wbmw,

If you buy shares at $3 and sell it $30, you make a profit of $27 per share.
If you buy shares at $15 and sell at $30, you make a profit of $15 per share.

Rate of return is how investments are compared, not the amount of return. Amount of return varies between investors (depending on how much they invested), the rate of return is a universal measurement. Nobody really measures things the way you outlined above.

Joe

wbmw,

Mark my words

I have marketed your words, which is why I linked to the post twice already.
Here it goes again: http://www.investorshub.com/boards/read_msg.asp?message_id=7905326

Incidentally, you are down 7% per week on that advice. That's an annualized return of either 98% loss (a wipe-out) or 5344% gain as a result of either following your advice or doing the opposite.

At least you are putting your money where your mouth is - oops, you have not sold or sold short last week.

there will be a time next year when AMD's stock will take a beating.

All I can say is: "Put Up Or Shut Up". Or are we going to be hearing "I told you so" a year from now, based on a retroactive trading from the stock's 52 week high?

Do I need to explain to you, though, how this claim is different from a claim that Intel will "bury" them?

It seems to me that you are predicting Intel performance lead, end of (or reversal) of AMD market share gains, a stock plunge. So, yeah, I would be curious how that it is different from saying tha Intel will bury AMD.

Joe

bobs,

Anyone notice the way the wind seems to blow harder when AMD is doing well?

You can use a windsock to measure it...

Joe

wbmw,

Nothing is going to change until Intel starts delivering on their promises, anyway.

Things can change while Intel is NOT delivering.

Joe

Combjelly,

AMD didn't get much use out of the D stepping masks either, but they provided some risk reduction.

As far as E-stepping, don't you think it is a must for 300 mm? How else would they conducting the testing?

Joe

combjelly,

Pacifica is due to be in the F stepping, which looks good for early Q106.

I must have missed the news (rumor). Where did you read about F stepping and Q106?

Joe

wbmw,

It's the only place where I can actually get engaged in an Intel and AMD competitive discussion

Could that be because the moderator (scratch that, ex-moderator) tells posters (with opinions contrary to that of Kool-Aid drinkers) that their posts are not welcome?

Joe

wbmw,

I have never said "bury", and I'm not saying it now.

Didn't you said just recently that AMD will fall from the 6% market cap share of the combined market cap (and therefore Intel will gain from 94% of combined market cap)?

All that's going to change, though, and soon. Find a good place to sell, you won't want to be part of the group holding the bag next year when valuations drop like a rock.
http://www.investorshub.com/boards/read_msg.asp?message_id=7905326

Joe

CombJelly,

Because he is a Kool-Aid drinker.

Plenty of Kool-Aid being serve on that thread...

Joe

wbmw,

Blackford and Dempsey will close the performance gap relative to dual core Opteron, while offering better RAS, manageability options via IAMT, I/O acceleration via IOAT, virtualization, and better scalability (with FBD memory and dual FSB). Opteron may have performance and power going for it now, but it's soon to get eclipsed.

If INTC stock gained a dollar for each promise of how Intel will bury AMD, INTC would be in the vicinity of the moon.

Joe

mas,

I was looking forward to these scores. I thought the DL585 with 2.4 GHz with software update would produce >205K score.
http://www.investorshub.com/boards/read_msg.asp?Message_id=7833634&txt2find=200k

202.5K is of course enough for a big lead.

Joe

Andy Grave,

He doesn't believe half of what he says and the talk of this all being about investments is just for consumption. In his 4 years here, Intel stock has done absolutely nothing despite how many thousand posts? How many real investors would waste that much time? Would you?

The same could be said about some AMD investors who post here, if AMD stock was stuck somewhere between low single digits and low teens.

Of course plenty of emotions would be flying here if that were the case. What I find interesting is a complete detachment of Intel fans (on their thread) from the fact that Intel the stock has been a dog for years. I don't see many questions of how Intel can make billions of dollars of profits per year and how it is not achieving anything for the shareholders. Well, we don't need to bother this thread about that.

Joe

Durl,

It may be a good idea to limit the number of tangents you go on in a single post, and to concentrate on the subject of the post (and thread) at hand.

Joe

Durl,

Ooh. That has really got to hurt AMD. Here they are 3 years late in mobile and now they introduce a product that is about the same or a little worse?

Kind of like Intel in desktop and server market isn't it?

Joe

wbmw,

Good points. And as you point out, it is the memory size that can make a big difference, as well as disk latency and bandwidth. So Tpm-C is actually 3 steps removed from SPEC CPU benchmarks as a CPU test. Spec CPU tests pretty much ignore memory size, disk latency and disk bandwidth.

In terms of comparing CPUs, you can certainly use TPC-C as a basis, given how well it scales with faster CPUs. You can count on vendors such as HP and IBM creating best-of-breed configurations with a given CPU, making it fairly apples-apples to compare two scores from the same vendor.

I will actually differ here. There are way too many variables in how the test systems are put together, and there is software as a further (often differend databases OSs and version of databases and OSs) removing it further from being apples to apples as far as CPU comparison.

Joe

Tenchu,

There are many facets of performance. SPEC is very useful when architecting a CPU and/or programming a compiler, but it falls short when trying to predict actual application performance. For that, TPC-C is what I've relied on for years in the server world.

It is a benchmark for system level database performance, which exercises all of the components. But since the disk access is an order of magnitude (actually several) slower than anything with CPU and memory, the results can be skewed heavily by the choice of I/O.

Spec CPU tests at least eliminate the disk out of the equation, making it much more of a pure CPU test (even though memory bandwidth and latency still play a role).

joe

wbmw,

Pentium II launched at 300MHz at $1981

Yup, thanks for the link. The exact number slipped my mind, but I remember distinctly that it was very high price in terms of x86 processor at that time.

Joe

wbmw,

Not in the beginning. It took years to promote PPro as a viable alternative in the enterprise. Intel couldn't do it overnight. And versus RISC, they had the benefit of huge economies with inexpensive Linux operating systems, vastly less expensive hardware, and huge 3rd party infrastructure.

x86 was in the enterprise for a long time, but mostly on the department and in the small to mid sized companies. The OS of choice in that time frame was of course Novell, with Windows gaining share. Linux in the enterprise is a much more recent phenomenon. Linux started out in web servers on x86.

Joe

wbmw,

Well it sounds like prices must have seriously dropped during that year, because I looked it up, and the 200MHz PPro (small cache version) launched in November, 1995 for $1325. Meanwhile, Pentium 166 MMX launched in January, 1997 for $407.

Pentium 2 launched at 300 MHz and somewhere between $2K and 3K, IIRC. But it dropped in price, eventually reaching Celeron. 128K L2 version was the Intel CPU I used for my main personal computer. (I got another P 2 or 3 server at home for some light testing work, but it is not my workstation).

Joe

wbmw,

Pentium Pro launched in November, 1995, and was a huge step towards putting x86 into the enterprise, where RISC was a formidable opponent.

True, but prior to P6, x86 were really only the low end of the servers, and it was P6 that allowed x86 to take over the mainstream of the server market.

Pentium II launched in May, 1997, and by this time AMD already had the K6 processor. They went pretty much tit for tat for the next couple years until AMD launched the K7.

On balance, I think Pii and Piii were ahead in absolute performance vs. K6. And it went head to head vs. K7 (Coppermive vs. Athlon classic) for some time in the very important Integer centered apps while clock speeds were roughly on par (700 to 800 MHz). Tbird gained clear leadership when Piii could not keep up with clock speeds, and when Tbird features started piling on - on die L2 (to reach parity in that department), higher bandwidth, PC-133, DDR-2100. But at that point, Piii was eol.

True, but I'm not expecting any miracles from AMD until their K10 micro-architecture, whenever that is.

I am not expecting miracles either, but sometimes, something that seems within grasp is not really there yet. Case in point Mentocito. When was the first time Intel flashed a waver with Montecito die on it? It seems like more than a year ago. It may be another year before full featured Montecito (2 GHz, 24MB L3) arrives.

We were told that OEMs have been working with Montecito for some time, and now it turns out that the feature set and the power dissipation numbers are still in state of flux.

I am not saying that the same will happen with Merom. But it is just the nature of the beast that there are delays and often unfulfilled promises.

The same applies to K10, BTW.

Joe

chipdesigner,

Sounds like you sold near the lows of 2005.

That was not my text. It was a quote of wbmw's text.

Joe

re: SGI

I think SGI will be the major victim of IPF being day late and dollar short. Some of their design wins will not be able to be delivered on time, and will come below expected spec, which may provide additional strain on their fragile balance sheet.

BTW, is the SGI plan you refer to the reorg / bankruptcy consultant they hired recently, or is there something newer?

Joe

wbmw,

I'm betting that the Israeli designed core will be the next really great thing since the P6 (or perhaps even rivaling it).

Banias/Dothan and when released, Yonah/Merom will likely avert the meltdown Intel was leading up to with their EPIC and Netburst micro-architectures. But it is yet to be seen if it will achieve performance leadership that P6 had during some periods of its life time.

One thing is pretty clear, which is that this series of cores will provide huge improvements (on balance of features such as performance, power consumption, manufacturability, die size) over legacy Netburst and EPIC microarchitectures.

It remains to be seen where the competition will be in the future, so comparisons vs. the competition will have to wait for a while.

Joe

wbmw,

First, I'm not attacking anyone, unless you count those people on the opposite end of the spectrum who are counting on AMD market cap surpassing that of Intel's. Besides, there's nothing wrong with believing there is NO aberrant difference between Intel's and AMD's valuation.

I was not talking valuation, which for Intel is fine, after coming down from the unrealistic days of the bubble (given thngs will not get worse profit-wise).

What I mean is that whenever a company had a dominant position in any industry, it invariably lost it over time. Microsoft and Intel had a pretty good run, but I think it is only question of time until they are taken appart by hungry competitors, especially in case of Intel. Intel is already suffering from complacency, and nothing really great came out of the US design teams since P6, and Intel averted a major meltdown only thanks to the Israeli design team.

Only Intel will have a marketing team capable of communicating their value proposition to the critical end users and IT departments for businesses around the world.

You mean like threatening OEMs who want to offer choice (and a value proposition) to their customers? True, only Intel can do that. But can it last? I think the law enforcement already officials already have Intel in their crosshair.

I sold my last AMD several months ago when I felt their gains were nearing their peak.

Ok. But since you keep telling people to sell, you can go further and short AMD, to keep up with your rhetoric.

As for my INTC, I already own many shares, as I believe I've said enough times before. My money has been where my mouth is for quite some time, in spite of hardly ever seeing the same from you hypocrites.

I posted my position on SI, which is long AMD, no position in Intel. You did not check, which makes me a hypocrite?

Joe

wbmw,

I'd be interested in hearing why you think this is an aberration. AMD deserves ...

It is not about what AMD deserves, or what Intel deserves. It is just that it is uncommon for a player to have such a dominant position, and that it is uncommon for it to last.

You are going on the limb saying that it will become even more aberrant in the future, and attack people who say the opposite will happen ...

BTW, I wonder if you are just saying all that, or if you actually believe what you say, and have invested some money along the lines of your hypothesis? The people you attack apparently did put some money on the line.

Joe

Tenchu,

You mean the very AMDroids who said SPEC wasn't a valid benchmark until Athlon 64 started leading in it (peak or base)?

It is certainly more valid as a CPU benchmark then TPC benchmarks.

Still, as a CPU benchmark it is not perfect, because of the compiler issues, making Intel's CPUs look better than they do in real life code.

Joe

wbmw,

You are obviously misreading if that is your interpretation. I have only said that Intel's upcoming products will be a lot more competitive.

No, you have been speaking of upcoming Merom generation products as current products, or in present tense. All I am saying is that this is a public board visited by outsiders, for whom Merom, Yonah, Montecito are future products.

On internal Intel BBS, these may be the current projects, and stuff that is being sold in stores the past. On this BBS, past, present and future refer to something different.

Joe

wbmw,

AMD is the kind of company that survives in tough times and thrives in good times. There's a lot of money to be made by catching them during their lows and betting against the optimists, such as what Elmer does when he sells calls from month to month. Anyone expecting Intel to fumble enough for AMD to catch up with them in market cap might as well invest in the lottery. It might happen, but it would be like

When 2 companies have a market cap in ratio of 15:1, it is a long shot to say that it will become 1:1.

But to say that 2nd biggest company in the merchant CPU business (where there are basically 2 competitors) can't have a market cap that exceeds 6% is as unprecedented.

The current state is basically an aberration. You seem to be predicting that the aberration will not only continue, but will become even more pronounced.

I don't know about too many instances of something like this in the past, other than Standard Oil. Maybe Microsoft.

Joe

UpNDown,

I'd give 1000 to 1 odds that Intel won't reach $37 in the next two years.

I think you would find that the option market will give you better odds on this bet.

Joe

wbmw,

I often see sour grapes AMD investors turn to attacking older generation Intel products when they are uncapable of finding fault with the current generation.

I notice that you very often (even consistently) refer to future Intel products as current ones. As far as AMD, it is what ships today is the current one.

You may be working on these Intel products, and they may be current to you, for the rest of us on the outside, current Intel products are Prescotts (and its shipping derivatives) and for AMD, it is K8 (and its shipping derivatives).

Joe

wbmw,

thanks for the run down of the technology issues. But I was wondering why it is that AMD can prosper, or merely exist only under when Intel slips up (your earlier reference of Muzo ending up holding the bag).

This is not true for HP / Dell, Coke / Pepsi, Sony / Philips etc.

Joe

chipguy,

Feel free to document it.

I posted 2 "Oopses". Isn't that enough?

Joe

wbmw,

You must be expecting another strikeout from Intel, something along the lines of a floor lined with bananas. If so, I'd love to hear what you have in mind; otherwise, it blows my mind that you cannot see what is so perfectly obvious.

Maybe you can spell it out for us.

Joe

chipguy,

Speaking of gumbies, have you figured out your problem
reading the SPEC web site yet?

I get Opteron leading both Xeon and Itanium across the board. What do you get?

Joe

chipguy,

So 131 to 104 SPECfp_base_rate_2k is "by a mile"?

Actually, it is 144 SpecFp_rate I was comparing. and leading by 38% is leading by a mile. Itanium seems to be leading 1870 Sparc V by that margin, and that's an old submission. Also interesting is that Opteron sells for a fraction of a cost while leading by 38%. So Itanium would need to sell for about 1/2 price of the price of Opteron to be competitive.

It is amazing how uncompetitive Itanium is across the board, and has to wait for the next Hail Mary, some 6 months out, to save it.

Given
we are comparing four 90 nm dual core chips compared to
four 130 nm single core chips in a system configuration where
the 90 nm chips have twice the memory bandwidth

I am not sure if you took all the variables into account in the past when comparing Tpm-C scores, such as 32 bit OS, disk subsystem, memory, cost of the system... I am not sure if you discounted Itanium scores appropriately. I mean from the scores showing Itanium to be behind to Itanium being way behind.

When Montecito ships...

How many posts did you make about all the things that would happen when Madison ships. How about doing some accounting of promises of what would happen when Madison shipped vs. reality.

Here is kind of interesting, where you called Opteron based system a row boat and Madison based system an aircraft carrier:
http://www.investorshub.com/boards/read_msg.asp?message_id=1637589
"A Category killer" - IBM and Dell killed that category.

Joe

mas,

HP's internal IPF memo rebuttal to the Inq

Sounds like the dead parrot scene. Or is it "Bring out your dead"?

Joe

Ooops is right.

It does not change the fact that Itanium is now behind Opteron in 4 socket configuration on SpecFP_Rate - by a mile.

Joe

Alan,

Yeah, I missed a bunch of I2 scores. I guess poor SpecMine skills on my part.

Joe

chipguy,

I thought you might enjoy this:
http://www.spec.org/osg/cpu2000/results/res2005q3/cpu2000-20050902-04578.html

4 socket Opteron is now beating 8 socket Itanium in SpecFP_rate. I think it was Itanium's last claim to fame. Who would have thought that?

Joe

chipguy,

Only someone deliberately trying to deceive would choose
to compare IDC to Gartner numbers instead of IDC to IDC
numbers.

???

Sounds about 30k to 50k processors too low for the sum of
2003 and 2004 and about 40 to 70k too low for all IPF MPUs
shipped to the end of 2004.

Too low compared to you assumptions. The claims figure is for actual shipments. Shouldn't that be a good reason for you to revisit you assumptions, rather let u know how much they are short of your assumptions? You are starting to sound like Adi on Aces.

Joe