Imperceptible
Junior Member
- Jun 7, 2005
- 16
- 0
- 61
Some more discussion enhancing material: http://image208.poco.cn/mypoco/myphoto/20100915/13/557742492010091513152203.jpg
You memory must be bad, as they only traded blows once, just around the "NetBurst" era, all the time before that and all the time after after that AMD has always been subpar to Intel in performance.
You memory must be bad, as they only traded blows once, just around the "NetBurst" era, all the time before that and all the time after after that AMD has always been subpar to Intel in performance.
Some more discussion enhancing material: http://image208.poco.cn/mypoco/myphoto/20100915/13/557742492010091513152203.jpg
Doesn't add up, 12 SIMDs at 960 SP, 1gb old gddr5, and 850mhz core clocks != >150W TDP. That's higher than the 5850 and its a much large GPU on the same node.
Barts XT at >150W means the full unlocked 16 SIMDs at 1280 SPs.
At least now ATI knows who leaked that screenshot, which is obviously an AIB briefing.
Speaking of bad memory, do you remember when the first Athlon CPU came out? Slot A? Direct competition to PIII 500 Coppermine thru PIII 1.2 Tualitan. And continued to fight Intel on even or better ground up until P4 Northwood 1.6a. This was the introduction of K7. Caught Intel by surprise and often manhandled Intels equivalent clockspeed offerings. This was around 1999.
And by the way, the "Netburst Era" lasted longer than any other incarnation of Intels processor technology. From 2000-2008. During that time, the only segment where Intel led AMD in performance was when Northwood was placed against T-birds. When Athlon64 hit in 2003, it was all over for Intel for about 5 years. All the way up until Core2Duo.
Now, apologize to Kenmitch.
This new naming scheme is unfortunately misleading.
Hopefully, this will drive the prices of the 5850s down to under $200 so I can grab another for Crossfire. That is, unless the performance 6900 series (6950?) will give me roughly the same 5850 Crossfire performance for $350 with lower power usage.
The problem is this rumour is stupid. If barts XT is about the same performance as a 5850 and they are going to call it the 6870 dont you see a massive flaw with that argument. There is no way that AMD's new 68xx generation is going to be slower than the 58xx generation.
Here is my view based on past launches and the current rumours.
Antilles, 2x cayman xt, 6970 --> faster than anything out at the moment
Cayman xt, 6870 --> about 25% faster than GTX480
Cayman pro, 6850 --> about GTX480 level
Barts xt, 6770 --> about 5850 level
Barts pro, 6750 --> about GTX460 level
Earlier I suggested barts was about 190-200mm^2 but after looking at the rumours again this is probably incorrect and I think a die size of about 240mm^2 is more likely. At this die size AMD will likely charge a bit less for a 6770 vs a 5850 to increase perf/£ but at the same time they can increase their own margins.
The only rebadge I can see is possibly 5770 --> 6670 but if AMD can make a smaller die than juniper at that performance level they will probably go that route instead.
So this is one of the falsified screenshots AMD released to find out who the leaks are?
You memory must be bad, as they only traded blows once, just around the "NetBurst" era, all the time before that and all the time after after that AMD has always been subpar to Intel in performance.
Nope, since Intel started making CPU's in 1972 and is still running strong..aka market leader.
That is ~38 years.
The period you are talking about is from ~1995 to ~2005.
You do the math
What you posted is not correct in the first quote. I was going to post what Keysplayr did more or less. AMD was faster than Intel in the P3 era back and forth. Remember Intel trying to catch AMD by releasing the 1.13GHz P3 then having to recall it? Then the P4 came out and the K7 kept up for a while but ran out of steam. But the K8 then pulled ahead pretty well until the C2D's were released. So they traded blows back and forth for a while.
With Bulldozer coming up they may jump ahead again, it's performance is yet to be seen though. At any rate, what you posted was incorrect. No one is arguing that Intel isn't the CPU leader, even when they didn't have the performance crown they were still the market leader. That doesn't change the fact that AMD and Intel were leap frogging each other for a while, and it may (or may not) happen again with Bulldozer.
I will hightlight what I wrote:
You memory must be bad, as they only traded blows once, just around the "NetBurst" era, all the time before that and all the time after after that AMD has always been subpar to Intel in performance.
I have owned CPU's since the i286 era, so my claim still stands.
AMD is now where they have been for the majority of time...chasing Intel.
And don't even get my started om the lackluster FPU performance of the first AMD CPU's...
Tualatin was from 2001, the P4 came in 2002...the "netBurst" era...but I guess most people here are younger and thus lack the perspective on the pre-"NetBurst" era...and thus not realize that the current situation is the norm...Intel ahead of AMD on performance.
I think what Lonbjerg means is something like this:
K7 went head-to-head with PIII, but PIII was essentially the PPro architecture that had been on the market for a few years, and was getting ready to be replaced by Netburst. The rest is all AMD vs Netburst, until finally Conroe arrived and restored the normal order.
By the way, P4 didn't come in 2002, it came in 2000 (Willamette).
Tualatin actually came after P4, it's a 130 nm PIII, where the original P4s were 180 nm (like Coppermine).
True we're just arguing 'spin', the facts are pretty clear. I just question what is meant by 'normal order' given that the 'PIII vs K7 plus netburst' time period makes up a large proportion of the entire time AMD and Intel have been competitors. I don't see you can say there _is_ a 'normal order' (other than that AMD will always be the underdog in terms of size and market share).
I will hightlight what I wrote:
You memory must be bad, as they only traded blows once, just around the "NetBurst" era, all the time before that and all the time after after that AMD has always been subpar to Intel in performance.
I have owned CPU's since the i286 era, so my claim still stands.
AMD is now where they have been for the majority of time...chasing Intel.
And don't even get my started om the lackluster FPU performance of the first AMD CPU's...
Tualatin was from 2001, the P4 came in 2002...the "netBurst" era...but I guess most people here are younger and thus lack the perspective on the pre-"NetBurst" era...and thus not realize that the current situation is the norm...Intel ahead of AMD on performance.
Most people will remember the points in time that Intel and AMD traded blows for the performance crown. Back and forth they went for some time until the point in time when Intel took the crown and hasn't looked back ever sense. This senario could be applied to the current point in time with gpus.
You memory must be bad, as they only traded blows once, just around the "NetBurst" era, all the time before that and all the time after after that AMD has always been subpar to Intel in performance.
Speaking of bad memory, do you remember when the first Athlon CPU came out? Slot A? Direct competition to PIII 500 Coppermine thru PIII 1.2 Tualitan. And continued to fight Intel on even or better ground up until P4 Northwood 1.6a. This was the introduction of K7. Caught Intel by surprise and often manhandled Intels equivalent clockspeed offerings. This was around 1999.
And by the way, the "Netburst Era" lasted longer than any other incarnation of Intels processor technology. From 2000-2008. During that time, the only segment where Intel led AMD in performance was when Northwood was placed against T-birds. When Athlon64 hit in 2003, it was all over for Intel for about 5 years. All the way up until Core2Duo.
Now, apologize to Kenmitch.
You memory must be bad, as they only traded blows once, just around the "NetBurst" era, all the time before that and all the time after after that AMD has always been subpar to Intel in performance.
yeah, I had 1 ghz tbird and a 1.6 ghz t-bred a, those were competitive with anything intel had at the time.
Scratches head and wonders....Do you live in Amsterdam as that would explain the memory lapse
but if the HD 6970 becomes 33% larger than the HD 5870 because its stream processors are smaller, its possible to have easily a performance boost of 50% with little effort,
Like I said AMD's strategy is Small-die. Enlarging 5870 33% would mean that 6970 (or whatever it will be called) will be ~ 440 mm2. I don't buy it. That's moving into the direction of NV, which is not what AMD wants.
Also if you think producing a 440mm2 GPU to get 50% performance is little effort, AMD would have done so with 5870 in the first place. These types of designs take 3+ years. You think HD6000 series was designed in 12 months? AMD already has probably finalized the specs for HD7000. In many interviews in the past with Dave Orton (at the time the head of ATI), it was clear that 2-3 generations of GPU generations were being worked on at the same time by different teams.
So it's most likely scenario that AMD's team which worked on HD6000 series was different from the team that worked on HD5000 series. That first team was aiming to launch the HD6000 series on 28nm/32nm. Therefore, it's more likely than not that HD6000 series won't be as fast as originally intended.
Having said that, I fully expect ATI to regain performance (where as now 5830<GTX460, 5850<GTX470, 5870<GTX480). All ATI needs to do is to be 20%+ faster than each of the NV cards with lower power consumption and they are in the lead for another 6+ months.