Athlon XP's naming scheme is getting out of hand.

[Kevin]

When the Athlon XP 1600+ made its debut, its XP rating was 200 mhz faster than its actual mhz rating. However, every 66 mhz speed jump these CPUs make, AMD boosts the PR rating 100. The Athlon XP+ 2200 will run at 1.8 ghz. That is a 400 mhz speed difference. In my opinion, this is getting way out of hand. I would rather see AMD extend the pipeline of the CPU like Intel did so they could have a CPU that performs less per mhz, but scales well to higher speeds. Oh well... just a rant.
-Kev

[Kevin]

oh and for the record, i originally supported AMD's idea to use a PR rating system vociferously, but I just think they are twisting it a bit too far now.
-Kev

[WebMasta33]

Hey, it sells joe sixpack more processors the more they bload the speed ratings....

In reality, Joe Sixpack is who AMD wants as their primary buyer, not us... We're used for name recognition, and then bye bye....

[Haddy]

im not a big fan of the rating crap cause its just hard to remeber...cause like u said they use like 100mhz jumps and then only like 40 or 60mhz jumps....the industry needs a standard

[Kevin]

In reality, Joe Sixpack isn't even buying it. Ever since the Athlon XP and Pentium 4 have butted heads, AMD has lost marketshare. I think Intel cheated to start with by extending the pipeline of the P4 and making their "new" processor clock for clock worse than the last one. However, AMD needs to find a way to beat them at their own game because marketing can't pull you out of everything.
-Kevin

[Haddy]

what they need to do is market more to people like dell and compaq/hp....they pretty much cornered the market for homebuilt stuff (bout as much as their probaby gona anyways)

[Brad]

Kev, sites have been saying this since the 2000+ came out, I wholely agree, worsening this is the P4 Northwood with the extra cache, and when they have 533mhz fsb the pr scheme will be even more inaccurate

[MeltMan]

I think processors should be advertised with a standardized benchmark rating. Or the average of two or something to even the playing field a little.

[Razor6]

The PR numbers are derived by taking the processor speed multiplied by 1.5 and subtracting 500. So the only reason the gap gets larger is because of that multiple.

Kevin, why should AMD lengthen the pipeline to get higher speeds that at the same time would lower performance? AMD still has room left in the core, all they do is lower the process size.

[phreenet]

It is all a marketing scheme to sell more processors, Intel obviously owns the MHZ market boasting 2.2 and 2.4 Ghz processors for the consumer, I know Mhz means jack crap to the informed consumer, so AMD feels that it needs to use this bogus and radomly created PR system to compeate. If you look at the releases of the Intel processors, AMD isn't short behind they always try to stay on par with the latest processor or 100 points of PR behind, for example when the P4 2 Ghz came out they had to have this 1900+ or 2000+. So when the 2.4 Ghz becomes more mainstream AMD will have to add another 66 Mhz to their current processor and make it a 2300+ or 2400+. Thats proof that its totally made up. Furthermore its a fact that AMD processors can do much more work per clock cycle, and Intel can compeate with that by adding more cycle's to their processors, but I wouldn't go as far to say its cheating
.

[Razor6]

Well the PR numbers aren't really made up. Every new AthlonXP boosts the speed by 66Mhz. When you use the equation the PR number goes up by 100 points and the difference between the actual speed and the PR number increases 33 each time. AMD states that the PR numbers try to equate Palomino performance to Thunderbird performance, ie: what speed TBird would I need to equal an AthlonXP.

[phreenet]

That has to be BS, why would they even worry about that, they called them XP so when consumers think of Windows XP they will think "Hey maybe I need an XP processor too". They have to be comparing it to the Pentium IV processors, they aren't worried about consumers getting mixed up over TBird/XP processors.

[Kevin]

Originally posted by Razor6
Kevin, why should AMD lengthen the pipeline to get higher speeds that at the same time would lower performance? AMD still has room left in the core, all they do is lower the process size.

There is not that much room in the core left at all, first of all. I think AMD should extend the pipeline and such merely for marketing purposes. But actually mebbe the hammer will be able to scale well.
-Kev

[jtroutma]

My own personal rant is this:

I am not very happy with the way Intel is (personally) cheating everyone by making a processor that could barely keep up with its own predessesor (back when the PIII could out perform a P4). Intel is playing a marketing stratagy and it sucks! They know that X.XGhz sell and so they made their processors so they could overlclock them to hell and back JUST so they could sell them! Personally that is BS!

AMD's design for the original Athlon processor was excellent and made absolute sence to me when I first read up about their architecture. Not to mention that you can do more with less is an excellent idea. I mean HELL, with the economy going the way it is, more with less is the way everything is going. (Side note, nothing more

I use to be an Intel only guy for several reasons but mostly because of RAW performace figures, where Intel beat AMD hands down. Then AMD came back and designed a CPU that was as good (if not better) than what Intel had at the same time. Then Intel and AMD were duking it out against eachother with identical clock speeds and after some revisions to Intels PIII, Intel was competing or comming up a close second to AMD. Could it be possible that Intel realized that they couldn't compete with AMD clock per clock so they desined a chip that would just go for all looks and decent performace at ultra high clock rates? HMMMMMMMMMMMMMMMMMM............... I think so!


GRRRRRR The idea that Intel is just screwing a great innovation from another company by tricking the customer just burns me........

Ok.........

In a nutshell, AMD has made an great attempt and has succeeded in making a good name for itself with the Athlon and Intel is trying to steal their glory. PERSONALLY, from a technical stand point, AMD's processor is superior and Intel knows this. It is not AMD's fault that they had to go to the stupid PR rating again but Intels becuase they played the marketing game and now has dragged AMD into it in order to say competative in the UNKNOWING CONSUMERS MIND!

Ok enough ranting for one day.........and I know that I am going to get some bashing for this but HEY! Its my opinion which I am basing off the FACTS that are presented in front of me. Have a problem with it? Your not the only one.

Have a great day!

[Razor6]

Originally posted by phreenet
That has to be BS, why would they even worry about that, they called them XP so when consumers think of Windows XP they will think "Hey maybe I need an XP processor too". They have to be comparing it to the Pentium IV processors, they aren't worried about consumers getting mixed up over TBird/XP processors.

I would say that AMD is definitely not comparing the AthlonXP to the P4's. Starting on this link you can see how a PR 2000+ AthlonXP beats the 2.0Ghz P4A, by either a little or a lot. It does loose in one test that does not use the latest WMA encoder and also in the web benchmark. It is worth noting that in several tests it is neck in neck with the performance of a 2.5Ghz P4A. So thats why I don't think the PR numbers try to compare the AthlonXP to the P4.

[Razor6]

...and so what if AMD is comparing the AthlonXP to the P4's with the PR numbers? If Joe puplic buys a machine with a 2000+ AthlonXP instead of that 2.0Ghz P4A machine he was looking at he will at least be getting either equal or greater performance in all but a few instances. However I do think that the buyer should first be educated to the fact that 2000+ does not mean 2.0Ghz. Of the machine stickers I've seen for Athlon based systems they state both the PR number and actual speed so I don't think it should be much of problem.

[ColdFlare]

Originally posted by jtroutma
My own personal rant is this:

I am not very happy with the way Intel is (personally) cheating everyone by making a processor that could barely keep up with its own predessesor (back when the PIII could out perform a P4). Intel is playing a marketing stratagy and it sucks! They know that X.XGhz sell and so they made their processors so they could overlclock them to hell and back JUST so they could sell them! Personally that is BS!

AMD's design for the original Athlon processor was excellent and made absolute sence to me when I first read up about their architecture. Not to mention that you can do more with less is an excellent idea. I mean HELL, with the economy going the way it is, more with less is the way everything is going. (Side note, nothing more

I use to be an Intel only guy for several reasons but mostly because of RAW performace figures, where Intel beat AMD hands down. Then AMD came back and designed a CPU that was as good (if not better) than what Intel had at the same time. Then Intel and AMD were duking it out against eachother with identical clock speeds and after some revisions to Intels PIII, Intel was competing or comming up a close second to AMD. Could it be possible that Intel realized that they couldn't compete with AMD clock per clock so they desined a chip that would just go for all looks and decent performace at ultra high clock rates? HMMMMMMMMMMMMMMMMMM............... I think so!


GRRRRRR The idea that Intel is just screwing a great innovation from another company by tricking the customer just burns me........

Ok.........

In a nutshell, AMD has made an great attempt and has succeeded in making a good name for itself with the Athlon and Intel is trying to steal their glory. PERSONALLY, from a technical stand point, AMD's processor is superior and Intel knows this. It is not AMD's fault that they had to go to the stupid PR rating again but Intels becuase they played the marketing game and now has dragged AMD into it in order to say competative in the UNKNOWING CONSUMERS MIND!

Ok enough ranting for one day.........and I know that I am going to get some bashing for this but HEY! Its my opinion which I am basing off the FACTS that are presented in front of me. Have a problem with it? Your not the only one.

Have a great day!

Dude if you are talking about competiting processor speeds, i bet intel could care less, as they already own most of the market. They still have the fastest processor now on the market, and amd is coming in second, by a medium margin. They have the speed crown now with raw mhz, and they own most of the market, so i dont think they have anything else to prove.

The whole pr rating thing messed everyone's mind up. Its confusing to them to have to calculate the speed.

Dont blame intel, cause intel set most of the mhz standard until AMD came along with the t-bird and added an extra .5 ipc to each clock. now you think who messed up the entire mhz speed rating.

Anyway, with all the money intel has, you can sure bet they can destroy amd with a flick of a finger. They wouldnt want to tho, as they can keep them in their little corner, while Intel practices their mastery in marketing and business, while protecting from the monopoly trust busters.

[jtroutma]

Hmmm

I will attempt to be polite with my reply however if I offend you; I am sorry but at the same time not

now....

Yes......Intel controls the majority of shares of the market in almost every field.
Yes......Intel currently hold the speed crown with their latest P4 2.5Ghz CPU
Yes......The PR rating that AMD is using is confusing (even to me)

However, I find a problem with your statement:

"Dont blame intel, cause intel set most of the mhz standard until AMD came along with the t-bird and added an extra .5 ipc to each clock. now you think who messed up the entire mhz speed rating. "
If anyone screwed up the Mhz rating, it would be Intel... Reasoning:
Among the Computer engineer community, it is a well known fact that Processor speed (measured in Mhz) is a very inaccurate way of describing how "fast" a CPU is. A CPU's true processing power is rated at how many MIPS (Million Instructions Per Second) the CPU can do in a single second NOT how many clock cycles it runs at per second.

The AMD CPU OBVIOUSLY has a higher MIPS rating that the P4 at an equal "Clock Rate" becuase if you were to take an Athlon 1Ghz (100FSB) and compare it to a P4 1Ghz (100x4FSB), the Athlon "would most likely" pound on the P4 and then ask for more! ***NOTE: I say "would most likely" becuase there has never actally been an actual bench test of this nature. Wonder why?

Now what you are saying is that AMD screwed up the Mhz standard by making their CPUs more effecient than just ramping up "clock speeds" like everyone else? The only reason WHY AMD HAD to go with a PR rating was becuase the majority of the customers only look at the "clock rate" and automatically means faster performace. Well OK............. Lets compare a Celeron 1Ghz (66FSB) to a PIII 733Mhz (133FSB). The average user would automatically assume the Celeron was the faster processor when the PIII clocked WAY lower would cremate the Celeron; not because the Celeron is slower but becasue the PIII has a better design and other advantages in order to bring its MIPS count way up. (Granted if the customer could also see the FSB rating, it may change their opinion; most likely not because most customers dont even know what FSB their CPU is running at, let alone what a FSB is.

Intel is counting on the "un-educated" buyer seeing a very high "clock speed" number and seeing the name INTEL INSIDE on the box an thinking that the $2000 they are spending is well worth their money. Sad, becuase you could get an AMD system that is faster for about $1000 (Based on the old I850 boards, Dual RIMMS, and high price tags of the P4)

PLEASE correct me if I am wrong in this department.........

Long and short of it..............Intel Screwed Up the Mhz rating before the Athlon XP PR rating did (not including the old 586 battles w/ IBM and Cyrix in the picture)

YES AMD never should have reverted back to the PR rating, BUT Intel forced them to do something in order to make their products "appear" on the ball with Intel to the "un-educated" consumer.

(........now how much flak am I going to get ...............

[jtroutma]

BTW you are doging AMD for trying to make something better than it already was and want them to stop?


Heh, too bad someone didn't do that to Gates back in the day and spare us the trouble of his wonderful gift to the computer society...............Windows95

[phreenet]

would say that AMD is definitely not comparing the AthlonXP to the P4's. Starting on this link you can see how a PR 2000+ AthlonXP beats the 2.0Ghz P4A, by either a little or a lot. It does loose in one test that does not use the latest WMA encoder and also in the web benchmark. It is worth noting that in several tests it is neck in neck with the performance of a 2.5Ghz P4A. So thats why I don't think the PR numbers try to compare the AthlonXP to the P4.

Actually man, if you look at AMD's wording on the subject, they usally say something along the lines of "Out performs competitive processors at this speed." All AMD is doing is using numbers to make their processors look as big as an Intel when the consumer is buying one. Why else would they pick such a random naming of processors? All their PR ratings do is keep pace with Intel mhz numbers. So what other competitive processor do you know thats 2000+ Mhz? Not that many, I think Intel is the only company that sells processors that go over 2.0 Ghz out the box.

[phreenet]

Lets compare a Celeron 1Ghz (66FSB) to a PIII 733Mhz (133FSB). The average user would automatically assume the Celeron was the faster processor when the PIII clocked WAY lower would cremate the Celeron;

Well not really because the consumers also know which make and model works best.

And one of the reasons Intel holds the PC Market is because when OEMs tried and still try to incorporate Athlon based solutions into their product line up they found this:

1) Powering an Athlon solution means needing something more powerful then a standard 250W PSU (back in the PIII days). This cost more money, which in turn causes the consumer more money.

2) Heat output, you can't argue that AMD's processors put out way more heat than any Intel does. That means they have to incorporate more cooling into their boxes, this cost more money again, and is louder!.. The average P3 system back in the day when OEMs were tring Athlon's was a heatsink on the P3 and then duct work from the heatsink to the PSU Fan! Even Dell with their latest P4 systems do something similar except they have some more smaller fans that are quieter but push less air. You can't do that with an Athlon. Furthermore this leads to instability in the processor when it gets to hot, and to fix this problem, more fans better heatsinks would be needed, more money there, AND.... most of all, Consumers HATE loud PCs, they like it when they even have to secong guess if the machine is on.

[ColdFlare]

Originally posted by phreenet


Well not really because the consumers also know which make and model works best.

And one of the reasons Intel holds the PC Market is because when OEMs tried and still try to incorporate Athlon based solutions into their product line up they found this:

1) Powering an Athlon solution means needing something more powerful then a standard 250W PSU (back in the PIII days). This cost more money, which in turn causes the consumer more money.

2) Heat output, you can't argue that AMD's processors put out way more heat than any Intel does. That means they have to incorporate more cooling into their boxes, this cost more money again, and is louder!.. The average P3 system back in the day when OEMs were tring Athlon's was a heatsink on the P3 and then duct work from the heatsink to the PSU Fan! Even Dell with their latest P4 systems do something similar except they have some more smaller fans that are quieter but push less air. You can't do that with an Athlon. Furthermore this leads to instability in the processor when it gets to hot, and to fix this problem, more fans better heatsinks would be needed, more money there, AND.... most of all, Consumers HATE loud PCs, they like it when they even have to secong guess if the machine is on.

Not to mention that in 90f summers, my athlon system makes my room 100f, forcing me to turn on the AC to reclaim my stability.

Back then when the p2, and p3s use to whip on AMD processors bad, intel already had the mhz standard up there. When the athlon was released, its when they added more IPC to the clocks, this in turn messed up how we figure out speed of a processor. And to add to that is the PR rating which is even more confusing, have you ever tried to explain this to someone? its nearly impossible, if you accomplish that, then try to explain it to someone whose english isnt that good... Since i live in new york city, its super diverse here, and its almost impossible.

Intel is a will always be the trusted brand, for home users. P4 quiet fan, RIMMs rock, intel chipset stable, .13 micron, also the ability to deliver.

What amd should of sat down, on the tv, watch intel commercials. There you will see what AMD is lacking, they should take a lesson on advertising and marketing instead of resorting to confusing bigger numbers.

You guys shouldn't bag on intel commercials, come on aliens use P4 processors, that has got to be advanced!! plus it sells, and AMD processors arent selling like intels are.

Also another thing to point out is AMD's earning is in the red now, even after they gain market share!! The price of an athlon is way toooo cheap!!

[bigben2k]

This has to be the most sensitive subject for a thread...

What it really comes down to, is the chips, and what they're used for.

It's no surprise to anyone, becauseof AMD's chip design, that the Athlon XP will outrun a Pentium 4 in FPU (Floating point). As a result, the Athlon XP, even though a little more restricted in memory bandwith than the Pentium 4, is a smarter choice for most office applications.

Office applications constitutes most of PCs purchased, and as it happens, the Athlon XP is cheaper than the Pentium. A double hit. (Look out the window boys and girls, there's a world out there!)

On the other hand, if you're looking for memory intensive applications or anything that isn't too mathematical (like games), then the Pentium 4 might be a better choice.

Now let's go in some more details...

If you ask me, I'd tell you that AMD would be better off marketing their chips at the real speed, and let their product be compared with Intel's as they are. Honesty is always the best policy. AMD went ahead with this (now querky) naming scheme for a couple of reasons:
1- There are still some people out there who will compare Intel and AMD CPUs on a MegaHertz level. (they should know better, but apparently they don't)
2-Intel's marketing machine is far, far bigger than AMD's, simply because Intel is a very large corporation with a multitude of products and hence, a very large budget. Intel is actually marketing itself through its processor, which is a very smart move. (Do you remember the last time that you bought an Intel product that was NOT a CPU?). So you see, instead of having a small marketing budget for each Intel product line, Intel has re-appropriated the budget from most divisions into their CPU products.

AMD is nowhere near being able to compete with Intel's marketing budget. Not now, not in fifty years from now. So AMD is doing the best it can: marketing themselves directly to the PC manufacturers, (wherever Intel hasn't interfered) who in turn will turn in a giant profit because their PCs are now cheaper than their "Intel inside" counterpart, and for general purposes, WILL ACTUALLY OUTPERFORM IT.

The key struggle that AMD has just surpassed is this:

Intel owns its own chip manufacturing plants, and has many of them. AMD never owned anything, except the design of their CPUs.

This year (2002), AMD entered into an agreement with a Japanese chip manufacturer, where AMD would share some of their technology, and the manufacturing plant would produce Athlon's at .13um, AND EVEN SMALLER in very large quantities.

As a result of this AMD has secured a supply source for it's design. In the past, AMD has always had difficulties meeting the demand with its product. PC manufacturers don't look at delays very kindly, in fact, they don't like them at all.

No more.

New facility, greater production, the latest technology.

AMD's Athlon will also use a smaller die (the actual chip itself, you know, that square thingy that sticks out from the top). This reduces the production cost of a chip.

Intel is actually going to use a larger die, increasing their production cost. But then, when you own your own facilities, why should you care?

Keep an eye on AMD for the next 2 years. You'll be shocked.

[ColdFlare]

I am actually pretty shocked about the whole .13 micron thing amd is doing. If you ask me, its pretty stupid, if not retarded to put an athlon core on a 80mm square area. We ALL know that athlons run super HOT compare to the pentiums, and having it dissapaite heat in only 80mm square is just plain dumb.

Another thing is that the T-Bred is suppose to be released in march, its now may, now they are saying June. They also changed the 1.5v specified standard for .13 micron, to 1.6v to run default speeds. Now whose changing standards.

Yes intel has the market structure and business practices that will give them a huge budget to work with. Well they deserve to have such a huge budget, they worked their asses off to get where they are now. They are the pioneers of the cpu industry and still are.

Most businesses pick intel cpus instead of amd is because of 1 major thing. its called stability!.

[bigben2k]

Stability of supply, yes.

A CPU manufacturing plant is MUCH more expensive than it was a few decades ago. Intel has been around for a very, very long time, and got in quickly on establishing themselves. AMD was actually licensed by Intel to produce Intel-like CPUs, like 20 other manufacturers.

AMD is the only one that is still around.

And you'd never believe who was responsible for that licensing agreement:

IBM.

That's right. IBM.

As per company policy, IBM will not do business with a supplier that is the only source of a product. So when they wanted to build a PC based on Intel's 8086, Intel simply licensed the technology out. Voila, multiple suppliers, IBM is happy, let's get down to business.

And the x86 IBM PC was born. The rest is history.


You might be shocked that AMD is going to go to .13, but you have to think about it this way: since .13 is relatively new (for mass production), there's only so many places that will manufacture anything in .13 . This means that AMD was facing a shortage of supply, again...

So they fixed it. They found a partner that is ready to manufacture in .13 right now, in a necessary quantity for AMD to stay competitive. As part of the agreement, this plant will use some of AMD's technology to improve the chips that they make for other companies, and in return, keep their plant up-to-date.

AMD could barely afford to build a new plant, and let's face it, it would be a very bad business decision: AMD would have all of its resources and capital invested in the production of only one product. This NEVER sits well with investors.

Welcome to Corporate America.