|
|
|||||||||||||||
 |
|||||||||||||||
|
|
|||||||||||||||
|
|||||||||||||||
|
|
|||||||||||||||
|
|||||||||||||||
|
|||||||||||||||
|
|||||||||||||||
|
|||||||||||||||
|
|||||||||||||||
|
|
|||||||||||||||
|
|||||||||||||||
|
|
|||||||||||||||
| 2015 © KickAss Gear | |||||||||||||||
|
|
|||||||||||||||
|
|
| KickAss
Gear News Archive: July 2001
July 31st You Can Help Stomp Out Rambus! As Intel's good ship Itanic leaves port, CEO Craig Barrett waves to Rambus Inc. Execs, left sobbing on the docks with hankies in hand. He bids them a fond farewell as he strains to lean over the railing, and he shouts "The consumer will decide!" as a new round of sobbing breaks out on the docks below. No this is not the opening scene to a surrealist movie, but it may be the closing scene for Rambus Inc. The latest twist in the tangled web of Rambus Incorporated's fortunes has Intel's CEO admitting that Rambus memory may have kept the Pentium 4 processor from reaching the market share it deserves. Barrett said, "Clearly if the 845 chipset allows the Pentium 4 family to go to these lower price points, and those are the high-volume price points in the market, then you'd expect it to ramp very quickly." Intel is not abandoning Rambus DRAM just yet, but they clearly recognize that the market will decide these things, and that they can't come as decrees from On High. So there you have it folks. You can now participate in the elimination of Rambus from the face of the planet if you so desire. How? Just buy i845-based P4 systems with SDRAM, rather than i850-based systems with Rambus DRAM. If the equations shifts in favor of the i845, then Rambus will be slowly phased out. Of course it could have been different. If Rambus hadn't gotten a severe case of the greedies, they could have become the high performance, higher priced memory 'object of desire'. That is if they hadn't tried to extort money from the entire SDRAM manufacturing industry, making all their potential friends into overnight enemies. Their reputation is so tarnished now that even performance enthusiasts find it difficult to buy machines branded with the Rambus moniker. If only they could have it to do over again, maybe things would turn out different...... Nah! July 30th Intel's Tualatin (O.13u) PIII Can't Catch the Athlon The Tech Report and Hot Hardware have Benchmarks of the new 0.13 micron Pentium III (Tualatin) versus the Pentium 4 and the Athlon. The Tualatin is currently at 1.2GHz, which puts it at a disadvantage vis-à-vis the Pentium 4 at 1.8GHz and the Athlon at 1.4GHz. Indeed, the newer, smaller, faster PIII seems to get left in the dust as the Athlon and P4 slug it out. The most interesting results are that the Athlon with DDR memory wins in most benchmarks, even when up against the higher clocked Pentium 4 with Rambus DRAM. The Tualatin lags behind in almost every case. When the Tualatin was overclocked to near 1.5GHz, it gave both the Athlon and the P4 a run for their money, beating them in several benchmarks. And speaking of money, the Athlon is by far the least expensive chip of the bunch, so in my opinion, it still deserves the crown. Loyal Intel enthusiasts will have a choice for a short time now, between a Tualatin PIII, or a P4. But considering that the Tualatin at 1.5GHz beat the Pentium 4 at 1.8GHz in benchmarks like 3D Mark 2001, Intel may not be anxious to flood the market with Tualatins, lest they cut into already lagging P4 sales. These results may explain why Intel is phasing the Tualatin out as quietly and quickly as it appeared. The Celeron will remain the low-end Intel CPU, safely slower than the P4 so as to not rock the boat. The P4 will be the remaining choice at the high end. The Tualatin will live on for some time in notebooks, and the 512KB L2 cache version will find it's way into small 2-way servers for a little while. But the death of the Pentium III draws nigh. July 27th Is the Windows XP Performance Hit Intentional? Benchmarks of Windows XP versus Windows 2000 have shown that even with the newest release of XP (RC1), the new operating system performs at least 25% slower than the previous operating system. Even with the new "look and feel" features, and new background tasks disabled, the performance of XP remains about 10% below that of Win 2K. Each time Microsoft releases a new operating system, the allegations fly that the new OS requires more CPU power as a devious method of driving new hardware sales. People buy the new OS and install it, only to find their machine now performs more poorly, not better. The end result is that many customers upgrade their hardware. So with benchmarks clearly showing potential customers they will not get as much performance from their machines after the upgrade, will this hurt XP sales, or help hardware sales? Only time will tell. Windows XP .Net Scaled Back for Now? Privacy Groups filed formal complaints yesterday with the FTC concerning the bundling of Microsoft's .Net services in their new XP operating system, which is scheduled for release Oct. 26th. Coincidentally, on the same day, MS announced it would not include the business to business portion of .Net (known as Blackcomb) in the initial release of Windows XP. The question now is, will the FTC let MS release XP on schedule with it's current version of .Net (including Hailstorm, Passport and Wallet)? Based on previous releases of Windows, it's my un-educated guess that MS will be allowed to release XP on time. There may be some stipulation about having Verisign, PayPal or other icons on the desktop to give customers a choice, but that would be about it. MS clearly intends to dominate both the business to customer as well as the business to business internet transaction market, but they may have realized it could be too much to take on all at once. Blackcomb is now scheduled for a later version of XP, to be released in 2003. July 26th Windows XP Release Challenged Privacy advocate groups EPIC and Junkbusters, among others, are filling motions with the FTC today to block the October 26th release of Microsoft's Windows XP operating system. The dispute involves Microsoft's bundling of the .Net service, which includes several components, including 'Passport', 'Hailstorm', and 'Wallet' (the latter being Bill's favorite). At issue is the fact that Microsoft will attempt to become both the keeper of personal information, as well as the gatekeeper for Internet secure transactions through it's dot Net services, which are an integral part of Windows XP. Hailstorm will be the information collector and distributor, under the guise of being your 'gateway to your personal information', regardless of where you are, or who's machine you are using. Passport and Wallet will be the secure transaction component of .Net, and they have been likened to MS providing each user with an "Internet drivers license". In the end, Microsoft intends to put itself and it's services between end users and all Internet transactions, which will lead to more of the PC's functions to be centralized and under Microsoft's control. The idea is to slowly wean the public on the idea that they will need to pay MS a monthly fee for it's services, which they hope will eventually become the only real choice Internet shoppers have. MS has decided that advertising on the Internet is not a money making proposition, so they want to turn the Internet into .Net, with monthly licensing fees for end users. How many of you want MS to have all your personal information, credit card numbers, and on top of that, charge you a monthly fee for using their transaction services?.... Raise your hands! CNET
link July 24th Intel/Inquirer Bad-Mouth AMD The processor wars are heating up and seem to be approaching critical mass. The savage processor price war has now developed further into a PR/propaganda war. Mike Magee over at The Inquirer rarely tows Intel's line, but seems to have done just that by reporting that "some AMD Athlon and Duron chips are duff" (Brit slang for busted). This is exactly what big-time Intel investor Paul Engel wanted reported. Intel, and those with large monetary stakes in the company's fortunes, have long sought to bad-mouth AMD chips as the cheap, unreliable alternative. And they have certainly had success in the corporate arena where AMD chips are chump. But the PC enthusiast group has found otherwise: that the AMD Athlon is a better (and less expensive) processor for most of today's applications and games than the Pentium 4. So what is this tale about AMD processors being "duff"? As long as nearly 2 years ago, scattered reports surfaced here and there on the Internet that some K6-2 chips did not display JPEG and MPEG compressed images correctly. This story was true, but pertained to a tiny fraction of the total number of chips that left the fabrication plants. In other words, it was a quality control problem with a small number of marginal chips (probably those near the edge of the wafers used to produce the chips). This is in sharp contrast to the several design flaws that Intel has let get though with it's processors, including the famous bug in the original Pentium processor. We have heard later reports that a very small number of Athlon chips also exhibited the same compressed image problem, but in our experience making computers for sale, we had never run into a single AMD processor that had this problem. Considering that we have sold many hundreds of Athlon and Duron processors over the last two years, I would have to guess that the percentage of AMD chips with this problem is far below 0.2%. And because AMD has said it will immediately replace any CPU with the problem, it seems to be much ado about nothing. It does point out one thing though. Companies should not try to save money during the "price wars" by skimping on quality control. It is a sad fact of the numbers game though that QC is one of the first things to go during an industry price war, and the end result is tarnished reputations and irritated customers. Even if the number of bad parts is minuscule, you can bet your bottom dollar the competition will make the most of the bad press, with the help of reporters, of course. July 22nd Warzone 2100 Rebirth Many 3D real-time-strategy games have come out over the last 3 years, but one of the first was one of the best. Warzone 2100 came out in 1999, but it is still played today by it's fans who find its mix of simplicity, strategy, weapons, vehicles and buildings to be unmatched by later, more flashy 3D strategy games. But Pumpkin Studios went out of business in early 2000, and version 2 of Warzone never materialized (see what happens when you pirate games instead of buying them??). But before they went out of business, they posted their map editor for download so that fans could make their own maps and scripted games. Well, the most loyal fans have done much more than that. They have organized a programming group to keep Warzone 2100 alive. As many as 50 volunteers have chipped in time and talent to make new Warzone campaigns. The name of the project is N.E.W.S.T. Currently they have several new maps, which are in fact mods as much as they are new maps. These mods give you new weapons and vehicles (including flying construction vehicles!), as well as new scenery. It's very refreshing to see new things added to Warzone after all this time. They are also working on new campaign mods for the game, which should be ready for download soon. On top of that, they have a startup utility which gives you more control over Warzone 2100 at startup. If you liked Warzone 2100, get ready for some more fun. For starters, I recommend just downloading a few of their new maps/mods (you'll need the official 1.1 patch, which is available for download). Later you may want to give the World Editor a shot, and see how good your map making skills are. There is lots more advanced stuff to download too, but be careful with the 1.11 patch/mod, I had some trouble with it. If you don't have a copy of Warzone 2100, you may be out of luck. It is not on sale anywhere anymore, so your only hope is to "borrow" a copy from someone who has it. Warzone 2100 started the 3D real time strategy genre, and is still considered one of the best. Now, with it's new facelift, and upcoming campaigns, it's got a second lease on life. If you like Warzone, I highly recommend NEWST. Let's give them all the support and encouragement we can. July 20th Windows XP Manifest Destiny? The Inquirer has a story today which states that there is a fundamental flaw in Windows XP’s new "User Experience Controls". The article was a little vague on the situation, so I did a little digging. The user interface in Windows involves a file called ComCtrl32.dll, which stands for common controls. This library handles the "look and feel" of Windows applications. Windows 98/Me and Windows 2000 use ComCtrl32.dll version 5.8, while Windows XP will ship with version 5.8 for backward-compatibility, and version 6.0 for the new "XP look". To get an XP-compatible application to have the new spiffy look of Windows XP, it will need an XML file included which tells Windows to use ComCtrl32.dll version 6. These new files are called manifest files (doesn’t the Mac use manifest files?). Here is an excerpt from Microsoft’s windowsxp.devx.com website. "Isolated XP applications use a manifest, which is an XML file containing information that self-describes an assembly or an application. All binding and activation metadata, such as COM classes, interfaces, and type libraries, is now stored in the manifest, rather than the registry. There are two types of manifest files: applications manifests, which describe isolated applications, and assembly manifests, which describe individual assemblies". What The Inquirer is reporting is the fact that if a manifest file is copied to the C:\Windows directory, Windows XP will no longer start. Only deleting that manifest file from the Windows directory will get the machine up and working again. If true, MS may have some serious work to do before XP is ready for release. And speaking of the release date, head over to the official XP web site for a countdown clock! This is an extremely late date to be hearing about serious design flaws in XP. It seems a little hard to believe that such a large bug could have been missed up until now, especially because MS has made such a big deal about software developers testing their applications thoroughly with ComCtrl32.dll version 6, to make certain that applications don’t get "broken" as they say. It will be very interesting to see how this story develops over the next few days and weeks. Links: DDR More Bandwidth Efficient Than Rambus? Van Smith has a new web site, and he's come out swinging with a new article comparing the bandwidth efficiency of DDR DRAM to Rambus DRAM. Surprisingly, DDR DRAM wins (77% efficiency vs. 64% for Rambus). Rambus still wins the straight memory bandwidth competition, but it uses that bandwidth less efficiently than DDR DRAM. Van has some other choice comments on Rambus Inc. as well. July 19th Is MS Backing Off Windows XP Activation? The rumors are flying from both sides of the product activation divide. Microsoft is simultaneously saying that it may loosen up the way XP activation works, while at the same time saying that the actual product activation scheme in the shipping version will be different than the current scheme. At the same time, more cracks have been announced for the existing scheme. It should be clear to Microsoft at this point that they are in a lose-lose situation here. Changing the activation scheme in any meaningful way this late in the product cycle is nearly suicidal. At the same time, keeping the system as is will certainly lead to clogged activation support phone lines, and angry customers. What's a software giant to do? If they are serous about getting XP right, they will tone the activation scheme down significantly, or remove it altogether. MS understands better than anyone that burdening your paying customers is no way to stop software piracy. Software pirates cost MS far more than home users with 2 computers and one copy of Windows. So if they want to go for the goldmine, rather than the piggy bank, they know that product activation isn't the way to do it. July 17th Windows XP Activation Smashed! If the report over at tecChannel.de is correct, then the Windows XP product activation (WPA) scheme is very weak, and is easily cracked. The entire Activation Code system is dependent on a single file called wpa.dbl. This is the file that stores information about your hardware setup and the state of product activation. Also, the date of XP installation is stored, so that if 4 hardware items are changed more than 30 days after XP installation, then re-activation becomes immediately mandatory. In other words, once the 30 day period is up, your grace period is over, and re-activation must be done immediately after making the hardware changes. The wpa.dbl file is located in the c:\windows\system32 directory, and is deleted by the system as soon as the 4th hardware change is detected. Without the file in the system32 directory, re-activation becomes necessary. However, the folks over at tec Channel found that saving your wpa.bdl file, and copying it back to the system32 directory negates the re-activation scheme! This is a childishly simple work-around, and makes me wonder how serious MS is about implementing a rock-solid product activation scheme. It may have been Microsoft's intention all along merely to stop inexperienced home users from installing a single copy of XP on multiple home machines. But that's not enough of a problem for MS to make such a controversial move with the most touted operating system they have released in years. Indeed, it may be the case that they originally had big plans for WPA, but had to scale it back as potential problems cropped up during early testing. Perhaps the thing that I noticed most was the product activation screen that was shown at tec Channel. They give you two phone numbers you can call to activate XP. One is toll free (which will be swamped), and the other is a toll number, which many customers will opt for after waiting endlessly and fruitlessly to get through on the 800 number. Why do I think there will be problems getting through to the toll free activation number? Because you need to carefully read off a 50 (that's a big 5-0) digit number to the operator, who must enter that code without any typos or transpositions. How many customers will miss at least one number out of 50? How much will this slow down the process and back up the phone calls coming into Microsoft's activation center? Plenty. The first few days after XP debuts will be a living hell for the operators at MS, and for the customers who just want to get the activation process over and done with. And considering how easy the existing WPA scheme is to crack, it makes me stop and wonder why MS would even consider creating such a potential nightmare scenario for their operators and their paying customers, when the benefits are so dubious. The extreme ease with which WPA is circumvented has led to speculation that MS has something tougher up it's sleeves. The logic here is that they may not want to tip their hand before the operating system is even released. I have my doubts that they will be making any significant changes though, because they would not have enough time between now and October to get the beta testing done on any new scheme they implement. July 14th Rambus Royalty Retribution So what has been going on with Rambus Inc. recently? There are court cases and judgments a plenty in Rambus’ future, but let’s take a look at where their business plan has gotten them so far. For those of you who are not up on recent Rambus scuttlebutt, they are an intellectual property Corp. that deals exclusively in patents for computer memory modules. They have gotten themselves into hot water recently when it was disclosed during their first patent infringement case against Infineon Technologies, that Rambus had used information garnered from memory standards meetings to amend their pending memory patent applications. The revelation of this misappropriation resulted in Rambus losing the court case and having punitive damages levied against them. It became clear during the court proceedings that Rambus had developed a business plan in the early ‘90s whose sole aim was to patent as much of the technology in modern computer memory as possible, with the intent of charging memory manufacturers royalties for every stick of memory sold in the World. The plan was working until Rambus actually got to court in an attempt to force Infineon to pay royalties on the SDRAM memory it was producing. That’s when the plan began to unravel from both ends. The first part of the unraveling was the high litigation costs that Rambus incurred as a result of pursuing multiple court cases against several larger, better funded corporations. This cut heavily into their coffers. The second part of the unraveling was the high cost of Rambus DRAM (their invention) relative to SDRAM (the memory design they misappropriated the patents for). This made the SDRAM royalties that Rambus was receiving from several corporations, including Samsung, more important for the moment that Rambus DRAM royalties. Then the third and most devastating factor cut in. SDRAM prices plummeted to historical lows! And because the royalty payments made by companies like Samsung are a percentage of the price, royalty payments to Rambus have also plummeted. Add to this the fact that large memory makers are now negotiating reduced royalty payments, in light of the poor state of the memory industry, and you've got a Rambus royalty meltdown in progress. So the end result is that the Rambus business plan has failed. Their attempt to patent by nefarious means memory they did not design has backfired in court. Their attempt to collect huge royalties from the entire SDRAM industry has dried up, and turned memory manufacturers against them. And their hopes of getting the cost of Rambus DRAM down close to the price of SDRAM has become a pipe dream, with ever-receding prices on SDRAM making it a moving target that is impossible to hit. All in all we can say that the computer memory industry is no longer the cash cow it used to be. The days of $600 sticks of memory have been resigned to history. This means that Rambus used the wrong business model to get into the wrong industry at the wrong time. Couldn’t have happened to a nicer bunch of folk. Links: July 12th Microsoft Caves on IE Integration and Licenses Microsoft has been fighting tooth and claw to keep Internet Explorer tied to the Windows operating system for years. They said it was an integral part of the OS, and could not be removed. It's mandatory presence in the OS was merely a matter of Microsoft exercising their "freedom to innovate". If true, their freedom just got taken down a notch. As part of the ongoing discussions between the Justice Department and Microsoft, the software giant agreed to make Internet Explorer an optional part of Windows XP, with a remove function in the "Add/Remove Programs" utility. This isn't just good news for Netscape and others who have suffered from Microsoft's anti-competitive practices, but also for consumers who almost certainly will appreciate the new flexibility. Internet Explorer is known to make the Windows operating system less stable (I can confirm this from much personal experience), so it's removal could help make computers more reliable. Microsoft also agreed to change their license agreement for the OS, which now makes it acceptable for computer makers to change the desktop appearance, or to add and remove icons at will. This is exceptionally excellent news for system integrators. But will these changes be enough to satisfy everyone, including the Justice Department? Only time will tell. It's a darn good start though, and hopefully, Microsoft will continue along this road of rationality, rather than belligerently cramming the "freedom to innovate" down everyone's throats. July 10th Windows XP Activation Code As
the debut of Windows XP nears, the discussions about it's product
activation scheme have grown exponentially. Rumors have ranged from fears
that Microsoft will get all sorts of personal information about each user,
to the possibility that slight changes in a computer's configuration will
force the user to acquire a new activation code. A group in Germany has
just published a paper detailing how the Windows XP activation system
works. They downplay the negative aspects of the activation system, but
the negative aspects are clearly present. The
good news is that very little information about each user will get back to
Microsoft Central. But the
privacy issues have never loomed as ominously as the fact that the
activation is tied to your specific hardware configuration.
Microsoft had to come up with a method that they felt would prevent
people from installing the same copy of XP on different computer systems.
The simplest way to do that would be to couple the activation code
to a hardware specific code which is generated based on your computer's
configuration. Then the
question arose, "how many hardware changes should we let a user make
before reactivation is required?" The answer they arrived at was
three. The fourth change kicks in re-activation. Below
is a listing of the different hardware components that the XP activation
system checks when generating a unique ID code for each computer system. double
word | offset | length | bit-field value based on:
H1 | 0
| 10 | volume serial number of system volume According
to the authors, the H2 31 value for dockable notebooks will give the user
a little more wiggle room with their hardware configuration. Otherwise
just plugging the notebook into the docking bay could trigger the
requirement for a new activation code. In
plain English, here are the things that are monitored by the activation
code: Hard drive volume number (generated every time you format a drive),
Network card (MAC address), CD ROM, graphics card, CPU serial number, hard
drive ID (distinct from volume number), SCSI adapter, IDE controller, CPU
make and model, amount of RAM, dockable vs. non-dockable system. While
the authors make the point that you can change a couple items in your
system without triggering re-activation, they clearly did not discuss the
many ways that re-activation could be invoked.
Let's
assume you have a desktop system that arrived with XP installed.
You go through the activation process so that you can determine if
your new system is working properly.
After the activation and testing phase, you decide to add another
stick of RAM from your old system, as well as the network card from your
older computer. Also, you install the CD burner from your old system. Then you realize that the CPU serial number feature is
enabled on your motherboard's BIOS. You
decide you don't want that feature enabled.
After disabling the CPU serial number and rebooting the system, the
XP product activation routine will kick in again.
This is because you just changed four of the items in the list. So
it's back to the MS phone lines to get another activation code. Microsoft
is clearly counting on two things. The
first is that most people don't change their computer's hardware
configuration at all, and the second is that everyone won't try to
activate their operating systems all at once.
If tens of thousands of people attempt to activate Windows XP on
the same day, many of them will have great difficulty in getting through
to Microsoft's activation center. That
will make even the most inexperienced Windows users very unhappy. Perhaps
the most onerous part of the activation scheme will fall upon PC power
users. Most PC enthusiasts
swap hardware in and out of their systems on the slightest whim.
This practice will become exceedingly tiresome with Windows XP
installed on their systems. Further,
companies that need to test and validate various hardware configurations
before selling them or posting hardware reviews may find Windows XP
unusable. For example, in
order to test motherboard and CPU combinations (as in the case of computer
upgrade kits), the tester needs to have a hard drive handy which has the
operating system loaded on it. However,
each time a motherboard and CPU are mounted on a test bench and connected
to the hard drive with Windows XP installed, product reactivation will be
required. That is an
impossible situation for hardware testers. Apparently,
Microsoft realizes this is an unworkable situation for corporations to
have to deal with. As such,
Windows XP Professional version will probably not incorporate the product
activation scheme. Instead,
only the home version of Windows XP will be saddled with this
inconvenience. In my mind it
is clearly part of Microsoft's strategy to entice as many home users into
buying the substantially more expensive ($300) professional version of the
operating system in order to avoid the activation scheme. It
is fully within Microsoft's rights to protect their software from piracy.
However, it is clear that the product activation codes incorporated
into the home version of Windows XP will do little to stop software
piracy. It may force users with extra cash on hand to buy two copies of
Windows XP to populate their two home computers, but that is not where
Microsoft loses most of their income. Organized software bootleggers cost Microsoft far more income
than folks loading one copy of Windows onto two machines at home.
And software bootleggers are much easier to catch than going through this incredibly complicated software activation scheme to stop small-scale license infringements in private homes. If Microsoft simply sent representatives to the hundreds of so-called "computer shows" that occur around the country each year, they would be able to prosecute virtually all software bootleggers. If you have ever gone to one of these computer shows, you almost certainly would have seen bootlegged Microsoft products on sale for a fraction of their actual cost. If Microsoft were actually serious about software piracy, they would not be concentrating their efforts on burdening their loyal paying customers, they instead would be after the bootleggers with a vengeance. It's my humble opinion that they have their priorities bass-ackwards, and are punishing their loyal paying customers, rather than the software bootleggers. Links: July 8th Price Wars Hurt Everyone But Customers The news throughout the last week has been dismal for computer makers, and computer parts manufacturers. AMD stock was hit particularly hard after they posted lower earnings. Intel also is not fairing nearly as well as they have historically. What's going on? The price war. I and many others have been warning that a continued price war would hurt all companies involved. And in fact, AMD and Intel sold quite a few processors in the last quarter, but profits were way down because of the price war (Link). In the short run, this will hurt all computer makers, as prices continue to slip on a weekly basis. It is already hurting the chip makers bottom line, and that will continue for the foreseeable future (Link). The erosion of memory prices in the last 2 months has reached unprecedented levels, forcing many memory makers to close fabrication plants. Everyone's profits are down, even if sales are relatively good, because of the intense competition and the price wars. Indeed, the lowered earnings of all computer-related companies, and poor PC sales figures in general, have resulted in slumping operating system sales for Microsoft. This OS sales slump may be in part responsible for Microsoft's new and aggressive moves to threaten companies into buying more site licenses. I suppose if enough new people aren't buying MS software, Microsoft figured they should just go after the people who are currently using it. It is interesting to note that intense competition has driven down the price of every part of a modern computer, except one. The MS operating system. Currently, the price of Windows can amount to as much as 20% of a budget computer's total price. This is another clear indication that MS has a monopoly in the OS arena. The high, deflation-resistant price of MS operating systems has not gone un-noticed. Currently MS is in talks with the government about reducing their pricing and licensing model as part of their concessions in the anti-trust case against them. Unfortunately, only large named accounts with MS will get the small break in price. All other OEMs (original equipment manufacturers) will have to pay the same or higher prices for MS operating systems. (Link) Fortunately for consumers, the competition in the hardware arena is very good news, but it can't last forever. Companies are dropping like flies in Winter, and eventually the amount of competition in computer-related businesses will be reduced. In the longer run, prices may begin to rise again if the price-meltdown drives enough companies out of business. So this Summer is a good time to buy. But beware, buying a budget computer from a company that is about to close it's doors can be a bit of a gamble. July 6th Microsoft Hate Mail Factory Nothing gets the ol' hate mail flowing like a negative editorial on Microsoft's business practices. It almost seems as though the hate mail comes from a Microsoft factory, churned out in vast shrink-wrapped quantities like their software. It's heartwarming to know there are so many people out there with free time and a bone to pick with anyone who criticizes Microsoft. After all, Microsoft is the underdog here, and they need all the defending they can get, right? Some folks even questioned whether the mailer was real, suggesting that MS would never stoop to such a tactic. Well, here is a picture of the back of the mailer for those non-believers.
Attention MS hate-mail machine! Read before you flame. We use MS software, we sell MS software, and for the most part we really like MS software. But we don't like being treated like criminals, by a company that assumes everyone is guilty until proven innocent. We also don't like being threatened by a company we have supported for years. My point for those who are too angry to think straight is this. Only a monopoly could ever think about threatening their loyal customers into buying even more software. Capitalism doesn't work that way. If MS had any serious competition in the OS market, they would be using the carrot approach, rather than the stick approach, to get more sales. MS has gone too far this time, and it's about time the public speaks out. You can flame all you like, but you won't change the fact that MS has gone off the deep end with this mailer. Flame Here => Dr. John Links: July 5th Microsoft Admits It's a Predatory Monopoly! A picture is worth a thousand optical character recognition words. And this picture is straight from Micro$oft themselves!
This friendly mailer was sent out to virtually all businesses that are considered "Microsoft Channel Partners". What kind of message is Micro$oft sending here? First, that your business is small and unimportant to Micro$oft, except as another tiny source of income in a sea of paying customers. Second, they are tacitly admitting that they are a monopoly, since no company in competition with others would send out such an insulting and threatening mailer. Third, the choice of images clearly indicates Micro$oft thinks of itself as a predatory company (you goldfish... me shark!). Finally, this mailer shows just how arrogant and unconcerned Micro$oft is about the ongoing court cases against them, and how they intend to continue to do whatever they want. Now keep this image in mind as you contemplate Micro$oft's upcoming XP operating system with it's "activation codes". This is the "feature" in the next operating system that will halt your computer after you change it's video card, and tell you that you must contact Micro$oft for a new activation number before you can proceed. I have already heard horror stories of this thing kicking in while one user was on a flight to a business meeting. Sorry pal, you're out of luck until you get your new activation code! Now that's what I call the "Freedom to Intimidate"! Forget about that freedom to innovate stuff. Sushi anyone? Happy July 4th Everyone! Desktop Athlon 4 Delay? For those of you hoping to get a new "Palomino" Athlon 4 for your desktop this Summer, you may need to think again. The Register has another in a long line of rumors about when the Athlon 4 for desktops will be available. If the rumor/roadmap is true, it will probably not be until at least September before these parts are available. And then it will be several weeks or even months after that before they become plentiful. The question is why is AMD delaying the Athlon 4 for desktops? First, they don't have a lot of fabrication capacity, so they need to concentrate on the higher speed 1.3 and 1.4GHz Athlons now, which are very popular. They are also trying to introduce the 1.5GHz Athlon, but so far that has been a slow starter. Also, with Pentium 4 sales in the doldrums, and Pentium III sales faltering, AMD may not feel any pressure to push the new Athlon forward until it has more inventory built up. Finally, AMD has a lot on it's table now for a smaller chip company. They are not just working on the Athlon 4, but also on the newer Clawhammer 64-bit processor, the updated Athlon 4 (Thoroughbred), the updated Duron (Morgan), and even the much anticipated "Barton" version of the Athlon with Silicon-On-Insulator technology. So it's my guess they want to take things slow, and get them right. But this means that customers wanting to upgrade this Summer will have to settle for a regular old Thunderbird Athlon CPU. July 3rd Pentium III Fire Sale Coming? The
Inquirer has a report today which suggests that Intel is getting ready to
kill the Pentium III line of microprocessors.
Mike Magee suggests that the Pentium 4 and Celeron processors will
round out Intel's offerings. But
what about the Tualatin Pentium III?
How could Intel think about discontinuing it before it even debuts? If these
rumors are true, (how good are your sources Mike?), then I have to assume
that Intel feels that their product line is getting too confusing.
But why keep the Celeron when the 0.13 micron Pentium III is such a
better processor? The answer
may lie in the benchmarks that have come from testing the Tualatin. This
new version of the Pentium III is capable of outperforming Intel's fastest
Pentium 4 processor. In many
ways the Tualatin shows how Intel made mistakes with the design of the
Pentium 4 processor. It's smaller cache and longer pipeline put the
Pentium 4 at a disadvantage relative to the Tualatin in many benchmarks. Intel is hoping to fill the gap between the Celeron and the Pentium 4 by introducing the i845 chipset with SDRAM support for the Pentium 4 processor. Intel wants to move the Pentium 4 from a low sales volume to a high sales volume product, and the new Pentium III is getting in the way. Many Intel loyalists have skipped the P4 for now, and are waiting for the Tualatin. Intel needs to nip this trend in the bud, and get P4 sales figures moving up. Hence, drop the PIII like a hot potato, and make the P4 affordable for the masses. But will enthusiasts want a P4/SDRAM system, or will they wait for P4/DDR systems? Intel has
some real cogitating to do here. Are we going to see a Pentium III
Fire Sale in the late Summer or early Fall? Link: The Inquirer July 1st NVidia Says No to Microsoft (and starts to act like Rambus!) In these times when nobody who is interested in making money ever says no to Microsoft, it's refreshing to see NVidia being as plucky with MS as they are with everyone else. It's a regular old clash of the Titans, with the exception that one titan if far more titanic than the other. Microsoft (the titanic titan), is holding talks (do they ever stop having talks?) with graphics industry leaders to determine what specifications Direct X 9.0 should have. MS wants members to sign the usual mum agreement, as well as an agreement not to try to make any of the stuff proprietary. NVidia (the not-so-titanic titan) has apparently declined to sign, which raises two questions. First, will this affect the relationship between MS and NVidia, who are working together feverishly on the upcoming X-Box (remember, the holidays are only 5 months away!)? And second, what has NVidia got up it's corporate sleeves that makes signing the agreement a no-go? Vertex and Pixel Shaders. That's correct. NVidia is apparently in the process of patenting things that MS has already included in Direct X 8.0 (oops), and is also working on patenting things that may go into Direct X 9.0! Holy Rambus Batman! Haven't we heard this tale before? A bunch of standards meetings attended by all in the industry, but one attendee is desperately trying to amend and submit patents faster than Speedy Gonzales can get a taco!? Yes, it's a play right out of Rambus Inc.'s snidelyish business plan. But this patent trap seems better laid than the Rambus plan. Microsoft has the resources to pay royalties till the cows come home, and there is no danger that the graphics chip/game console market will crumble the way the RAM market did. Now we can see a clever plan unfolding, with Microsoft stuck between a rock and a very very hard place. They can't abandon the NVidia X-box, and they have no rights to the technology in it. Therefore, by the time the X-Box comes out this Fall, you can bet your bippie that MS will have been forced to sign licensing deals with NVidia to use the Pixel and Vertex Shaders. Oh what a tangled web we invent, when first we practice to patent. Link: The Inquirer June 29th Rambus Inc. Hindering New Memory Standards! An article at EBN states that current IP litigation, in particular the legal action brought by Rambus against several SDRAM makers, is having negative effects on current standards committees. According to David Balto, a former policy director at the FTC, "increasing litigation and claims for licensing fees has dramatically increased uncertainty in high-tech standard-setting bodies, undermining the effectiveness of these bodies to facilitate technological change." So now we can not only thank Rambus Inc. for throwing the whole memory industry into turmoil, but also for suppressing the development of new memory standards in the future. Balto suggested that standards committees must adopt much stricter disclosure requirements for all members of the committee. No more "silent partners" like Rambus Inc., thank you very much. I'd personally like to see some teeth put into those committees to punish members who break the rules. But that's not going to happen, because standards committees have no authority over anyone. So it's always up to the courts to decide these issues. Hence we can expect long rocky roads for future standards committees of all types, especially if more IP companies adopt the Rambus "business model". June 28th And What of Rambus? Fortune.com has posted an article nailing Rambus Inc. to the wall. Here is a telling excerpt about Rambus' attendance at the JEDEC memory standards committee. "They listened as their industry colleagues discussed an element known, in typically impenetrable techno-gibberish, as "programmable CAS latency." A week after the meeting, one of the two Rambus staffers, Richard Crisp, met with a company attorney to talk about amending Rambus' pending patent applications. Among the new technologies that Crisp wanted to add: programmable latency. A few months before, Crisp had recommended adding patent claims for "mode registers"--right after they were debated at the same standards committee. Only a few months before that, it was "low-voltage swing." All those technologies made the same quiet journey from the standards committee agenda into Rambus patent applications." The article is an excellent read. Perhaps now that Fortune has come out with this information, making it more widely known, Rambus will have to reconsider it's "business model" if you can call it that. You may recall that the last company which Rambus took to court, Infineon, not only won the case, but actually had a jury award them damage payments from Rambus. It is unlikely that Rambus will fare much better in any of their other upcoming court cases. But at least it keeps the lawyers busy, and well fed. Copyright 2001, KickAss Gear |
