• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

Question: Nvidia vs AMD shader count.

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

itsmydamnation

Platinum Member
Feb 6, 2011
2,247
1,911
136
Because I want to know why and how things are achieved. If Amd has better performance per sq mm, then why do they not have a single gpu card that can match the 580?
because they have performace targets, you know 580/G110 is around 550mm sq and cayman is around 380mm sq. AMD use a VLIW architexture NV uses some kind of MIMD desigin. its justs design trade offs. you then have things like bandwidth/memory MTU's and ROP's which make a big difference not just shaders.

so the big differenece yuo can see here is that VLIW relies on a good compiler for performance, MIMD relies on a good scheduler for good performance.
 
Last edited:

Borealis7

Platinum Member
Oct 19, 2006
2,811
147
106
rather than FPS/mm^2 i think you should look at FPS/Watt, while its true that engineers have to stay within reasonable limitations in both categories. (make a big chip = make less money. make a hot,power hungry chip = have to limit performance in other areas, spend more on cooling, etc)
the category you should be interested in is FPS/$ and FPS/Watt combined. it means how much are you getting for your buck (unless money is no object), and midrange cards are FPS/$ kings. i think the GTX 460 and 6850 offer the best value right now.
 

Arkadrel

Diamond Member
Oct 19, 2010
3,681
1
0
Because I want to know why and how things are achieved. If Amd has better performance per sq mm, then why do they not have a single gpu card that can match the 580?
Because costs rise exponetionally with bigger sq mm sizes.

Its called profits... they make smaller cheaper chips, that are more effecient.
AMD use 2 of these smaller cheaper chips, to fight nvidias bigger 1 chip's.

AMD uses more money on designing cards, to make them smaller/more effecient.
Then they earn back profits, by makeing the chips cheaper than Nvidia does.

Nvidia use less money on designing cards, they make them big to compensate.
They dont earn back profits, by makeing them smaller/faster, but they use less money on R&D.

They have differnt statagies as companies, thats all.
In the end all that matters is their profitable.
 
Last edited:

wahdangun

Golden Member
Feb 3, 2011
1,007
146
106
Because I want to know why and how things are achieved. If Amd has better performance per sq mm, then why do they not have a single gpu card that can match the 580?
because its easier and cheaper to make two small GPu and combine it than build big, hungry and hot GPU
 

tincart

Senior member
Apr 15, 2010
630
0
0
Because I want to know why and how things are achieved. If Amd has better performance per sq mm, then why do they not have a single gpu card that can match the 580?
Because they are not targeting that market. AMD's strategy involves making smaller chips, which cost less to manufacture and have higher yields (e.g. more chips per manufactured wafer are good if you are making smaller chips) and then using two of those chips to make an enthusiast class product (4870x2, 5970, 6990).

Presumably, AMD could throw a bunch more transistors on the Cayman architecture and come up with something to beat the GTX 580. The question someone in a business position would ask is "Sure we could, but why should we?" As die size increased, you run up against the law of diminishing returns. Your chips become more expensive to produce and you have a much smaller market of people who want to buy them. AMD's approach is simply to avoid the headaches and try to focus on spots just below the top single-chip.

It is a business decision. I imagine they think they can make more money by avoiding building the best single-GPU card than they can if do try to build it. Whether that is a good business decision is a different matter.
 

Idontcare

Elite Member
Oct 10, 1999
21,118
57
81
Folks, just a few reminders:

Posting in a thread to refer to a fellow forum colleague as a troll or to refer to their post as a "troll post" is a personal insult as well as a thread-derail. It is not acceptable.

If you feel a post violates the forum posting guidelines then you are expected to report the post as a matter of community stewardship and then leave the matter to the hands of your moderator staff without broadcasting the fact you reported the post/poster.

Likewise making a post to broadcast that you have put another member on your ignore list is a thread-derail as well as a personal insult. You have the right to put members on your ignore list, you do not have the right to derail a thread by posting about it in the technical forums.

Just a reminder, excerpts taken directly from the AnandTech Forum Guidelines:
We want to give all our members as much freedom as possible while maintaining an environment that encourages productive discussion. It is our desire to encourage our members to share their knowledge and experiences in order to benefit the rest of the community, while also providing a place for people to come and just hang out.

We also intend to encourage respect and responsibility among members in order to maintain order and civility. Our social forums will have a relaxed atmosphere, but other forums will be expected to remain on-topic and posts should be helpful, relevant and professional.

We ask for respect and common decency towards your fellow forum members.
Moderator Idontcare
 
Last edited:

ASK THE COMMUNITY