The Game of Leapfrogging Needs At Least 2 Frogs

Do you recall the numerous times when disillusioned AMD die-hards repetitively say “well if it wasn’t for AMD, there won’t be any competition… ”. With the release of Penryn, Intel's 45nm QX9650 ahead of Phenom whilst adding another level of gap in performance, I really think it's time for the "fanatics" to stop talking such nonsense.

The moment we see Intel sand bagging its “Extreme Edition” to levels just enough to keep a noticeable distance from competition that is the point when we can say that real competition is dead. AMD has stopped pushing Intel to release what’s at the edge of what it can offer and it appears that things won't change much on this front. As AMD tries to squeeze water from a rock with it's soon to be released Spider platform, Intel on the other hand launches a product that beats everything in the market without even trying. The sheer amount of interest in overclocking Yorkfield only points to the untapped capability of Intel's newest 45nm CPUs, ready and waiting when real competition do come around.

Benchmark results on Intel's next generation CPU family is all over the Internet and it continues to look bad for AMD:

These are just sample benchmark results randomly taken from the TechReport.

I think the key message that Intel wants to come across with the Penryn launch is the noticeable improvement in power efficiency. Performance is only secondary since it is obvious by now that Intel has a tight grip on this crown. Launching the new processors at the existing speed grade of 3.0Ghz is only meant to ensure that everyone takes notice of what 45nm brings to the table. The power efficiency improvements are quite impressive and any performance gains achieved while transitioning to an industry leading process node is just icing on the cake.

As if things couldn't get any worse for AMD, There is another Phenom benchmark review using Crysis that shows exactly what it means for AMD. It's indeed a crisis when you're soon to be released processor is already beaten before it even gets out the door. In the last earnings call when AMD said that Phenom will allow them to enter markets where they couldn’t normally participate, by looking at these latest benchmark they would be lying if they meant the performance market.

Ah Yes! That DTX Again

Once hailed by some blogger to be a potential source of revenue, AMD's DTX surfaces yet again to remind everyone that it's still alive. In fact this time around it has a complete set of specifications. I've already given my verdict on the outcome of this AMD endeavour and it appears that theInquirer completely agrees with my opinion.

by Wily Ferret: "The DTX spec appears to bring nothing new to the table, bar some compatibility with ATX which, to us, seems contradictory in purpose if not actual physical reality. Chalk this one up for another veto by the Taiwanese board-making mafia, then."

Like I said in the past, DTX is facing an uphill battle. The fact that the dominant player in the industry isn't supporting the standard already creates enough doubt for mobo makers not to waste time on such a risky platform. The second and also the biggest problem is the fact that SFF's already exist in proprietary formats . Large OEMs already have them in their own unique way and it is that differentiation that generate healthy margins in an already saturated desktop market. An open standard means open competition and clearly DTX only means low margins to OEMs. To illustrate further, this is just like AMD coming in and creating an open standard on Mac PC's. Clearly if everyone can now make Mac's, guess what that does to Apple's business?

As to the future of DTX? The TechReport spells it out even better:
"Nearly every motherboard and chassis maker we contacted for this story had a similar position on DTX. Publicly, they have all announced support for the standard, usually in the form of a single product or project, in conjunction with AMD's efforts. Privately, they are hedging their bets, waiting to see whether DTX gains any momentum in the market before committing to producing anything in volume".

Other than to remind someone that we told him so, I must admit to not really having any substantial reason to write about this topic -- and continue beating this already dead horse. Apologies.

AMD's Q3 2007 Report - How Deep Is The Hole?

In a market that saw Intel make $1.9B in profit, AMD continued to post another $400M loss. But everyting isn't gloomy for the Sunnyvale company. If anything can be said about AMD's luck, it appears that the market seem to be trying its very best to save it from dissolution. Exceptionally higher demand in Q3 helped provide significant gains in AMD's topline. The consistent gross margin increases from AMD and Intel shows that there is indeed a shortage of processors to go around. There seem to be an armistice in the price war at the moment which limits it only on the consumer mobile space. This allowed both companies to significantly increase revenue and ASPs sequentially.

AMD’s Q3 2007 numbers:
Revenue at $1.63B up $300M from last quarter
Operating loss cut in half to $226M from $457M in Q2
Gross margin is up at 41% from 33% last quarter.

Some key notes from the conference call include:

AMD planning to “ramp” 45nm in the 1st half of 2008. This means we could probably see AMD’s first 45nm mid Q3’08 the very least. This contradicts the rumour that came from fabtech that suggests a delay. AMD did mention a 6% increase in CAPEX to support the ramp.

AMD shipped “tens of thousands” of Barcelona CPUs in Q3. If I were to guess exactly how many while basing on how AMD plays around with words while never making understatements, I would say they managed to ship about 20,003 units. This is because you need at least two 10,000’s to be able to at least say “tens of thousands”. The 3 extra units were the ones sent out free to the 3 reviews sites.

AMD plans to ship “hundreds of thousands” of Opteron and Phenom Quad-cores in Q4. If you guessed that to be about 200,003, then you should look into becoming a physicist.

When asked to explain the abysmal Barcelona speeds and poor 65nm yields, AMD said that Barcelona’s problems have nothing to do with manufacturing or their 65nm process. I'm not sure how good their trouble shooting skills are, but either what they said is true or they really haven’t figured out that that is indeed the problem. AMD said that Barcelona’s yield is in-line with 65nm yields (There you go. A clue!). Instead, they suggested that K10 has an issue with "tuning the design to the technology". From an engineering point of view, that doesn’t really sound less frightening than "manufacturing issues".

Overall, AMD did beat the market estimates and is truly heading towards the right direction. Serious. If the trend holds and this market conditions remains, there is a very slim possibility of AMD breaking even (which i personally doubt). But again, the dynamics of a market with healthy demand is considerably different from one where they have to fight tooth and nail for a sale. It remains to be seen how much of the gains AMD made this quarter are bottom line improvements rather than gains acquired by simply riding out the current upside in the market.

Intel's Q3 2007 Report

Intel today announced very strong financial results for the 3rd Qtr of 2007:
• Revenue $10.1 Billion, up 15 Percent Year-over-Year
• Operating Income $2.2 Billion, up 64 Percent Year-over-Year
• Record Microprocessor, Chipset and Flash Unit Shipments
• Net Income $1.9 Billion, up 43% from last year, 46% from last Qtr
• EPS 31 Cents (expected: 30 cents)• Gross margin: 52 percent

Even more impressive is the Q4 2007 guidance:
• Revenue: Between $10.5 billion and $11.1 billion. (expected: $10.4B)
• Gross margin: 57 percent plus or minus a couple of points.

The atmosphere at the conference call sounded like it was 1995 once again. Record revenues, healthy margins, incredible growth, laughter and congratulations were in order. Not a word about the competition as if it wasn't even worth mentioning. The closest thing that would remind you of AMD is the comment made by Intel about how they walked away from some of the low-end businesses and having the unique ability to cherry pick markets due to its strong product portfolio.

The global demand for CPUs was healthy and we can expect AMD benefit similarly. To paint a complete picture of the battle for market share between AMD and Intel, we need to watch AMD's margins closely. Market share gains in servers and higher ASP in desktops for Intel means AMD is under pressure from all 3 platforms especially in mobile where Intel's ASPs were also down. We'll just have to wait and see if it's significant enough to cancel the upside in the market.

The Fundamental Law of Progess: NEW > OLD

We once claimed that major OEMs are ignoring Barcelona and the report today from ZDNET confirms this to be true:
"AMD may have a bigger problem since you can’t even order these systems from any of the major server makers and none of them can tell you when their Barcelona servers will be available."

None of the major OEMs seemed to be rushing out with their new K10 systems. The fact that the Big 5 cannot commit on server availability can mean several things mostly which can only be bad for AMD. Normally when you have a ground breaking product every OEM wants to be the first to market. But when AMD's new product violates the fundamental law of progress that demands all NEW PRODUCTS MUST BE BETTER THAN OLD PRODUCTS, it’s no surprise many are willing to wait, if not too afraid to try.

It is increasingly alarming how AMD seemed to be moving backwards. We've already given up hope that AMD would once again beat the competition (Intel, Nvidia) but at the very least we expect them to beat their older generation products. The introduction of slower 65nm CPUs and now K10's inferior performance to K8 are just some of the bad habits AMD seemed to be developing.

The article also tackles the question about Tigerton’s availability. The result of George's investigation showed that anyone can order Intel’s new server MP but with a product shipment of within a month or so. Whereas compared to Barcelona, orders are not allowed without even a promise when one can order. In the business world there is a big difference between an order taken and an order declined. Guess which one involves making money.

Phenom Problem

The good news is that Phenom is confirmed to launch next month (probably Nov 30 2007). The bad news is 2.4Ghz is the best AMD can come up with while it gets beaten to market by Intel's 45nm SKUs. Setting it up against Intel's QX6850 at 3Ghz, anyone would think it's a bit inhumane and sadistic to run a side by side comparison when AMD is in a big and embarrasing disadvantage.

2.6GHz is expected to ship on the last working day of this year (Dec 2007) while anything higher including the FX chips will arrive sometime in spring. In a manner typical of AMD, its Desktop parts appears to be delayed and expected to be underwhelming even before launch. We can only expect another round of horrible benchmark comparisons for AMD, in most cases left trying to compete with its own Athlon FX CPUs, never mind trying to beat Intel's Core2. Phenom's poor showing should stifle the hope anyone may yet have, thinking that K10 will retake any kind of meaningful leadership.

Based on our experience with the Barcelona launch and the relatively low speed of Phenom, we can expect this to be another paper launch, only meant to save face due to competitive reasons. If you're one of those loyal enough to buy inferior products to support the underdog, don't expect to get hold of these until next year.

Phenom 9 Series
9500 - 2.2Ghz - 89W - Nov 2007
9600 - 2.4Ghz - 89W - Nov 2007
9700 - 2.6Ghz - 125W - Dec 2007
9xxx - >2.6Ghz - Q2 2008 <- performance wall?

Phenom FX
FX82 - 2.6Ghz - Q1 2008
FX8X - Q2 2008

AMD Finally Admits K10's Inferiority

AMD submitted scores to SPEC.org that finally puts an end to the debate. AMD’s K10 is inferior to Intel’s Core2 microarchitechture. At it’s current state Barcelona cannot even compare with its own higher clocked K8. Even if K10's clock speed is successfully improved to 3GHz, it still wouldn't be enough to overtake Intel.

We've warned on this blog how K10 will be dead on arrival and discussed how in its current broken state creates a conundrum of competing with its own product line at the low end of the performance scale. It seems that a lot of you who agreed are correct. The scores are worrying for AMD.

SPECint2006 (base/peak) - 2P system
AMD 1.9GHz - 9.97/11.3 (K10)
AMD 3.2GHz - 14.1/15.2 (K8)
Intel 2GHz - 14.2/15.6
Intel 3GHz - 18.9/20.8

SPECfp2006 (base/peak) - 2P system
AMD 1.9GHz - 10.7/11.2 (K10)
AMD 3.2GHz - 14.2/14.5 (K8)
Intel 2GHz - 14.5/16.9
Intel 3GHz - 18.4/21.4

You can make quite a lot extrapolation and analysis from the scores but you will ultimately arrive at the same conclusion. That the capability of AMD's new processor falls short and flat on its face. They may be single threaded benchmarks but they highlight the core potential of the design in terms of per-core unit processing ability. When Intel's 45nm processors arrive, Penryn next month and Nehalem less than a year from now, we can only expect the gap to increase.

Thanks to axel for raising the awareness.
RTW discussion here.

RIAA Wins Landmark Case

From Ars Technica:

Side stepping a bit from the usual subject and focusing instead on the landmark case Capitol Records v. Jammie Thomas. I certainly don't want to inflate the implications of this case although it is precedent setting and affects the future behaviour of the RIAA against the 'pirates'. The rest of the world seemed to be polarised on this issue as if it is black and white, Record labels versus music fans, them against us. This isn't always the case.

In fact there is something unique about this legal battle that may not necessarily be applicable to most piracy cases. The evidence against Jammie Thomas is overwhelming. The problem was that she's quite tech savvy enough to work herself around the Internet but at the same time tech stupid to have used the same username wherever she went. Even when using Kazaa! (In fact using Kazaa speaks volume on what kind of Internet user she is). The question of willful file sharing is pretty much evident in this case. The big problem for the RIAA is that everyone is not as blatant as Ms Thomas. Gathering evidence that corroborates wouldn't be as easy.

For the RIAA, its always the case of bad publicity whenever it comes out as sues its customers. In fact I blame the RIAA and the MPAA for making everyone like Jammie Thomas aware of the possibility of getting music or movies online for free. They're the biggest piracy advertising movement on this planet locking in anti-piracy ads on everything they touch showing people that it can be done.

The record labels are corporations made up of people like you and me. The tactics they're employing annoys the hell out of everyone including music fans. We need to send the message to the record labels that we own MP3 players not CD players. We stopped buying CD's 5 years ago and they have yet to wake up and align their business to our needs. I don't care if you get your music illegally or not, but I do believe that blatant actions by the defendant in this particular case is quite over the top. Again, it's not always black and white.

The Story Behind Intel's Breakthrough Transistor

Here's an interesting article about the development of the halfnium based high-k plus metal gate transistor that will debut next month on Intel's 45nm processors. Written by those deeply involved with the research, it shows a glimpse of how Intel's R&D diligently work on finding solutions to problems that are predicted to arrive only in a decade or so. The research for the replacement of the SiO2 gate oxide started ten years ago and was completed in 2004. The decision to implement the technology for 45nm in 2005 is a testament to Intel's lead with the rest of the world. Leading the 45nm transition is a feat by itself. But doing so with a key technology advancement makes it even more frightening.

The High-K Solution

The article may also leave you with a conclusion that without this new high-k technology, AMD's 45nm processors will be disadvantaged in terms of power efficiency and performance customisation. As Nehalem arrives at a time when 45nm matures, a strong microarchitechture standing on top of a highly advanced transistor technology is probably the last thing AMD needs.

"The invention of high-k plus metal gate transistors was an important breakthrough. Although we could have continued to shrink transistors to fit the dimensions needed for the 45-nm generation without this breakthrough, those transistors would not have worked much better than their predecessors, and they certainly would have expended more watts. We're confident this new transistor can be scaled further, and development is already well under way on our next-generation 32-nm transistors using an improved version of high-k plus metal gate technology". - MARK T. BOHR

Intel's WiMAX Push

Intel demonstrated during its recent Developer Forum a WiMAX enabled laptop. The laptop based on the Montevina platform which is scheduled to be released next year contains within it "Echo Peak", a WiFi and a WiMAX compliant chipset. These next generation Centrino notebooks will enable seamless Internet connection - assuming of course that there is a WiMAX Internet service provider close by. The problem at this point in time there's not so many of them around.

Being the founding member of the WiMAX Forum, Intel is trying to accelerate implementation of this open, IP based standard by driving end user availability by integrating WiMax with Centrino. 2007 is coming to a close and while the list of spectrum licenses and infrastructure deployment are growing, still it may not be enough to guarantee universal adoption. The struggles and the financial burden inflicted by 3G are all too fresh to make anyone less skeptical with new communication standards. Especially ones that promises the all-too-familiar "increased bandwidth for media rich content".

None the less, Intel is forging ahead and will do its part to ensure that end-to-end implementation goes according to plan (i.e., Nokia). 2008 should be the year the WiMAX user base starts to trickle, increase and hopefully create enough demand for telecoms to take notice. Intel understands that it won't be its Network & Communications Group that will ultimately benefit from the initial SoC sales (comms silicon on an open standard = low gross margins). Instead it believes that an anywhere, anytime Internet connectivity will ignite more demand for mobile computing. Intel's roadmap for mobile and UMPC products clearly align with its WiMAX strategy and along with it its predicted market growth - a clear emphasis on its pro-active business approach.

In contrast, if anyone wishes to know what AMD is doing to promote ubiquitous Internet connection, they created this brochure to make perfectly clear what their products are. So far all they have is this brochure! Naturally AMD will be benefiting from the increased mobile demand, but Intel once again will be associating Centrino with WiMAX in the coming years like what it did with WiFi. Intel's marketing approach appeals to end user needs and again this will be very successful. AMD's brochure talks about AMD Connect (TM) but I need someone to explain to me what exactly is AMD selling.

Sharikou's Journal Loses the Argument

This doesn't mean Sharikou had a case for argument in the beginning. Anyone claiming Intel's bankruptcy any time soon (Q2'08) or AMD's 30% market share "run-rate" only attracts ridicule followed by the immediate loss of credibility.

After the controversial existence in the blogosphere we can finally conclude once and for all that Sharikou has lost the argument. Godwin's law states that whoever mentions Nazi's or Hitler in a discussion loses the debate.

Sharikou said: "I once concluded that Intel's exclusionism has its roots in NAZIsm."

End of discussion.