A shy Intel

Intel has no problem pumping its chips to gullible journalists and script kiddies. But Tom Yager at infoworld is wondering why Intel pulled a no-show at the Spring Processor Forum, which chip designers use to demo their know-how. Actually, Intel is playing hide-and-seek with script kiddies too, the latter are just too overjoyed by the attention Intel giveth them.

Chinese WAPI standard body accuses Intel of unethical behaviour

In a 38 page complaint, China's WAPI consortium accuses Intel and IEEE of unethical, unfair, irresponsible and unacceptable behaviour in rigging the ISO voting for the China proposed WAPI wireless standard.

In other news, China's telco equipment giant Huawei released a Telecommunications Standard Service Platform based on AMD Opteron technology.

Intel readying global benchmark fraud

The fraud continues, June 4th will be the date.

AMD to expand to 45000 wspm

AMD announced plans to refit FAB30 with 300mm equipement, and expand total 300mm production to 45000wspm. This is capacity for producing 75 million dual core dies per quarter. By converting FAB30 instead of building a new one, AMD saves about 18 months of time.

AMD generated 566 million cash from operations in 1Q06. It should have no problem funding the conversion. (Bulldozer)

Intel finally making some moves as I proposed

Back in 2005, I made the following suggestions to Intel, which I sent to Intel execs via email:

It seems that the only way out is for INTEL to drop the most favoured treatment for DELL and increase prices cross the board, and make prices the same for everyone, making it possible for every one (instead of just DELL) to profit in the INTEL space and create a more balanced INTEL market.

My suggestions to INTEL:

0) Face the reality that AMD will take 25% share.

1) Abolish the volume based or AMD quota based rebate system.

2) Establish an uniform pricing scheme, set the same equal global price for everyone, just like McDonalds sell cheeseburgers at the same price. Get rid of all discounts. Such as an uniform pricing system will make small whitebox players competitive against big guys like DELL and foster a broader market presence.

3) Reduce the production of low end chips, this will lead to higher ASP.

I was trying to find a way for Intel to survive the AMD onslaught. Later, I further explained why Intel should cut rebates to Dell. I projected that Dell had to go AMD by the end of 2Q06.

Now, it's reported that Intel is finally doing 1), 2) above. However, Intel failed to follow my advice to reduce production in anticipation of AMD's capacity increase and market share gain. As a result, Intel's inventory piled up to the neck. This has led to massive oversupply of Intel's legacy chips and Intel's pricing collapse. Intel is now trying to cut prices by up to 60% to gain back 5% of unit share, which is kind of wishful thinking. Intel probably thought that they could halt sales of AMD64 by hyping Conroe. Of course, that was naive. People still need computers, so they go AMD64, because Intel told them Pentium 4 is indeed junk. Intel's slashing of the P4 price further validates the notion that P4 is junk. The continuing steep price reduction will only cause buyers unwilling to pay whatever reduced prices today, in expectation of further price drop as Conroe launch quickly approaching. As customers wait, Intel sees demand drop; to stimulate demand, Intel cuts prices even more, causing more customers to wait for further price cut. The cycle continues. The pricing collapse also causes Intel to lose credibility and customer confidence. Those who bought Intel earlier definitely regret. This may even push more customers to AMD, as they feel AMD's prices are more stable and fairer. As I observed here, people are not buying Intel's Pentium Ds, even though their prices have already been cut to rock bottom levels.

I projected operating losses for Intel from 3Q06 onward (2Q06 GAAP loss is also pretty much in the bag due to impairment of goodwill). As the launch of Conroe will cause total collapse of pricing on 80% of Intel's chips. Intel may be able to ramp Conroe to 40% of its volume in 1Q07. However, by then, AMD's superior 65nm CPUs will start to flood the market, Conroe will again become obsolete.

Intel has made the following irreversible errors that ruined its only chance of survival:

1) Intel hyped Conroe too much and too early
2) Intel failed to reduce production to hike price and keep revenue and profitability constant
3) Intel triggered a cycle of domino effect type pricing collapse of its CPUs
4) Intel has squandered its cash, leaving it out of ammo in the death struggle against AMD

Inventory write-downs, pricing-collapses, goodwill impairments, cash burn will cause a vicious cycle of spiraling down. Fighting capacity and morale will get lower and lower...

The story has a sad end for Intel.

More on Intel's Woodcrest performance claim fraud

Examining Intel's Woodcrest performance claims on TPC-C, Floating point, Integer, Java, Web, HPC and application

Today, former Enron execs were found guilty on charges of fraud, false statements and conspiracy. Let's revisit Intel's Woodcrest performance claims. I pointed out that Intel's changing of the Opteron TPC-C benchmark description from 32 bit x86 to 64 bit x64 was a fraud.

Some of the readers said that Intel simply picked up the highest reported TPC-C results for two way servers, Woodcrest and Opteron, regardless of the operating system used. Let's test this assumption on other benchmarks. Let's look at floating point performance.

For SpecFP_rate_2000, the highest reported score for 2P 2.6GHZ Opteron 285 was 85 under Solaris 10. Guess what? Intel ignored this result, instead, it uses a lower Opteron result for Linux with a score of 72.9. The 3GHZ Woodcrest scored 83 under Linux. The 3GHZ Woodcrest (Linux) was 3% slower than 2.6GHZ Opteron (Solaris). Also, notice that Intel chose the SPECfp_rate_base2000 scores for comparison. The SPECfp_rate_base2000 is for conservative optmization of the benchmark, so it's always lower than the SPECFp_rate_2000 score. For some strange reason, the DELL 2950 Woodcrest server's optimized SPECfp_rate_2000 score was the same as the conservative SPECfp_rate_base2000 score, which may indicate that there were some issues with how the benchmark was done. Any way, Intel was shopping for the lowest Opteron scores. This clearly shows that Intel knew different configurations lead to different results. Had Intel chosen the highest score regardless of OS, the 2.6GHZ Opteron would outperform 3GHZ Woodcrest in SpecFP_rate_2000.

According to this report, the 3GHZ Woodcrest (Xeon 5160) will be the topmost chip, the next will be the 2.66GHZ Xeon 5150. Intel's topmost desktop chip the Conroe XE will be at 2.93GHZ. This indicates that a 3GHZ Woodcrest will be a cherry-picked chip. According to this page, the 2.8GHZ Opteron x90 has been in the wild for quite a while now.

For SpecInt_rate_2000, Intel again changed the OS description of the systems. The Woodcrest benchmark was done in 64 bit. The Opteron benchmark was done in 32 bit. This time, Intel changed the Dell PowerEdge 2950's benchmark description from "Microsoft Windows Server 2003 Enterprise x64 Edition" to just "Microsoft Windows Server 2003", making it look like the same as the Opteron test. This is just like Intel's Woodcrest TPC-C performance claim fraud.

Let's look at another example: Intel's page on Java performance. Intel used an unpublished Woodcrest test result on a Fujitsu Siemens PRIMERGY server running Windows Server x64 with BEA JRockit 5.0 P26.4.0 JVM. But for Opteron, Intel decided to use the score from a Tyan S2895 server with two 2.6GHZ Opteron and a SATA drive, the score was only 54490. However, from www.spec.org, we can find a Fujitsu Siemens PRIMERGY server with two 2.4GHZ Opteron 280 (running Linux, JRockit 5.0 P26.0.0) scoring 61155. Again, Intel was shopping for the lowest Opteron scores.

Let's look at yet another example: Intel's page on web performance. An IBM 3GHZ Woodcrest server got a SpecWeb2005 score of 9182. Mysteriously, there is no Opteron scores on this Intel page. However, going to www.spec.org, we quickly found this 2.4GHZ Opteron 280 server achieving a score of 8394. The 3GHZ Woodcrest has a 25% clockspeed advantage but only 9% performance lead over the 2.4GHZ Opteron.

Let's look at one more example: Intel's page on application performance. For the SunGard ACR test, Intel sent two servers to a company called Principled Technologies. One was an Intel built Opteron server and one was a Woodcrest server. Not surprisingly, the Woodcrest won the benchmark. The details of the benchmark is in this PDF file. The motherboard Intel chose for the Opteron was an UNIWIDE SS232_128_03 model using Nvidia NF4 chipset. One has to ask why Intel built the Opteron server themselves instead of using a proven server such as SUN's X4200 or HP DL385. We know server performance does vary from system to system. Not only Intel built and configured the Opteron server, it also provided the Intel compiled test application "SunGard ACR Intel Demo 2.5". It is unclear how Intel optimized this test application, but in a previous report (later removed), it was reported SunGard ACR is significantly faster for Xeon when compiled with Intel C++ compiler.

The more we examine Intel's presentations, the more problems we find. Looking at Intel's HPC performance page, pay attention to the fluid dynamics results (Fluent). Intel used a Woodcrest 3GHZ (2530.44) against an IBM 2.2GHZ LS20 Opteron blade (2014.34) , with the Woodcrest having 36.4% clockspeed advantage and 26% performance lead. However, if you go to the Fluent full results page, you can see there are quite a few Opteron results better than the 2.2GHZ IBM LS20 Opteron blade. In fact, there is a 2.6GHZ IBM LS20 Opteron blade scoring 2404.72. Using this result for 2.6GHZ Opteron, the 3GHZ Woodcrest would have only 5% performance advantage, despite 15% clockspeed advantage. Actually, both results show that Woodcrest being 10% slower than Opteron clock for clock, in agreement with our previous analysis. One can imagine Intel tabulated the Fluent benchmark results, and decided to use AMD's entry level 2.2GHZ Opteron 275 for comparison against the topmost Woodcrest 3GHZ (Xeon 5160). On the same HPC performance page, for "Finite Element Analysis for Crash Simulation", Intel also picked a low score for Opteron, despite existence of better Opteron results (see user comments).

So, why did Intel change the Opteron TPC-C description from x86 (32 bit) to x64 (b4 bit)? Why did Intel consistently choose the lower Opteron scores for comparison?

The answer is obvious, to create a false impression that the Intel CPU is much better.

Fraud: Any act, expression, omission, or concealment calculated to deceive another to his or her disadvantage. (Merriam-Webster’s Dictionary of Law, 1996).

Intel's behaviour satisfies the above legal defintion 100%.

Intel to slash Pentium D prices by another 60%

Intel schedules to slash prices by another 60%. Notably, the Pentium D 820 dual core CPU will cost $93. Right now, this chip is selling at $175 and very few people want it. The expected price cut will further reduce sales of these Intel chips as people are unwilling to pay for something that will depreciate 60% in a couple of months.

DELL will have a hard time keeping its revenue from falling due to the collapse of Intel's pricing. DELL has already lost its edge over its competitors, as everyone can now get Intel dirt cheap.

As I pointed out here, since AMD's ASP is about $95, Intel needs to reduce its ASP to $80 to win back 1% unit share. But I also pointed out that INTEL will suffer a ($0.40) EPS loss as a result.

I now project 2Q06 GAAP loss for INTEL due to charges associated with impairment on goodwill. I project operating losses for Intel's 3Q06, 4Q06, 1Q07 and beyond.

Socket AM2 gaming performance gets 19% boost

LegitReviews found that by upgrading to 4GB PC2-6400 memory with CL3 timings, socket AM2's DOOM3 performance increases by 18.7%.

firingsquad.com found that increasing L2 cache size by 512KB led to 5% gaming performance increase.

This test includes Athlon 64 AM2 performance under Windows x64, very impressive.

Opteron has 48.1% of the multi-processor server market in US

Gartner reports that AMD has taken nearly half of the multi-processor server market in US. AMD's share in overall x86 server space is 26%. Opteron server revenue topped $1 billion. If you take out those $299 Celeron and Pentium D "servers" out of the stats, AMD's share will be substantially bigger.

SUN's server revenue surged 8% year/year. SUN gained across the board, from x64 to big-iron. IBM, DELL and Fujitsu's server revenue dropped 4%, 2%, 14%, respectively, while HP's server revenue was flat.

Intel Woodcrest performance claim a fraud

Update: Please check this page on closer examinations of Intel's claims on FP, Web, Java, HPC and Application performance.

Intel setup a web page showing TPC-C result of Woodcrest comparing to that of an Opteron. The 3GHZ Woodcrest server with 64GB FB-DIMM memory was running Microsoft Windows Server 2003 Enterprise x64 Edition, Microsoft SQL Server 2005 Enterprise x64 Edition. It achieved a score of 169,360 tpmC, $2.93/tpmC. The Woodcrest machine will be available by November 22, 2006.

In comparison, Intel gave the following details of an AMD system:
"Dual-Core AMD Opteron* Processor Model 285 based platform details: HP Proliant DL385 G1* server platform with two Dual-Core AMD Opteron* processor 2.60GHz, 32GB memory, Microsoft Windows Server* 2003 Enterprise x64 Edition, Microsoft SQL Server* 2005 Enterprise x64 Edition. Referenced as published at 113,628 tpmC; $2.99/tpmC; Availability Date as listed in the submitted report is May 5, 2006. Results at http://www.tpc.org/tpcc/results/tpcc_result_detail.asp?id=106032001"

Intel's message to potential customers is clear: Woodcrest is 50% faster than Opteron in database performance. Remember, database is about reading from/writing to disks.

If you go to the executive summary of the HP DL385 at the TPC-C site, you can see that the Opteron machine was running Windows Server 2003 Enterprise x86 Edition SP1 and Microsoft SQL Server 2005 Enterprise (x86) Edition SP1. These are 32 bit versions of Windows Server 2003 and SQL server 2005. If you have any doubt on the OS and SQL server used, read the full disclosure of the DL385 TPC-C results. If it was a 64 bit OS, you would see information on Windows\SysWoW64\, but on the HP DL385 page, you only find Windows\System32\. So, the HP DL385 was running in very inefficient 32 bit PAE mode.

Why did Intel change x86 to x64 for the HP DL385 benchmark description? Let's first rule out the possibility that Intel did this accidentally. 86 and 64 are too distinctive from each other to be confused with one another.

If Intel told you that the Opteron result was actually from 32 bit, you would ask immediately: wait a minute, how about running Opteron in 64 bit and also with 64GB ram (instead of 32 GB PC2700 RAM)?

We know that Opteron performs 20-40% better in 64 bit mode -- which will immediately wipe out Intel's claimed performance lead. So, cunny Intel changed 86 to 64.

By changing x86 to x64 for the Opteron tests, Intel was trying to create an impression that it has a big lead over AMD under similar conditions. This is similar to Intel's previous acts of putting a faster harddrive in its own server and claiming its CPU is faster.

In fact, not only the Woodcrest was running 64 bit OS, it costs $158,000 or 47% more than the Opteron system. The Woodcrest system has 64GB RAM and 48 MSA 60 StorageWorks with 576 SAS drives at 15K RPM. The Opteron system has 32GB ram and 18 MSA 30 StorageWorks wtih 380 SCISI drives. There is a huge difference in storage performance between the two. TPC-C is a database benchmark. A database reads data from disk to memory and write data from memory to disk.

Independent results show Intel's Woodcrest 10% slower than Opteron at the same clock under 64 bit mode.

Fraud: Any act, expression, omission, or concealment calculated to deceive another to his or her disadvantage. (Merriam-Webster’s Dictionary of Law, 1996).

Intel satisifies the above legal definition 100%. More on this (SpecFP_rate, SpecWeb2005 ...).

Charlie perplexed

Charlie at INQ was deeply puzzled by Intel's PR, but he was wrong to call Intel dumb. Intel was actually trying to be smart. The game is called bait and switch.

You see. First, Intel pumps up the Xeon 51XX with numbers. Hooked so far? Thinking about spending a grand? Now, at the end of the PR, there is a Xeon 50xx price list. Not $2000, not $1000, not $500, not $200, but $177, yes, $177. "Dual-Core Intel Xeon 5050 3GHz, 2x2MB L2, $177". "Wow!", you say, so good yet so cheap, reaching for your wallet.

Remember, Intel still has 103K people to feed and it needs to sell some chips today instead of 4 weeks later.

A cheap solution to frag Conroe in single threaded 32 bit apps

Intel's Conroe/Woodcrest/Clovertown doesn't show any IPC advantage over Athlon 64 under 64bit mode. However, Conroe's 4MB cache is definitely making a difference in 32 bit single threaded loads.
AMD can't do 4x2MB cache, too expensive.

One possible solution is this
1) dual core, independent clocks like the Turion X2
2) one core has 256KB L2, the other has 6MB L2.
3) clock the smaller core at 1.8GHZ (about 15 watts).
4) clock the core with larger cache at 4GHZ (75 watts)

I will call this AMM (Asymmetric Multi-core Marchitecture).

This is a cheap solution that can frag Intel to death in most gaming and single threaded 32 bit loads. In 2007, the problem will go away as Windows Vista enters the scene. Under 64 bit mode, K8 is about 10% faster than Conroe at the same clock.

3GHZ Woodcrest against 2.6GHZ Opteron performance duel

Under the watchful eyes of Intel employees, TechReport did some benchmarks of the Woodcrest. The results show that Woodcrest 3GHZ is only slightly faster than 2.6GHZ Opteron 285 in most tests. In multi-threaded applications, Woodcrest is often only 5% faster, despite having a 15% clockspeed advantage. Please note that the tests weren't even server benchmarks where bandwidth is crucial, they were Windows desktop applications.

My previous analysis reached similar conclusions. As you can see from TechReport's POV Ray benchmarks, the Woodcrest/Conroe has quite some advantage in single threaded loads, where one core can use the whole 4MB shared cache. However, once you start 2 threads for 2 cores - in other words, each core can only use 2MB of cache, Woodcrest's performance lead reduces to 6%. Since the Woodcrest engineering sample has a 15% clockspeed advantage, this indicates that the CORE2 is 10% slower than K8 at the same clock. This is in agreement with my previous analysis, that CONROE/Woodcrest/Clovertown only has an advantage for single threaded loads that have a working set comparable to 4MB.

TechReport also reports that the Woodcrest server platform isn't entirely stable yet. Woodcrest suffers BSOD (Blue Screen Of Death) under certain conditions. Woodcrest is set to launch on June 19. Intel better hurry.

Realtech.com also performed a set of benchmarks on the 3GHZ Woodcrest, the results on Cinebench 9.5 and POV Ray were in agreement with TechReport's. Realtech.com obtained a SPECjbb2005 result of 56,835 BOP/s. In comparison, a 2.4GHZ Opteron obtained a score of 62074 BOP/s (under Linux).

In March, answering the question on whether DELL would go AMD, DELL CEO Kevin Rollins said:

"There are multiple decision points. Is Intel going to meet the technology needs-–server performance and thermals–-where AMD does have a lead? That will answer the question. If they don't, that will also answer the question."

Intel is kinda weird. Today, it launched the Dempsey server CPU, just 3 weeks ahead of the Woodcrest. The 3GHZ Xeon 5050 dual core is only $177. Dempsey is dirt cheap, actually, cheaper than dirt. I admit that even I am tempted by this part. But once you look at the total system cost, it is actually quite expensive. The motherboard alone costs about $800. Adding the electricity bills, the equation changes completely.

AMD won Israeli government PC tender

According to this report, "despite Intel's long presence and large investment in Israel, the tender will not include Intel processors, instead opting for those by AMD." It seems to me that the Israelis speced 64 bit and Intel was automatically disqualified.

My guess was that AMD did postpone Turion 64 X2 launch to May 17 in anticipation of Dell's May 18th announcement. But our Mickey chickened out at the last minute, unsure how the Street would react.

Osborne effect on Intel

Newegg.com is a very popular site for computer parts, fast delivery, fair prices. The best part of the site is user comments. Look at this Pentium D 820 2.8GHZ dual core, the list price is only $174.99. Cheap. But, not many people are buying it, only 100 comments. The last comment was left on May 14, 2006, nine days ago. The message was "Cons: ummm wish it was a Conroe?? ;)"

The Athlon X2 3800+ is selling at $297. Three comments were added today.

The top three CPU sellers at newegg.com are:
AMD Athlon 64 3200+ 2000MHz HT Socket 939 Processor - Retail $135.00
AMD Athlon 64 X2 3800+ 2000MHz HT Socket 939 Dual Core Processor - Retail $297.00
AMD Athlon 64 3000+ 2000MHz HT Socket 939 Processor - Retail $106.00

AM2 party time. Frags readied for CONroe

DELL (DBA Alienware), HP, Lenovo, Microsoft, VIA, MSI, ASUS, NVIDIA, ATI, ECS, FUJITSU SIEMENS, GIGABYTE, KINGSTON, CORSAIR, SiS... all the big names you can find, are in for one big party for AM2 launch. The FX60 was 20-40% faster than Intel's fastest chip, the performance gap just got bigger.

One June 1st, AMD will demo some next generation futuristic technologies that will frag Intel amateurs like hell. In July, Intel is going to launch the Conroe. My suggestion to AMD is this: at the time of Conroe launch, release a quadcore AM2 chip built on 65nm with the K8L core. It doesn't matter if 65nm yield is only 10%. Just produce 5000 such chips. Send 100 to review sites, make the 4900 forsale at $3500 each. The goal is simple, show what AMD's true quadcore is capable of and push Intel further down the performance curve. This will create a halo effect on the whole AM2 product line.

AMD-DELL alliance signifies a dyansty change

Back in January, I predicted that DELL would spit at INTEL and kneel to AMD by the end of 2Q06. Since then, more changes in the computing landscape led to this inevitablity. Both Intel and Dell have become financially fragile. For DELL, the situation is getting more alarmingly precarious.

As of May 3, DELL's net assets totalled $3.37 billion, or a bookvalue of $1.43/share. This means if one liquidates DELL today, all he can get back is less than $1.43 per share. DELL's lofty 17X price/book valuation can only be observed with small bio-tech companies. Unless DELL sustains a level of hyper-growth, its market cap will suffer a total crash. We are talking about 2-4x bookvalue, or $3-$5 share price here. As we have seen from the case of Enron, crash of stock price may lead to loss of borrowing power and eventually bankruptcy.

Without the AMD annoucement, DELL's stock could have collapased on May 18, 2006, in the wake of DELL's 18% year/year profit fall and HP's strength in a growing PC market. Because of the AMD announcement, most of the attention were diverted to the AMD alliance. But a mere AMD announcement doesn't mean DELL's problems have gone away. In fact, DELL will face even greater danger in the days ahead, as investors try to identify real growth prospects in DELL's vague AMD plans in an effort to justify DELL's valuations.

DELL is a scared kid. It's scared because its stock is already nose-diving and it knew that it stuck with Intel for too long. It shedded tears only because it saw its coffin. It was even more scared of going AMD. It didn't know how the Wall Street would react: would the Street dump DELL because of the fear of losing Intel's favourite treatment, or would the Street greet its decision to go AMD as another growth oppurtunity?

That was why DELL choose to limit its commitment to AMD in its earnings release statement. It was trying to test the Street's reaction. And the reaction was quite positive. Instead of crashing below $20, DELL's stock went up a few percent. In fact, the Street expects that four way servers are just the beginning, DELL is going to do more AMD and get more growth.

The math is simple. AMD has 22% of the x86 market. Going AMD could potentially lead to 20% growth for DELL. DELL should look at the AMD market this way: it's like a nation bigger than China and India combined. DELL keeps bragging about its growth in China, but its sales there is very small. DELL only sold 1.5 million units in China last year (2005). In comparison, the AMD market size is 50 million units/year. There is a lot of growth oppurtunities in the AMD market. That's the kind of growth the Street expects to keep DELL valued at $55 billion.

Intel is a sinking ship. Conroe/Woodcrest will not change Intel's fate of doom even if they perform faster than current K8. They are too few, too late and too archaic. They will only bleed Intel more on majority of Intel's units -- P4 and Core Duo. When Conroe eventually ramp up and Woodcrest finally gets validated, AMD will leap further ahead. We expect AMD to demonstrate major innovations in its June 1 technology conference.

Even if DELL enjoys a preferential pricing from Intel, such favoured status has little residual value at this moment. Intel's pricing has already collapsed. Getting 20% more discount on something dirt cheap is not big differentiation. For DELL to sustain the expected revenue growth, it must adopt AMD platforms across the board, it must grab its share of the AMD pie. Failure to adopt AMD in a wider market will lead to an additional Osborne effect on DELL's products: in addition to the Conroe Osborne effect, DELL's other products would also suffer from AMD64 Osborne effect. On the other hand, DELL going AMD across the board will inspire investor confidence in DELL's long term growth prospects and even forgiveness of DELL's short term misfortunes.

Intel has lost the technonology battle, it's also morally bankrupt. Those guerilla benchmarketing ploys totally lack intellectual integrity and are bordered on fraud. It's no longer cool to boast about products based on Intel's outdated FSB technology. It's foolish to defend Intel primitive and outdated FSB technology against AMD64. Increasinly, Intel is trying to use new branding and bundling to eat the lunch and dinner of all others in the PC food chain.

Intel is self-inflicting massive wounds and bleeding in a feeble attempt to slow down the advance of AMD64. DELL will inevitably suffer severely from Intel's pre-death struggle in the next three quarters and beyond.

AMD64 is the only thing that can save DELL from total collapse. A Pax AMD gurantees a profitable future for the PC industry to all of its participants. Graphics, chipset, networking, wireless, server, notebook and desktop vendors can all benefit.

I hope DELL's Opteron move will put more pressure on SUN to deliver more Galaxy designs. SUN should produce something like HP DL385, a 2U small business workhorse with 2 CPUs and 6 hard drives. SUN should also take advantage of its Solaris 10 OS to create turn-key integrated web and storage solutions for the mass market.

For IBM, my suggestion is to recoup the Hurricane chipset development cost by selling it to Intel and move to Opteron.

Intel 80386 a victim of downsizing, Chartered FAB7 cranks for AMD

Intel plans to cease production of 80486 after 28 September 2007. This must be part of Paul O's restructuring plan, some jobs will be cut or relocated for sure.

In other news, Chartered FAB7 has started AMD64 production in May at 1000wspm. Shipment is scheduled for July. A 1000wspm 300mm rate translates to about 1 million dual core Athlon 64/quarter.

Woodcrest will suffer the fate of Athlon MP

Athlon MP, introduced in 2001, had an architecture similar to that of Woodcrest/Bensley. Just like Intel's 2006 design, the 2001 Athlon MP had dual independent bus, which allowed 2P operation but no more than that. Although its performance was quite good at 2P level compared to the Xeon DP at the time, Athlon MP failed to generate much interest. AMD was not considered a worthy server CPU vendor, because it didn't have a four way solution.

Today, Intel is in a very similar situation. AMD scales up to 8P 16 core, while Intel is stuck at the cheapo 2P level.

Originally, Intel planned to follow up the Sossman 32 bit CPU derived from Core Duo with Dempsey/Bensley in Q2, then Woodcrest in Q3. Intel released Sossaman32 in Q1, but it was ignored by major Intel server vendors, including HP and DELL. Intel prepared to launch Dempsey in Q2, but it has apparently been cancelled to lack of performance and performance/watt improvement. As a result, the whole Intel server line is being shattered right now, as the hot, slow and antiquated Xeon Paxville has only 50% of the Opteron's performance.

Intel's plan is then to move Woodcrest/Bensley launch ahead to June 19, 2006. Due to Intel's manufacturing inefficiencies, I can safely bet these CPUs will be delivered from Intel's development FABs. In March, Intel showed Conroe and Woodcrest samples, these were from Intel's development FABs. With Intel's so called copy-exact methodology, it would take a couple of months to duplicate the process at its production FABs. This is why we are seeing a slow Woodcrest ramp (75% by end of of 2006), despite the fact that server CPU volume is only a couple million units per quarter.

So, Intel is brushing its Sossaman32, Dempsey plans aside, and rushing to push Woodcrest out of the door, presumably out of its development FABs. I think that one of the main purposes of the rush was to stop DELL from defecting to AMD. But, Woodcrest is already too late. Intel has lost the strategic initiative even before Woodcrest is launched.

The problem is AMD has been sitting on its next generation designs for quite a while now. AMD can deliver on these designs at the time of its own choosing. Right after Intel's IDF Woodcrest demonstration, AMD decided to move its launch schedule of Rev F parts ahead. Turion X2 was launched on May 17, 2006, Socket AM2 is to be launched on May 23, 2006. AMD will demonstrate its technologies on June 1st, 2006. AMD may launch Rev F Opteron (socket 1207) on June 6, 2006.

The impending launch of Socket F (1207) is significant, as it establishes the road to true quadcore technology. As for core revisions, K8L has been taped out, AMD should be able to pull in the new core with relative ease. This gives Opteron users a comfortable upgrade path.

AMD is producing significant number of 65nm test chips right now, in an effort to tune yields at the 65nm node. Due to the limited server CPU volume, AMD should have no problem releasing 65nm server CPUs, even before the process is fully tuned for general production. We may expect to see quadcore Opterons sooner than expected.

Going forward, we can see AMD always at least one step ahead of Intel. Preliminary results show that Intel's CORE2 CPU performance will be very close to AMD K8 at the same clock speed, even according to Intel's own guerilla benchmarketing*. AMD enjoys massive advantage on areas of mutli-CPU, multi-core scalability and throughput. Just as Intel plans to catch up with AMD in 2P performance, AMD moves the bar even higher.

One of AMD's mistake was that it talked about dual core Opteron too early, which alarmed Intel into creating its dual-die solution. This time, AMD has been very tight-lipped about its next generation plans. The only hint we got was from AMD's CEO Dr. Ruiz: "It will be interesting to see the things that we're going to do later, which will again continue to force them (Intel) to react and figure out what else to do next." Dr. Ruiz also stated in a recent Jim Cramer's CNBC show that AMD will again challenge Intel to another duel. These words from Dr. Ruiz indicate the existence of some major AMD innovations that will put Intel further behind.

Compared to Opteron users, Woodcrest users will see diminshed returns from their investments. Like Sossaman and Dempsey, Woodcrest is just another stepping stone in Intel's hope to catch up with AMD. In 6 months, before enterprises can fully validate the Bensley platform, Woodcrest will become totally obsolete again, as Intel tries to copy more of AMD64's Direct Connect Architecture.

Woodcrest, like Sossaman, will have the shelf life of a banana.

*Even with Intel installing faster hard drives in its servers, the Woodcrest 3GHZ is ony 37% faster than 2.4GHZ Opteron in Intel's SunGard ACR demo. This translates to 1.37/3 * 2.4 -1 = 9.6% clock/clock performance improvement over K8 in Intel's favourite benchmark. Sun Microsystem's John Fowler commented on Woodcrest that "Five percent (performance improvement) is not particularly easy to explain."

Phone conversation "transcripts"

I will write some in-depth analysis of the AMD-DELL deal and fundamental shift of power in the computing landscape, or the dynasty change.

For now, let's review some past "transcripts". To Intel, it's a series of unfortunate events.

We will see INTEL and DELL blaming each other soon October 27, 2005

Paul and Mike phone conversation 29 Nov 2005

Paul and Craig conversation 23 Dec 2005

Hector calling Malaysia 25 Dec 2005

Eric and Paul conversation 30 Dec 2005

Michael calling Paul 25 Jan 2006

Mike calling Paul 17 Apr 2006

Mike calls Paul again 29 Apr 2006

Paul called Mike two months ago 3 May 2006

Paul calls Mike on May 18, 2006 <--- we are in the proceess of asking NSA to secure the recordings in the name of national security, stay tuned

Mike calling Henri recorded 18 May 2006 (Well, this happened in the near future, we used a time machine)

Merrill Lynch votes AMD strong buy, Intel strong sell

Joe Osha re-iterated his "sell" rating on AMD today after the landmark DELL-AMD alliance. However, Joe Osha's firm, Merrill Lynch is dumping massive amount of INTC stock and on an AMD stock buying spree. Merrill Lynch increased its AMD holdings by 475.58% in the March quarter, when its analyst Joe Osha kept a "sell" rating on the stock.

Increasing a stock holding by 475.58% is super strong buy.

Similarly, Sumit Dhanda gave a $25 target for INTEL, but Bank Of America is strong selling INTC at a huge loss. Bank of America was strong buying AMD, increasing its holdings by 76%.

DELL goes AMD

The Register reported here. Why? DELL has been hurt enough. Its results today show massive decline on server and desktop revenue. DELL needs growth to sustain its ridiculously high 17x price/book valuation ($3.37 billion stock holder equity, $55 billion market cap).

Previously, I projected that DELL has to go AMD before end of Q2 or go BK.

A few days ago, when I saw the news that AMD pushed Turion X2 launch from May 9 to one day before Dell's earnings release today, I conjectured that there might be a correlation.

The significance of the DELL-AMD deal needs no emphasizing. As I pointed out, Intel's guerilla benchmarketing on Conroe and Woodcrest can only dupe amateurs. DELL's going AMD proves that AMD64 rules now and in the future.

With the AMD announcement, DELL's stock went up 4% after hours despite its terrible results. Intel stock fell almost 5% after hours. I expect INTC to fall below $15 in the next few weeks, as its hope of growing 18% in 2H06 is now shattered. Perception is everything. AMD's launch of Socket AM2 on May 23, 2006 and launch of Socket 1207 Opteron in June will push AMD up and Intel down more. The June 1 AMD technology conference will put the final nail on Intel's coffin.

The DELL-AMD deal will put pressure on SUN and IBM to accelerate their Opteron plans.

Mike calling Henri

Mike: Henri, you happy now? We announced Opteron support, and your stock shot up $15.

Henri: Well, you are not doing so bad either, yours went up 15%.

Mike: Yes, it was a win-win situation. But, I am calling you for some help.

Henri: Really? Anything I can do, name it.

Mike: You see, we got these big customers, tens of thousands of four way Opteron servers, a lot of money...

Henri: Good for you, congratulations.

Mike: But the customers want them soon, and we don't have enough Opteron parts.

Henri: Let me check your allocation... Sorry, Mike, our supply is kind of tight...

Mike: Henri, you got to help me this time, we are losing a lot of Xeon sales, these customers are very important...

Henri: Have you heard of our Turion 64 X2 chip? It 's 64 bit and dual core, much much better than Core Duo. That IA32 bit chip has the shelf life of a banana. Why don't you try Turion X2 out and ship a product or two with that?

Mike: If I do, you will get me the Opterons?

Henri: I will try my best, Mike. We are partners now, right? I think Turion will be good for you, as people will be wondering why you are doing the highest end servers with Opteron but not PCes derived from it.

Mike: I see. I will phone ASUS CEO now, he has Turion X2 notebook designs. I will send him 5 million DELL stickers right now...

Henri: Good. Give me a ring when you are done.

Socket AM2 Athlon 64 X2 3800+ fanless opreation

This site tested the AM2 Athlon 64 X2 3800+ dual core CPU operating without a fan ! Core temperature was at 42C with the fan, without the fan, temperature rose to 49C.

Intel share holders upset

At Intel's annual stock holder meeting, share holders aren't happy. One share holder likes to see someone fired, another wants more dividend because he can't sell the stock, one asks why Intel increased its head count by 20,000 and now needs to downsize. I can give the answer to the dividend question, Intel doesn't have any money left for more dividends.

One share holder asks about power efficiency. Intel CEO Paul Ottellini says that he sees that AMD billboard everyday on the way to work and he hopes that the clock will reverse when Woodcrest launches in June.

However, judging from AMD's recent release of energy efficient desktop CPUs, AMD is not sitting idle and Socket F Opteron will be introduced in June with system availability in July. BTW, the Opteron HEs are at 55 watts today.

AMD readies Socket F 1207 Opterons

Our INQ friends tell us that the long awaited Socket F (1207) Opterons are only a few weeks away, just in time to frag the Woodcrests. AMD's situation is so enviable, it's like fighting a battle with 100 divisions of reserve forces. The enemy re-enforces with one fresh regiment? No problem, bring in a few reserve divisions and mass frag the opponent.

AMD launches Turion 64 X2, Intel Core Duo fragged

AMD just launched its socket S1 Turion 64 X2 dual core 64 bit mobile processors, available immediately. The top model Turion 64 X2 TL60 (2GHZ) costs $354, the lowest model Turion 64 X2 TL50 is only $184. In comparison, Intel's 32 bit Core Duo T2600 costs $700. The outdated 32 bit Core Duo has a shelf life of a banana. Very soon, people will be mad to find out their Core Duo won't be able to run 64 bit applications.

Year over year, Intel's desktop and server CPUs suffered a massive $1 billion drop of revenue in 1Q06. Intel had $4.944 billion revenue from server and desktop CPUs back in 1Q05, in 1Q06, the number fell to $3.892, a 21.3% year/year decline in a growing PC market. The only thing that showed revenue growth was mobile, increasing from $1.917 billion to $2.347 billion, but was still too small to compensate the losses in server/desktop . Now, Intel's last profit engine will be cluster fragged. You got an AMD 64 bit future-proof dual core chip against an outdated IA32 chip. Merom won't be available until three months later. By then, AMD will have moved forward yet again. K8L has taped out.

AMD K8L ready to blood Intel

INQ reports more detail about K8L here. It is also reported that AMD's next generation will consume 43% less power than Intel's Conroe/Woodcrest.


A force multiplier for Opteron is co-processor: media processing, JVM/CLR acceleration, TOE, XML or SSL processing, you name it. I envisioned an industry food chain for these co-procs based on ccHT previously. I hope there will be a Parrot co-processor that can speed up PERL6, Python and other dynamic languages to 2x C++ speeds.

Intel, ready for cluster frags!

Bank of America, Bear Stearns and Merrill vote AMD strong buy

We talked about Goldman Sachs James Covello's pumping of INTC previously, at that time I recorded Goldman Sachs' INTC holdings at 53.5 million shares. Today, checking www.nasdaq.com, we can see that Goldman Sachs dumped 19.544571 million shares, or 36.47%, at a huge loss.

I think the government should lock such frauds in jail for the rest of their lives.

The following is a list of prominent INTC dumpers :

FMR CORP: dumped 128.854 million shares or 60.66%
WELLINGTON MANAGEMENT: dumped 33.987 million shares
PUTNAM INVESTMENT: dumped 30.918 million shares
MELLON FINANCIAL: dumped 22.124 million shares
James Covello's GOLDMAN SACHS: dumped 19.544 million shares
...
Sumit Dhanda's B of A: dumped 14.512 million shares , or 37.98%
Joe Osha's Merrill Lynch: dumped 40.52% of its INTC, or 9.652 million shares
Legg Mason, who foolishly bought 60 million INTC in 4Q05 when it was $26, started unloading with a loss


For AMD, Joe Osha's Merrill Lynch incerased its AMD's holdings by 475.58%,
Gurinder Karla's Bear Stearns increased its AMD by 758.69%, Sumit Dhanda's B of A increased its AMD by 76.33%.

* All data above are based on reports for the March quarter.

HP posts strong quaterly results

HP, armed with AMD64 technologies, is fragging DELL all over the place. HP's x86 server business grew 12%, and blade business grew 56%. Those DL585, DL385, DL145 rack servers, BL45p, BL35p, BL25p blades, Turion 64 notebooks, Athlon 64 and Sempron desktops give HP a decisive edge over Intel's obsolete IA32.

DELL just warned a few days ago. When I first predicted that DELL is the next Enron, it stock was at $38. Today, it 's at $23 and falling.

A retard and a crook running wild and naked

Eric Ross at Think Equity claimed that his Asian friends told him that AMD prices crashed by as much as 50%.

But, on Forbes.com, Sumit Dhanda at Banc Of America claimed that AMD has only lowered Sempron prices by 8-10%.

I think Eric Ross was confused about AMD and Intel. It was Intel who slashed its prices by 50%. Those hot and slow Pentium Ds are selling dirt cheap. AMD prices are quite stable. Considering that Intel's topmost chip, the 65nm Pentium Extreme Edition XE 965 3.73GHZ dual core is slower than AMD's entry level Athlon 64 X2 3800+ in benchmarks, I think AMD CPUs are fairly priced.

Sumit Dhanda maintained his buy rating of Intel with target price of $25, while Bank of America is taking huge losses by dumping Intel like crazy at below $20. Bank of America also increased its AMD holdings by 76%.

Conroe will kill both DELL and Intel due to its limited availability until 4Q06. AMD has announced its low power Socket AM2 CPUs, but it is silent on high speed ones. In any case, current AMD CPUs are fragging current Intel ones. Intel is hyping its Q3 Conroe like carzy. AMD has yet to show its hand.

AMD is a server CPU company

I stated this long time ago, some still seem surprised by this notion.

All AMD CPU designs are based on server class architectrues. Even Semprons are downgraded Opterons -- it has crossbar, HT, IMC, all the stuff you found in Opterons.

AMD camp has the Grand Masters of all major server architectures: Opteron, DEC Alpha, MIPS, IBM Power, IBM RS/6000, Sun UltraSparc, HP PA-RISC and Itanium. And these Grand Masters are working together with a mission -- to crush Intel once and for all.

No company on earth can match AMD in terms of server CPU design.

When was the last time Intel designed a server CPU? Itanium, 1992.

Xeon, Woodcrest, Colvertown are just desktop chips with a different label.

AMD introduces dual core desktop CPU at 14 watts

A whole set of low power desktop chips, consuming 14 watts while running SysMark 2004. AMD also published the pricing list.

Conroe Extreme gaming performance analysed

On Slashdot, geeky nerds are having loads of fun on the report that Conroe Extreme Edition will be 40% faster than the 3.7GHZ Pentium XE 965 in gaming performance. Of course, there are debates about the impact on AMD. Some readers point out that the old 2.8GHZ Athlon FX 57 was 36% faster than the Pentium XE 965.

Previous benchmarks showed that even the low end Athlon 64 X2 3800+ is often faster than the 65nm Presler Pentium XE 965.

OLPC will launch in the summer

According to this report, all OLPC production will be made in Shanghai by Quanta. Right now, motherboards are being manufactured. Finished OLPCes will be delivered in July to August time frame.

AMD next generation architecture will be super powerful

Nebojsa Novakovic is complaining that next generation AMD Deerhound quad-core Opteron will have only 2MB of shared L3.

Think it in another way, AMD won't need huge caches (like Intel's 16MB) because its architecture will be so advanced, memory latency will be so low, it will have no need to waste die area on caches!

Look at another optimized multi-core architecture, the 8 core UltraSparc T1, it has only 3MB shared L2, and the instruction set is RISC (larger code).

Andy Bechtolsheim on HPC

Andy Bechtolsheim gave a talk (<--click link for video) on High Performance Computing in a Stanford EE Computer Systems Colloquium. He reconked that Opteron + infiniband is the best clustering solution today. He also talked about using HyperTransport as the interconnect.

HP is fragging DELL big time

According to this CNET report, Alberto Bozzo, vice president of HP's personal systems group, claims that HP is "displacing Dell in most of the bigger and strategic accounts."

It seems that DELL's Enron time may come sooner than expected. A screw driver company such as DELL can be replaced overnight.

Even IBM has to admit that HP and SUN have an edge.

Intel the chapter 11 candidate

"Are you nuts?"

"No, check out Intel's balance sheet."

Cash+cash equivalent: $3.948 billion
Short-term investment: $3.906b
Trading assets: $1.265b
Accounts receivable: $3.912b
Deferred tax assets: $1.032b

Total: $14.063 billion

Bills and other money that must be paid now: $9.843 billion

$14.063 - $9.843 = $4.22 billion, that's all the money Intel can count in the bank. Now, the IRS has ruled that Intel owes Uncle Sam $2.2 billion in taxes from 2001 to 2005. If Intel fails to prevail over IRS, then $2.2 billion + interest is about $2.5 billion Intel must pay.

$4.22 - $2.5 = $1.72 billion would be all that is left...but, Intel also has $2.6 billion of debt.

With a few bad loss quarters (which should start from 3Q06 onward), Intel may have to liquidate some of its FABs to come up with the cash needed for workers' salaries and debt payments.

Then comes 2008, Intel has 97.5% chance of losing the anti-trust lawsuit...

HP readies Turion 64 X2 notebooks

Turion 64 X2 enables dual core mobile 64 bit computing for the American middle class, while DELL pushes dirt cheap 32 bit Core Duo to the poor. Since Intel Merom won't be ready till August 2006, AMD will own 100% of the dual core 64 bit computing for a whole quarter.

With new architecture, Intel will be four generations behind AMD

After three years of hard work, Intel seem have made a major improvement in its next generation Conroe/Merom CPUs: its implementation of AMD64 instruction set is looking good, unlike EM64T in Pentium 4, which is about 10-20% slower than 32 bit mode.

However, 64 bit was just one of the five disruptive technologies AMD introduced with Opteron. Intel's new architecture will still be four generations behind AMD.

An AMD CPU consists of two major functional parts: execution and communication. AMD CPUs have circuits for Core-Core communications (XBAR), Processor-Processor communications (ccHT), Processor-I/O communications (HT) and Processor-Memory communications (IMC). The communication channels are dedicated and separate from each other. Also, since these communication circuits run at CPU's clockspeed, they have very high performance and consume little power. The communication circuits are also very intelligent, for instance, ccHT establishes a single physical memory space from multiple memory banks controlled by different CPUs.

Intel CPUs have none of the above. In Intel architecture, all communications happen on an external shared bus controlled by an external chipset manufactured on 130nm process. The fastest future Intel bus has a bandwidth of 10.6GB/s, less than the bandwidth required for DDR2 800MHZ(12.8GB/s). If you add a couple of GbEs and SATAII drives to an Intel system, you are jamming the bus. When you add more Intel cores, you are choking the bus. An Intel quadcore system will be like an IBM XT connected to a 2400baud modem.

We expect AMD to continue to innovate on both execution and communication with its next generation processors.

Intel bullying the judge

According to this report, Intel is asking the judge to dismiss AMD's anti-trust lawsuit. Intel argues that the US Court doesn't have subject jurisdiction over the matter. Basically, INTEL told the judge: "Hey, judge, our boys in Japan screwed AMD real good, so what? That's beyond the reach of damned US laws. That is none of your business. Stay away."

I think Intel's motion will be denied. Intel Japan is a wholly owned subsidiary of Intel US. Furthermore, SONY and other Japanese companies' PC products were sold in the US, affecting US consumers and US companies. US Courts definitely have jurisdiction over the parts which affected US market. All these matters are one case and should be tried as one.

Intel's argument may have some merit during the penalty phase. But throwing out a dissmissal motion at this stage based on jurisdiction will only upset the judge. This is especially true after Intel already answered AMD's complaint.

Remember, a judge is a human being. Bullying a judge does no good to your case.

AMD answers Intel's guerilla benchmarketing

Tired of Intel's hit and run guerilla benchmarketing, AMD setup billboards all over the world in an effort to draw Intel out of the woods for a hand-to-hand duel. The billboards (see picture by BusinessWeek) display the electricity cost wasted by Intel servers. Currently, the amount is at over $1 billion dollars.

Previously, I estimated that Google can save $140 million a year on power by going AMD.



According to this report, DELL has joined the Green Grid, an AMD led effort for environmentally sustainable computing. Incidentally, a former DELL exec added 25,000 AMD shares to his portfolio.

Recently, Intel became an EMC reseller, competing against DELL. Previously, AMD exempted Michael Dell from testifying against Intel. Earlier, Michael Dell made the following comments on Intel's Conroe architecture.

Paul called Mike two months ago

Based on this report of Intel stuffing the channel to make 1Q06 numbers.

Paul: Mike, how is your business going?

Mike: Paul, business is slow as hell, you know that. Fewer and fewer people want your outdated chips. Last month, you asked me to take delivery of a huge pile of chips. We haven't gone through half of those yet. And you started to compete against us on EMC business.

Paul: Mike, I know, thanks for accepting those chips, which allowed us to book revenue for Feburary. We promise our EMC reselling business is harmless. BTW, I still have a small favour to ask from you.

Mike: Paul, no problemo, we are buddies for years. Whatever, name it, just don't ask me to take more of your chips, our warehouses are full.

Paul: Mike, I am afraid that I have to ask you to do this for me one last time. I am gonna miss the reduced guidiance by a mile if you don't help me.

Mike: Again? We still got millions of your chips rusting in our warehouses. We have to guard them. The chips worth hundreds of millions, and my insurance company won't pay a dime if someone steals them.

Paul: No worry there, Mike. I heard thieves only want AMD chips these days, they are selling very hot on ebay.

Mike: Paul, my friendly advice, you can't just stuff the channel to make your numbers like this. What about your next quarter?

Paul: Mike, I will deal with next quarter later.

Mike: You said the exact same thing last quarter.

Paul: Mike, I promise this won't cost you anything. Just take our inventory, name your price.

Mike: But, our warehouses are really full.

Paul: We don't have to ship the chips to your place. We just need to change the owner of those chips to you in our database, it takes just one second. No hassle there.

Mike: OK, I will pay you 20% of the list price.

Paul: Mike, at that price, I will have to give you 5 times the units to make our quarterly number.

Mike: What if I can't sell them?

Paul: We will give you rebate for unsold chips one year later, no risk.

Mike: Sounds agreeable to me. Now I can sell Core Duo notebooks for $499 to strike HP and Lenovo hard.

Paul: Mike, sorry to tell you, but they already took delivery of our chips ....

Mike: So, we are going to have a global garage sale from Asia to US?

Financial Times (FT.COM) spins around

Today, there was a Financial Times artcile titled "Intel's strategic review sparks debate as rival grows" authored by Chris Nuttall. The title reflected some sharp analysis of Intel's plans. Mysteriously, after a while, the page showed some 503 HTTP errors, and later, FT.com updated the article with a new title: "Intel strives for new route to domination". The title now spins the view of the pump&dumper Sumit Dhanda .

I copied the two articles into the comments, so you can compare them yourself.