Thats like saying a car with 2 V6's is the same as a car with a V12. Not saying that the 2 V6 car would be slower, just that it is not the same because of the design.
Why do you think that V12 would be faster than V6?
What number of cylinders have to do with the speed of a car?
First you need to put CHEVRON gas in it .....
While I don't really have a preference for either, I really don't expect AMD to come out and say that their yet unreleased core will not be able to beat Intels chips (present or future). Also, most of this is purely marketing since they have no answer for the Core2 any time soon. They need to make sure people know that they are still in the game.
Don't get me wrong, everyone should be hoping K8L kicks some serious &*^. Competition works in the consumers favor but words are just that, words.
Don't get me wrong, everyone should be hoping K8L kicks some serious &*^. Competition works in the consumers favor but words are just that, words.
GlacierFreeze
Distinguished
- Sep 22, 2006
- 131
- 0
- 18,680
Correct me if I'm wrong but:
Does that look like four native cores to you?
Does that look like K8L to you? NO! That is not K8L. Those are 4x4 (4 CPU cores = 2 DUAL CORES on 2 SOCKETS and 4 gpu cores = SLI or CROSSFIRE) CPUs.
4x4 is coming out later this year. K8L dual-core 1st half '07 and K8L QUAD-core 2nd half '07. There is some talk that you'll be able to put 2 K8L Quads in a 4x4 which will actually make it an 8x4 (8 cpus cores and 4 gpu cores).
Got it??
chrislax20
Distinguished
- Jul 6, 2006
- 75
- 0
- 18,630
Perhaps you need a lesson in reading, or you just need to do some reading before you type. Either way, cylinders and cores can be related, as in 2 2 core processors verses a 4 core processor. Same number of cores, different design, yet to be seen which has better performance.
BaronMatrix
Splendid
- Dec 14, 2005
- 6,655
- 0
- 25,790
From the specs and comparing the Core 2 over Core increase, I would say that Barcelona will be 80% faster than Opteron. I think that will propel them ahead because of HT and Direct Connect.
But then I think that a slow FSB will cause a swapping bottleneck so what do I know.
BaronMatrix
Splendid
- Dec 14, 2005
- 6,655
- 0
- 25,790
From the specs and comparing the Core 2 over Core increase, I would say that Barcelona will be 80% faster than Opteron. I think that will propel them ahead because of HT and Direct Connect.
But then I think that a slow FSB will cause a swapping bottleneck so what do I know.
MrCommunistGen
Distinguished
- Jun 7, 2005
- 1,042
- 0
- 19,310
Usually these are placed side by side and are designated as "W" engines. In this case a W12. 😀
-mcg
Regarding 4 native cores Vs. two dual cores, please take note of some of the problems IBM has had with their CELL processor. Increasing the number of cores on the chip, increases the number of ways your yield can drop. So while 4 native cores MAY have some performance advantage (and nobody has demonstrated that it is an absolute advantage that can not be overcome), 2 dual cores MAY have a yield advantage which could translate to a cost advantage (which hopefully would be reflected as a price advantage).
Everyone sounds up in arms about K8L vs Core, but the way I read the original article was that clock for clock, the Intel core will continue to perform better than the AMD core, most likely due to the number of instructions per clock, but that AMD plans to counter that (for the time being) with a faster platform. The Intel chips will probably remain the performance leading CPU, but the AMD platform will be the performance winning platform.
The question remains whether any of us mere mortals will be able to afford this stuff. Platforms like 4x4 are bound to be very expensive. Those of us trying to make the most of more meager budgets want to see affordable stuff that don't require tons of other stuff to be competitive.
Everyone sounds up in arms about K8L vs Core, but the way I read the original article was that clock for clock, the Intel core will continue to perform better than the AMD core, most likely due to the number of instructions per clock, but that AMD plans to counter that (for the time being) with a faster platform. The Intel chips will probably remain the performance leading CPU, but the AMD platform will be the performance winning platform.
The question remains whether any of us mere mortals will be able to afford this stuff. Platforms like 4x4 are bound to be very expensive. Those of us trying to make the most of more meager budgets want to see affordable stuff that don't require tons of other stuff to be competitive.
ltcommander_data
Distinguished
- Dec 16, 2004
- 997
- 0
- 18,980
You truly expect to see an 80% performance increase in real world? Perhaps is a (very) few optimized applications, but there is no way the real world average increase is 80%. In terms of Core to Core 2, it was a major architecture shift yet the real world average performance increase is really only in the order to 20%.
My feeling is that K8L will definitely be stronger in its traditional pure FP field, it may have a small lead in SSE, but Core 2 will still have a slight lead in integer calculations. Overall, K8L will probably be 10-15% faster than Core 2, which means it's all down to final implementation, software optimization, and clock speeds. Obviously, if K8L really brings a Earth changing 80% increase I'm not going to complain, but I'm going to be a bit more realistic if not pessimistic for now.
In terms of 4x4, I'm really disappointed. The use of Socket F implies it's just a Opteron knock-off and also implies the use of registered DIMMs which is terrible. High latencies, lower bandwidth. The high 125W TDP also places a 2 processor 4x4 system much higher in terms of heat and power than Kentsfield. Also a clocking of 3GHz also isn't impressive against Kentsfield. A 2.67GHz Conroe can already hold it's own and generally lead a 3GHz FX. A 2.67GHz Kentsfield may not have perfect scaling to 4 cores, but the inherent strength of even just 2 of its cores will make it difficult for 4x4 to catch up. We'll have to see if dedicate memory through IMCs, and dedicated communications links through HT, in otherwords AMD's superior platform, can really make up the difference.
ElMoIsEviL
Distinguished
- Jan 10, 2006
- 2,363
- 1
- 20,015
ches111
Distinguished
- Feb 4, 2006
- 1,958
- 0
- 19,780
Actually Lordpope,
I HOPE (not think) that AMD totally thrashes the Core 2 Duo platform with the release of the K8L. I "feel" that it may not be not likely.
First looks, and analysis from others here (much more in the know than I like JJ) have pretty much determined that it is a substantial upgrade over the existing platform but may in fact fall short or JUST meet the capabilities of the soon to be released and already benched Kentsfield.
My statement above is even irrelevant since NO Engineering Samples have been made available for true first looks... At this point we really can not draw any concrete conclusions.
Just like with the Conroe, as soon as ESs or retail product ships for review we are all speculating.
Should the Intel folks tell you (if an ES is released of the K8L) that it is only an ES and not reflective of what will actually be released.
Should Intelites complain that the tests were rigged?
Should they say that the benchies were synthetic and therefore no where near accurate?
Of course they should also state that some benchies favor the K8L arch too and therefore are skewing the results?
Did I catch nearly all of the comments that have been made during the last Conroe release?
Oh wait the wrong memory was used in the test systems and the Intel now needs DDR2-8888 to fully realize its potential.
I am sure there are more. Mrs and Lord did I catch them all?
If these things have not yet been said about the future release let me say them now and get them all out of the way 😉
Oh and to be thorough Please apply all to the next release of anything from Intel as well....
I HOPE (not think) that AMD totally thrashes the Core 2 Duo platform with the release of the K8L. I "feel" that it may not be not likely.
First looks, and analysis from others here (much more in the know than I like JJ) have pretty much determined that it is a substantial upgrade over the existing platform but may in fact fall short or JUST meet the capabilities of the soon to be released and already benched Kentsfield.
My statement above is even irrelevant since NO Engineering Samples have been made available for true first looks... At this point we really can not draw any concrete conclusions.
Just like with the Conroe, as soon as ESs or retail product ships for review we are all speculating.
Should the Intel folks tell you (if an ES is released of the K8L) that it is only an ES and not reflective of what will actually be released.
Should Intelites complain that the tests were rigged?
Should they say that the benchies were synthetic and therefore no where near accurate?
Of course they should also state that some benchies favor the K8L arch too and therefore are skewing the results?
Did I catch nearly all of the comments that have been made during the last Conroe release?
Oh wait the wrong memory was used in the test systems and the Intel now needs DDR2-8888 to fully realize its potential.
I am sure there are more. Mrs and Lord did I catch them all?
If these things have not yet been said about the future release let me say them now and get them all out of the way 😉
Oh and to be thorough Please apply all to the next release of anything from Intel as well....
ElMoIsEviL
Distinguished
- Jan 10, 2006
- 2,363
- 1
- 20,015
You mean AMD's new K8L will be faster then Intel's 1 yr old Core 2 Duo architecture.. 😛
Yeah wtf? Thats comparing AMD's new tech to intels old tech. No fair!!!!
Nawww.. it's fair.. but K8L is a Quad Core CPU and well by this time next year Intel will be readying 45nm native Quad Cores.
BaronMatrix
Splendid
- Dec 14, 2005
- 6,655
- 0
- 25,790
You truly expect to see an 80% performance increase in real world? Perhaps is a (very) few optimized applications, but there is no way the real world average increase is 80%. In terms of Core to Core 2, it was a major architecture shift yet the real world average performance increase is really only in the order to 20%.
My feeling is that K8L will definitely be stronger in its traditional pure FP field, it may have a small lead in SSE, but Core 2 will still have a slight lead in integer calculations. Overall, K8L will probably be 10-15% faster than Core 2, which means it's all down to final implementation, software optimization, and clock speeds. Obviously, if K8L really brings a Earth changing 80% increase I'm not going to complain, but I'm going to be a bit more realistic if not pessimistic for now.
In terms of 4x4, I'm really disappointed. The use of Socket F implies it's just a Opteron knock-off and also implies the use of registered DIMMs which is terrible. High latencies, lower bandwidth. The high 125W TDP also places a 2 processor 4x4 system much higher in terms of heat and power than Kentsfield. Also a clocking of 3GHz also isn't impressive against Kentsfield. A 2.67GHz Conroe can already hold it's own and generally lead a 3GHz FX. A 2.67GHz Kentsfield may not have perfect scaling to 4 cores, but the inherent strength of even just 2 of its cores will make it difficult for 4x4 to catch up. We'll have to see if dedicate memory through IMCs, and dedicated communications links through HT, in otherwords AMD's superior platform, can really make up the difference.
The speed increase is the word from AMD and like I said the specs seem to bear it out. As far as 4x4, I don't care which socket but they have to let it use nonECC as that is a MAJOR expense.
I don't see this being a watered down Opteron, because that would definitely make product differentiation more difficult. What I want is two sockets with NO ECC, NO SCSI 320, NO PCI-X, NO SODIMMS.
That would differentiate perfectly as mission critical thinsg want the reliability of SCSi and data protection of ECC.
As a dev, I just need a fast box with lots of RAM. I posted a link to voodoopc.blogspot where I suggested that to the CEO in March.
Anyway, enthusiasts will buy it no matter the power use. I figure I could get two FX 7000 and one 7900GTX and that would be like getting one FX62 and QUAD SLI power-wise.
I think it will be close for Kentsfield because AMD already extracts more bandwidth so this may allow them to use more of it. I think when mem usage is low Kentsfield will win but when you load past physical RAM, then 4x4 will more than likely outshine it because of the FSB.
PS
I load more than 2GB RAM at times and I need something that won't choke. When I go over to Server 2003 x64, I will load it even more. i hope that the FSB won't cause problems when overloading the physical RAm but FSB is uni-directional IIRC and for 4x4 disk IO will go over the South Bridge and memory IO won't. That may turn out to be huge.
At least for us megataskers.
AddictiveHerring
Distinguished
- Sep 21, 2006
- 168
- 0
- 18,680
The Bugatti Veyron is a piece of junk
ever wonder why you never here of any record breaking track times coming from the wonderfull veyron? Because it is only fast in a sraight line, not in any meaningful way. When the enzo or the mclaren F1 came out, they broke track times AND set landspeed records. The mclaren is still the fastest naturally aspirated car out there hitting about 240 and it's from like 91. Unbelievable I can name several ferraris most of which cost less than half as much than a veyron, would put it to shame on a track, don't suffer from turbolag, and are plain classier.
360 Challenge Stradale
430 coupe OR convertible
Enzo
599 gtb
Oh, and it would be cool if the new Amd platform was faster than intel, not because it benefits me, since it doesn't, my system is set up for kentsfield not k8l so I wouldn't be able to upgrade to it. I want the AMD platform to be faster because I simply want cpu's to get faster and faster. I want improvement, I think any moron who is pissed because they don't have the fastest anymore at the expense of progress is a fucking idiot fanboi piece of shit. Same goes for people who spend more on an inferior product because they love their brand and ignore benchmarks and performance figures.
ever wonder why you never here of any record breaking track times coming from the wonderfull veyron? Because it is only fast in a sraight line, not in any meaningful way. When the enzo or the mclaren F1 came out, they broke track times AND set landspeed records. The mclaren is still the fastest naturally aspirated car out there hitting about 240 and it's from like 91. Unbelievable I can name several ferraris most of which cost less than half as much than a veyron, would put it to shame on a track, don't suffer from turbolag, and are plain classier.
360 Challenge Stradale
430 coupe OR convertible
Enzo
599 gtb
Oh, and it would be cool if the new Amd platform was faster than intel, not because it benefits me, since it doesn't, my system is set up for kentsfield not k8l so I wouldn't be able to upgrade to it. I want the AMD platform to be faster because I simply want cpu's to get faster and faster. I want improvement, I think any moron who is pissed because they don't have the fastest anymore at the expense of progress is a fucking idiot fanboi piece of shit. Same goes for people who spend more on an inferior product because they love their brand and ignore benchmarks and performance figures.
ltcommander_data
Distinguished
- Dec 16, 2004
- 997
- 0
- 18,980
I wonder if Socket F's Register DDR2 memory controller is also backwards compatible with unregistered DDR2 memory. Otherwise they would have to design a separate chip which isn't very convenient. Opteron 1xxx which use unregistered DDR2 are use AM2 not Socket F so they don't really have an example of that yet.
In terms of disk IO, 4x4 isn't a saviour either. In Kentsfield's case, both processors are just 1 hop from the NB. However, in 4x4s case, the second chip is 2 hops away having to go through HT to the 1st chip and then HT again to the NB. That adds latency and even worse, differential latency between the chips.I load more than 2GB RAM at times and I need something that won't choke. When I go over to Server 2003 x64, I will load it even more. i hope that the FSB won't cause problems when overloading the physical RAm but FSB is uni-directional IIRC and for 4x4 disk IO will go over the South Bridge and memory IO won't. That may turn out to be huge.
At least for us megataskers.
Similarly, in terms of memory bandwidth through 2 IMCs, if it happens that the 1st chip needs data that is stored in the 2nd chips memory banks it'll have to go over HT to get it. That adds latency and also starves the 2nd processor of bandwidth needed to access the NB. People call for "true" multithreading, but in the event that 4 cores are running a single program this problem is actually increased since each core will need data from each other's memory banks more frequently. In this case, 2 or 4 completely separate programs each having a dedicated chip and it's memory banks would be better at avoiding conflict.
Both 4x4 and FSB have their inherent problems. So we'll have to see. I guess a factor is also the cost of 4x4. It is definitely looking more expensive as a complete system than Kentsfield right now. If 4x4s price is too high, Cloverton may become another option in addition to the cheaper Kentsfield.
(Reply to BM)
You idiot!
How many times do I have to slap you, the FSB will have no impact on the performance once you use all of the physical RAM and are swapping
(reply to ltcommander_data)
Don't bother trying to argue that the FSB has nothing to do with swapping performance, BaronBS has not the reading comprehension to understand. In his case, why didn't his AXP 3200+ perform "worse" when swapping compared to his Craptiplex?
You idiot!
How many times do I have to slap you, the FSB will have no impact on the performance once you use all of the physical RAM and are swapping
(reply to ltcommander_data)
Don't bother trying to argue that the FSB has nothing to do with swapping performance, BaronBS has not the reading comprehension to understand. In his case, why didn't his AXP 3200+ perform "worse" when swapping compared to his Craptiplex?
TRENDING THREADS
-
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
-
Question Windows is using Microsoft Basic Display Adapter ?
- Started by Corbin8756
- Replies: 89
-
Facebook
Twitter
Instagram