mickyb - "How is this any different than multi CPU SMP? It isn't, except for compressing them to a smaller space"
I agree, at least for the Intel model. The AMD dual core is quite different however, in that the 2 cores are connected locally, and the MOESI protocol AMD uses allows for easy cache snooping.
#16 My statement was in context of the performance issue, not cost. The graphs imply that multi-core is the savior, when we have been experiencing what multi-cpu will do. Any graph that implies "exponential" growth for multi-core vs. single core is just a lie. I have been creating tools that analyze system performance for a while. It is far from the truth. It is still SMP.
#9: "How is this any different than multi CPU SMP? It isn't, except for compressing them to a smaller space. SMP has its problems as well and the number of CPUs does not create an exponential graph like Intel is implying."
Actually, there is one major difference: one socket is sufficient. Designing motherboards with two CPU sockets increases costs a lot, and so the SMP market in the desktop space is extremely small. There are so many applications that *could* use multiple threads that don't, mostly because programmers would end up spending tons of effort in improving performance on a small percentage of systems.
Just like the move to 64-bits will have more benefits in the future rather than in the short term, multi-core is looking to the future rather than the present. Software will have to be coded properly, but once that is done, multi-core will start to give us a lot of improvements in performance. Now that programmers have an incentive to support threading (probably almost all CPUs sold by late 2007 are going to be SMP), they will spend the time.
Cygni - "Im also wondering whether Intel will allow the Nforce4 to use the 1066fsb?"
Good question...
I would ASSUME that the answer to this is yes..."In for a penny, in for a pound" as they say. Considering all the trouble Intel has been having with 3rd party developers lately, I would assume that they will support Nvidia completely (if not, why support them at all?).
I thought the Pressler info was pretty shocking too, Viditor... especially considering its slated to replace the single die PD's. Wouldnt two phsyical cores over FSB be far slower? Really strange.
Im also wondering whether Intel will allow the Nforce4 to use the 1066fsb?
We all said it, and I think recent events and comments at IDF prove it beyond a reasonable doubt: Intel and Microsoft were in cahoots on XP 64-bit all along. XP x64 wasn't ever going to be released before Intel had 64-bit capable P4's and even Celerons ready to go.
I hadn't realised how very different Intel's dual-core is to AMD's until now...
Just this one line:
"The two cores in Pressler are totally independent, meaning that they must communicate with each other over the external front side bus and not over any internal bus"
That will be a HUGE latency problem when compared to AMD's dual-core! It takes 1 clock for AMD to communicate from one core to another because of Direct Connect (the on-die "switcher" between cores), I would guess that Intel will require at least 3-5 clocks (based on it taking 6-8 clocks for their SMP)...
even if intel did implement x86-64 at the "right time" at the launch of Win64, how is this the right time? Anyone who wants to use Win64 that intel now harks as the future, will have to buy a new chip, while A64 users already have everything they needed.
I dont even have an athlon64 but its just silly to say you are at the forefront of some new technology (64bit) when everyone has to go buy a new CPU to use it, amd had it right by equiping users before the "right time" so they were prepared.
Anyway. Yeah FBDIMMs look tight, so does rambus, the DDRx shit is so lame. by the time they get DDR2 up and running they find 100 ways to do it better and no one wants DDR2 anymore, its stupid.
Doesn't Intel's GOAT (Eh hmm IOAT) sound a lot like what nVidia was doing with their I/O chips? Intel should put "Not to scale" or "No real data was found" on every single one of those extrapolation graphs. I find it kind of funny how multi-core is the panacea to all performance problems. How is this any different than multi CPU SMP? It isn't, except for compressing them to a smaller space. SMP has its problems as well and the number of CPUs does not create an exponential graph like Intel is implying.
I am interrested in this FBDIMM and will need to do some checking around on that one. It looks interesting. RAMBUS is still at it. We'll see how things shape up.
I love the graph on page 4, where multicore just jumps ahead by leaps and bounds, with the "Performance" being exponential growth. I'm sorry but the last time we saw something like this it was the NetBurst graph taking us into "10 GHz Space" and lo and behold, well you know the story. I'm so tired of Intel just putting a band-aid on a bad idea for a chip (not a bad chip mind you, just designed by marketing people, not engineers). Multicore without onboard memory controllers, tacking on an extra meg of slower cache to Prescott...why are we not seeing samples of a new chip that aims to correct the problems of NetBurst rather than just adding more and more to Prescott like it really is going to change the fact? Until games get really multi-core oriented, this last generation of single-core products is going to be the best thing out there until probably late 06.
Guess what, none of these are anywhere near store shelves so CTFD (calm the F down).
The Nforce4 board finally adds some appeal to the latest Pentium 4s. I'd say that i875/865 were the last two exciting chipsets. 9xx has fallen short on innovation.
The SpinMeisters from Intel are realing blow smoke up the azzes of journalist, as usual. Only the gullible would belive the nonsense these folks peddle when they can't even deliver a P4 without a fire extinguisher. No one with a clue would touch any of Intel's current or short term products. Maybe by '07 Intel will have something worth considering but that remains to be seen.
The idea of VT is a good one I think -- it may be possible to run a small office on only 1 multicore, multithread system? (spreadsheets and email aren't exactly taxing)
and intel's idea of split dual cores may be future possibilities of selling cpu's scaled with cores rather than clock speeds?
eg. Extreme edition will have 8 cores, regular will have 4-7 cores ... celerons will have 1-3 cores... according to how the cpu's are binned?
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
19 Comments
Back to Article
sprockkets - Thursday, March 3, 2005 - link
FDimms, the way we get around having each processor have its own memory controller but still have a lot of slots.Of course the right time for you to enter 64 bit for the desktop is now, cause you managed to kludge it on your processors oh BFD.
Sorry Patty, but the AMD/Linux crowd didn't need to wait on the Wintel crowd for 64 bit.
Viditor - Thursday, March 3, 2005 - link
mickyb - "How is this any different than multi CPU SMP? It isn't, except for compressing them to a smaller space"I agree, at least for the Intel model. The AMD dual core is quite different however, in that the 2 cores are connected locally, and the MOESI protocol AMD uses allows for easy cache snooping.
mickyb - Wednesday, March 2, 2005 - link
#16 My statement was in context of the performance issue, not cost. The graphs imply that multi-core is the savior, when we have been experiencing what multi-cpu will do. Any graph that implies "exponential" growth for multi-core vs. single core is just a lie. I have been creating tools that analyze system performance for a while. It is far from the truth. It is still SMP.JarredWalton - Wednesday, March 2, 2005 - link
#9: "How is this any different than multi CPU SMP? It isn't, except for compressing them to a smaller space. SMP has its problems as well and the number of CPUs does not create an exponential graph like Intel is implying."Actually, there is one major difference: one socket is sufficient. Designing motherboards with two CPU sockets increases costs a lot, and so the SMP market in the desktop space is extremely small. There are so many applications that *could* use multiple threads that don't, mostly because programmers would end up spending tons of effort in improving performance on a small percentage of systems.
Just like the move to 64-bits will have more benefits in the future rather than in the short term, multi-core is looking to the future rather than the present. Software will have to be coded properly, but once that is done, multi-core will start to give us a lot of improvements in performance. Now that programmers have an incentive to support threading (probably almost all CPUs sold by late 2007 are going to be SMP), they will spend the time.
Viditor - Wednesday, March 2, 2005 - link
Cygni - "Im also wondering whether Intel will allow the Nforce4 to use the 1066fsb?"Good question...
I would ASSUME that the answer to this is yes..."In for a penny, in for a pound" as they say. Considering all the trouble Intel has been having with 3rd party developers lately, I would assume that they will support Nvidia completely (if not, why support them at all?).
Cygni - Wednesday, March 2, 2005 - link
I thought the Pressler info was pretty shocking too, Viditor... especially considering its slated to replace the single die PD's. Wouldnt two phsyical cores over FSB be far slower? Really strange.Im also wondering whether Intel will allow the Nforce4 to use the 1066fsb?
johnsonx - Wednesday, March 2, 2005 - link
We all said it, and I think recent events and comments at IDF prove it beyond a reasonable doubt: Intel and Microsoft were in cahoots on XP 64-bit all along. XP x64 wasn't ever going to be released before Intel had 64-bit capable P4's and even Celerons ready to go.Viditor - Wednesday, March 2, 2005 - link
I hadn't realised how very different Intel's dual-core is to AMD's until now...Just this one line:
"The two cores in Pressler are totally independent, meaning that they must communicate with each other over the external front side bus and not over any internal bus"
That will be a HUGE latency problem when compared to AMD's dual-core! It takes 1 clock for AMD to communicate from one core to another because of Direct Connect (the on-die "switcher" between cores), I would guess that Intel will require at least 3-5 clocks (based on it taking 6-8 clocks for their SMP)...
skiboysteve - Wednesday, March 2, 2005 - link
even if intel did implement x86-64 at the "right time" at the launch of Win64, how is this the right time? Anyone who wants to use Win64 that intel now harks as the future, will have to buy a new chip, while A64 users already have everything they needed.I dont even have an athlon64 but its just silly to say you are at the forefront of some new technology (64bit) when everyone has to go buy a new CPU to use it, amd had it right by equiping users before the "right time" so they were prepared.
Anyway. Yeah FBDIMMs look tight, so does rambus, the DDRx shit is so lame. by the time they get DDR2 up and running they find 100 ways to do it better and no one wants DDR2 anymore, its stupid.
and BTX is a joke
sphinx - Wednesday, March 2, 2005 - link
I agree with mickyb. I am more interested in the FBDIMM than NF4 for Intel and dual cores.mickyb - Wednesday, March 2, 2005 - link
Doesn't Intel's GOAT (Eh hmm IOAT) sound a lot like what nVidia was doing with their I/O chips? Intel should put "Not to scale" or "No real data was found" on every single one of those extrapolation graphs. I find it kind of funny how multi-core is the panacea to all performance problems. How is this any different than multi CPU SMP? It isn't, except for compressing them to a smaller space. SMP has its problems as well and the number of CPUs does not create an exponential graph like Intel is implying.I am interrested in this FBDIMM and will need to do some checking around on that one. It looks interesting. RAMBUS is still at it. We'll see how things shape up.
glennpratt - Wednesday, March 2, 2005 - link
That changes his point very little. And, YOU probably won't be buying crap out of pocket...Questar - Wednesday, March 2, 2005 - link
When will you guys realize how small the gaming market is?I'll buy more corporate systems this year than every gamer on this site will buy in the next five years.
Pete84 - Wednesday, March 2, 2005 - link
^^ Not just games, but every app too . . .ZobarStyl - Wednesday, March 2, 2005 - link
I love the graph on page 4, where multicore just jumps ahead by leaps and bounds, with the "Performance" being exponential growth. I'm sorry but the last time we saw something like this it was the NetBurst graph taking us into "10 GHz Space" and lo and behold, well you know the story. I'm so tired of Intel just putting a band-aid on a bad idea for a chip (not a bad chip mind you, just designed by marketing people, not engineers). Multicore without onboard memory controllers, tacking on an extra meg of slower cache to Prescott...why are we not seeing samples of a new chip that aims to correct the problems of NetBurst rather than just adding more and more to Prescott like it really is going to change the fact? Until games get really multi-core oriented, this last generation of single-core products is going to be the best thing out there until probably late 06.raskren - Wednesday, March 2, 2005 - link
^^You^^Guess what, none of these are anywhere near store shelves so CTFD (calm the F down).
The Nforce4 board finally adds some appeal to the latest Pentium 4s. I'd say that i875/865 were the last two exciting chipsets. 9xx has fallen short on innovation.
Beenthere - Wednesday, March 2, 2005 - link
The SpinMeisters from Intel are realing blow smoke up the azzes of journalist, as usual. Only the gullible would belive the nonsense these folks peddle when they can't even deliver a P4 without a fire extinguisher. No one with a clue would touch any of Intel's current or short term products. Maybe by '07 Intel will have something worth considering but that remains to be seen.xsilver - Wednesday, March 2, 2005 - link
The idea of VT is a good one I think -- it may be possible to run a small office on only 1 multicore, multithread system? (spreadsheets and email aren't exactly taxing)and intel's idea of split dual cores may be future possibilities of selling cpu's scaled with cores rather than clock speeds?
eg. Extreme edition will have 8 cores, regular will have 4-7 cores ... celerons will have 1-3 cores... according to how the cpu's are binned?
bersl2 - Wednesday, March 2, 2005 - link
--quote--Microsoft’s Jim Allchin came on stage and echoed Gelsinger’s statements with the simple line “it’s time.”
--quote--
Duuuuuuuu... no, really?
"And Wintel said, 'Let there be light!' But they were too slow to realize that the light had been on for quite some time."