The Consequence of Waking Up a Sleeping Giant: Intel Roadmaps Inside
by Kristopher Kubicki on January 25, 2005 7:44 PM EST- Posted in
- CPUs
Dual Core on the Horizon
So we lied originally - we have even better news. Dual core Smithfield processors, which are really nothing more than two Prescotts slapped together with independent caches, are scheduled to launch a little earlier than we originally claimed in previous roadmap articles. In fact, part of the push to launch so early seems to be to coincide with the 945/955 launch as those chipsets are the only ones to support the multiple core processors. Recall AMD's dual core launch strategy is to enable existing hardware (nForce4, K8T890, 8xxx) to run multiple cores. So while you can't plug a Smithfield into your existing 925X motherboard, it may be for the better. DDR2 has plenty of bandwidth to offer, but as we have seen in server benchmarks, multiple Pentium 4's competing on the memory bus can be quite slow. Dual core Pentium 4's might be horribly inefficient without DDR2-667, however that is another theory we can put to the test on launch day. If you look carefully, you'll see the Smithfields launching only at 800FSB. We find it slightly unusual that the entire 945/955 platform supports a front side bus speed that two $1000 SKUs utilize.
| Intel Dual Core PerformanceDesktop Lineup LGA775 | ||||
| Processor | Speed | L2 Cache | FSB | Launch |
| Pentium 4 840 | 3.20GHz | 2x1MB | 800MHz | Q2'05 |
| Pentium 4 830 | 3.00GHz | 2x1MB | 800MHz | Q2'05 |
| Pentium 4 820 | 2.80GHz | 2x1MB | 800MHz | Q2'05 |
Also note that the dual core processors on the desktop do not support HyperThreading. The server implementation of Smithfield, "Dempsey," has HyperThreading enabled. For database applications, this makes sense - although we have known for a long time that single threaded applications take a performance hit when a HyperThreading processor exclusively runs that program. Interestingly enough the Smithfield lineup has some very competitive price points according to the launch data. The 820, 830 and 840 models will launch at $241, $316 and $530 respectively - compare that to the Pentium 4 lineup today [RTPE: Pentium 4 775]. At today's prices that's only an $80 premium on the second core.
Facebook
Twitter
RSS
74 Comments
View All Comments
Postoasted - Sunday, January 30, 2005 - link
Heat dissipation and power consumption are my main concerns.GiantPandaMan - Sunday, January 30, 2005 - link
Oops,Anyhow I was just gonna say, who cares who comes out on top?
Without AMD we'd be using Pentium 166's?
Without Intel we'd be using Apples.
I hope both AMD and Intel are around for a very long time and they keep pushing each other to make better processors faster and cheaper. Same as Nvidia and ATI.
After all, what's gonna power Doom 6 and Half-Life 4?
The PS3 Cell "4.6" Ghz processor?
GiantPandaMan - Sunday, January 30, 2005 - link
JarredWalton - Sunday, January 30, 2005 - link
^^ Don't we all, AtaStrumf?Don't underestimate Intel, though. AMD's power usage on 90nm was substantially less than their 130nm. Part of that is almost certainly due to the lower clock speeds, but regardless we can be sure that Intel isn't going to be underestimating power/heat concerns.
Prescott 2M is really the first major redesign of a 90nm core since the release of the Prescott. Dothan has shown that Intel can get very good power and heat results, and I'm certainly curious to see how Prescott 2M compares to the original. We might assume that it will use more power and produce more heat, but depending on what other tweaks have been made, it may not be so bad.
There's a LOT going on at Intel right now behind the scenes (including some serious consideration of Pentium M on the desktop, I think). Don't just blindly stick with a company because of one or two good products, but wait for actual availability before coming to a conclusion. Also, don't just blindly hate a company for one or two mistakes. If we all did that, AMD would have never even lasted this long. K6 and earlier CPUs were thoroughly outclassed by Intel chips. Only K7 and K8 have been competitive designs, and only K8 has truly beaten Intel's counterpart.
To draw a parallel, look at the GPU market. The FX line was a disaster for NVIDIA, and the R3xx ATI parts completely outclassed it. The latest NVIDIA chips, however, are arguably the better chip - not in small part due to their actual retail availability. These are two competitors that are very close together in price, performance, and market share. The AMD/Intel match-up is not nearly as close. If NVIDIA could make a "comeback" if sorts, how much easier would it be for a company with the resources of Intel to address some of their problems?
As we've said before, only time will truly answer that question.
AtaStrumf - Sunday, January 30, 2005 - link
OMG, how much hype over even more power leaking, room heating, inefficient transistors. If you think 65 nm is gonna be any better/easier/faster transition than 90 nm was, you're, well, khm, how do I put it ... WRONG. It'll be even worse! I see nothing for me from Intel in 2005 or 2006. If they offered me Dothan in desktop, then maybe I’d think about it, but MORE preshott, no thanks!!!Nice to see Intel in action once again though. Hopefully it will get AMD off their arses and start putting out something NEW. They seem to have fallen asleep since Intel got themselves in Prescott trouble. I wonna see high volume 2nd generation 90 nm SS SOI chips with SSE3, with dual core following shortly thereafter.
fitten - Saturday, January 29, 2005 - link
Actually, AMD doesn't have plenty of headroom on frequency as seen by the major overclocks only reaching 2.7GHz. If you think that clock frequency is tied to process technology (as in 90nm means that the Athlon can hit 3.8GHz just like the P4) then I'd suggest some study in processor/circuit design. Just a little food for thought is that some P4s have double clocked adders (2.6GHz P4 has a 5.2GHz adder, for example).karlreading - Saturday, January 29, 2005 - link
BTW, scuse the typos, thats cause me mum braught me a nasty TRUST keyboard for xmas, ill have to sneek a decent one in the house soon. as for the spelling, poor british education LOLkarlos
karlreading - Saturday, January 29, 2005 - link
I use mainly AMD processors, i would almost call myself a slight fanboi!! but i cant help but deel thers some amd fans on here showing themselves up.personally i dont understand why intel do continue with netburts. 65nm might help them get some more frequency from it, prhaps 4.2ghz? but generally i dont see that helping much. AMD64 (em64T) is welecome, but i doubt we will see the benefits of 64bit till all the software developers pull ther fingures out. IMHO pentium M is very strong, overclock one a bit on a oldskool platform i.e AGP4x / pc2700 / 400fsb) and it can frighten a athlon fx55. If i was intel id be binning netburst and getting the pM acheteture ramped up and on the desktop ( althouhg i know the design dosent scale well per core variation, its not ment to )
one last thing, yes, intel going for non compatibility on current chipsets regarding smithfield dose allow them to bring platform / performance benefits, whils amds upgradable stance does allow for a cheaper alternative. this dosent have to be seen as a weekness for AMD. Remeber, Intel wanted the whole world and his dog to go 64 bit there way, via Itaniums EPIC archetecture. EPIC on optimised code can probably cain x86-64 hands down. But AMD pushed x86-64 and eventually one the day. Why, ease of upgrade, cost effectivness. so by allowing opteron / a64 users a cheap dual core upgrade, all AMD are doing is repeating a act that has previously done them well.
Lets face it, to lead the desktop market into x86-64, make the 800IB gorilla conceed and bring out a product, push a product ( EM64T ) that a few month ago they even denide a said they would never consider, is a achievment AMD should be proud of.
as for these road maps, very good, certainly better than i expected for 2005, thaught it would be stagnent, but, more Pentium M actio is whats needed form intel.
karlos
justbrowzing - Friday, January 28, 2005 - link
I read these roadmap articles and gosh they always sound like intel/amd is just about to launch sliced bread in Q-X of 200Y--and occasionally they do. Northwood and Athlon 64 were great products. But this time you kinda wonder just when (if ever)it'll pay off like that with all these prescott iterations.Seems like waiting for the sweetspot--when to upgrade an intel platform for performance & longevity from the intro of Prescott & into the foreseeable future--is & will be a very long, confusing wait indeed.
StriderGT - Friday, January 28, 2005 - link
What really strikes me is why intel keeps trying to keep the clearly sinking P4/netburst afloat... ? More cache, more FSB, more DDR2 more laughable stuff, that would really benefit database servers... and power bills.I can not understand the reason they do not promote dothan/derivatives as their desktop solution and put there all their latest hype/duo letters. The only reason I can think of is that they might be working on something bigger for dothan, like on-die controller (ala AMD) and keep it under wraps for a later launch
PS1 Dual core in, HT out for the desktop or to put it otherwise: our HT did fail miserably...
PS2 Those lads that bought the 925, well you need to upgrade in less than 6-9 months (dual), because intel is always thinking with the customer in mind and changes the chipsets/sockets like T-shirts. Don't you love to change T-shirts?:-)
PS3 Same applies to DDR2, why don't they launch tech stuff when they are ready to be launched?!? What is the point of having a memory tech we do not take advantage of.
What would be the difference if they launched DDR2 with the 955 or the next chipset, where it could actually make some difference? Answer:
Pure marketing hype...