AMD: Smoke and Mirrors?

[TechnologyCoordinator]

Very interesting if true. First the smoke and mirrors with product releases, and now perhaps financials? Intel may be ruthless to competitors, but Intel does show extreme loyalty to its stockholders; an area where AMD has left much to be desired which is evidenced by AMD's sub $5 stock price.

http://www.overclockers.com/tips01360/

More Smoke and Mirrors . . .
Ed Stroligo - 7/12/08

As we mentioned yesterday, AMD is writing off another $880 million off its investment in ATI.

What should you make of this?

In all likelihood, AMD has been shopping around the consumer products part of what used to be ATI and found that it isn't going to get much money for it, nowhere near what they paid for it.

So what they did was the accounting equivalent of slashing the price for the division before actually selling it.

Why would they do that? Well, if they hadn't done this, if they sold the division next week or month, they would have to show a huge loss on the sale of the division. So instead, they're taking the loss now, so when the sale happens, it will look like AMD broke even or maybe even "made" money.

Of course, if you have a memory that lasts more than a week and math ability greater than a gnat's, you'll realize that six is one-half dozen of the other, and it is, but many have neither and will get fooled by the smoke and mirrors.

Another game AMD appears ready to play involves the sale of the fabbing equipment from Dresden.

Quick accounting lesson: If you're a business and you buy equipment for the purpose of making money, you are generally allowed to deduct a portion of what you paid for the equipment against your profit/tax each year as you use. This is usually called depreciation, the idea being that you gradually lose the value of the asset as you use it.

Depreciation is only a very rough measurement of this loss, and very often, a business can depreciate the entire cost of an item, but can still sell it and get some money for it. When that happens, the business has a gain on the sale of the property, and must add that gain to their profit/tax.

So, for instance, if AMD bought $1.5 billion of fab equipment for Fab30, depreciated $1.2 billion of that over the course of years, then sold it for $500 million, they would have a gain of $200 million, which is roughly what AMD actually did.

Such an event is perfectly normal and natural. It only becomes funny business when it comes to characterizing the nature of the gain/profit.

Say you own an auto repair shop, and you own a lot of expensive equipment. Making money by using the tools is one thing, making money by selling the tools is quite another. For one thing, once you sell the tools, you aren't going to be making any more money by using them in their business. You had a one-time gain, as opposed to a regular day-in, day-out income.

It is a general, basic accounting principle that you are supposed to separate any one-time gains and losses from your regular, ongoing income/losses. This is to let people easily see how the regular ongoing business is doing.

To put it mildly, AMD has been playing fast and loose with these rules in their public presentations the last couple years, calling ongoing expenses one-time and vice versa, and it looks like they're going to do it big-time again next week.

This report indicates that AMD plans to add the gain from the sale of its Fab30 equipment to its gross margin. That's accountantese for saying that they're going to treat the sale of fab tools like it were the sale of CPUs, treating a one-time gain like it were regular ongoing income.

Why would they want to do that? Well, a few quarters ago, the executives said that they would reach "operational breakeven" point fairly shortly. First, it was around the middle of the year, then it was 3Q.

To make a long story short, there's no way AMD can hit "operational breakeven" for Q2 legitimately, but they probably "would" if they included the fab equipment gain as regular income. I would bet dollars to donuts that this is exactly what AMD will do when it releases its earnings next Thursday, and they'll try to spin this into a great triumph (and maybe hand themselves some extra bonuses and stock options afterwards).

Now I'm sure that at least some, maybe most of those reading this have had their eyes glaze over. Unfortunately, at least some, maybe most of those who will end up writing about this, even the "financial" reporters, will have their eyes glaze over, too.

Unfortunately, that's what some people in certain high positions are counting on, and even more unfortunately, they'll be assisted by some who can see, but are more than happy to help to pull the wool over their own eyes and those of others.

The Big Bias

The very sad fact is that with not many exceptions, there's a bias in favor of AMD in the computer hardware press. Some don't even pretend otherwise. This is mostly because there a bias in favor of AMD in the computer hardware press readership.

Why? The argument is as follows: It is in our best interest to have AMD around to compete against Intel, so we will slant the news in favor of AMD, or repeat whatever they tell us without question or thought, or ignore/minimize the bad news or deceptions.

I'm sure many of you reading this are saying something like, "What's wrong with that?"

Well, what was wrong with Enron doing what it did? What was wrong with Worldcom? What was wrong with all the financial institutions that have gone belly-up or are halfway there lately? They all played the same kind of smoke-and-mirror games we see from Green, and eventually, they all got burned.

What's wrong with that is that when you live by smoke and mirrors, eventually it catches up with you, no matter how complacent those watching are. And people get hurt, hurt badly as a result. Ask the former Enron and WorldCom and Bear Stearns employees and stockholders about their pensions and 401K plans and life savings.

You see, when people find out they can get buy and even prosper with smoke and mirrors rather than real success, they get hooked on them. It's just so much easier and rewarding than doing the hard job, especially if the hard job means firing yourself.

But if it stays all smoke and mirrors, eventually it catches up with you and suddenly blows up in your face, and then all the people who let you get away with it for so long jump up and down and say, "I never dreamed this would happen!"

Sometimes, it's cruel to be kind. You've heard of "enablers," people who allow others to do bad or destructive to themselves or others.

Isn't that what we're really talking about here?


Ed

 

[Just_An_Engineer]

Very interesting if true. First the smoke and mirrors with product releases, and now perhaps financials? Intel may be ruthless to competitors, but Intel does show extreme loyalty to its stockholders; an area where AMD has left much to be desired which is evidenced by AMD's sub $5 stock price.

http://www.overclockers.com/tips01360/

You're kidding yourself if you think most companies don't play these kinds of accounting games when they have to take a loss. Once upon a time I worked for Tyco International (a company that definitely knows how to lose money), and this kind of stuff happened all the time. Whenever a company has a acquisition or a contract on which it loses a tremendous amount of money it is usually dubbed a "legacy project" and steps are taken to spread the losses out over several quarters instead of taking a huge hit in the quarter the loss was actually incurred. This is very common for companies that deal with construction projects.

As for adding the revenue gained from the sale of equipment to the gross margin, this makes perfect sense. Every company does this. Depreciation is applied to all company assets so that it can be claimed on the tax return as a capital loss. If assets are sold for a higher price than their depreciated value on the company's asset list then the difference is taken as a profit. Pretty standard stuff.

The author of this article is trying to make it sound like a big conspiracy but really this is nothing unusual.

 

[royalcrown]

Since you asked, I think that is incredibly boring and couldn't care less.

 

[macer1]

Technology Coordinator this thread is very trollish


back to your bridge!!!!

 

[Proximon]

Someone is foaming at the mouth.

The "press bias" bit is especially good. Apparently, the OP hasn't read a magazine or visited a tech website since 2006.

 

[croc]

Very interesting if true. First the smoke and mirrors with product releases, and now perhaps financials? Intel may be ruthless to competitors, but Intel does show extreme loyalty to its stockholders; an area where AMD has left much to be desired which is evidenced by AMD's sub $5 stock price.

http://www.overclockers.com/tips01360/

More Smoke and Mirrors . . .
Ed Stroligo - 7/12/08

As we mentioned yesterday, AMD is writing off another $880 million off its investment in ATI.

What should you make of this?

In all likelihood, AMD has been shopping around the consumer products part of what used to be ATI and found that it isn't going to get much money for it, nowhere near what they paid for it.

So what they did was the accounting equivalent of slashing the price for the division before actually selling it.

Why would they do that? Well, if they hadn't done this, if they sold the division next week or month, they would have to show a huge loss on the sale of the division. So instead, they're taking the loss now, so when the sale happens, it will look like AMD broke even or maybe even "made" money.

Of course, if you have a memory that lasts more than a week and math ability greater than a gnat's, you'll realize that six is one-half dozen of the other, and it is, but many have neither and will get fooled by the smoke and mirrors.

Another game AMD appears ready to play involves the sale of the fabbing equipment from Dresden.

Quick accounting lesson: If you're a business and you buy equipment for the purpose of making money, you are generally allowed to deduct a portion of what you paid for the equipment against your profit/tax each year as you use. This is usually called depreciation, the idea being that you gradually lose the value of the asset as you use it.

Depreciation is only a very rough measurement of this loss, and very often, a business can depreciate the entire cost of an item, but can still sell it and get some money for it. When that happens, the business has a gain on the sale of the property, and must add that gain to their profit/tax.

So, for instance, if AMD bought $1.5 billion of fab equipment for Fab30, depreciated $1.2 billion of that over the course of years, then sold it for $500 million, they would have a gain of $200 million, which is roughly what AMD actually did.

Such an event is perfectly normal and natural. It only becomes funny business when it comes to characterizing the nature of the gain/profit.

Say you own an auto repair shop, and you own a lot of expensive equipment. Making money by using the tools is one thing, making money by selling the tools is quite another. For one thing, once you sell the tools, you aren't going to be making any more money by using them in their business. You had a one-time gain, as opposed to a regular day-in, day-out income.

It is a general, basic accounting principle that you are supposed to separate any one-time gains and losses from your regular, ongoing income/losses. This is to let people easily see how the regular ongoing business is doing.

To put it mildly, AMD has been playing fast and loose with these rules in their public presentations the last couple years, calling ongoing expenses one-time and vice versa, and it looks like they're going to do it big-time again next week.

This report indicates that AMD plans to add the gain from the sale of its Fab30 equipment to its gross margin. That's accountantese for saying that they're going to treat the sale of fab tools like it were the sale of CPUs, treating a one-time gain like it were regular ongoing income.

Why would they want to do that? Well, a few quarters ago, the executives said that they would reach "operational breakeven" point fairly shortly. First, it was around the middle of the year, then it was 3Q.

To make a long story short, there's no way AMD can hit "operational breakeven" for Q2 legitimately, but they probably "would" if they included the fab equipment gain as regular income. I would bet dollars to donuts that this is exactly what AMD will do when it releases its earnings next Thursday, and they'll try to spin this into a great triumph (and maybe hand themselves some extra bonuses and stock options afterwards).

Now I'm sure that at least some, maybe most of those reading this have had their eyes glaze over. Unfortunately, at least some, maybe most of those who will end up writing about this, even the "financial" reporters, will have their eyes glaze over, too.

Unfortunately, that's what some people in certain high positions are counting on, and even more unfortunately, they'll be assisted by some who can see, but are more than happy to help to pull the wool over their own eyes and those of others.

The Big Bias

The very sad fact is that with not many exceptions, there's a bias in favor of AMD in the computer hardware press. Some don't even pretend otherwise. This is mostly because there a bias in favor of AMD in the computer hardware press readership.

Why? The argument is as follows: It is in our best interest to have AMD around to compete against Intel, so we will slant the news in favor of AMD, or repeat whatever they tell us without question or thought, or ignore/minimize the bad news or deceptions.

I'm sure many of you reading this are saying something like, "What's wrong with that?"

Well, what was wrong with Enron doing what it did? What was wrong with Worldcom? What was wrong with all the financial institutions that have gone belly-up or are halfway there lately? They all played the same kind of smoke-and-mirror games we see from Green, and eventually, they all got burned.

What's wrong with that is that when you live by smoke and mirrors, eventually it catches up with you, no matter how complacent those watching are. And people get hurt, hurt badly as a result. Ask the former Enron and WorldCom and Bear Stearns employees and stockholders about their pensions and 401K plans and life savings.

You see, when people find out they can get buy and even prosper with smoke and mirrors rather than real success, they get hooked on them. It's just so much easier and rewarding than doing the hard job, especially if the hard job means firing yourself.

But if it stays all smoke and mirrors, eventually it catches up with you and suddenly blows up in your face, and then all the people who let you get away with it for so long jump up and down and say, "I never dreamed this would happen!"

Sometimes, it's cruel to be kind. You've heard of "enablers," people who allow others to do bad or destructive to themselves or others.

Isn't that what we're really talking about here?


Ed

While normally I treat Ed's articles with the same large grain of salt that I reserve for Fuad's, Theo's or Charley's articles, I do think that that in this case he makes some valid points. Whether anyone listens may be the issue, every bit as much of an issue as was Worldcom or Enron, or Bear Stearns. One can only do 'creative accounting' for so long, then there is nothing left to account for. What does AMD use next quarter? Another write-down on ATI? Seems impractical as ATI may very well be making money. Sell another FAB? How many do they really have left?

When does their next financial report come out? End of this week? Patiently waiting...

I only hope that they don't take ATI down with them....

 

[iocedmyself]

Is this for real? AMD...smoke and mirrors? Playing fast and loose surviving off marketing hype and feeding false specs to review writers?

1) AMD has almost no marketing at all
2) they are very very very close mouthed about any technical details or performance comparisons even comparing current AMD hardware against upcoming, let alone comparisons against competition
3)No takes AMD's performance estimates at face value. Not the tech journalists, the hardware manufactures, or any consumer semi-knowledgable in the hardware world.

There have been countless times that AMD has released performance stats, power consumption and overall price/performance ratio only to be ignored, taken out of context or straight out sabotaged.

Intel didn't make a chip that possessed any kind of worthwhile design advancements for a 6-7 year period (1999/2000 being the PIII series launch releasing the P4 in the end of 2000 up until the release of the core 2 series in the end of 2006)

How has Intel always managed to keep such a huge market share regardless of quality, performance, or even progress? Marketing smoke and mirrors and strong-arming vendors.

AMD began stomping intel going alll the way back to the PIII/thunderbird days in 99/2000. Thus began the megahertz war. Throughout which AMD was consistent in being the better chip. While intel went from 1.3ghz - 3.6 ghz. AMD needed only to go from 1.2ghz -2.2ghz in the same time frame, staying on the same socket design, with the only major changes being die shrinks and cache increases with higher clocks and lower power consumption.

Intel's only brief lead was in releasing the 800mhz FSB flavor of the P4, which enjoyed minor performance gains in the 4 or 5 months between it's release and the release of the Athlon 64 line. But the overclocked bartons still stomped the 800mhz fsb p4's.

But what was Intels delightful claim about Netburst....oh yes "netburst will scale to 10ghz" It topped out at...3.6? hmmm seems to be a small margin of error there.

It took a 5.2ghz clocked P4 to tottally out perform a stock AMD 2.6ghz FX-55.

Now lets see, the core 2 chips, are based off the p6 series (pentium - pentium 3) So intel pretty much admits that the 6 year span of netburs was just them putting out crap that would have been better off never existing. But intel prefers to phrase it as "having the performance king laying around for 7 years without knowing it"

Hmm what other smoke and mirror magic has intel worked.

after the conroe was released intel said

There is no reason for an on-die memory controller. They attempted it and came to the conclusion that it created to many problems with little benifit for the increased production cost and in turn an increased cost to the consumer

In reality, they attempted an IMC for what was supposed to follow the PIII series, years of research and development, hundreds of millions spent and it was scrapped after numerous failed production runs. Intel thought it was a good idea 10 years ago. They just couldn't make it work. Still can't make it work, 5 years after AMD did it. So in reality, it's not a good idea until intel can make it too.

The same can be said for native dual-core and quad-core chips. Intel couldn't get 2 cores on a single die so they did two seperate cores in a single package, then upon getting 2 cores on one die, couldn't get 4 cores to work thus the two dual-cores in a single package for their "quad-core" chips. Note that they have been working on the IMC, and native quad core's while telling the world, there's no good reason to do it.

But the best of all. Their biggest con, 64bit? Who needs it?

They had to license the 64bit extenstions from AMD to get in the 64bit game. Once they got there, no one wanted them. Intel is horrible when it comes to 64bit computing.

So three years after the 64bit desktops come out from AMD, 11 years after the first 64bit processor was made by the Alpha team who also worked on the AMD barton's and 939 opterons.

Intel says who needs it?

It had nothing to do with their server chips not being reconized as 64bit compatible despite the claim of it being so. It had nothing to do with Opterons stomping the Xeons. Nothing to do with AMD performace gaining an addition 15-20% in 64bit applications.

They just thought it was silly. Microsoft was silly for making 64bit OS's, all those software developers were just wasting everyones time adding 64bit support to little things like 3D studio max, Cinema 4D, maya and softimage., file compression, video/audio encoding and of course The game developers. All just silly.

Because Core 2 does best in 32bit Xp, just like the PIII chips it's based off of. They will shoddily support 64bit software. So the motherboard makers can sell boards with 4, 8 or 16gig supported RAM capacity. Even though only 2gigs is addressable in 32bit.

Then of course Intel never really seems to get benchmarked in 64bit, despite supporting, thus niether does AMD because how would that be fair? Intel chips are reviewed against completly unbalanced AMD chips. But they'll boast about the $1500 intel quad having a 15-30% performance gain over an $180 AMD dual core, or the $1800 intel quad's 20-30% gain over the $200 AMD quad in real world benches.

But they never seemed to focus to much on AMD's $60 chips having 25-30% better memory performance over the $1500+ intel chips.

The fact that intels synthetic mathmatic benches are so high only because of the huge cache sizes which have no actual performance benifit in real world computing as rendering/encoding/photo editing requires a minimum of 80 or 90 megs cache.

AMD bought ATI 2 years ago, they doubled their company workload, and by golly they have a pair of video cards out that stomp the competitions and they do it at 1/4 - 1/2 the price.

Of course there is going to be an adjustment period. But hey, i guess intel and nvidia are just knocking 25% -60% off their prices after AMD/ATI launches because it's fashionable.

 

[nowwhatnapster]

nice rant iocedmyself.

much more interesting than the original article you posted in reply to

i concur, amd is still kickin ^^

 

[aznguy0028]

+1 iocedmyself

 

[takealready]

PREACH IT iocedmyself

I agree with you. If AMD spent HALF as much money as INTEL on marketing. Then AMD wouldn't have to relie on "Word of Mouth" to sell their chips (they forget word of mouth spreads all the good and the bad about something).

People who know NOTHING about computers know the name INTEL. It's a household name.

And iocedmyself, I could almost here the church organ playing "that song" in the background, AMD fanboys getting worked up and someone who's not Familiar with the Gospel of AMD looking confused.

 

[dragonsprayer]

wow i had a vision - i saw sugar plums and intel buying ati!
i dreamed of spider renamed scorpion and the all intel gaming platform with nehalem and triple 4870's running 50k 3dmark

it flashed before my eyes when i saw the tech c post - what if? what if amd sold ati who would buy it? intel!

i really hate nvidia! i never forgave them for my x975 systems with out sli

 

[dragonsprayer]

Is this for real? AMD...smoke and mirrors? Playing fast and loose surviving off marketing hype and feeding false specs to review writers?

1) AMD has almost no marketing at all
2) they are very very very close mouthed about any technical details or performance comparisons even comparing current AMD hardware against upcoming, let alone comparisons against competition
3)No takes AMD's performance estimates at face value. Not the tech journalists, the hardware manufactures, or any consumer semi-knowledgable in the hardware world.

There have been countless times that AMD has released performance stats, power consumption and overall price/performance ratio only to be ignored, taken out of context or straight out sabotaged.

Intel didn't make a chip that possessed any kind of worthwhile design advancements for a 6-7 year period (1999/2000 being the PIII series launch releasing the P4 in the end of 2000 up until the release of the core 2 series in the end of 2006)

How has Intel always managed to keep such a huge market share regardless of quality, performance, or even progress? Marketing smoke and mirrors and strong-arming vendors.

AMD began stomping intel going alll the way back to the PIII/thunderbird days in 99/2000. Thus began the megahertz war. Throughout which AMD was consistent in being the better chip. While intel went from 1.3ghz - 3.6 ghz. AMD needed only to go from 1.2ghz -2.2ghz in the same time frame, staying on the same socket design, with the only major changes being die shrinks and cache increases with higher clocks and lower power consumption.

Intel's only brief lead was in releasing the 800mhz FSB flavor of the P4, which enjoyed minor performance gains in the 4 or 5 months between it's release and the release of the Athlon 64 line. But the overclocked bartons still stomped the 800mhz fsb p4's.

But what was Intels delightful claim about Netburst....oh yes "netburst will scale to 10ghz" It topped out at...3.6? hmmm seems to be a small margin of error there.

It took a 5.2ghz clocked P4 to tottally out perform a stock AMD 2.6ghz FX-55.

Now lets see, the core 2 chips, are based off the p6 series (pentium - pentium 3) So intel pretty much admits that the 6 year span of netburs was just them putting out crap that would have been better off never existing. But intel prefers to phrase it as "having the performance king laying around for 7 years without knowing it"

Hmm what other smoke and mirror magic has intel worked.

after the conroe was released intel said

There is no reason for an on-die memory controller. They attempted it and came to the conclusion that it created to many problems with little benifit for the increased production cost and in turn an increased cost to the consumer

In reality, they attempted an IMC for what was supposed to follow the PIII series, years of research and development, hundreds of millions spent and it was scrapped after numerous failed production runs. Intel thought it was a good idea 10 years ago. They just couldn't make it work. Still can't make it work, 5 years after AMD did it. So in reality, it's not a good idea until intel can make it too.

The same can be said for native dual-core and quad-core chips. Intel couldn't get 2 cores on a single die so they did two seperate cores in a single package, then upon getting 2 cores on one die, couldn't get 4 cores to work thus the two dual-cores in a single package for their "quad-core" chips. Note that they have been working on the IMC, and native quad core's while telling the world, there's no good reason to do it.

But the best of all. Their biggest con, 64bit? Who needs it?

They had to license the 64bit extenstions from AMD to get in the 64bit game. Once they got there, no one wanted them. Intel is horrible when it comes to 64bit computing.

So three years after the 64bit desktops come out from AMD, 11 years after the first 64bit processor was made by the Alpha team who also worked on the AMD barton's and 939 opterons.

Intel says who needs it?

It had nothing to do with their server chips not being reconized as 64bit compatible despite the claim of it being so. It had nothing to do with Opterons stomping the Xeons. Nothing to do with AMD performace gaining an addition 15-20% in 64bit applications.

They just thought it was silly. Microsoft was silly for making 64bit OS's, all those software developers were just wasting everyones time adding 64bit support to little things like 3D studio max, Cinema 4D, maya and softimage., file compression, video/audio encoding and of course The game developers. All just silly.

Because Core 2 does best in 32bit Xp, just like the PIII chips it's based off of. They will shoddily support 64bit software. So the motherboard makers can sell boards with 4, 8 or 16gig supported RAM capacity. Even though only 2gigs is addressable in 32bit.

Then of course Intel never really seems to get benchmarked in 64bit, despite supporting, thus niether does AMD because how would that be fair? Intel chips are reviewed against completly unbalanced AMD chips. But they'll boast about the $1500 intel quad having a 15-30% performance gain over an $180 AMD dual core, or the $1800 intel quad's 20-30% gain over the $200 AMD quad in real world benches.

But they never seemed to focus to much on AMD's $60 chips having 25-30% better memory performance over the $1500+ intel chips.

The fact that intels synthetic mathmatic benches are so high only because of the huge cache sizes which have no actual performance benifit in real world computing as rendering/encoding/photo editing requires a minimum of 80 or 90 megs cache.

AMD bought ATI 2 years ago, they doubled their company workload, and by golly they have a pair of video cards out that stomp the competitions and they do it at 1/4 - 1/2 the price.

Of course there is going to be an adjustment period. But hey, i guess intel and nvidia are just knocking 25% -60% off their prices after AMD/ATI launches because it's fashionable.

TIME TO FAN THE FAN BOYS FIRES!

WHAT KIND OF AMD SYSTEMS DO YOU HAVE? your opertrons are worthless!

AMD got lots a of false media postives on single tasking cpus - pentiums blow pre-amd dual cores. the FX-60 was the start of the amd lead in may 05, it ended in july of 06 with core2

a pentium 965 running 4.5ghz with hyperthreading is bad boy mutlitasker that still gets $250 on ebay, i have seen dual cpu opertron systems selling for the same price!

lol! amd fan boy's - all the amd fan boys left to some low life extrme cpu site

IFB #1 amd is always #2 (accept for may 06-july 06)

 

[takealready]

And FYI, my last INTEL chip was the Pentium III @ 700mhz.

Ever since then I've been with AMD, made sure my wife has AMD, my dad, my sisters, my cousins, my mom and some of my easily influenced friends. So I'm a Uber-Fan boy.

 

[takealready]

wow i had a vision - i saw sugar plums and intel buying ati!
i dreamed of spider renamed scorpion and the all intel gaming platform with nehalem and triple 4870's running 50k 3dmark

it flashed before my eyes when i saw the tech c post - what if? what if amd sold ati who would buy it? intel!

i really hate nvidia! i never forgave them for my x975 systems with out sli

And that's why AMD has to pull it together. They have to put up some sort of fight. I can admit just saying "skulltrail" makes me lightheaded to think of that much INSANE power. You'd need a WMD just to power that thing :lol:

If AMD dies tomorrow, INTEL will start charging $400 just for a 2.0 Ghz chip. Without AMD's presence INTEL will start defecating on consumer's bank accounts.

If ATI became available, INTEL will buy it two seconds after AMD let's the company go. I know Intel has spies working for AMD at corperate (like hector Ruiz ).

 

[JDocs]

Will somebody please tell me where all those AMD Native quad design adds came from again, they ain't cheap and they weren't paid for by Intel.

 

[apache_lives]

Is this for real? AMD...smoke and mirrors? Playing fast and loose surviving off marketing hype and feeding false specs to review writers?

1) AMD has almost no marketing at all
2) they are very very very close mouthed about any technical details or performance comparisons even comparing current AMD hardware against upcoming, let alone comparisons against competition
3)No takes AMD's performance estimates at face value. Not the tech journalists, the hardware manufactures, or any consumer semi-knowledgable in the hardware world.

There have been countless times that AMD has released performance stats, power consumption and overall price/performance ratio only to be ignored, taken out of context or straight out sabotaged.

Intel didn't make a chip that possessed any kind of worthwhile design advancements for a 6-7 year period (1999/2000 being the PIII series launch releasing the P4 in the end of 2000 up until the release of the core 2 series in the end of 2006)

How has Intel always managed to keep such a huge market share regardless of quality, performance, or even progress? Marketing smoke and mirrors and strong-arming vendors.

AMD began stomping intel going alll the way back to the PIII/thunderbird days in 99/2000. Thus began the megahertz war. Throughout which AMD was consistent in being the better chip. While intel went from 1.3ghz - 3.6 ghz. AMD needed only to go from 1.2ghz -2.2ghz in the same time frame, staying on the same socket design, with the only major changes being die shrinks and cache increases with higher clocks and lower power consumption.

Intel's only brief lead was in releasing the 800mhz FSB flavor of the P4, which enjoyed minor performance gains in the 4 or 5 months between it's release and the release of the Athlon 64 line. But the overclocked bartons still stomped the 800mhz fsb p4's.

But what was Intels delightful claim about Netburst....oh yes "netburst will scale to 10ghz" It topped out at...3.6? hmmm seems to be a small margin of error there.

It took a 5.2ghz clocked P4 to tottally out perform a stock AMD 2.6ghz FX-55.

Now lets see, the core 2 chips, are based off the p6 series (pentium - pentium 3) So intel pretty much admits that the 6 year span of netburs was just them putting out crap that would have been better off never existing. But intel prefers to phrase it as "having the performance king laying around for 7 years without knowing it"

Hmm what other smoke and mirror magic has intel worked.

after the conroe was released intel said

There is no reason for an on-die memory controller. They attempted it and came to the conclusion that it created to many problems with little benifit for the increased production cost and in turn an increased cost to the consumer

In reality, they attempted an IMC for what was supposed to follow the PIII series, years of research and development, hundreds of millions spent and it was scrapped after numerous failed production runs. Intel thought it was a good idea 10 years ago. They just couldn't make it work. Still can't make it work, 5 years after AMD did it. So in reality, it's not a good idea until intel can make it too.

The same can be said for native dual-core and quad-core chips. Intel couldn't get 2 cores on a single die so they did two seperate cores in a single package, then upon getting 2 cores on one die, couldn't get 4 cores to work thus the two dual-cores in a single package for their "quad-core" chips. Note that they have been working on the IMC, and native quad core's while telling the world, there's no good reason to do it.

But the best of all. Their biggest con, 64bit? Who needs it?

They had to license the 64bit extenstions from AMD to get in the 64bit game. Once they got there, no one wanted them. Intel is horrible when it comes to 64bit computing.

So three years after the 64bit desktops come out from AMD, 11 years after the first 64bit processor was made by the Alpha team who also worked on the AMD barton's and 939 opterons.

Intel says who needs it?

It had nothing to do with their server chips not being reconized as 64bit compatible despite the claim of it being so. It had nothing to do with Opterons stomping the Xeons. Nothing to do with AMD performace gaining an addition 15-20% in 64bit applications.

They just thought it was silly. Microsoft was silly for making 64bit OS's, all those software developers were just wasting everyones time adding 64bit support to little things like 3D studio max, Cinema 4D, maya and softimage., file compression, video/audio encoding and of course The game developers. All just silly.

Because Core 2 does best in 32bit Xp, just like the PIII chips it's based off of. They will shoddily support 64bit software. So the motherboard makers can sell boards with 4, 8 or 16gig supported RAM capacity. Even though only 2gigs is addressable in 32bit.

Then of course Intel never really seems to get benchmarked in 64bit, despite supporting, thus niether does AMD because how would that be fair? Intel chips are reviewed against completly unbalanced AMD chips. But they'll boast about the $1500 intel quad having a 15-30% performance gain over an $180 AMD dual core, or the $1800 intel quad's 20-30% gain over the $200 AMD quad in real world benches.

But they never seemed to focus to much on AMD's $60 chips having 25-30% better memory performance over the $1500+ intel chips.

The fact that intels synthetic mathmatic benches are so high only because of the huge cache sizes which have no actual performance benifit in real world computing as rendering/encoding/photo editing requires a minimum of 80 or 90 megs cache.

AMD bought ATI 2 years ago, they doubled their company workload, and by golly they have a pair of video cards out that stomp the competitions and they do it at 1/4 - 1/2 the price.

Of course there is going to be an adjustment period. But hey, i guess intel and nvidia are just knocking 25% -60% off their prices after AMD/ATI launches because it's fashionable.

Heh remind me who's on top atm?

Remind me who's platform is more reliable and respectable for corporate/business?

Remind me which company sells more cpus in the end?

AMD aint this glorious company, there as bad as Intel, and dont get my wrong, both companies can kiss my a$$, ill buy whats best at my price range.

 

[croc]

Is this for real? AMD...smoke and mirrors? Playing fast and loose surviving off marketing hype and feeding false specs to review writers?

Yes, to some extent it seems to be true. I’d refer to all of the Phenom hype prior to the launch, and the debacle that came after.

1) AMD has almost no marketing at all

Less than Intel, but certainly ‘almost no’ marketing is a bit of hyperbole.

2) they are very very very close mouthed about any technical details or performance comparisons even comparing current AMD hardware against upcoming, let alone comparisons against competition

What? They may be a bit close-to-the-chest when it comes to detail, but they certainly shoot off their collective mouths when it comes to hyping new technologies. Intel, OTOH, tends to give out too much detail.

3)No takes AMD's performance estimates at face value. Not the tech journalists, the hardware manufactures, or any consumer semi-knowledgable in the hardware world.

AMD is a member of the SPEC org., and as such ‘should’ submit samples for testing. Check out the SPEC charts.

There have been countless times that AMD has released performance stats, power consumption and overall price/performance ratio only to be ignored, taken out of context or straight out sabotaged.

Refer to above.

How has Intel always managed to keep such a huge market share regardless of quality, performance, or even progress? Marketing smoke and mirrors and strong-arming vendors.

OEM’s seem to like stability of supply… Intel had fabs, AMD did not. Intel was concerned with over-supply, AMD was concerned with fulfilling contracts, and the resulting penalties.

after the conroe was released intel said

There is no reason for an on-die memory controller. They attempted it and came to the conclusion that it created to many problems with little benifit for the increased production cost and in turn an increased cost to the consumer

Speculating here, but many of the Alpha team that they inherited from Compaq (along with the IP and the DEC engineers that they hired prior to the fire-sale) had reached that conclusion from their efforts on the EV8 (?).

In reality, they attempted an IMC for what was supposed to follow the PIII series, years of research and development, hundreds of millions spent and it was scrapped after numerous failed production runs. Intel thought it was a good idea 10 years ago. They just couldn't make it work. Still can't make it work, 5 years after AMD did it. So in reality, it's not a good idea until intel can make it too.

The IMC issue needed to be re-visited once Intel started thinking about 8 core and 16 core chips. 4 cores just didn’t gain (or in AMD’s case, lost) performance

The same can be said for native dual-core and quad-core chips. Intel couldn't get 2 cores on a single die so they did two seperate cores in a single package, then upon getting 2 cores on one die, couldn't get 4 cores to work thus the two dual-cores in a single package for their "quad-core" chips. Note that they have been working on the IMC, and native quad core's while telling the world, there's no good reason to do it.

What a beautiful engineering solution! Cheap, and less lost product. A disabled core is still worth something, even AMD finally got on that bandwagon. Phenom 3 cores, anyone?

But the best of all. Their biggest con, 64bit? Who needs it?

They had to license the 64bit extenstions from AMD to get in the 64bit game. Once they got there, no one wanted them. Intel is horrible when it comes to 64bit computing.

So three years after the 64bit desktops come out from AMD, 11 years after the first 64bit processor was made by the Alpha team who also worked on the AMD barton's and 939 opterons.

Intel says who needs it?

Any of the original Alpha team that I knew would either be laughing, or looking for a new rope. Their main goal was to replace the 32 bit VAX family of processors with a 64 bit RISC processor that could emulate the VAX instructions. They also managed the same feat, same processor family, running NT. EV7, I believe.

It had nothing to do with their server chips not being reconized as 64bit compatible despite the claim of it being so. It had nothing to do with Opterons stomping the Xeons. Nothing to do with AMD performace gaining an addition 15-20% in 64bit applications.

Neither AMD not Intel make a 64 bit chip, unless you count the IA-64 based on the Alpha processor. A 64 bit extension for memory, and some extended instructions does not a 64 bit chip make. Now you’ve got the SPARC, PowerPC etc. engineers looking for new rope.

They just thought it was silly. Microsoft was silly for making 64bit OS's, all those software developers were just wasting everyones time adding 64bit support to little things like 3D studio max, Cinema 4D, maya and softimage., file compression, video/audio encoding and of course The game developers. All just silly.

Because Core 2 does best in 32bit Xp, just like the PIII chips it's based off of. They will shoddily support 64bit software. So the motherboard makers can sell boards with 4, 8 or 16gig supported RAM capacity. Even though only 2gigs is addressable in 32bit.

Then of course Intel never really seems to get benchmarked in 64bit, despite supporting, thus niether does AMD because how would that be fair? Intel chips are reviewed against completly unbalanced AMD chips. But they'll boast about the $1500 intel quad having a 15-30% performance gain over an $180 AMD dual core, or the $1800 intel quad's 20-30% gain over the $200 AMD quad in real world benches.

Please show me an AMD chip that can run a 64 bit OS. Why do you think Apple went to Intel x86-64 chips? Lack of supply from IBM for the PowerPC? Cost? How about lack of coders that were willing to write 64 bit only programs….???

But they never seemed to focus to much on AMD's $60 chips having 25-30% better memory performance over the $1500+ intel chips.

The fact that intels synthetic mathmatic benches are so high only because of the huge cache sizes which have no actual performance benifit in real world computing as rendering/encoding/photo editing requires a minimum of 80 or 90 megs cache.

OK, once again I suggest a look at SPEC.org.

AMD bought ATI 2 years ago, they doubled their company workload, and by golly they have a pair of video cards out that stomp the competitions and they do it at 1/4 - 1/2 the price.

Of course there is going to be an adjustment period. But hey, i guess intel and nvidia are just knocking 25% -60% off their prices after AMD/ATI launches because it's fashionable.

Kudos to ATI. About the only thing in your whole rant that you have somewhat correct. But Intel does not dabble these days in the discreet graphics market, and their IGP chipsets don’t seem to be taking much of a hit.

 

[Ycon]

This should be no surprise to anyone since AMD has ALWAYS been all about smoke and mirrors ever since it was founded.
Best example of this is the infamous Athlon64 that seemed to do a nice job at benchmarking on sterile test systems but drowned against the HT-enabled P4 & PEE processors when used in a regular users everyday PC, because they were less able to handle all the parallel workloads (most importantly internet security software).

 

[Reynod]

TC fanning the flames after starting a thread encouraging people to rate people down who deliberately engage in this sort of behavior.
You can't rate the OP down and make the thread disappear ... PITY.
That would be interesting ... don't you think?

This typical TC post is why nobody took him seriously with the last thread.


Ed's rant does have some interesting comments ... though you have to realize Ed is one of those disgruntled original overclocker crowd who the world passed by several years ago.

Why ??

Well he had nothing to say ... bar bashing the old blue drum ... over and over again.

Bit like the old guy on the side of the road with the sign that says "The end is Nigh ... for AMD".

After a while nobody looks at the sign.

But they do notice the old guy smells and hasn't changed in several weeks ... and has a beard Moses would envy ...

That's Ed.

Plus his site layout is as boring as batsh!t.

I do recommend reading what he has to say ... along with Anad, Sander, and Scott ...they just make a lot more sense.

Ed is just a bit out on the fringe these days ... and you can tell from the style.

He has an account here too.

Hi Ed !!

 

[TechnologyCoordinator]

TC fanning the flames

I'm not fanning anything. I thought it was an interesting article and acknowledged that I had some doubts about some of it and put it on the table for discussion.

The first words of my post:

Very interesting if true.

Unlike AMD fanboys (who link to posts from other forums and quote them as gospel truth), I don't believe anything out on the Internet.

I thought it was very interesting information from Ed, especially with AMD's earnings coming out this week. AMD's going to continue to hurt until they can make a better processor and thus charge more for it.

 

[Reynod]

I was referring to the following statements you made TC.

First the smoke and mirrors with product releases, and now perhaps financials? Intel may be ruthless to competitors, but Intel does show extreme loyalty to its stockholders; an area where AMD has left much to be desired which is evidenced by AMD's sub $5 stock price.

These were your words.

Your just stirring up the sh!t as usual.

Like I said before .. pity we can't rate a thread down so it disappears.

 

[jimmysmitty]

Will somebody please tell me where all those AMD Native quad design adds came from again, they ain't cheap and they weren't paid for by Intel.

More than likely IBM. Like where AMD gets the bulk of its technology including the IMC and their 64Bit.

Funny thing is when I built a PC back in 2002/2003 (just before the release of Athlon 64) I built a system with a Pentium 4. My reasoning? Well at the time Athlon XP ran very hot, wasn't as fast as Pentium 4 (performance wise) and my friends who had them had the worst time getting them to just run at the stock speeds they were supposed to run at due to horrible chipsets and mobos. Heck I had to overclock a friends so it would be seen as a 2700+ (think it ran at like 1.4GHz) but it would only be seen at 700MHz if you left the BIOS as is.

I don't even want to say this but Terascale. 80 cores naitive. Intel has a buisiness plan. It turned out that 65nm naitve quad would not have be profitable for them. Considering the problems that AMD has gone through so far with it it seems Intel made a smart move waiting till 45nm where it will benefit them and the consumer, especially since they used Core 2 to get most of the kinks out of their new HK/MG 45nm process so that means they will be able to produce more per waffer and cut costs as well.

Its kind of funny how that post y iocedmyself is so anti Intel. Its funny because he doesn't realize how many of those nice features he enjoys on his mobo (PCIe, SATA, WiFi ect) that Intel helps create. Either way fanboyism is just stupid. You limit yourself from experiencing a whole world of different abilities and you also limit yourself to the best price/performance. But meh.

If I were iocedmyself and all the AMD fans I would take this a bit more seriously. If this is true its a very big problem as unfortunately not all problems revolve around the product but also the financials. If AMD does not run things right financially and looses too much money you soon will not have an AMD to praise.

 

[BaronMatrix]

Much of the write-off is because Intel canceled AMDs chipset license. That was a large part of ATis income. They are doing pretty good with product launches. 9950BE is available at Newegg as is the 3GHz Brisbane (5800+). Puma is everywhere. Between just HP and Toshiba there are about 12 models, though the 2.4GHz chips aren't around yet.

I remember saying that AMD should have stopped dropping prices before they lost 80% of their X2 prices. That's their major problem. But they had to respond to a $183 E6300. Even funnier is that, according to Newegg, the 9850BE is the best selling AMD chip.

 

[NMDante]

How exactly is that stirring or fanning the flames? TC's opening statement is clearly his opinion of what the article he link might be referring to, and the latest hits on AMD stocks are simply his way to confirm his opinion.

So, what did he say that was so "anti-AMD"? Where is the "stirring up the sh!t as usual"?

Sounds like another case of "it's against AMD, so it's got to be untrue or bad" syndrome. How is this any different than linking an article from Fudzilla or the Inquirer?

 

[jimmysmitty]

OMG.......................................................................................................................................................................................

:O :O :O