AMD Cayman info (or rumor)

Quote:

---

Originally Posted by TheBreezyBB

So, Tomorrow finally this endless thread will END?!

---

And more threads about the actual Cayman will start :) .It's actually an endless GPU circle and we are trapped in it :p:

If 6950 is close to GTX 570 and is at $299, I will impulse buy or otherwise...

So I guess we're going to get cayman sucks or cayman rocks threads on the 15th onward

Quote:

---

Originally Posted by TheBreezyBB

So, Tomorrow finally this endless thread will END?!

---

Yeah ,and the new one will begin:
AMD Antilles info (or rumor)
:ROTF:

Quote:

---

Originally Posted by TheBreezyBB

So, Tomorrow finally this endless thread will END?!

---

I wouldn't be so sure. Look at the signs:

1. It Won't die
2. It displays aggressive rhetoric
3. It eats brains

The diagnosis is clear, it's a zombie thread. Aim for the head guys.

Quote:

---

Originally Posted by SpuTnicK

Yeah ,and the new one will begin:
AMD Antilles info (or rumor)
:ROTF:

---

This will happen if Cayman is a success :D

or else it's doom and gloom

Quote:

---

Originally Posted by SpuTnicK

Yeah ,and the new one will begin:
AMD Antilles info (or rumor)
:ROTF:

---

Here we go again!! :ROTF: I'll keep a close eye on this one :shocked:

How come we don't have any sold benchmarks a day before launch??? is it still under NDA?

Yes, indeed. Tomorrow everything will change.;)

Quote:

---

Originally Posted by kam03

How come we don't have any sold benchmarks a day before launch??? is it still under NDA?

---

Actually, we have: link1, link2 ;)

A very basic calculation has come to my mind... As you know, Cypress was absolutely RAPING GF100 in Performance/mm2. Which should mean that Nvidia should have been able to increase Perf/mm2 much more than AMD did, since they started from horrible levels.

However, when you do the math...

HD5870 -> HD6970: 320mm2 to 390mm2 = 22% die size increase, 40% performance increase => 15% increase in Performance/mm2
GTX 480 -> GTX 580: Same die size = 0% die size increase, 20% performance increase => 20% increase in Performance/mm2

AMD has been able to keep up with the ratio of efficiency increase of Nvidia... and keeping in mind they have started from a much much efficient level already, this is quite a feat. Actually they had a couple of months more time to do this than Nvidia did, yeah, but still.

Quote:

---

Originally Posted by SpuTnicK

Actually, we have: link1, link2 ;)

---

Let me be the first to say denied. Those benchmarks as of this moment in time cannot be verified.

Quote:

---

Originally Posted by hurrdurr

A very basic calculation has come to my mind... As you know, Cypress was absolutely RAPING GF100 in Performance/mm2. Which should mean that Nvidia should have been able to increase Perf/mm2 much more than AMD did, since they started from horrible levels.

However, when you do the math...

HD5870 -> HD6970: 320mm2 to 390mm2 = 22% die size increase, 40% performance increase => 15% increase in Performance/mm2
GTX 480 -> GTX 580: Same die size = 0% die size increase, 20% performance increase => 20% increase in Performance/mm2

AMD has been able to keep up with the ratio of efficiency increase of Nvidia... and keeping in mind they have started from a much much efficient level already, this is quite a feat. Actually they had a couple of months more time to do this than Nvidia did, yeah, but still.

---

Isnt 580 a smaller die, becuase of the lost HPC performance. That would increase thier performance/mm2.

Quote:

---

Originally Posted by Kristers Bensin

Isnt 580 a smaller die, becuase of the lost HPC performance. That would increase thier performance/mm2.

---

I don't think there are official values about die size released by Nvidia. According to Anandtech GF100 is 530mm2 and GF110 is 520mm2, so a 2% decrease.

Quote:

---

Originally Posted by kam03

How come we don't have any sold benchmarks a day before launch??? is it still under NDA?

---

launch == NDA lift

Quote:

---

Originally Posted by hurrdurr

A very basic calculation has come to my mind... As you know, Cypress was absolutely RAPING GF100 in Performance/mm2. Which should mean that Nvidia should have been able to increase Perf/mm2 much more than AMD did, since they started from horrible levels.

However, when you do the math...

HD5870 -> HD6970: 320mm2 to 390mm2 = 22% die size increase, 40% performance increase => 15% increase in Performance/mm2
GTX 480 -> GTX 580: Same die size = 0% die size increase, 20% performance increase => 20% increase in Performance/mm2

AMD has been able to keep up with the ratio of efficiency increase of Nvidia... and keeping in mind they have started from a much much efficient level already, this is quite a feat. Actually they had a couple of months more time to do this than Nvidia did, yeah, but still.

---

we still don't know the performance of the 6970.

the 580GTX was not 20% faster. it is more 2-17%. average is more around 15% than 20%. (although it is more semantic wise since we don't know the numbers of the 6970).

Quote:

---

Originally Posted by LordEC911

Chil?

---

V for Volterra :cool:

http://home.coolpc.com.tw/upon/vga/a...-hd6950-17.jpg

Quote:

---

Originally Posted by hurrdurr

A very basic calculation has come to my mind... As you know, Cypress was absolutely RAPING GF100 in Performance/mm2. Which should mean that Nvidia should have been able to increase Perf/mm2 much more than AMD did, since they started from horrible levels.

However, when you do the math...

HD5870 -> HD6970: 320mm2 to 390mm2 = 22% die size increase, 40% performance increase => 15% increase in Performance/mm2
GTX 480 -> GTX 580: Same die size = 0% die size increase, 20% performance increase => 20% increase in Performance/mm2

AMD has been able to keep up with the ratio of efficiency increase of Nvidia... and keeping in mind they have started from a much much efficient level already, this is quite a feat. Actually they had a couple of months more time to do this than Nvidia did, yeah, but still.

---

5870 is 334mm² and 6970 still unknown

isnt guru3d.com going to accidently leak a review?
or did amd spank them for thier (cough) accidental 6800 serie leak so they never got any cards?
16hour or so to go.

this is one of the thightest nda ive seen :O

http://img801.imageshack.us/img801/9893/48822272.gif

http://img339.imageshack.us/img339/6619/72838504.gif

http://img219.imageshack.us/img219/338/54754688.gif

http://img375.imageshack.us/img375/1289/49923852.gif

http://img152.imageshack.us/img152/3373/35527787.gif

Quote:

---

Part from the performance, it seems that HD6970 is not out of the GTX580, but some say the driver is not perfect.

However, the price point of view, these two original houses offer a special edition are as follows

HD6970 ... 11998 元
HD6950 ... 9198 元

From the price side view, HD6970 GTX570 is indeed positioned to rival!

---

http://www.coolpc.com.tw/phpBB2/view...276669#p276669

What's up with Lost Planet 2?!

And also how much are those 2 in US$ ?

HD 6970 - 420€ here in France, In stock

Quote:

---

Originally Posted by TheBreezyBB

What's up with Lost Planet 2?!

---

Its been said earlier in this thread, nvidia sponsored the development of lost planet 2 and thus favoring geforce cards. Its allways been like this, one time its AMD then Nvidia and so on.

Yes in deed but Amd's new arhitecure doesn't count for nothing?

Quote:

---

CPU bottleneck in Crysis at 1920x1200 8xAA? The shape of the graphs don't show any of the typical signs. It's more likely the driver not enforcing AA settings.

---

Have you forget that is just the AVG FPS, what about min/max. Probably MIN FSP is lower at 8AA than 2AA or 4AA.

Quote:

---

Originally Posted by Kristers Bensin

Its been said earlier in this thread, nvidia sponsored the development of lost planet 2 and thus favoring geforce cards. Its allways been like this, one time its AMD then Nvidia and so on.

---

This i know But, i thought the new card would fare better :p: