AT News Update: Removing the MHz Myth, Again
by Kristopher Kubicki on March 12, 2004 10:03 PM EST- Posted in
- CPUs
News broke earlier today that Intel will most likely change its current "Megahertz" strategy in favor of a more subdued "Model Name" approach. This does not necessarily mean Intel will change its processors to a PR rating, like "3000+". Rather, the new model system sounds very familiar to AMD's Opteron approach, with three or four digit numbers replacing the product name.
Further sources claim that Intel isn't dropping the Megahertz convention entirely. Rather, it is simply adding a model name to other specifications as part of a new approach to customer awareness. Front Side Bus (FSB) speed, cache size and core name will play a larger role in differentiating one product from another. Unfortunately, this will not be an easy task as the last 10 years of microprocessor marketing has been almost exclusively "MHz" oriented.
Even when AMD began to push its processor "ratings" on the Athlon XP and MP CPUs, the ratings were designed to correspond with the clock speed of an equivalent Pentium 4. Undoubtably, this new name change may force AMD to rethink their labelings as well. Although AMD was originally correct to shy away from MHz labeling, now it will be more difficult for them to inform consumers on which processors are equivalent in performance.
Hurting AMD is probably not one of the main concerns driving the new model name approach. Take the mobile market, for example. Even though Banias (Centrino) is a brilliant processor (we have covered many times), customers are more apt to buy higher clocked Pentium 4 notebooks rather than a lower clocked, "low power" Pentium M notebook. Unfortunately, this is a case of when the trained "MHz" rating system is confusing and poor for the consumer. This ultimately becomes poor for the notebook manufacturer, as consumers complain about low battery life.
Intel's decision to move towards the new MHz less model name coincides with several new technologies, including Dothan (Banias's replacement) and Socket T (Socket 775). With an even more capable Pentium M, a mobile Pentium 4 killer, Intel and the consumer have a lot to gain by picking up the new lingo. We will have more Dothan information later this week.
How will this affect us, the consumer? Well, whether or not you agree with Intel's strategy, there is a more mundane advantage to adding a model number (other than OEM SKU) to each new processor; simplicity. If you just saw a new Intel commercial and want to buy a new computer, isn't it easier to remember "Intel 550" rather than "Intel Pentium 4 3.2GHz, 1MB Cache, 800MHz Front Side Bus, Socket 775."
Facebook
Twitter
RSS
31 Comments
View All Comments
Xentropy - Wednesday, March 17, 2004 - link
"Wow, it took CRAMITPAL more than 17 hours to respond to a thread just begging for his usual dumbass mutterings. Must be a new record."Abraxas was already here to cover him.
"AMD processors (aside from HT ability) are faster than intel CPUs at everything."
TrogdorJW - Tuesday, March 16, 2004 - link
The real problem with determining IPC is that it isn't set. It can vary wildly between different programs. For example, Quake 3 might end up with an IPC of 1.6 on P4 and 1.8 on Athlon, but then Return to Castle Wolfenstein might have IPCs of 1.3 and 1.5. Throw in the fact that certain applications are completely bottlenecked in areas other than the CPU (i.e. on RAM, hard drive, graphics, etc. performance), and IPC becomes even more difficult to quantify.Then you throw in the problem of chipsets, as Pumpkin pointed out, and it becomes almost useless. A P4 with only 256K L2 cache running at 2.8 GHz on an 800 MHz FSB with dual-channel RAM would likely beat out a P4 with 512K cache running at 3.06 GHz on a 533 FSB with single channel RAM by a decent margin. What are the "model numbers" going to be, were such chips to actually exist?
In Intel's defense, at least it sounds like they're saying the model names are really there as a guideline for the "Dummies" out there looking at computers, and they will still continue to list the GHz, Cache, FSB, etc. of their CPUs. Thankfully, most resellers of AMD CPUs also list this information. AMD almost went the route it sounds like Intel is going to take with their Opteron and Athlon FX lines, and I find the "model names" to be much more preferrable to "3200+".
Oxonium - Monday, March 15, 2004 - link
phantom505: Not to sound like a lameass Intel fanboy, but you are wrong on a few things. Intel beat AMD to 90nm. AMD still has no 90nm chips. Intel cpu's could use DDR chipsets from VIA for the Pentium III roughly at the same time they were introduced for the Athlon (again from VIA). Historically (until Prescott), Intel chips have had a lower power consumption than equivalent AMD chips, especially in the mobile arena where power consumption really matters. Yes, lately it seems like Intel has been following AMD with x86-64 and now the new processor rating system. It was pretty much inevitable that Intel would have to do those things. Whether they continue to follow is anyone's guess.phantom505 - Monday, March 15, 2004 - link
Face it, there is not good way to describe processor speed with the exception of some sort of somewhat hokey benchmark like AMD uses. People are entirely too critical about that. They are doing their best to give consumers a pretty good idea of where their processers lie.Unlike AMD, Intel that decided to ride out the lie as long as possible now finds itself having to get line with AMD, yet again. Sort of interesting how that keeps happening to follow AMD on almost EVERYTHING, DDR, transition to 90nm, x86-64, power consumption, the list goes on. AMD follows, what? SSE, SSE2, SSE3, just because otherwise Intel with their highly optimized software would run almost as fast. So as long as you make the software specifically for Intel then of course it will run better on say a P4 vs a K8. Personally I can't wait for them to declare that they were the first to use SOI and on chip memory controllers.
Pumpkinierre - Saturday, March 13, 2004 - link
Eman #24, I didnt say I liked it('MHz is liked because it is an absolute index'). In fact I've never seen Intel push it either nor AMD rubbish it (although indirectly they promote IPC). I've only heard on the web hearsay and anecdotal evidence that higher Mhz is better for sales. Personally I am not sure of the truth of this. But, like you, I prefer the MHzxIPC index (first proposed to my knowledge by Abraxas). However as others have stated in these posts, IPC is more difficult to quantify than clockspeed which is an absolute (WYSWYG- non-debatable). As you suggest the FTC could set the standard. I think it should be taken up and pushed by AT (known P4 lovers) initially.Pumpkinierre - Saturday, March 13, 2004 - link
Expornentially- a measure of size increment- definitely another well understood absolute.MrEMan - Saturday, March 13, 2004 - link
Pumkinierre, you stated "MHz is liked because it is an absolute index". How exactly is that the case?About the only thing which can be said for MHz/GHz is that it is accurate for similar architectures since one half of the IPC x GHz product is the same for both processors (the same would be true if the clockspeed was the same and the IPC was varied).
The only other thing that can be said for clockspeed is that Intel marketing loved clockspeed when that is all their processors could claim (certainly in many cases it wasn't performance).
As I have posted in the past, would you brag about your processor that it needed a 30% clockspeed advantage to perform as well as your main competitor... I sure wouldn't.
MrEMan - Saturday, March 13, 2004 - link
AMD tried to come up with a more meaningful performance metric with their True Performance Initiative, but since Intel didn't want a real measure of performance it didn't get anywhere.I find it interesting that many who defend(ed) the totally bogus "clockspeed is everything" (when they know/knew in fact that it was a bullshit way to measure performance of different CPU architectures), are now slamming IPC as a bad measurement metric.
Why not come up with performance test suite based on something much more accurate: the product of clockspeed and IPC? Maybe the FTC should step in like in the case of audio amplifier power specs.
It would show high GHz x low IPC (Intel) = low GHz x high IPC (AMD).
Oxonium - Saturday, March 13, 2004 - link
Expornentially: A naming system based on how fast a cpu can encode a porno from DVD to divx.Pumpkinierre - Saturday, March 13, 2004 - link
Sorry about the 'expomemtially' - exponentially. At least I'm consistent.I thought it was expornentially at first. That would have been a real Freudian slip. Hell, I've got to use my P4 for something!