Sign in with
Sign up | Sign in
Your question

INTEL MULTICORE APPROACH NEEDS PSYCHOANALYSIS

Last response: in CPUs
Share
March 1, 2006 9:15:09 AM

Actually everyone can understand what they did.
If there is no need to get rid off the FSB yet, then they simply wont do it, simple as >_>
March 1, 2006 9:52:08 AM

Interesting.
Related resources
a b à CPUs
March 1, 2006 10:01:34 AM

Whatever that dang article was bout

Intel instead of ditching there fsb doubled it - DIB (dual inline bus) and it acutually managed to outperform an AMD system with similar specs, altho the power and heat was extreme as ever (shows the aging P4 fsb).
March 1, 2006 10:45:45 AM

Correct, with DIB they EASILY have enough headroom to make it with quad-cores until 2008 then the FSB many people seem to hate even though it still does provide sufficient bandwidth will die.
March 1, 2006 11:24:01 AM

H 8O LY SM 8O KES!!! 1 8O WH 8O 8O PING GHZ!!!.. it will seems too impossible for a desktop processor but we'll see how it will going to look like.
March 1, 2006 11:52:04 AM

WHAT DO YOU KNOW ANOTHER ANTI INTEL AD FROM THEINQUIRER
March 1, 2006 12:40:45 PM

huh? why did you post this 9-inch everyone knows intel lies like sonnavab****
March 1, 2006 1:05:56 PM

Calling all AMD Fanboys! Calling all AMD Fanboys!

Technology Coordinator present and accounted for.

The article makes a great point. I can understand Intel using the FSB in the first dual cores, but moving forward they must change their approach. Intel's design gives you the benefits of two processors, but AMD's design gives you the advantage of Dual Core, there is a difference.

With Intel's current design they will need to make drastic improvements to the FSB, improvements that AMD has already implemented with HT.
March 1, 2006 1:18:42 PM

How's it going today Mr. Ruiz? :wink:
March 1, 2006 6:38:13 PM

Quote:
Whatever that dang article was bout

Intel instead of ditching there fsb doubled it - DIB (dual inline bus) and it acutually managed to outperform an AMD system with similar specs, altho the power and heat was extreme as ever (shows the aging P4 fsb).


Only benches I've seen are of 3.46GHz 1066MHz Bently Xeon w/ DDR2 533MHz FB-DIMM's and Dual INDEPENDENT bus's, compared to 2.4GHz Opterons w/ DDR400. Their FB-DIMM's are their saving grace, and even then the Xeon's are only about 5-12% faster. Once Socket F comes out, and you throw in DDR2-800, the Xeon will go right back under the table where it belongs.

I think Intel is in a temper tantrum and the reason they won't resort to On-Chip Mem Controller, is because AMD already uses it. It's kinda like the reason Intel won't touch SOI, because IBM has so many patents on SOI and Intel isn't too fond of IBM.

I concur, wtf did that article say?

~~Mad Mod Mike, pimpin' the world 1 rig at a time
March 1, 2006 7:11:32 PM

Intel has the on-chip memory controller while AMD only has an on-die memory controller (yes, there is a difference).
The Dempseys I think you are talking about are faster even at non-memory-depending benchmarks...
March 1, 2006 7:20:27 PM

Quote:
Intel has the on-chip memory controller while AMD only has an on-die memory controller (yes, there is a difference).
The Dempseys I think you are talking about are faster even at non-memory-depending benchmarks...


If you're referring to "On-Chip = Northbridge and On-Die = CPU", keep that sh*t to yourself. As for Dempsey, every application is "memory depending", regardless of how little or great. And as I said, Socket F > Dempsey.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
a b à CPUs
March 1, 2006 9:44:45 PM

Quote:
Whatever that dang article was bout

Intel instead of ditching there fsb doubled it - DIB (dual inline bus) and it acutually managed to outperform an AMD system with similar specs, altho the power and heat was extreme as ever (shows the aging P4 fsb).


Only benches I've seen are of 3.46GHz 1066MHz Bently Xeon w/ DDR2 533MHz FB-DIMM's and Dual INDEPENDENT bus's, compared to 2.4GHz Opterons w/ DDR400. Their FB-DIMM's are their saving grace, and even then the Xeon's are only about 5-12% faster. Once Socket F comes out, and you throw in DDR2-800, the Xeon will go right back under the table where it belongs.

I think Intel is in a temper tantrum and the reason they won't resort to On-Chip Mem Controller, is because AMD already uses it. It's kinda like the reason Intel won't touch SOI, because IBM has so many patents on SOI and Intel isn't too fond of IBM.

I concur, wtf did that article say?

~~Mad Mod Mike, pimpin' the world 1 rig at a time

INTEL TIMINA - A COPPERMINE CELERON WITH A RAMBUS MEMORY CONTROLLER ON CHIP - YEARS BEFORE AMD (NOTHING NEW)

CSI - COMMON SYSTEM INTERFACE (REPLACES FSB)

SOI - NOTHING WRONG WITH PENTIUM M'S AND THEY DONT USE IT AND THERE COLDER THEN THE AMDS, INTEL DONT NEED IT

SOCKET F - YEAH AND CONROE WILL BE OUT THEN TO WHIP THEM WITH DDR2-800, DIB AND QUAD CORE.
March 1, 2006 9:52:03 PM

Apache, I sense you're a nicely cooked Intel Fanboy, cheers for your effort though. BTW: I already knew about your great CSI, and no, Conroe will not beat Socket F or Socket AM2, sorry to burst your happy bubble fanboy. BTW: CSI is only planned for Xeon's and Itanium's as well as released in 2008, that leaves your Conroe out of the fun, :( .

P.S.: Several revisions to HyperTransport are in development, which is going to put CSI to shame, another sad moment :( .

~~Mad Mod Mike, pimpin' the world 1 rig at a time
March 1, 2006 10:03:14 PM

Quote:
Intel has the on-chip memory controller while AMD only has an on-die memory controller (yes, there is a difference).
The Dempseys I think you are talking about are faster even at non-memory-depending benchmarks...


If you're referring to "On-Chip = Northbridge and On-Die = CPU", keep that sh*t to yourself. As for Dempsey, every application is "memory depending", regardless of how little or great. And as I said, Socket F > Dempsey.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
Once again you show how small your knowledge is, sorry.
March 1, 2006 10:06:14 PM

dont insult my fellow amd fanboy or ill make sure you suffer :twisted: lol just kiddin or am i :?:
March 1, 2006 10:08:31 PM

Wow, Ycon, you stupid Intel Fanboy Newb. Please refrain from posting, I think I lost a few IQ points by reading your google.com knowledge posts.

Piddy: I am not an AMD Fanboy, I simply promote the technology that is better right now.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
March 1, 2006 10:24:27 PM

Quote:
Wow, Ycon, you stupid Intel Fanboy Newb. Please refrain from posting, I think I lost a few IQ points by reading your google.com knowledge posts.

Piddy: I am not an AMD Fanboy, I simply promote the technology that is better right now.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
ok sorry man :cry: 
March 1, 2006 10:33:56 PM

Quote:
I am not an AMD Fanboy, I simply promote the technology that is better right now.


Excuse me to barge in but... you really got this funny, fan...ny way of promoting it.
Aren't you being a little biased by any chance... on your own right, of course?! :) 


Cheers!
March 1, 2006 10:47:25 PM

I don't mean to disrespect you joset, and I won't, but I want to know what post I said that was a fanboy one, because stating FACTS and FANBOYISM are 2 different things.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
March 1, 2006 10:57:52 PM

Why would we need someone to "promote" "better" technology to us?!?
That evades me all the time. If the technology is _better_ it doesn't need promoting anyway.

About how Intel Dual-core is "bad", a simple test on my Pentium D 930:

D:_ISO_>gogo -b 256 -m s -q 0 test.wav test.mp3

encode test.wav to test.mp3
GOGO-no-coda ver. 3.11 (Jan 10 2003)
{ 100538/ 100538} 100.0% ( 78.14x) re:[00:00:00.00] to:[00:00:33.61]

test.wav is 43min 46sec (441.8MB)

Encoding speed was 39x on my Pentium 630 with the rest of the system unchanged -- I just upgraded the CPU. Speedup? Figure it out yourself.
March 1, 2006 11:15:13 PM

By "Promote" I simply mean to express my views, I am not flaunting it off (at least I hope I am not). I didn't say Intel Dual-Core was bad, I said that it was inferior to that of the Athlon 64 X2's.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
March 1, 2006 11:23:24 PM

I'm stayin out of here bye mike hope you have fun pwning these noobs
March 2, 2006 12:04:33 AM

Quote:
Once Socket F comes out, and you throw in DDR2-800, the Xeon will go right back under the table where it belongs.

I think Intel is in a temper tantrum and the reason they won't resort to On-Chip Mem Controller, is because AMD already uses it.


Well, there you have it... for instance.

And, should I thank you for not disrespecting me?


Cheers!
March 2, 2006 12:07:48 AM

Okay, but 1 is a fact and 1 is speculation. Xeon is inferior to Opteron 64, that is a FACT, Intel not using onboard mem controller was a personal opinion and could be a fact, but it is impossible to know for sure.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
a b à CPUs
March 2, 2006 1:14:14 AM

Quote:
Apache, I sense you're a nicely cooked Intel Fanboy, cheers for your effort though. BTW: I already knew about your great CSI, and no, Conroe will not beat Socket F or Socket AM2, sorry to burst your happy bubble fanboy. BTW: CSI is only planned for Xeon's and Itanium's as well as released in 2008, that leaves your Conroe out of the fun, :( .

P.S.: Several revisions to HyperTransport are in development, which is going to put CSI to shame, another sad moment :( .

~~Mad Mod Mike, pimpin' the world 1 rig at a time


Everyone on the forums is "AMD this" and "AMD that" and whatever, like AMD fanboys, im not an Intel fanboy (im a fanboy for P6 tho, they never let us down :D  good old days) and i know there is alot that AMD did before Intel aswell - DDR (ram AND fsb) and 64bit standard (for desktops) anyone? Just sick of the "Perfect AMD" idea, no offence MadModMike.
March 2, 2006 1:35:40 AM

No offence taken. But one thing you must agree with, is that on a CLOCK FOR CLOCK basis, the Athlon 64 is FAR superior to that of the Pentium 4 :) . Only Conroe will tell, but the Common System Interface has been facing some problems in development and is now in view to be on par with current HyperTransport performance.

But, it is also a fact that HyperTransport is expecting major revisions and these will increase speeds. HyperTransport 3.0 is supposed to be Double that of HyperTransport 2.0, use DDR3 and FB-DIMM's. HyperTransport 4.0 is supposed to be using DDR4 and FB2-DIMM's. This is all expected in the next 5 years, whereas Intel's CSI is not going to even come about for another 2-3 years. It will be interesting to see the next battle, but in my opinion, the Opteron 64 CPU will be the pinnacle of performance for the next 10 years.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
a b à CPUs
March 2, 2006 5:29:22 AM

Actually i heard CSI was cancled (maybe HTT came out and was better, or too many pins?), but yeah.

Intel is prolly going to keep there FSB and clock it up more within/for a year or two check this out http://www.computerbase.de/bild/news/11597/15/ - note the 600mhz FSB (QDR 2400mhz or 2.4ghz) - maybe thats intels answer? and the original article http://www.computerbase.de/news/hardware/prozessoren/intel/2005/august/idf_benchmarks_sossaman_yonah/

I think AMD's A64/K8 performance will only increase with clock speed rather then HTT and RAM - THG used DDR600 with an A64 and the same clock speed and the performance didnt really gain much, so 65nm and beyond might be there answer.
March 2, 2006 6:04:32 AM

Quote:
Actually i heard CSI was cancled (maybe HTT came out and was better, or too many pins?), but yeah.

Intel is prolly going to keep there FSB and clock it up more within/for a year or two check this out http://www.computerbase.de/bild/news/11597/15/ - note the 600mhz FSB (QDR 2400mhz or 2.4ghz) - maybe thats intels answer? and the original article http://www.computerbase.de/news/hardware/prozessoren/intel/2005/august/idf_benchmarks_sossaman_yonah/

I think AMD's A64/K8 performance will only increase with clock speed rather then HTT and RAM - THG used DDR600 with an A64 and the same clock speed and the performance didnt really gain much, so 65nm and beyond might be there answer.




WAIT a minute here.

How can the Core run at 1500MHz and the BUS @ 2400MHz?????????


If the BUS really runs @ 2400 MHz and the core only runs @ 1500MHz the BUS will easily flood the CPU.

That does not make sense. A 1:1 Core clock to bus ratio would make more sense but even then a 1500MHz BUS sounds a little far fetched, especially for a P6 derived architecture.

I suspect CPU-Z is showing incorrect specs for this CPU.
a b à CPUs
March 2, 2006 8:55:26 AM

one thing is for sure the north bridge heatsink and fan is as big as the processors, and there dual core per chip giving the cpus 1200mhz each, i dont think its read wrong - the fsb is common nothing new (unless its using DIB and cpuz counts it as 2x1200?)
March 2, 2006 11:44:42 AM

Quote:
(unless its using DIB and cpuz counts it as 2x1200?)


Correct (see previous slides...)

It's also a BTX form factor, for what matters...


Cheers!
March 2, 2006 5:48:44 PM

Wow, a FSB that is faster than the CPU Core. Let's see here....hmm...no performance gain? Yep...that pretty much sums it up.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
March 2, 2006 11:57:22 PM

Quote:
If the BUS really runs @ 2400 MHz and the core only runs @ 1500MHz the BUS will easily flood the CPU.

I suspect CPU-Z is showing incorrect specs for this CPU.



If it's Intel's "Sossaman", CPUz's recognizing a "Yonah"; in any case (and with/without DIB), it'a two core chip, 1.5GHZ each (3.0GHz aggregate). Hence, no prob if the bus runs at 2.4GHz. After all, it's a parallel BUS, not a serial interconnect like HTT.
a b à CPUs
March 3, 2006 5:57:41 AM

before i forget - its 2x dual core 1.5ghz vs 1x dual core AMD, and the * for some speeds is calculated (not real) and where it "should be" if it was at that speed.

The whole system ran using under 100w of power!
March 3, 2006 5:20:38 PM

So 2 PHY Intel CPUs with 2 cores each = 4 cores?

vs.

1PHY AMD CPU with 2 cores = 2 cores?

Is that right?
March 3, 2006 6:08:31 PM

Quote:
So 2 PHY Intel CPUs with 2 cores each = 4 cores?

vs.

1PHY AMD CPU with 2 cores = 2 cores?

Is that right?


Yea, that's a fair test. Let's compare 2 CPU's to 4.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
a b à CPUs
March 3, 2006 10:11:23 PM

Yeah read the benchmark "2x Dual Sossaman 2,0 GHz*" (* meaning what it COULD be, using the other benchmarks and calculating the performance at 2ghz) and its "beating"1x Dual Athlon 64 X2 4800+" - sort of doesnt mean much.

Ignoring the 2ghz fake, Intels using ~1ghz less per core and twice the cores (4x1.5ghz vs 2x2.4ghz) so fair but not, im impressed at the total system power consumption - under 100w for once!
March 3, 2006 10:13:52 PM

Quote:
Yeah read the benchmark "2x Dual Sossaman 2,0 GHz*" (* meaning what it COULD be, using the other benchmarks and calculating the performance at 2ghz) and its "beating"1x Dual Athlon 64 X2 4800+" - sort of doesnt mean much.

Ignoring the 2ghz fake, Intels using ~1ghz less per core and twice the cores (4x1.5ghz vs 2x2.4ghz) so fair but not, im impressed at the total system power consumption - under 100w for once!


So basically you're telling me these are Benchmarked Benchmarks....yea....I'm laughing right now.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
a b à CPUs
March 3, 2006 10:29:17 PM

i also concider the benchmark app to be intel favoured
March 3, 2006 10:34:16 PM

Quote:
i also concider the benchmark app to be intel favoured


SPEC is about the only Non-Intel-Favored benchmark.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
March 4, 2006 1:30:21 AM

Although the review reports "August 2005", it makes me wonder about the BTX form factor; It seems that Intel is well commited to it, with the next CPU generations...

Bad news, if it's to go forward; it has some strong points [few, in my opinion] but leaves a lot to desire...
a b à CPUs
March 4, 2006 2:21:45 AM

the link shows a sossman system for servers not for desktop cpus, and btx i think for now is more like a oem (eg dell) only idea to make it easier for em?
March 4, 2006 3:05:55 PM

Quote:
the link shows a sossman system for servers not for desktop cpus, and btx i think for now is more like a oem (eg dell) only idea to make it easier for em?



Well, perhaps Intel will start with the BTX server market, first; but, if you care to have a look at their [website] white papers on the subject, BTX form factor IS intended to replace ATX in all fronts, asap.

The overall market (and OEMs) have been dragging ATX as far as possible, due to its well established versatility, cost & pervasiveness. However, there's yet no consensus among the relevant OEMs although the original proponent, Intel, should have the last & decisive word.

One thing's for certain: We've seen this turn before and we'll see it again. ATX is going to be replaced, sooner than later. I just don't think BTX is (will be) the consensual ultimate form factor.

In my humble opinion, any form factor platform to be, must rely upon the basic principles of thermodynamics; all else must adapt to comply with these principles and, I don't think Intel BTX is (will be) that platform...


Cheers!
March 4, 2006 3:10:22 PM

Quote:
the link shows a sossman system for servers not for desktop cpus, and btx i think for now is more like a oem (eg dell) only idea to make it easier for em?



Well, perhaps Intel will start with the BTX server market, first; but, if you care to have a look at their [website] white papers on the subject, BTX form factor IS intended to replace ATX in all fronts, asap.

The overall market (and OEMs) have been dragging ATX as far as possible, due to its well established versatility, cost & pervasiveness. However, there's yet no consensus among the relevant OEMs although the original proponent, Intel, should have the last & decisive word.

One thing's for certain: We've seen this turn before and we'll see it again. ATX is going to be replaced, sooner than later. I just don't think BTX is (will be) the consensual ultimate form factor.

In my humble opinion, any form factor platform to be, must rely upon the basic principles of thermodynamics; all else must adapt to comply with these principles and, I don't think Intel BTX is (will be) that platform...


Cheers!

IMO, the reason BTW hasn't taken off yet, is because there really is no need. Nowadays, enthusiast users know how to cool their rigs down to 20c's idle, you can't get much better than that on AIR and WC won't care what formfacter you have. Average users, such as those that buy Dell's, eMachines, etc., they aren't really concerned with heat because they'll already have a crappy machine inside that thing anyways. I see Intel selling new BTX CPU's, this must mean you need a new CPU for a BTX motherboard, I have not read alot on BTX as I do not care to, but if it means new components and such, it will never take off.

~~Mad Mod Mike, pimpin' the world 1 rig at a time
March 4, 2006 6:37:04 PM

Quote:
IMO, the reason BTW hasn't taken off yet, is because there really is no need. Nowadays, enthusiast users know how to cool their rigs down to 20c's idle, you can't get much better than that on AIR and WC won't care what formfacter you have. Average users, such as those that buy Dell's, eMachines, etc., they aren't really concerned with heat because they'll already have a crappy machine inside that thing anyways. I see Intel selling new BTX CPU's, this must mean you need a new CPU for a BTX motherboard, I have not read alot on BTX as I do not care to, but if it means new components and such, it will never take off.


No, it doesn´t require any new components, apart the cooling... "THING"!

http://www.hexus.net/content/item.php?item=991&page=2

Most first-tier case manufacturers already have optional BTX tray adapters (just in... case!).

I believe the compromise between a more (?) efficient cooling layout (horizontally set-up) and a proprietary, HUGE & expensive (?) HSF, doesn't appeal at all, either to me (irrelevant, anyway), either to first-tier OEMs (definitely relevant!). The option for the same slot orientation, is obvious in any future form factor implementation; and, to me, it seems more thermodynamically effective if implemented vertically (heat goes up, even if it requires some 'help'). Of course, it's not easy to figure out how vertical PCIe slots could do it, anyway...


Cheers!
March 4, 2006 7:50:51 PM

Quote:
I am not an AMD Fanboy, I simply promote the technology that is better right now.


Excuse me to barge in but... you really got this funny, fan...ny way of promoting it.
Aren't you being a little biased by any chance... on your own right, of course?! :) 


Cheers!

Well it looks like consumers are now talking with their dollars. So the only ones holding intel up right now are none other than the fan boys.....

http://yahoo.reuters.com/financeQuoteCompanyNewsArticle...
March 4, 2006 8:00:43 PM

Let me get something straight... if a PC has an FSB that's faster than the CPU, that's a bad thing??? I always thought the FSB was a bottleneck, and going for the 1:1 approach as mentioned earlier was the way to go... CPU's seem to go higher than the FSB can handle, so a reversal at this point could be a good thing.

Let it be known (again): I'M AN AMD FANBOY!!! INTEL SUCKS!!!

If any of you eggheads are listening, get us some faster hard drives, stop worrying about the rest of the components!!!

PDH-NicFury
March 4, 2006 8:43:06 PM

wow, does anyone still do real comparisons of systems or do they just read benchmark scores? Neither of the next generation processors are released, and when they are, one site will say something, and the other will say something completly different. As it stands, HT is Great... so is HT tho... (hypertransport and hyper threading respectivly). My favorite thing is my MSI p4n board, which used Intel chips with Nforce 4 chipset, so, it has intel processor with AMD stuff includeing Hyper transport on every device, save the cpu (and it is kinda faked on the ram aswell... since AMD didnt see DDR2 as a big thing).

Which brings me to my next point, WHO CARES about the chip wars. All we need to do is get some accurite testing done (as in multiple boards, multiple chips, multiple ram configs etc...) and see which PLATFORM does best.

BTW AMD AND INTEL SUCK!!!!! THEY ARENT RISC!!!!
March 4, 2006 8:53:21 PM

Quote:


If any of you eggheads are listening, get us some faster hard drives, stop worrying about the rest of the components!!!

PDH-NicFury


not all applications need faster disk drives, but they do need faster memory and processing power.
!