ABT Forum

It is currently Fri Oct 31, 2014 5:31 am

All times are UTC




Post new topic Reply to topic  [ 113 posts ]  Go to page Previous  1 ... 3, 4, 5, 6, 7, 8  Next
Author Message
PostPosted: Sat Dec 24, 2011 7:08 pm 
Offline

Joined: Mon Nov 22, 2010 9:43 pm
Posts: 1143
The card will probably be 4gb. the 7980. LOL.


Top
 Profile  
 
PostPosted: Sat Dec 24, 2011 7:12 pm 
Offline
User avatar

Joined: Fri Jul 04, 2008 5:26 am
Posts: 26052
Location: 404 - Not Found!
vRAM can be 1.5G or 3GB for the 384-bit interface without using mixed RAM as Nvidia did on some of their lower end cards

-- next stop would be 4.5GB or 6GB
-- about 2016 as far as i can tell
:)


Top
 Profile  
 
PostPosted: Sat Dec 24, 2011 7:16 pm 
Offline

Joined: Mon Nov 22, 2010 9:43 pm
Posts: 1143
There goes that.


Top
 Profile  
 
PostPosted: Sat Dec 24, 2011 9:36 pm 
Offline
User avatar

Joined: Mon Nov 22, 2010 9:22 am
Posts: 3516
apoppin wrote:
Not true. AMD is almost TWO YEARS LATER than Nvidia in giving us the answer to Fermi and GPU computing
- Don't you think they might have had a little extra time to work on it? And it is still not impressive as it should be for a generational change on a full node lower. As Ryan said, performance wise it is "acceptable". Not stellar.
:tease:



All true. My review would have shown some DX9 games where the HD 7970 does not look quite so impressive. AMD could not afford that.

However, AMD Graphics will not talk to me any longer. And i now realize that i provoked them by posting at ATF.
:blush:


No doubt AMD is late in that light. They are late by two yrs if they were trying all along but they wasnt. They were on a different path and considered Nvidia's "the wrong direction". AMD were slow to see the light. Its obvious Nvidia took the leap first, they lead the way. But AMD shamefully was wrong. They are goin the fermi path, following nvidia's lead and by this they are late to the game.

What i was getting at is: although they were late to pursue, their transition wasnt terrible at all. Their execution is impressive to me. It wasnt smooth, and i am surprised they got it together so quick. They got a product to sell in very short order. In this i commend them.

BUT-

It will become apparent how much AMD is behind once nvidia drops kepler. There is no way around this. AMD dragged their feet too long and nvidia is yrs ahead in this GPU/compute path. The 7970 will not be enough to stand up against kepler.

For now, AMD has the crown. For now they have a decent product that competes well with nvidias gf110. This is what they have. Their 7950 will compete under the 580 and their 6970 is a notch above it. Kepler will be out of reach. But what can you really expect when AMD sit of the fence and waited sooo long before they jumped in.

That said, I think AMD has gained some ground impressively fast. Of course its not gonna be enough. But instead of 2yrs behind like they should be, when the smoke clears they have gained ground. They have gained a lot of ground real fast.


Top
 Profile  
 
PostPosted: Sat Dec 24, 2011 10:30 pm 
Offline
User avatar

Joined: Mon May 10, 2010 7:46 pm
Posts: 5879
Location: Earth
It's interesting that the 7970 seems to really shine at 2560x1600 vs the GTX 580 (probably due to stellar memory bandwidth) and at eyefinity resolutions with FXAA. It's a bit unexpected to see so many review sites using FXAA in their benchies this soon, already! But when 8x MSAA is used, the performance plummets to levels closer to the 580 - probably due to only 32 ROPs.

What I'm a bit disappointed in is that FP64 performance remains at only 1/4 the FP16 performance, like that of the VLIW4 cards from AMD. I was kinda expecting GCN to do it at least 1:2 ratio or better, like the Fermi-based "unlocked" Tesla cards from NV.

Have anybody done approximations on the 4.3 billion trannies with all of the statistics to logically deduce if parts of the 28nm chip could have been disabled? A 384-bit bus shouldn't take that much die space at all, plus there is still only 32 ROPs. The increase of shader count only went up by 33% but I'd assume that the GCN arch takes about 15-20% more trannies per shader (with the resulting performance boost of around 10% per shader). Who knows how much more trannies are needed for the new DX11.1 features? The doubling of L2 cache does seem to take up some space, but heck it's still ONLY a measly 768K (compare that against 12+MB on the CPUs that take up far less trannies). 2048 shaders and 128 TMUs seem like a fully-rounded number, so it's a bit hard for me to imagine the unlocked number of shaders if there were more... perhaps it's the ROPs (really being 48, advertised as only 32) since they take up a LARGE amount of die space/trannies? I'm afraid that AMD could be playing us once again like they did with the # of Bulldozer trannies (first announcing 2B, then a few months after the launch admitting the "error" and correcting it down to 1.2B). Especially as the HD 5830 and 6790 show some "impossible" stats, along with the Barts series claiming VLIW5 when they behave like VLIW4, I've taken a healthy dose of skepticism this time around.

Perhaps behardware.com would be doing some synthetic tests on the bandwidth and pixel fillrate capacity to see if the specs hold true (like they did on the 5830 and then the 6790 http://www.behardware.com/articles/827- ... -6790.html )....
Quote:
As AMD has decoupled the ROPs from the memory controllers in its GPUs it can deactivate ROPs without affecting the memory controllers. The ROPs do however take up a lot of memory bandwidth and also represent an important pathway for this memory. Deactivating half of them therefore does have an impact on the GPU’s capacity to exploit fully the memory bandwidth made available by the 256 bit Barts bus.


Maybe we'll be seeing an unlocked 48 ROP part soon that allows the 384-bit bandwidth to be used to its fullest?? My guts say no.. that we'd merely be seeing a better-binned process like the RV790XT (4890) refresh of RV770XT.

_________________
Which one do your intentions belong to: Heaven, or Hell?
If you DIE tonight, do you partake in good days, good days, good days, good days, good days, good life in Heaven, forever and ever?

To be forgiven by Jesus, He rose us from the sinful dead with bonding love.

______

Dear GOD, My GOD, I Will Live With YOU In YOUR BONDING LOVE and YOUR Heavenly Faith And Love Forever And Forever, Amen.


Top
 Profile  
 
PostPosted: Sat Dec 24, 2011 11:57 pm 
Offline
User avatar

Joined: Fri Jul 04, 2008 5:26 am
Posts: 26052
Location: 404 - Not Found!
i am guessing that we shall hear if there are disabled parts when the new chip launches. In the meantime, we need to do some testing on it. Since the architecture has changed and much of the die is now devoted to GPU computing, we can't make an easy guess now.

And i am also guessing that Nvidia will be taking it apart to see if they can expect anything better than what AMD just served up in the respin. They would know (better than anyone but AMD) if there are disabled parts - but i doubt we will ever hear that kind of inside information from them
:think:


Top
 Profile  
 
PostPosted: Sun Dec 25, 2011 5:09 am 
Offline
User avatar

Joined: Mon May 10, 2010 7:46 pm
Posts: 5879
Location: Earth
Found one.. http://www.hardware.fr/articles/848-12/ ... ixels.html

Image
Image
Image

It's showing some considerable gains over the 6970, confirming the increased efficiency of ROP render back-end capacity (largely due to the doubled L2 cache). I was expecting a minimum of around 5% (plus another 5% due to 925Mhz vs 880Mhz) but we're seeing a minimum of no less than 20-25% increase over the 6970 (except for non-blending 32-bit fillrate which is only like 3% or so)... not bad at all!

Oh well, I don't know!! Nvidia, please let me know if you find anything fishy about Radeon chips!!! >:) :hello:

Also, I still don't like the IQ--after looking at the other site.. while it's improved some over the Cayman series, there's still that distinct "brilinear / trylinear" AF line unlike with the NV cards.

_________________
Which one do your intentions belong to: Heaven, or Hell?
If you DIE tonight, do you partake in good days, good days, good days, good days, good days, good life in Heaven, forever and ever?

To be forgiven by Jesus, He rose us from the sinful dead with bonding love.

______

Dear GOD, My GOD, I Will Live With YOU In YOUR BONDING LOVE and YOUR Heavenly Faith And Love Forever And Forever, Amen.


Top
 Profile  
 
PostPosted: Sun Dec 25, 2011 1:25 pm 
Offline
User avatar

Joined: Fri Jul 04, 2008 5:26 am
Posts: 26052
Location: 404 - Not Found!
i'll quote Russian Sensation over at ATF
Quote:
The real reason HD7970 is 40% faster than HD6970 is not because GCN is magically better in games than VLIW4, but because:

- HD7970 has 40% more shader performance (925mhz x 2048 SPs vs. 880mhz x 1536 SPs)
- HD7970 has 50% more memory bandwidth (264 vs. 176)
- HD7970 has 40% more texture performance (128 TMUs @ 925mhz vs. 96 TMUs @ 880mhz)

There is no evidence at all that shows that each one of 2048 GCN shaders is more efficient than 1536 SPs of VLIW-4.

The performance increase is almost exactly linear with increases in specifications. That means GCN ~ VLIW-4 for games today, with a significantly larger boost in GPGPU compute. Basically GCN provided no loss in gaming efficiency, but it didn't improve efficiency for games either, at least not in today's game engines. Perhaps in the future DX11 engines, that may become true.

Of course, AMD's viral over there has now labeled him a Nvidia fanboy.
:rolleyes:


Top
 Profile  
 
PostPosted: Sun Dec 25, 2011 2:25 pm 
Offline
User avatar

Joined: Sat Apr 24, 2010 3:19 pm
Posts: 7290
That is irony indeed given how much he recommends buying Radeon cards in his posts generally. Then again, the fanATIcs have never been noted for their logical prowess or their intelligence ........

_________________
"At NVIDIA we know that all shredders are green." --Jensen Huang
Adam knew he should have bought a PC, but Eve fell for the marketing hype. >:)
Having a lot of people buy your product doesn't do you much good if they subsequently decide they don't like it. There's a reason snake oil salesmen moved from town to town every few days. ;)


Top
 Profile  
 
PostPosted: Sun Dec 25, 2011 2:52 pm 
Offline
User avatar

Joined: Fri Jul 04, 2008 5:26 am
Posts: 26052
Location: 404 - Not Found!
The *point* is that ANY SANE person who is not a dyed-in-the-wool AMD Graphics fan, can see this is a very underwhelming launch while AMD PR is doing the distraction dance (but look how well our cherry-picked samples overclock!! It's new architecture, the drivers will improve!!) and yet the FACT of the matter remains - a next generation chip is only 25% faster than their current competitor's STOCK offering and they overpriced it so that it will come back to bite them in the butt.

(i didn't think i could get that all in one breath)
:D


Top
 Profile  
 
PostPosted: Sun Dec 25, 2011 3:12 pm 
Offline

Joined: Mon Nov 22, 2010 9:43 pm
Posts: 1143
It's only $49 more bucks than the 580.

http://www.techpowerup.com/reviews/AMD/HD_7970/


Top
 Profile  
 
PostPosted: Sun Dec 25, 2011 3:14 pm 
Offline
User avatar

Joined: Fri Jul 04, 2008 5:26 am
Posts: 26052
Location: 404 - Not Found!
It hasn't launched yet. :P

You think Nvidia won't respond with a price adjustment on the GTX 580?
:)

When i predicted the $550 HD 7970 price in my preview, all the fanboys at HardOCP and ATF video forums were saying i was nuts - that it was going to be much cheaper while performing much better
http://alienbabeltech.com/main/amds-upc ... card/all/1
Quote:
We are also hearing that across the board, the stock HD 7970 is a bit disappointing as it only scores about 20-30% faster than the stock GTX 580 – well within reach of an overclocked GTX 580 and priced similarly at about $550.


It would have been very different if AMD had got their yields better and could bring out the stock-clocked HD 7970 at 1GHz (same thing for Bulldozer; all they need was 5GHz). Hence the need for a respin
:hello:


Top
 Profile  
 
PostPosted: Sun Dec 25, 2011 3:20 pm 
Offline

Joined: Mon Nov 22, 2010 9:43 pm
Posts: 1143
So it's forcing Nvidia to lower its prices. A win for everyone. That's what it's supposed to do. I'm cool with it.


Top
 Profile  
 
PostPosted: Sun Dec 25, 2011 3:24 pm 
Offline
User avatar

Joined: Fri Jul 04, 2008 5:26 am
Posts: 26052
Location: 404 - Not Found!
Not in the long run; Nvidia can take away much of the HD 7970's sales by pricing the overclocked GTX 580s below $500. A relatively small price cut for them of 10-30%.

It also means that Nvidia can RAISE prices on the new GTX 680 (were just calling it that, btw) and maybe also bring it out early - without raising the bar of performance by more than 20% over the HD 7970; in which case gamers lose.

This a reversal of a noteworthy strategy that AMD has employed since 2900xt. AMD has just started a reverse price war that they will lose (UNLESS they have disabled parts of Tahiti)
:hello:


Top
 Profile  
 
PostPosted: Sun Dec 25, 2011 3:40 pm 
Offline

Joined: Mon Nov 22, 2010 9:43 pm
Posts: 1143
That's if Kepler is indeed that much faster. We don't know. There's this expectation of a 20%-40% increase over the 7970.

I anticipate AMD will have a 3-4 month leeway before we see any real response from Nvidia (sans price drops). I won't tell anyone to quote me on that, but I believe that will be the case. If they can meet demand upon release (unlike the GTX 580 when it first appeared), then I anticipate this card will sell well. You also have to take into account the arrival of over-clocked 7970'. You also have to add in the benefit of the 7970 being new tech. With support of a new version of directx11 and some other new features.

Crap like that sells cards. Even if it doesn't really benefit most users. "Shiny new card!" "Oh, and it's the first 28mm!". "28mm? What the hell does that mean?". "Don't know, but it must be good!"

Plus, they can always lower the price. Also, as we've already established, this card (depending on how Kepler does) will likely be short lived. Then again, Kepler may only be marginally faster-plus more expensive-than the 7970 (this is possible), and AMD will already have been sitting in a nice position for some months.

There's likely a valid reason that AMD released this card so early. I cant believe it's entirely desperation, since they had record profits last quarter.

We'll have to see. It'll be fun. :D


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 113 posts ]  Go to page Previous  1 ... 3, 4, 5, 6, 7, 8  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group