Thanks for the response. Nov 22 was only a rumor. AMD (as far as I know) never made a public announcement as to a release date for the 6900 series.
AMD: R9xx Speculation
- Thread starter Lukfi
- Start date
Thanks for the response. Nov 22 was only a rumor. AMD (as far as I know) never made a public announcement as to a release date for the 6900 series.
Yeah, the gloom and doom surrounding the HD69xx is getting ridiculous, which intensive AMD has for launching HD69xx in a hurry, they are clearing the channel of HD58xx, they have a product which compared favorably to the Nvidia GT460 line. They should be concerned about the GT580? A 500$ product which mass availability is likely to remain questionable for a while?
Amd concerns should be about the very likely GT570 products but if their HD69xx are not up to the task to fight them both on price and perfs then we will be able to state that something went wrong...
Amd concerns should be about the very likely GT570 products but if their HD69xx are not up to the task to fight them both on price and perfs then we will be able to state that something went wrong...
Last edited by a moderator:
As I said, the arguments given so far can't be true. It simply doesn't work that way.
I would suspect some issue with the compute shaders for the texture decompression they use. Without exactly knowing what they are doing it is really hard to tell. One possibility would be that they try to be too clever and use some adaptive (to the assumed speed of the used GPU) scheme messing things up. Or it is some strange synchronisation problem with the compute shader (they try to run it asynchronously and miss when it finishes and effectively wait too long before moving on). I simply don't know.
But what I do know is that no increased rasterizing, shader, or culling efficiency can increase the framerate when inreasing the resolution.
Agreed about first part, also dont see the reason to worry about GTX570.
Lets think for a second, AMD refreshed 5770 with ~30% faster 6870, they are planning to refresh 5870 with 6970 very soon, ~30% faster is reasonable expectation. GTX580 is ~30% faster than 5870. See where it is going?
If anything GTX570 will have a hard time dealing with Caymans, if GTX580 will be able to beat them at all. Thus AMD is safe with either 6950 or 6970 being faster than GTX570, and the Top end will be secured by 6990.It might just be like last gen: 6990 > GTX580 >= 6970 > GTX570 >= 6950.
I take it that AMD will simply remain silence until they decide to launch the card? I don't think a lack of response from AMD is a sign of doom and gloom. However, if AMD wants consumers to be interested in their product they have to make an effort to reach out to us. Saying very little doesn't really convey the message of "we want you to buy our product(s), not their products".
They don't have to rush out and buy a $600 competing card. But they can abstain from buying if they don't think the information is there to draw interest. It's because consumer don't find a need to have to wait when the competition is showcasing what they have. Remember, it's not the consumers who are asking AMD to buy something here. Big difference when taken from another view point.
There is a difference between then and now though. The competition didn't launch a new card back.
Last edited by a moderator:
Agreed about first part, also dont see the reason to worry about GTX570.
Lets think for a second, AMD refreshed 5770 with ~30% faster 6870, they are planning to refresh 5870 with 6970 very soon, ~30% faster is reasonable expectation. GTX580 is ~30% faster than 5870. See where it is going?
It might just be like last gen: 6990 > GTX580 >= 6970 > GTX570 >= 6950.
Already the HD6850 is ~40% faster than the HD5770; at least according to CB:
http://www.computerbase.de/artikel/...70-und-hd-6850-update-2/22/#abschnitt_ratings
The HD6870 is ~15 faster
It's a $500 card and in this niche of videocards, mindshare is as critical as marketshare. Nvidia managed a refresh of their flagship card in a short time and beat AMD to the launch of their flagship which is coming a year after the initial launch.
They have officially announced that it will launch before the end of the year.
If I recall correctly, when Barts launched, one of the sites got an interview with someone from AMD and I believe he said next month, meaning Nov. I don't remember which site though unfortunately, though it does seem like common knowledge that it will be launching before the end of the month.
And AMD isn't going to rush their planned launch because some kid is whining like a 5 year old on some message board.
A few weeks of mindshare isn't going to add up to a whole bunch at the end of the quarter if AMD releases a competitive product on time and in volume. Releasing before the product is ready is exactly the wrong way to do business and opens you up to all sorts of criticism that can actually hurt your product more than a slight later release.
I don't agree, ad hominem aside, if AMD announced (for example) that they were releasing their product on Nov 20, 2010 would what you say make any sort of sense. However, we are still guessing as to when the product will be released. Other than to say before the end of the year
.Has this been dismissed yet? :
http://www.nordichardware.com/news/71-graphics/41433-amd-radeon-hd-6000-roadmap-leaked.html
http://www.nordichardware.com/news/71-graphics/41433-amd-radeon-hd-6000-roadmap-leaked.html
IMHO, apart from the much more understandable numbers in Civ 5, Damien is presenting, there's one possible explanation to this I think.
With Civ 5 being quite a CPU intensive application as well, there is the chance it probably wastes some cycles in lower resolutions trying to be clever on Tessellation usage in order to lower the graphics load. As I've pointed out here, there are some techniques suggested to limit tessellation factors at lower resolutions (screen space adaptive tessellation like „Consider the screen space patch edge length as a scaling factor”). Maybe those computations are just omitted at some point or „target screen res”.
Agreed about first part, also dont see the reason to worry about GTX570.
Lets think for a second, AMD refreshed 5770 with ~30% faster 6870, they are planning to refresh 5870 with 6970 very soon, ~30% faster is reasonable expectation. GTX580 is ~30% faster than 5870. See where it is going?
It might just be like last gen: 6990 > GTX580 >= 6970 > GTX570 >= 6950.
The 68xx are indeed 30% faster then the previous gen mid range but only due to its clock speed increases, and Cayman has less ALU's then the 58xx, so by what you posted can we get a chip that is clocked higher and higher effeciency to fill that 30% gap.
How did you come to conclusion that this is a performance chart?
Similar threads
