-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
ATI Radeon HD 6000 Series ''Southern Islands'' Graphics Cards For Sale from November
- Thread starter PraxxtorCruel
- Start date
Soldato
- Joined
- 8 Aug 2010
- Posts
- 6,453
- Location
- Oxfordshire
^^^
Yes, but yields reduce exponentially as you go up the die size, so they don't have that much room.
Yes, but yields reduce exponentially as you go up the die size, so they don't have that much room.
Soldato
- Joined
- 8 Aug 2010
- Posts
- 6,453
- Location
- Oxfordshire
It only needs to beat the 480.
Hmmm I don't know why it's so difficult to grasp... if the process isn't there (like it should have been) it isn't possible, so why expect the impossible.
Simples...
I understand what you're saying, but my point is, this is the norm. The HD4870 was around the speed of the HD3870X2. Admittedly the 3870 was created to "right the wrongs" of the HD2900 so it wasn't twice the speed but before that, the X1800 was around X800 crossfire, X800 around 2x9800 Pro and so on...
Basically I believe both nVidia and ATI try to double the speed of the previous gen's flagship card, or in other words match the previous dual-GPU card with a single one.
But realistically, without a process shrink, I don't think we'll see HD5970 speeds in a single GPU until NI. Not unless ATI really have stumbled on a miracle.
Of course, this could well be the "HD4890 to the 4870" and we'll only see a bump of 10-15% on average. Nothing amazing, but enough to take top spot and whet the appetite for NI.
Higher clock rates and a more efficient process says it is possible. I don't care what others would prefer or expect performance wise from new cards, it's my opinion only, from my experiences with current tech at stock and overclocked.
Last edited:
Soldato
- Joined
- 8 Aug 2010
- Posts
- 6,453
- Location
- Oxfordshire
That's allot of variables not in the hands of amd necessarily...
Soldato
- Joined
- 25 Sep 2009
- Posts
- 10,242
- Location
- Billericay, UK
If your looking for what could be the potential speed increase of the 6000 series cards you really ought to be looking at the jump from the 3000 series to the 4000 series as both were on the same manufacturing process.
Associate
- Joined
- 10 Nov 2009
- Posts
- 1,099
- Location
- London
I'm a lay-man in this, but I've been following the news pretty closely. I actually think there is a good chance the 6000 series could be a pretty substantial increase due to a number of factors:
- Process improvements: When Cypress was released AMDs first priority was getting a decent card out at profitable yields on a crappy process and they were extra careful to do it. Now you have an improved process that they are much more knowledgable about, so they can up the transitor density in the chips.
- New core: One thing I've heard is that a bottleneck in the 5000 series is caused by the core not being able to feed the shaders fast enough to fully utilise them. IMO this is backed up by AMD changing the uncore to the version that was destined for 28nm. This should give an additional speed boost, though how much is impossible to guess.
- Bigger die size: AMD won't want to increase this too much. drunkenmaster's prediction of ~25% is reasonable.
Taking all these factors together I think there is a decent chance that the improvement could be 50%, maybe even more. But it's all guesswork. AMD have been very quiet, but that could mean anything
- Process improvements: When Cypress was released AMDs first priority was getting a decent card out at profitable yields on a crappy process and they were extra careful to do it. Now you have an improved process that they are much more knowledgable about, so they can up the transitor density in the chips.
- New core: One thing I've heard is that a bottleneck in the 5000 series is caused by the core not being able to feed the shaders fast enough to fully utilise them. IMO this is backed up by AMD changing the uncore to the version that was destined for 28nm. This should give an additional speed boost, though how much is impossible to guess.
- Bigger die size: AMD won't want to increase this too much. drunkenmaster's prediction of ~25% is reasonable.
Taking all these factors together I think there is a decent chance that the improvement could be 50%, maybe even more. But it's all guesswork. AMD have been very quiet, but that could mean anything
Last edited:
Soldato
- Joined
- 25 Sep 2009
- Posts
- 10,242
- Location
- Billericay, UK
Soldato
- Joined
- 25 Sep 2009
- Posts
- 10,242
- Location
- Billericay, UK
Fine here's the math Raven.
They have lots of room.
5870 die size is 334 (mm squared)
480 die size is 529
So even if they shoot for a die size of 450, to be safe on power and heat, that's still plenty of potential.
Add in the more mature process, and that this is a new architecture, and it's looking possible they can produce a 6870 that can get close to a 5970 and certainly beat the 480 hands down.
The only potential issue I see is if they have to increase the memory bus to 384 bit if 256 creates too much of a bottleneck.
Last edited:
Associate
- Joined
- 24 Jul 2005
- Posts
- 482
This thread is giving me a lot more confidence about SI. Damn those clever engineers... I have debts to pay off, bass guitars to buy, other things in my PC that need upgrading, and now SI is looking very interesting! Daaamn them! xD
they're going for under 400mm2 die size iirc, and there will be 1920 shaders total iirc as they're going from 5 per shader to 4 per shader, which should increase the performance a little bit
can't remember the clocks but I think they were aiming for 900MHz-950MHz
Last edited:
Tessellation is one of the areas they have talked about boosting performance in with these cards.
Don't expect a massive performance boost - tho you might see it in some areas/benchmarks - this round is more about making things work more efficently, better balancing the card so that some areas aren't held back by others, boosting performance in some DX11 and eyefinity scenarios and so on.