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
DerekWilson - Saturday, March 13, 2004 - link
#10The problem with assigning a product number or rating based on average IPCxMHz is just as bad as AMDs method of rating against a reference processor.
In AMDs case, performance of a current processors is not always going to be exactly equal to the performance of a thunderbird at whatever MHz: Different programs will be more or less efficient in different situations.
In the case of average IPCxMHz, any given processor will have a completely different IPC for different programs in different situations.
Averaging IPC over all programs has essentially the same "goodness" as averaging relative performance of a specific processor over another.
Not only that, but IPCxMHz is still not the be all end all of performance in processors ...
if processor 1 is faster at instuction A and slower at B than processor 2, even if both processors have the same IPC and MHz, processor 1 will win benchmarks to which instruction A is critical and lose benchmarks that rely on instruction B.
It may seem like average IPCxMHz is the best solution, but even actual IPCxMHz for given code will only be a good indicator of relative performance between processors of the exact same architecture. going by average IPCxMHz is just as good (or bad) as AMDs scheme ... Though, I suppose both methods are a little better than pure MHz if you want to argue that point.
The bottom line is that the *only* real indicator of performance is the actual runtime of the programs you want to use.
I appologize if I didn't explain all that very well ... the point is a subtle one, and its pretty late so I might not be making much sense. :-)
Pumpkinierre - Saturday, March 13, 2004 - link
Somebody in an earlier post (abraxas?) suggested an IPCxMHz rating. An average IPC (instructions per clock cycle) could be established with a suitable test program. This would at least solve the horsepower side of the cpu spec. (sorely lacking in present descriptions and benchmarks). It would also stop the bleating about MHz or IPC being more important depending on the manufacturers product range. Cache size and FSB would still have to be quoted separately as their performance is subjective to application.Mangler - Saturday, March 13, 2004 - link
wow... I remember icomp!I did a google search and brought up this page about it.
http://www.ideasinternational.com/infofile/iicomp....
As they say the future is just the past re-organized!
johnsonx - Saturday, March 13, 2004 - link
Doesn't this remind anyone of the Intel iComp rating system from the way-back time? I still remember buying servers (from more than one vendor) equipped with "Intel Pentium 735/90" or "810/100" processors (in other words, the 90Mhz Pentium had an iComp rating of 735, while the 100Mhz had a rating of 810). The original iComp system gave a rating of 100 to an i486SX @ 25Mhz as a baseline. iComp 2.0 came out a couple years later, and I think a Pentium 75Mhz got the 100 point baseline rating then. Intel had some documents carefully explaining that Mhz wasn't everything. But the market didn't really bite on iComp, so it faded away...CrystalBay - Friday, March 12, 2004 - link
thier fabs cant keep upCrystalBay - Friday, March 12, 2004 - link
part of thier agreement...CrystalBay - Friday, March 12, 2004 - link
AMD sold out thier memory controller for SSE3 and thisViRGE - Friday, March 12, 2004 - link
nutxo, that's AMD's official line, but it's pretty obvious what they're really baseing it on.nutxo - Friday, March 12, 2004 - link
It was my understanding that the rating system employed by AMD was based on the thunderbird scaling higher ( like a 1700+ would be how fast a tbird would have to run to do the same work)and not to correspond with intels speeds.klah - Friday, March 12, 2004 - link
http://pc.watch.impress.co.jp/docs/2004/0312/ubiq5...7xx series - High end
5xx series - Main stream
3xx series - Value