Slashdot Mirror
Intel "East Fork" Technology Migration
Hack Jandy writes "When Intel's Centrino platform first unveiled, industry experts were surprised to see such great performance of the Pentium M, based off Intel's P6 (Pentium III) architecture. According to sources in the industry, Intel has officially adopted the approach to migrating Pentium M to the desktop (hence, "East Fork") to offset some of its Pentium 4 processor sales. Cheaper, slower, cooler, but higher performing processors are on the way to an Intel desktop near you!"
The Pentium-M was a seperate processor design (from the Israeli Intel team).
:p
It was more power efficient, and higher performing than the existing P4 line.
The processor was originally designed for Mobile applications but they've upped the clock speeds and retooled it a bit to bring them to the desktop.
They're faster and better engineered so everyone is a winner
Not really surprising, as the PIII has been faster than the PIV for awhile. Curiously at the same time, people were also noticing that NT4 was faster than 2000 at server tasks, yet most who reported such at the time were gagged by the no-publishing-benchmarks EULA fine print...
Because the Pentium-M does more per clock cycle than a destktop Pentium 4 this means that a 2.0 GHz Pentium-M is effectively as fast as a desktop Pentium 4 running at 3,2 GHz, and at the same time, it runs cooler and uses less power than the desktop processor.
In what way is the Pentium M "dumbed down?" Quite frankly, I'm firmly of the opinion that it's the best processor that Intel has produced to date, and I'm not alone in that view point.
The Pentium M is based on the old P6 core, with things like SSE added it to bring it up to current standards, and power saving circuitry of its own added in to suit the mobile role. The one major complaint about the chip is the fact that it's somewhat bottlenecked by a 400MHz FSB, but there's speculation that that's partly related to it currently being a mobile part. Even so, a relatively low clocked Pentium M compares very favorably to a much higher clocked P4.
Basically, the Pentium M is a move back to a P3 type design philosophy, away from the 30-stage pipeline madness Intel's gotten themselves into with Prescott. I fail to see how going with a more intelligent design is going with a dumbed down processor.
It seems the company is trying to go in a significantly different direction to retain its market dominance.
:0 411 151128.asp?S=Career%20Moves&A=MOV&O=FRGN
4 /intel_kill s_4gh/
s /display/2004 1111133206.html
1) New Non Engineer CEO
http://www.itweb.co.za/sections/business/2004/
2) GHz No longer a big deal after marketing it for so many years as the only major thing you need to know about the performance of a computer.
http://www.theregister.co.uk/2004/10/1
3) Shift to Better if not necessarily newer technology - see article above: oh who am I kidding....
http://www.xbitlabs.com/news/chipset
After reading for a while yesterday (after checking yesterday's k=note about the latest intel processor). I found that VIA (who bought Cyrix, I think the best processor at the 4x86 era, after National semiconductors almost broke it) has been working on it for a while. Perhaps the increase on "speed" (power consumption) was the strategy to take AMD and Cyrix out of the market? Now they want to come back because their processors are that inefficient?
Unfortunately, it seems like VIA is not focused on the PC market. Why? If anyone has some "fair" benchmarks, etc about this processors, it would be nice to read the results.
No - BTU may measure heat or energy, but heat output needs to divide this by time to give a rate, also known as power. So 1 GW is about 947817 BTU/s.
The Inquirer article concerned the 1st Gen P4 Xeons with 1MB L3 vs P3 Xeons with 2MB L2. The article is 2 years old.
I can understand why they're keen to experiment with different architectures, but I think such ideas are often panic measures.
Intel knowing that it's 64-bit offering is a lame duck and seeing AMD's opteron cleaning up in many areas is panicing and hoping to produce something radically better.
It was the worry that 32-bit CPUs were going to deliver that gave birth to the whole transputer concept (in the UK of all places).
Have a good read about the concept, it's not too disimilar to what is being proposed today (except the cores are more advanced).
http://en.wikipedia.org/wiki/Transputer
On the other hand, it's the slower clock speed that allows them to put more cache on the Pentium M chip. Basically, the larger a memory array you have, the slower it must be to access. This is why the Pentium 4 family has been lagging in cache sizes, as compared to other chips on the market.
Much of these speed increases are mostly a result of shrinking die sizes. Most archetectural changes revolved around the introduction of new instructions (SSE). A lot of work was also done to improve the effeciency of the PIII for the coppermine release (which saw a signficant speed increase). The PIV project, which worked in parellel and was doing a much more radical redesign, wasn't able to benefit from this work. The archetecture became different enough that new and much more thorough R&D would have to go into improving PIV effeciency. Unfortunately, the PIV design is one of brute force and these types of design improvements have limited returns for such designs.
.25 micron core. With efficiency improvents and a drop to .18 micron, the Coppermine was able to achieve excellent results with a max speed in the 1-1.13 ghz range (although 1.13 required nice cooling). Finally, the Tualatin didn't offer many changes other than moving the die size down to .13 and adding some improved heat dissapation technology. These babies got up to 1.4 ghz off the shelf and could clock up to 1.6 ghz in practice.
.13 micron die. For these reasons, many view the PIII coppermine/tualatin as one of the best made/designed CPUs of all time. Shrink them down to .9 micron and they would beat the crap out of PIV!
Just focussing on the PIII: the first to be introduced was the Katmai, which had a
The reason for the 1.6ghz ceiling? No it wasn't the CPU! Memory bandwidth was the reason these things couldn't go past 1.6. A PIII running at 1.6 ghz can effectively compete with 2.4-3.0 ghz PIVs!! If you could couple it with some high speed RAM, these things could have easily soared past 2.0ghz while remaining on a
Alas, the Pentium M is a PIII with MORE efficiency improvements. The capabilities of this design have to be WAY beyond the PIV. It's a discredit to Intel's leadership that they aren't marketing their best product!
I'm thinking of the power per acre coming off the INSIDE of a smelter, not the outside!
The Pentium Pro wasn't too popular because ... it's L2 cache was on the mainboard and thus Intel hand no QA over it and the L2 was often the cause of problems.
This is wrong.
PPro had its L2 cache integrated in the CPU package. It was the socket 7 chips which had L2 seperate and located on the mainboard. (Though, the K6-III SS7 CPU came with integrated L2, hence turning the cache on mainboard from L2 to L3).
The PentiumPro was *hugely* popular in terms of workstation and server sales - it was intel's first credible server/workstation CPU.
The P-II moved the PPro's cache from the CPU package to a seperate PCB, which meant they could use cheaper PCB surface mount of standard DRAM, and hence deliver P6 to the mass market.
MMX means nothing, firstly, MMX didnt exist until the P166MMX/P200MMX and secondly, PPro was never intended for mass consumer market, to which MMX was marketed, due to (in part) to aforementioned packaging cost and cost of the L2 cache (S7 boards cache was clocked at host bus 66MHz usually - PPro 's more expensive cache was clocked at CPU).
I use Friend/Foe + mod-point modifiers as a karma/reputation system.