Phenom video leaks from Lake Tahoe

[Silent_Buddha]

Personally the spider platform isn't even a consideration for me until they get the SB700 south bridge.

The rest of it sounds very nice however, but the fact that the SB600 is holding it back is just painful. To have all that work put into the 790 NB and then to pair it up with a relatively ancient and outdated SB?

Regards,
SB

 

[AlexV]

This is another not very enticing thing to see:http://www.legitreviews.com/article/597/10/. If that's how a 2.6 part behaves, where will the mythical 3Ghz on air one be?

 

[WaltC]

My point being, since there were some people who thought R600 was a worthwhile purchase, there will be people who, despite having expected Core 2 killer will talk themselves into still getting Phenom, actual results notwithstanding.

Actually, my point of view is somewhat different. After having perused the pack of Phenom reviews circulated today, I can't think of a time when I've been more disappointed with the lack of thought and clarity in the reviews I've read today for the Phenom/Spider product introductions. Every single one of the reviews I've read so far base their core assumptions (pardon the pun) on various benchmarks compiled and optimized primarily to run on Intel cpus. That in itself isn't so bad, since Phenom is only today coming out of the starting gate and only now beginning to ship. So naturally, things like compilers and optimizations specific to Spider platforms in general, not to mention motherboard bioses for these brand new AMD core logic chipsets, are currently under development--which means they're being debugged and optimized even as we speak.

What's so disappointing is the lack of recognition of any of these factors--factors which affect all system introductions--not by any means only AMD systems. The people writing the reviews I've read so far seem almost gleefully oblivious to the ins and outs of all product development and how those in and outs might specifically be affecting the Phenom introduction at this precise point in time. Basically, if the Phenom/Spider/790 product introductions follow the same pattern as all previous Athlon and Intel product introductions, things are going to improve very quickly over the next few months, and improve dramatically.

But this seems to be something today's (as in today, literally) crop of reviews simply think is utterly unimportant. Even so, this by no means keeps many of them from making grand predictions of doom & gloom for AMD--based merely on the running of literally a *handful* of Intel/and-or-nVidia-compiled and intel/and-or-nVidia-optimized benchmarks. Sandra, in particular, has always heavily tilted toward Intel product support, and obviously something in Sandra isn't quite reading the Phenom right with respect to cpu memory bandwidth--and probably lots of other things, too. Yet, most of the reviews I've read today proclaim that Phenom's bandwidth is half what the A64's was simply because that's what the current version of Sandra tells them. Yes, where would some of these "reviewers" be without canned benchmarks like Sandra to guide them along to the conclusions that the parties behind Sandra, parties hidden from view, wish them to reach?... Kind of a frightening thought--but there it is.

So, you might want to think about what the term "actual results" really means... Point is, today's "actual results" may not mirror the "actual results" even a month from now, let alone 3-6 months from now. I've seen exactly that scenario unfold many times in the past, and I surely have a hard time believing I'm the only one. That's why my greatest difficulty at the present time is talking myself *out* of a Phenom-based system in the next few months. Possibly it's hard for me because even now at the moment running my Athlon x2 4200/HD2600 XT 512/ Asus A8n5x/22" LCD system I'm having no difficulty running the latest games at smooth and fluid framerates and enjoying them without difficulty--despite the fact that this or that hardware reviewer keeps trying to tell me that if I'm not running all my games at 1900x1200 or higher then I'm not really gaming, if you know what I mean...

Right now, I see nothing to dissuade me from buying some version of a "Spider" system in the next few months, probably one I'll put together myself rather than a bog-standard OEM version, and the fact that current benchmarks do nothing to flatter Phenom isn't of any concern to me. At this point in time I simply would not expect them to, as for the past year they've been refined with Intel optimizations in mind--nothing sinister or underhanded about that at all, because Core 2's been shipping all that time while Phenom has not.

So If I buy a Phenom system in the next few months it will be because of lots of things beside canned benchmark frame-rates that few of today's reviewer's have bothered to write about, let alone think about, it seems to me. And, if I can get 80% of the performance of the highest-end Core2 shipping at the time I buy, and pay 30%-50% of that Core 2 platform's cost while doing so, then I'm going to consider myself and Phenom to be big winners...

Basically, I think that today's reviewers who have concluded that the contest between Phenom and Core 2 is over and done with have failed to apprehend that that particular contest is, in fact, only just beginning. I would have thought that today, on the day of Phenom's launch, that nothing could be more obvious.

 

[trinibwoy]

Basically, I think that today's reviewers who have concluded that the contest between Phenom and Core 2 is over and done with have failed to apprehend that that particular contest is, in fact, only just beginning. I would have thought that today, on the day of Phenom's launch, that nothing could be more obvious.

Sometimes reality is a hard pill to swallow. You are NOT going to see vast optimizations for Phenom that will bring it on par with Intel's Core architecture no matter how long you wait. It's just slower, plain and simple. And that's before accounting for Intel's clock speed, power consumption and process advantages.

 

[Freak'n Big Panda]

Basically, if the Phenom/Spider/790 product introductions follow the same pattern as all previous Athlon and Intel product introductions, things are going to improve very quickly over the next few months, and improve dramatically.

Similar things are said about almost all under performing parts upon launch. When have we seen a CPU architecture become competitive after launch? I can't remember any such event. The only thing that's going to save phenom is fast clock speed ramps (which doesn't seem very probable at this point), nothing else is going to make it competitive.

On another note: What the hell is AMD going to do? Phenom isn't ever going to compete for the performance crown, and ATI isn't doing any better against NV's offerings... I don't really see how AMD is going to survive. I guess there was a time back in the k6-2 days when intel dominated the performance arena and amd dind't die then so maybe they'll hold out, but the market is a lot different now and there's a lot more money at stake.

 

[hoho]

Shangi and R700? Or you just skipping over those?

As was said Shanghai is a simple 45nm shrink of K10. When was the last time AMD released its die shrinks at higher clock speed than its older ones? It certainly wasn't like so with 90 and 65nm. Also a fun fact is that during the last year AMD K8's have gone from 3->3.2GHz on 90nm and from 2.6->2.7GHz on 65nm.

Word is that the Phenoms OC pretty decent actully(Sami OC a 2.3 to 3ghz, and Kyle B seeing 3ghz modals scale to 3.6 helps support this).

Anyone else besides those two? I remember reading one Phenom review that said AMD had a special box for the journalists set up for OC'ing, the ones they were using for benching at stock clock speeds weren't really able to OC. One reported that from 2.3->2.35GHz resulted in crash.

What I saw and consider fairly scary is an AMD slide suggesting that the TDP for the 9900 2.6Ghz chip will be friggin 140W!

Interesting, do you happen to have a link to that? Legitreview did say it has 140W but I couldn't see else but their power usage test that would confirm it. Testing an ES chip for power usage is not exactly too well comparable to retail chips.
I wonder how much would that be in Intel ratings, >175W?

Sandra, in particular, has always heavily tilted toward Intel product support, and obviously something in Sandra isn't quite reading the Phenom right with respect to cpu memory bandwidth--and probably lots of other things, too. Yet, most of the reviews I've read today proclaim that Phenom's bandwidth is half what the A64's was simply because that's what the current version of Sandra tells them

You can thank split memory controller for that. I think in most real-world applications things should be better, especially when you have multiple tasks running in parallel.

 

[AlexV]

Walt, I'm glad to see that you've retained at least some of your verbosity, and the writing itself is pleasing enough, but your argument has no legs to stand on. Even if it were as you say, with the ubicuous software compiled for Intel, what makes you think that's not the reality, and will be the reality, simply due to the fact that Intel provides a lot of tools that devs like and use?Do you honestly believe that there are gajillions of software products that will be compiled with an AMD compiler(of which, if it even exists, I've seen scarce advertising of) that'll magically compensate for the lower IPC and sucky clock-rates?

Do you also believe that the R600 will start stretching its wings once devs start coding for it?Newsflash:it's not going to happen. Ever. First because even under such a scenario it wouldn't blow the doors of of anyone due to its own failings, and second because AMD isn't a driving force in any marketplace, be it CPUs or GPUs. They're hardly trend-setters, and they're hardly aggressive enough(nowadays I doubt they even have the resources to be) to become trendsetters. The much maligned TWIMTBP program illustrates this quite well, it shows which is one of AMDs shortcomings(yes, I know this could be considered as particular to ATi, but AMDs behaviour was fairly similar to ATi's anyhow):they all too often opt for a reactive stance, rather than a proactive one, and they all too often leave their products out to dry instead of pushing them like there's no tomorrow.

I fail to see the similarities between Phenom's launch and all other CPU launches in history. Phenom seems to have missed goal after goal after goal. They've missed their initially aimed release date, missed their target-clocks, seem to have missed the pricing niche they might've aimed for,as the pricing is not that competitive considering what you get for it.

Hoho, try this:http://www.extremetech.com/article2/0,1697,2218304,00.asp. I can't remember if I have also seen a slide floating around quoting that huge TDP.

 

[WaltC]

Sometimes reality is a hard pill to swallow. You are NOT going to see vast optimizations for Phenom that will bring it on par with Intel's Core architecture no matter how long you wait. It's just slower, plain and simple. And that's before accounting for Intel's clock speed, power consumption and process advantages.

I'm sure that's what you think. Sure, let's just discount the fact that Core2's been shipping for a year and has undergone numerous supporting core logic changes, supporting system bus changes, fsb memory controller changes, core bios revisions, compiler revisions, and benchmark optimizations that play to Core 2's strengths---not even, as you say, counting process changes and refinements to subsequent revisions, or even clock speeds. Let's fantasize that all of those changes over the course of the last year just don't matter to Core 2's performance today. After we accept that fantasy, it's much easier to accept the fantasy that Phenom will "never do this or that" as well as Core 2, "no matter how long you wait." Here we are on Day One of Phenom's launch, and you're telling me what it will *never* be... Think about that for a moment.

 

[AlexV]

I'm sure that's what you think. Sure, let's just discount the fact that Core2's been shipping for a year and has undergone numerous supporting core logic changes, supporting system bus changes, fsb memory controller changes, core bios revisions, compiler revisions, and benchmark optimizations that play to Core 2's strengths---not even, as you say, counting process changes and refinements to subsequent revisions, or even clock speeds. Let's fantasize that all of those changes over the course of the last year just don't matter to Core 2's performance today. After we accept that fantasy, it's much easier to accept the fantasy that Phenom will "never do this or that" as well as Core 2, "no matter how long you wait." Here we are on Day One of Phenom's launch, and you're telling me what it will *never* be... Think about that for a moment.

It's kindof pointless to illustrate to a guy that some other guy managed to get in and shag the cheerleader in a number of imaginative ways simply because he got there earlier, and they left together in his Porsche, whilst the aforementioned guy still drives a Pinto...with perspectives of maybe getting a Porsche someday...if the stars align. The first guy is still left with jacking off, and if he's constantly second best the likelihood of him becoming a monk is ever increasing. The IT business is just like a woman-not that good with waiting for a player to get his stuff in order, whilst there's another one already doing all the right things.

 

[hoho]

Hoho, try this

Thanks. That does look kind of scary and when I before thought AMD could have 2.8GHz out in Q1 I now highly doubt that.

Sure, let's just discount the fact that Core2's been shipping for a year and has undergone numerous supporting core logic changes, supporting system bus changes, fsb memory controller changes, core bios revisions, compiler revisions, and benchmark optimizations that play to Core 2's strengths---not even, as you say, counting process changes and refinements to subsequent revisions, or even clock speeds.

So are you claiming that e6600 released last summer is much slower than e6600/e6550 in shops now? How big is the difference? Sure, Intel had a revision that lowered TDP by quite a lot but speed of CPUs hasn't really changed all that much. FSB did increase but that doesn't help all that much either as most things aren't really limited by it.

 

[Geeforcer]

Walt, I'm glad to see that you've retained at least some of your verbosity, and the writing itself is pleasing enough, but your argument has no legs to stand on. Even if it were as you say, with the ubicuous software compiled for Intel, what makes you think that's not the reality, and will be the reality, simply due to the fact that Intel provides a lot of tools that devs like and use?Do you honestly believe that there are gajillions of software products that will be compiled with an AMD compiler(of which, if it even exists, I've seen scarce advertising of) that'll magically compensate for the lower IPC and sucky clock-rates?

Of course. Based on today's results, AMD market share will skyrocket. Will hundreds of millions Phenoms out there within weeks, everyone will be forced to recompile their code on the cutting-edge AMD compiler. Within weeks, Phenom will go from -5% to +60% VS Core2, per clock.

BTW, I love how Walt mentions Core2 over and over again. When it was launched, it did not need ridiculous "wait 6 month" excuse. It was dominant out of the gate.

 

[swaaye]

I've always thought that AMD tries to make their CPUs perform well with Intel-optimized code. Hence the fact that they implement the latest SSE they can and gave up on their own SIMD tech. And that Athlon's FPU was designed to beat P6's at its own game. I don't think AMD is in a position to dictate the direction of the industry at all. x86-64 was about the only exception, but it hasn't really been a huge competitive advantage obviously.

 

[hoho]

Erm, AMD has no compiler. It has helped GCC a bit and is helping Portland Group with their compiler. Though I bet ICC+MSVC have much bigger share of the market than everything else combined.

 

[WaltC]

Similar things are said about almost all under performing parts upon launch. When have we seen a CPU architecture become competitive after launch? I can't remember any such event. The only thing that's going to save phenom is fast clock speed ramps (which doesn't seem very probable at this point), nothing else is going to make it competitive.

You don't remember the K7, the P4, or the A64? Each of those cpus, for a wide variety of reasons, improved dramatically after its launch in relation to its competition at the time of its launch. What do you mean you can't recall that ever happening with a cpu? Such improvements for all launched cpus of any success are commonplace.

Let me also ask you a logical question...if AMD really didn't think that Phenom could be "competitive" from a price-performance perspective, then why would AMD release Phenom in the first place? To short circuit you answering with "AMD has nothing else," let me just assure you that we don't really know how competitive Phenom is going to be with Core 2 from here on out. That's as much up to Intel as it is up to AMD, certainly. It's also in the future and as such is a question that has yet to be resolved. In large measure the answer will depend on both companies being able to deliver that which they have promised. AMD needs to ramp Phenom up in MHz and yields, aside from doing all of the other things it needs to do to properly support its new cpus. OTOH, even though Intel has a year's head start with Core 2, Intel needs to ramp up Core 2 to be able to ship its 45nm Core 2 products in a sufficient volume so as to remain competitive.

On another note: What the hell is AMD going to do? Phenom isn't ever going to compete for the performance crown, and ATI isn't doing any better against NV's offerings... I don't really see how AMD is going to survive. I guess there was a time back in the k6-2 days when intel dominated the performance arena and amd dind't die then so maybe they'll hold out, but the market is a lot different now and there's a lot more money at stake.

Heh...join the small but active crowd who said precisely and exactly the very same things you've said about AMD just after the reportedly late launch of AMD's A64 cpu several years ago. It was said by the know-nothing pundits at the time that the A64 was "too little, too late" and so on, to ever make even a dent in the x86 server cpu market that Intel had effectively monopolized up to that point in time with Xeon. All the know-nothing pundits actually proved by those statements, however, was that they knew nothing at the time... On the eve of the A64 launch, speculation was rampant as to "who was going to buy" the flailing AMD--IBM? nVidia, even? The know-nothing pundits "knew" that AMD was doomed no matter what it did. Again, they knew nothing, didn't they?

It amazes me how lenient people can be in being willing to lend Intel all the time it needed to catch up to and/or surpass the A64--and we are talking year(s) here, in the plural... During those years when Intel was definitely playing second fiddle to AMD on the high-end of the x86 server cpu marketplace, did the know-nothing pundits emerge to ask when Intel was "going under"? Where were the prophets of doom and gloom? Yet, here we are on the day of Phenom's launch--the cpu isn't technically 24-hours old yet--and already people are telling me what it will "never" be, just as surely as they know what Core 2 "will be" in the future.

Is the fact that AMD is 10% the size of Intel news to anyone? AMD has never had the money to burn that Intel has burned over the last decade, particularly when the success of the A64 forced Intel to abandon it's "You don't need 64-bits on the desktop" PR campaign that tried unsuccessfully to bury AMD in the rubble of x86 which Intel unsuccessfully tried to bury by cajoling the world into IPF. The markets didn't want IPF because they'd have had to throw out the baby with the bathwater just to make things easy for Intel's continuing monopoly stake. Unsurprisingly, the markets elected to look to their own interests instead of Intel's. Ironically, Core 2 is the result of Intel's failure to persuade the markets that they "didn't need 64-bits on the desktop"--unless, of course, that desktop was Itanium...

So to answer your question, "What is AMD going to do?" I can think of only one answer, and that is that AMD is going to continue to *compete* as it has always done. Look, if pre-K7, Intel didn't have enough clout to make AMD dry up and blow away, why on earth would you assume Intel has a better chance of succeeding today, when AMD is so much more competitive in every way than it was then? I mean, what do you think--is AMD's acquisition of ATi and the subsequent Spider platform that launched today an indicator that AMD is becoming less competitive, or more competitive?

 

[AlexV]

Precisely the fact that AMD isn't Intel is the key to understanding that you avoid:they don't have the cash, resources and weight to afford a "slump" similar to Intel's Netburst day. They don't have the time to recover from floppiness. They can't afford to be re-relegated to second tier-edness.

 

[Bludd]

They don't compete at prestigious high end, that's for sure.

Oh, and I know huge and long posts is "your thing", WaltC, but it is getting kinda ridiculous.

 

[hoho]

To short circuit you answering with "AMD has nothing else," let me just assure you that we don't really know how competitive Phenom is going to be with Core 2 from here on out

Let me ask agian, when was the last time IPC of a CPU has risen with revisions? If there ever has been such a case, how much faster did it get? Remember, K10 is around 8-12% slower than Core2, compared to Penryn the difference is bigger. Now, Intel sells 45nm 80W 3GHz quadcore Xeons for a couple of weeks. AMD will have 140W 2.6GHz quadcore in four months.

What exactly can AMD do to become competitive? They are at least 1-2 years behind Intel in processing technology, that surely won't help them.

OTOH, even though Intel has a year's head start with Core 2, Intel needs to ramp up Core 2 to be able to ship its 45nm Core 2 products in a sufficient volume so as to remain competitive.

I think Intel said that by the end of Q2 50% of mobile CPUs will be on 45. On desktop it'll take one more quarter.

On the eve of the A64 launch, speculation was rampant as to "who was going to buy" the flailing AMD

Back then AMD didn't have a massive dept.

It amazes me how lenient people can be in being willing to lend Intel all the time it needed to catch up to and/or surpass the A64--and we are talking year(s) here, in the plural

During all the Netburst time Intel still continued making healthy profits. Since Core2 launched it got back most of the marketshare AMD had stolen from it during its supermacy.

During those years when Intel was definitely playing second fiddle to AMD on the high-end of the x86 server cpu marketplace, did the know-nothing pundits emerge to ask when Intel was "going under"?

Intel had huge piles of cash and continued making profits while it wasn't in the lead. Same cannot be said for AMD.

Yet, here we are on the day of Phenom's launch--the cpu isn't technically 24-hours old yet--and already people are telling me what it will "never" be, just as surely as they know what Core 2 "will be" in the future.

Well, we know quite well that no magic will happen during next four months. Also nothing is known about AMD 45nm. Usually a new CPU is demonstrated a year before launch. Has anyone seen anything about Shanghai? Sure, AMD has talked about "ramping" 45nm H1 next year but it "forgot" to mention that it is talkig about installing the tools. It takes quite some time from tools to working CPUs.

I mean, what do you think--is AMD's acquisition of ATi and the subsequent Spider platform that launched today an indicator that AMD is becoming less competitive, or more competitive?

To me it only means that there will be AMD logos all over the place. It is not like pairing the things together would actually increase performance. I don't expect much interesting from them until some 3-5 years when Fusion really kicks in.

 

[3dilettante]

Actually, my point of view is somewhat different. After having perused the pack of Phenom reviews circulated today, I can't think of a time when I've been more disappointed with the lack of thought and clarity in the reviews I've read today for the Phenom/Spider product introductions.

That's too bad, since AMD controlled most of the vague and unclear reviews.
Coincidence?

Every single one of the reviews I've read so far base their core assumptions (pardon the pun) on various benchmarks compiled and optimized primarily to run on Intel cpus.

Then it's business as usual.

That in itself isn't so bad, since Phenom is only today coming out of the starting gate and only now beginning to ship. So naturally, things like compilers and optimizations specific to Spider platforms in general, not to mention motherboard bioses for these brand new AMD core logic chipsets, are currently under development--which means they're being debugged and optimized even as we speak.

That should tell you something.
The product is days away from supposedly shipping in quantity, and the platform is in shambles.
It's a repeat of the Athlon days when Intel pressured motherboard makers to not make Slot A motherboards.
Only this time, the board makers would kill to make an AMD board, only AMD's failed execution has stopped them.

There is no way in hell that a final product that is shipping for real money should be in that position.
Opteron and A64 had running boards and near full-speed processors months prior to launch.

Phenom's status--as a product people pay real money for--is unacceptable, and the fact that AMD and partners are going through with it is indicative of desperation.

Let's not forget that the lousy chipset situation and motherboard issues really hurt AMD's gains with K7.

What's so disappointing is the lack of recognition of any of these factors--factors which affect all system introductions--not by any means only AMD systems. The people writing the reviews I've read so far seem almost gleefully oblivious to the ins and outs of all product development and how those in and outs might specifically be affecting the Phenom introduction at this precise point in time. Basically, if the Phenom/Spider/790 product introductions follow the same pattern as all previous Athlon and Intel product introductions, things are going to improve very quickly over the next few months, and improve dramatically.

No, Phenom's status is significantly worse.
There are questions about stability and functionality days prior to launch.
How often have either Intel or AMD ever admitted that there is a critical show-stopping crash bug that basically limits CPU clock speed to less than the top bin of chips from two process generations before?

But this seems to be something today's (as in today, literally) crop of reviews simply think is utterly unimportant. Even so, this by no means keeps many of them from making grand predictions of doom & gloom for AMD--based merely on the running of literally a *handful* of Intel/and-or-nVidia-compiled and intel/and-or-nVidia-optimized benchmarks. Sandra, in particular, has always heavily tilted toward Intel product support, and obviously something in Sandra isn't quite reading the Phenom right with respect to cpu memory bandwidth--and probably lots of other things, too. Yet, most of the reviews I've read today proclaim that Phenom's bandwidth is half what the A64's was simply because that's what the current version of Sandra tells them. Yes, where would some of these "reviewers" be without canned benchmarks like Sandra to guide them along to the conclusions that the parties behind Sandra, parties hidden from view, wish them to reach?... Kind of a frightening thought--but there it is.

Memory results on a number of programs (including AMD's) do not correctly detect the memory controllers.

That being said, AMD's greatest success didn't come when its products ran AMD-optimized software best.
Remember K7 vs P3 and K8 vs. P4?
AMD made money by making processors that beat Intel at its own game.
They ran Intel code better than Intel did.

Now, AMD is having a hard time making a chip that beats K8 in some situations.
I guess we should blame Sandra for being K8-centric.

So, you might want to think about what the term "actual results" really means... Point is, today's "actual results" may not mirror the "actual results" even a month from now, let alone 3-6 months from now.

Just how long from now is "too late"?
Let's decide on a date, and we'll mark it on our calendars.
When that date comes to pass, we can all get together and then we can discuss things over tea.

Phenom has been released in 4Q 2007, with a top bin half a GHz below the originally planned speed on a motherboard a revision back from what it should have been.
It has chips that cannot be trusted to overclock, because even if they did clock high, they'd freeze the system.

I've seen exactly that scenario unfold many times in the past, and I surely have a hard time believing I'm the only one. That's why my greatest difficulty at the present time is talking myself *out* of a Phenom-based system in the next few months.

I'm sure the one Phenom AMD manages to build in the next quarter will be there for you to buy.

Possibly it's hard for me because even now at the moment running my Athlon x2 4200/HD2600 XT 512/ Asus A8n5x/22" LCD system I'm having no difficulty running the latest games at smooth and fluid framerates and enjoying them without difficulty--despite the fact that this or that hardware reviewer keeps trying to tell me that if I'm not running all my games at 1900x1200 or higher then I'm not really gaming, if you know what I mean...

So AMD's latest and greatest can't even trump its old value lines?
Sounds fantastic.

Right now, I see nothing to dissuade me from buying some version of a "Spider" system in the next few months, probably one I'll put together myself rather than a bog-standard OEM version, and the fact that current benchmarks do nothing to flatter Phenom isn't of any concern to me. At this point in time I simply would not expect them to, as for the past year they've been refined with Intel optimizations in mind--nothing sinister or underhanded about that at all, because Core 2's been shipping all that time while Phenom has not.

I'm sure Phenom-optimized programs will come out.
Going by the current crop, the primary optimizations will be to downclock the chip to 2.3 GHz and only run on one core.
Either that or enable the hidden time machine to bring the Phenom release up 6 months, or do you really think (edit: not) beating a (edit: lower bin of) chip that is been out for over a year a staggering accomplishment?

So If I buy a Phenom system in the next few months it will be because of lots of things beside canned benchmark frame-rates that few of today's reviewer's have bothered to write about, let alone think about, it seems to me. And, if I can get 80% of the performance of the highest-end Core2 shipping at the time I buy, and pay 30%-50% of that Core 2 platform's cost while doing so, then I'm going to consider myself and Phenom to be big winners...

You might be a winner by buying an overly large and underperforming die at fire sale prices.
However, what makes WaltC a winner is not the same thing that makes Phenom a winner.

I'm also somewhat unsure as to where that 30-50% factor is coming in.
Seriously, I haven't seen anything to indicate that a decent motherboard and processor from either platform has that large a disparity.

Basically, I think that today's reviewers who have concluded that the contest between Phenom and Core 2 is over and done with have failed to apprehend that that particular contest is, in fact, only just beginning. I would have thought that today, on the day of Phenom's launch, that nothing could be more obvious.

The contest they're declaring over right now is Phenom versus 65nm Core2.
And given Intel's schedule for production, they're pretty much right.

 

[WaltC]

I've always thought that AMD tries to make their CPUs perform well with Intel-optimized code. Hence the fact that they implement the latest SSE they can and gave up on their own SIMD tech. And that Athlon's FPU was designed to beat P6's at its own game. I don't think AMD is in a position to dictate the direction of the industry at all. x86-64 was about the only exception, but it hasn't really been a huge competitive advantage obviously.

Is that also why Intel went with x86-64 support in Core 2 long after AMD had established it as a standard with the A64? If you'll recall Intel didn't want to go to Core 2 in the beginning--it wanted everybody to move off of x86 and go to Itanium to get their "64-bits on the desktop." As well, I believe I read a few weeks ago of AMD proposing SSE5...and making plans to implement it in '09...? I agree with this much of what you say: AMD simply wants a level playing field on which to compete. AMD isn't necessarily devoted to having the world follow it, but at the same time AMD seeks to serve its markets where Intel has traditionally tried to steer and manipulate them. I think that one major difference in approach is what has allowed AMD to survive and to grow despite every single obstacle Intel has been able to mount against such growth and survival.

Basically, I think the influence of AMD on the direction of the markets since the introduction of the k7 in 1999 has been huge--so huge it is difficult to estimate, imo. Even Microsoft agreed with AMD that there were much better ways than Itanium to do "64-bits on the desktop" and thus Windows x64 was developed and marketed, but not unfortunately until after a lengthy delay designed to allow Intel to get on board the x86-64 train.

As far as competitive advantage goes, that's exactly why AMD's in court with Intel at the moment. During the years where AMD was clearly ahead Intel used its money, which it had accumulated by virtue of its decades-long monopoly hold on the x86 cpu marketplace, to *buy* a market position for itself not commensurate with the performance and the value of the cpus it was selling. Look at Dell and the ensuing scandals revolving around the company--not to mention the company being forced finally into good enough judgment to stop relying on Intel subsidies to maintain its position. I mean, with Dell the company learned the hard way that Intel didn't have enough money to prop it up in the #1 market position indefinitely--so long as Dell agreed to buy only Intel cpus. AMD is the sole surviving x86 cpu competitor remaining of the companies who tried and failed to compete with Intel during those years.

Let's talk about about motherboard manufacturers around the world--like Asus, for instance--who did not like it at all when Intel started saber-rattling and threatening them with sky-high prices if they so much as marketed a single AMD-chipset board. It was companies like Asus and MSI and nVidia and others who pushed AMD to the forefront in those early years of the k7, which was such a vulnerable time for AMD in general. Without these companies and their active and overt aid, the K7 would have gotten very little play indeed, and the recent past might have been very different. I often speculate as to what speed P4 I'd be using on my 32-bit desktop today had AMD folded immediately after the K7 launch as was Intel's fondest wish at the time. Do you really think all of these companies want to return to an era where Intel ruled with an iron hand and to a time where if they wanted to play their only hope of winning was to live by the Intel rulebook? Nah, I really, really don't think that any of those companies pine for a return to the "good ol' days"...

Then there's the story of Rdram and DDR and Rambus and all the rest of it, and Intel was right there behind the scenes pushing and foisting the whole thing on the market, and spending hundreds of millions of dollars in wild abandon while doing it. In the end Rdram lost out and, like Intel would do later with Itanium and "64-bits on the desktop," Intel failed again to see its schemes come to fruition. But Intel has always had so much money to burn that it scarcely notices the billions of $ it has squandered in its various bids over the years to decimate and destroy all who would compete with it, or even defy it in some fashion.

On reflection, I think the influence of AMD on the markets for most of the last decade has been profound, as even the giant, sluggish Intel has found itself forced to do many things that left to its own devices the company would likely never have done.

 

[AlexV]

AMD's major influence on tech is bull. It's not like we wre in the flaming dark ages and they, our lords and saviours, came in and taught us how to light a fire and cook our meat, for crying out loud. I'd really like to learn how to cast these mass brain-washing spells that turn a money wanting company into some charity guided angel-it would make me earn more money with less effortm and I'm all for that.