I think this guy makes a good point

http://weblog.infoworld.com/yager/a..._fud_versu.html

I'm kinda tired of hearing the pundits yack about AMD and their issues. I just think this guy makes a good point about AMD.

Rather interesting and i do agree with what he says.
but regardless, i think the consumers are more concerned with what affects them directly, that is, price/performance in the present tense, not in the future. which is also why amd A64s were pretty popular before core 2 duo came out. but as it stands, AMD doesnt exactly have stuff on te market that can really compete with intel and even if they have nice stuff in the warehouse, it would be pretty useless if they cant get the $ rolling in to fund the nice research that thy are doing...

kinda sux and i hope AMD manages to be top dawg again for a longer period of time, or else intel might simply grow too big...(to handle) and then amd might go the via/ibm route - as in - not very popular anymore. (im talking average consumer)

Trust me, 80% of the market wouldn't notice the slightest difference between an AMD or an Intel chip in their desktops.

As the reporter says, Intel has made a giant leap with the Core series, but if they milk it out as long as they did with their previous families, AMD will roll gently by them over the next year and take a lot of distance, forcing Intel to make a similar leap 5 years from now.

It's a difference in strategy. Intel spends a fortune to do everything at once, and then has to milk it out to get their return on investment. AMD spreads their investments more gradually and brings out something new every year.

Not anymore. After getting hit hard with the K8, they've moved to a new release cycle - simultaneous development of two microarchitectures, alternating release every two years. That's something that I doubt AMD simply has the resources to keep up with.


Any way, I'd like to like that article, but I really can't. It'd be nice to explain away all AMD's problems like that, but there's no qualifications for any of it and it all seems too convenient. I don't doubt that AMD is going to do well with their new launch, as the last bit of the article alludes, but the rest of it? All unqualified.

The article posted above is good. It is very creative and certainly keyword rich content. I respect the author for his ingenuity.

I have no doubt of his writing success, his writing style is awesome, I am sure he is a successful writer and gets paid pretty well. Good job! But I respectfully disagree with his content. The way he explained and described the whole "fiasco" is obviously and unfortunately not true. It is highly fictional, maybe he started out as a sci-fi writer? Jokes aside, it is an okay article. He is pretty confident about his writing, which is again a huge pro, but yet again he just doesn't deals with "facts" even though, that keyword does exists in the title of the article. What he has written about could be real, but it is just not. Also, any experienced reader can feel his altered subjective perspective/view from a mile away; the author of that article for whatever reason(s) is a true believer and fan (better way explaining the overly common 'fanboy' term) of the company called Advanced Micro Devices.

It is okay to have certain opinions. It is okay to write articles about virtually any subject, it is also okay to make money via marketing and advertisement. That's what life is all about, to make money and get successful. It's easy. It's happening all over the world-- competition rules every area of life: 2 kids competing in the backyard who can jump higher, who can do better in school, score more in a video game, get a higher score than the next-door buddy, drag on the autobahn to see who's car is more powerful, pick up the gorgeous blonde at the pub, get that degree and that job, succeed in life and live the American Dream. Simple, right?

It's called competition. It happens everywhere, from the age zero and kids up to multi-billionaire companies. That's the way it is.

Like I said dozen of times already Intel doesn't competes only with AMD, neither AMD only competitor is Intel, the picture is just a notch bigger, at least. They are at first competing with themselves too; the technology of computing must advance and evolve. Millions of ingenious engineers are busting their asses of at thousands of labs all over the place. They are all striving for achievements and breakthroughs-- innovations and not only inventions. There is a day and night difference between an invention and innovation -[¹]-. Also, of course, they are competing with each other too. It's perfectly normal.

What people don't tend to realize is that most of the “real truth” that is obtainable over the Internet or in newspapers is not always true, especially in these cases where there is a no way to validate the content. It is either a ‘marketing gimmick’ or just a plain creativity and ingenuity of an author that might qualify for a fiction writer, too. Confidential informations will never be released until and not even after a processor or a certain product has been released. Keep in mind that these technological techniques, methods, solutions and backgrounds are well-kept secrets that won't sneak/leak out just like that.

So called previews of a certain product (processor in our case) if they are official have some bases on truth, though. They might be news on 'steroids' to pump up the expectations and advise people to start people saving up their money already! Think of those "40% increase in performance" the way as you take an advertisement or TV commercial about, say, "PEPSI" as a soda or "ARIEL" as a washing powder. It's just like that. Advertisements are made to "claim" that their product is not only better, is purely the best.

AMD and INTEL's situation is very delicate and sensitive since there are very few official technical numbers/values out there about their architecture that aren't just shameless ads. For example, Penryn's -[²]- 1600MHz FSB, 410mil of transistors, SSE4 with 47 new instructions (huge extensions compared to previous transitions), up to 6MB L2 Cache and over 3GHz of clock speed are not marketing gimmick. These are values, technical specifications that will be maintained and currently there are engineering samples with these specs tried and tested thoroughly in their labs. These have bases on the reality.

But if one goes that far to claim "+40%+ performance increase" without saying at least the good ol' YMMV (Your Mileage May Vary, from the advertisements used for EPA mileage ratings for car) is basically wrong-- marketing gimmick and we shouldn't "fall" for that. We should be enlightened enough to know that performance increase happens only in ideal cases, perfect situations where the bottleneck is low. Performance is relative.

Getting back on topic, both AMD and Intel are doing fairly well. They had, have and will always have issues, that's what technology is all about and computing is not an exception from this rule. Get used to it. Companies will have ups and downs; they will be on the top then fall back on the ladder, then repeat. Bottom-line is, it is not up to the fan-boys to keep them supporting. There is no need to feel empathy for AMD or Intel, they are doing very well and have experienced, knowledgeable employees, executives to keep 'em on the top (symbolically speaking).

My advice therefore is that if one wants to realize the whole picture and understand what's happening with both of these companies but not only, must have a deep knowledge and understanding of computing, microprocessor architectures and research a lot. By a lot, I really mean a lot. No, I don't mean typing in Google: "Intel vs AMD" and reading the threads from a gamer board where gamers are posting how much their FPS increased by changing their processors, neither do I advice reading highly creative articles. The first one are just a waste of time, odds are that those average posters have never even skimmed through a data-sheet or understands what IPC is; yet he claims how much his rig omgwtfpwnz. You won't gain knowledge from them, no offense. The latter ones are alright if you will read them from an objective point of view and realize that what you are reading might not be 100% truth. You will enhance your reading speed, your vocabulary and become a better person. If you are able to read an article and also having the analyst's mindset it is amazing. It will help a lot to realize tons of different opinions and ultimately you will form yours.

On the other hand, if you want to really research you need to dip inside the heavy stuff-- technology journals and datasheets. It will require a lot of time (which I don't have nowadays and I ain't that much up-to-date anymore). Technology journals and datasheets are the golden fountains of real-world knowledge on this area. Understanding these requires at least a basic understanding of microprocessor architecture. It is all about their architecture. That's where the bunnies are hidden, secrets are well-kept (even though, certain 'parts' will be undocumented but it is understandable).

Or, you might find very rarely real articles that are written by experienced authors in the field and provide real quality, value. Check this out for example. Unfortunately, as we discussed already, reviews/articles like these aren't required anymore; thus, the number of them gone down significantly and close to almost nonexistent. Reading a phenomenal article like that will get blown your mind away, fill up the gaps and fit the pieces of puzzle all together. An article like that has tons of research in background, starting from dozens of datasheets, technology journals, eventually interviews too and personal opinions linked with proof and explained in architecture terms. Writing an article like that can't be compared with one that goes like-- let's write an article, I sit down, think for a little bit, hmmm, here's a great topic.

A long time ago very few understood what's wrong with AMD's 65nm architecture problems. It appeared everywhere, virtually on every online computing-related forums and online boards there were thread(s) discussing AMD's "issues." Articles appeared over the Inquirer -[³]- on a regular basis about that topic. It has gone beyond one of the most searched keyword in online search engines (in computing/hardware areas since it certainly didn't beat porn). Whilst, very few people indeed realized at least the "tip" of the iceberg, one of their main problem. It was transistor leakage. Very few people knew the numbers: DC Idsat =700/1137 uA/um for PMOS/NMOS, and AC Idsat =735/1259 uA/um for PMOS/NMOS. All of these on Ioff 200 nA/um at 1V for IBM-Sony-Toshiba-AMD (ISTA alliance) @ 65nm -[]-. On the other hand, values declared by Intel back then were as follows: Idsat = 710/1210 uA/um PMOS/NMOS at Ioff = 100 nA/um -[]-. Intel avoided SOI, they have used "bulk Si" instead, whilst the AMD/IBM used SOI and they did not included SiGe. SiGe left a space for further possible improvements.

If an author says A > B then I need numbers to see why A > B, I want to see a complete logical or mathematical induction, deduction using statements or any other theory to back up his statement. Otherwise, I will respectfully state and classify that specific article as a fiction.

All in all to sum this up, the article that was posted in this thread is alright, the author makes a "good point" about the whole fiasco but I for one, respectfully disagree with its content. I therefore state that I do not believe it, period.

References and future reading:

-[¹]-: How to Tell an Invention from an Innovation and Why Should You Know
-[²]-: Penryn.
-[³]-: Charlie Demerjian- one of TheInq's authors; his articles have bases on truth and is very knowledgeable.
IEDM 2005: Selected Coverage. One of my personal favorites, I consider it as a bible. I just can't explain how phenomenal it really is...
-[]-: W. Lee et. al., “High Performance 65 nm SOI Technology with Enhanced Transistor Strain and Advanced-Low-K BEOL”, Technical Digest of the International Electron Devices Meeting, IEDM 2005, paper 3.3, pp 61-64.
-[]-: M. Horstmann et. al., “Integration and Optimization of Embedded-SiGe, Compressive and Tensile Stressed Liner Films, and Stress Memorization in Advanced SOI CMOS Technologies”, Technical Digest of the International Electron Devices Meeting, IEDM 2005, paper 10.5, pp 243-246.

DISCLAIMER: While I, the author of this post, have made considerable effort to ensure that the information contained in this post is accurate, the author is not responsible for any errors or omissions, or for the results obtained from the use of this information. All information contained in this post is provided "as is", with no guarantee of completeness, accuracy, timeliness or the results obtained from the use of the information, and without warranty of any kind, express or implied, including, but not limited to warranties of performance, merchantability and fitness for a particular purpose or product safety of any kind. In no event will the author, his related partnerships or corporations, or the partners, family members, agents, employees or employers thereof be liable to you or anyone else for any decision made or action taken in reliance on the information contained in this post, or for any indirect, consequential, special or similar damages including, without limitation, permanent loss or corruption of data, physical or emotional sufferings, moral standard corruptions, damages for loss of goodwill or self-esteem, loss of customers or market share, breach of security, permanent or temporary loss of employment, permanent or temporary loss of connectivity to the Internet, work stoppage, loss of income, computer related fires or other mishaps, computer failure or other malfunctions.

You really missed the point there, smart guy. That's okay, you can drop this quiz grade. It won't have any effect on your final grade. But I suggest that you re-read the article and find out what it actually was about.

I'll give you a hint . Its not about factsheets, datasheets, and technical journals at all.

Okay, it can happen quite easily since English is my 3rd language. Your "smart guy" sarcasm I take as a compliment. Thank you very much.

Your hint didn't helped me enough to realize the article's point. All I could figure out deducting from your statement that it has to do something along these following lines:

“In assessing these facts, three ancient axioms come to mind: Nice guys finish last, slow and steady wins the race, and haste makes waste.”

I still don't get how these aforementioned ancient axioms can be applied to computing. The last one might be true. Haste indeed can make a waste. The first two on the other hand are just beyond me. How can you qualify one of the companies as ‘nice guys’ (in this case, I think, Intel) and say that the ‘slow and steady wins the race.’ This isn't quite true. Because if one is ahead of another competitor, then its profit gets higher, total budget increases (due to increased sales). Therefore, will have more money to spend on future projects and investments.

I am sorry but it seems like I don't get the specified article in question. It might contain certain popular myths or be related to American legends or popular beliefs, which I am not aware of. There might be a hidden secret, or something along lines that I can't just find out. Forgive my ignorance. The article might contain symbols or poetic language that is too advanced for me. Or, there being the possibility of me being too stupid for that kind of article. Regardless, please ignore my previous post along with this one and continue this discussion.

To reinforce and emphasize on my aforementioned conclusion: I did not said that the article is crap, all I said was that I personally didn't agreed with its content and I have backed my statement with proof.

Have a nice day.