMD and NVIDIA were highly angle-dependent, resulting in a limited improvement to image quality from such filtering. The angle-dependent nature lead to shimmering and other artifacting that was not ideal.
As of the previous generation of cards, the quality of anisotropic filtering had become pretty good. NVIDIA’s best filtering mode was pretty close to angle-independent, and AMD’s only slightly worse. Neither was perfect, but neither was bad either.
The Radeon HD 4890
The GeForce GTX 285
However so long as no one had an angle-independent implementation, there was room to improve. And AMD has gone there. The anisotropic filtering algorithm used by the 5000 series is now truly and completely angle-independent. There are no more filtering tricks being used.
The Radeon HD 5870: Perfection
As you can see, the MIP maps in our venerable D3D AF Tester are perfectly circular, the hallmark of an angle-independent implementation. With angle-independent filtering, this effectively marks the end of the filtering arms race. AMD has won, and should NVIDIA catch up in the future the two would merely be tied. There’s nowhere left to go for quality beyond angle-independent filtering at the moment.
AMD tells us that there is no performance hit with their new algorithm compared to their old one. This is a bit hard to test since we can’t enable the old algorithm on the 5870, but certainly whatever performance hit there is, is similarly minor. In all of the testing we’re doing today, you will see results done with 16x anisotropic filtering used.
What you won’t see however is a difference, particularly with our static screenshots. When discussing the matter, AMD noted that the difference in perceived quality between the old algorithm and the new one was practically the same. After looking at matters we find ourselves in agreement with AMD; we were not able to come up with any situations where there was a noticeable difference, beyond the obvious AF quality tests that are designed to identify such changes.
Regardless of the outcome, AMD deserves kudos for making angle-independent anisotropic filtering happen. It’s demonstrably perfect filtering with no speed hit versus the previous generation of filtering; making it in essence a “free” improvement in image quality, however slight the real-world results are. We’re always ready to get better image quality out of our video cards, after all.
Facebook
Twitter
RSS
327 Comments
View All Comments
erple2 - Wednesday, September 23, 2009 - link
I think that you're missing the point. AMD appeared to want the part to be small enough to maximize the number of gpu's generated per wafer. They had their own internal idea of how to get a good yield from the 40nm wafers.It appears to be similar to their line of thinking with the 4870 launch (see http://www.anandtech.com/video/showdoc.aspx?i=3469">http://www.anandtech.com/video/showdoc.aspx?i=3469 for more information) - they didn't feel like they needed to get the biggest, fastest, most power hungry part to compete well. It turns out that with the 5870, they have that, at least until we see what Nvidia comes out with the G300.
It turns out that performance really isn't all people care about - otherwise nobody would run anything other than dual GTX285's in SLI. People care about performance __at a particular price point__. ATI is trying to grab that particular sweet spot - be able to take the performance crown for a particular price range. They would probably be able to make a gargantuan low-yield, high power monster that would decimate everything currently available (crossfire/SLI or single), but that chip would be massively expensive to produce, and surprisingly, be a poor Return on Investment.
So the comment that Cypress is "too big" I think really is apropos. I think that AMD would have been able to launch the 5870 at the $299 price point of the 4870 only if the die had been significantly smaller (around the same size as the 4870). THAT would have been an amazing bang-for-buck card, I believe.
Doormat - Wednesday, September 23, 2009 - link
[Big Chart] and suchfaxon - Wednesday, September 23, 2009 - link
page 15 is missing its charts guys! look at it, how did that happen lmaoGary Key - Wednesday, September 23, 2009 - link
Ryan is updating the page now. He should be finished up shortly. We had a lot of images that needed to be displayed in a different manner at the last minute.Totally - Wednesday, September 23, 2009 - link
the images are missingdguy6789 - Wednesday, September 23, 2009 - link
You very clearly fail to mention that the cheapest GTX295 one can buy is nearly $100 more expensive than the HD 5870.Ryan Smith - Wednesday, September 23, 2009 - link
In my own defense, when I wrote that paragraph Newegg's cheapest brand-new GTX 295 was only $409. They've been playing price games...SiliconDoc - Friday, September 25, 2009 - link
That "price game" is because the 5870 is rather DISAPPOINTING when compared to the GTX295.I guess that means ATI "blew the competition" this time, huh, and NVidia is going to get more money for their better GTX295.
LOL
That's a *scowl* "new egg price game" for red fans.
Thanks ATI for making NVidia more money !
strikeback03 - Wednesday, September 23, 2009 - link
lol, did they drop the price while they had 5870s in stock, then raise it again once they were gone?SiliconDoc - Wednesday, September 23, 2009 - link
Oh, so sorry, 1:46pm, NO 5870's available at the egg...I guess they sold 1 powercolor and one asus...
http://www.newegg.com/Product/ProductList.aspx?Sub...">http://www.newegg.com/Product/ProductLi...1&na...
---
Come on anandtech workers, you can say it "PAPER LUANCH !"