Doesn't option 1 basically invalidate the trend toward microkernels and preventing direct hardware writes? The whole point of the NT kernel was to prevent programs from accessing the hardware directly, potentially crashing the entire system.
It's great you enjoy playing Quake3 Arena under 640x480 at 30fps. However, the standard for demanding gamers dictate that 30fps is barely adequate. In order to get really smooth action, you need 40+ fps.
Also, you didn't mentiont the color depth you were running at.
Half-Life is a great game. Fantastic detail as well as good game play. But you lose a lot of the fun if you can't see the game the way it was intended to be seen. Imagine admiring a Monet painting through a telescope or through a TV broadcase. Just not the same thing.
I think Intel gave themselves a lot of time in their original timeline to develop Willliamette. Given the trouble Intel is having with Merced/Itanium, that is probably a good idea.
But AMD's unexpected strong release of the Athlon means Intel can't be complacent any more. Most reviewing sites agree that the PIII-600 was rushed into production as a defensive measure. The fact that AMD promptly added Athlon-650 and -700 neatly demolished that Intel move to claim at least parity.
Let's look at some of Intel's moves lately...
Introducing the 820 chipset in order to sell MBs[diversify] Dramatic price cuts in PIII and Celeron[price war] which cuts into Intel's profits Premature release of PIII-600 which resulted in some errata Hyping up future products like Williamette
Six months has brought Intel and AMD into a whole different situation. Early this year, Intel was also on the defensive as AMD chips flooded the low end market, even making in-roads to Gateway. In response, Intel cut their margins to the bone and flooded the market with cheap Celerons and PIIIs. But now Intel faces a situation that money alone won't solve.
Besides enjoying the fruits of competition, I'd be interested to see whether AMD can finally reach profitability and parity with Intel.
In terms of computational power, I agree that a better architecture counts for more than sheer Mhz ratings. That is where Athlon is beating Pentium-III, after all. At the same clock rating, the Athlon is a handy winner by 10%-30%.
At the same time, speed ratings sell. Motorola recently revised their PPC roadmap to include G4-II. The new development line is meant to push the existing G4 architecture to higher speeds by lengthening the pipeline. In short, trading away architectural advantages to increase the Mhz ratings. The triumph of image over substance.
That's an awfully confident posture to take about the price hike. Michael Dell confidently predicted 2 weeks ago that Dell wouldn't be impacted by the earthquake because of their volume buying power.
Now, Dell is admitting a negative impact from the earthquake because of interruptions in DRAM and LCD production. Apple operates on an even slimmer inventory than Dell, plus outsources more of their manufacturing to Taiwan. The iBook is already shipping in smaller numbers than expected, after all.
Finally, on September 3rd [before the G4 shortage and Taiwan earthquake], the First Call estimate was for $0.89 earning for Q1 '00. So the estimate of $0.86 quoted above is already discounted which is what I have predicted previously.
1. Apple introduces several new lines of kick-butt products; iBook, iMac DTV, G4 Professional Boxes.
I'd point out the saga of G4 tends to show that a good PR show doesn't always lead to a successful product launch. There's no doubt in my mind that Apple's G4 sales won't meet their internal projections. Not only do they have slimmer margins from DRAM prices, they won't sell many $3500 G4s.
2. Apple announces higher than expected earnings and projections and that because of massive backorders they are going to have a blowout fourth quarter.
Very true. Most analysts who commented did so by focusing on the coming quarter.
3. Apple announces unit shipments are targetted to be 40% higher next year, thus increasing market share.
I would disagree with that, since Apple's sales numbers are actually in decline.
Apple Financial Results show a steady decline in international sales for the last 3 quarters. I'd agree we should see a big bump in unit shipment, much as there was a big jump with the introduction of the iMac. But true to history, Apple's sales just tapers off after a strong start.
It is by no means assured that sales growth will continue.
4. Apple announces G4 supply problems - cannot deliver as many machines as users want to buy.
I guess this pretty much speaks for itself, but let's remember this point as we move to point #6.
5. Apple announces Mac OS X preview due in spring.
I don't see how this would excite investors. OS X is a clear move away from the old MacOS. The selling point of the consumer devices like iMac and iBook is ease of use. It hasn't been proven yet that OS X would offer the same features. The big benefits of OS X would have no impact on the consumer market at all, being more geared toward high power users.
Don't get me wrong, it's an important development. But the market which would benefit from OS X is the one Apple retreated from a long time ago.
6. Apple annonces inventory pipeline averaged TWO DAYS in 1999. Best in industry, better than even Dell.
This folds right back into point #4. When you can't even supply your existing orders, then of course you have almost no inventory in the pipeline.
The point of the build-to-order model is two-fold.
1) Minimize inventory to free up working capital. 2) Minimize delivery time
Apple has achieved objective 1 at the cost of failing objective #2. Not a successful story, I think.
Given a choice, would you prefer to have 2 days inventory and $700 million backlog or 2 weeks inventory and $195 million [$700 million * 28% margin] in your pocket??
First of all, the Kihei iMacs were never slated for delivery in Q4 '99.
Secondly, the iBook was originally scheduled for delivery on Sept. 25-26th. 4 days before books close on Q4. Even had the expected 20,000 iBooks been shipped, it would not have made that much of a difference. The iBook is not a Q4 '99 product. It was meant to impact Q1 '00.
The last major product line, G4, shipped a bit over 64,000 units. That's about 40% of expected shipment, so it's not bad. In short, Apple's supply problems for Q4 '99 wasn't as bad as you'd think.
But to address your point directly, the pent up demand doesn't invalidate these two points:
1) Apple will sell fewer of its highest margin machines.
2) DRAM price hikes will put a crimp in EVERY computer maker's profits
The analyst estimates were made before the G4 repricing thing hit the newswires. Once they've had the chance to digest the fact that Apple's product mix will be even more tilted toward the low-end, I'm sure the earning estimates will be revised down a couple of cents.
Finally, the $0.86 per share earnings is only a 60% increase over the reduced earnings estimates. I remind you that Apple was originally projecting $0.76 per share earnings for Q4 '99. The projection of $0.86 isn't nearly as impressive when viewed in perspective.
Now that the dust has settled, what are the lessons to be learned?
1) Apple did the right thing in the end, neatly avoiding the fiasco of upsetting its customers and inviting a horde of lawsuits.
2) Apple has re-learned a basic economic lesson. You can't renegotiate orders you already accepted.
3) Higher prices = upset customers. I hope other computer OEMs learn this lesson as well, since news snippets indicate Gateway may be considering raising prices in the aftermath of the DRAM price crisis.
4) Don't depend on a single supplier. In 3-9 months, IBM will be ready to sell G4 chips, which would be a great improvement for Apple.
5) Christmas won't be as merry as expected in Cupertino, CA.
One, there won't be any revenues from those $3,500 G4-500's. Sales of $3,500 G4-450s will probably be lower. Add in CFO Andersen's warning about increasing advertising spendings, and you can smell the gross margins falling.
Two, rising DRAM prices will put even more pressure on those profit margins.
Three, Apple is still stuck with supplying the entire backorder of G4s from Motorola alone. For Apple, Q1 of 2000 will depend on how well Motorola delivers.
And the most important lesson to take away from this debacle? The rumor mill isn't always right, but they do have an impact. Apple could have crushed all these rumors with a press release on Thursday or Friday. They didn't. I'm willing to bet Apple was gauging public reaction over the last 4 days to decide on a course of action.
That 12% market share being cited is a classic example of both of the cliches. Apple does have a 12% market share in U.S. Consumer market. But that totally excludes international market, corporate market, etc...
Apple's overall market share in the computer market is estimated to be 6%. That's down from the highs of ~15% in the late 80's and even down from the 7% market share of 1996.
You can put the blame on Motorola for the G4-500 fiasco.
But Apple is solely to blame for 2 actions, which will anger customers far more than not having the 500Mhz processor.
1) Cancelling pre-orders 2) Downgrading processors without reducing the prices
The rationale for #1 is directly connected to #2. In order to protect their profit margin, pre-orders must be cancelled in order to make people order into the 3 price tiers.
The pathetic excuse of DRAM prices is ridiculous. Apple carries a 30-35% margin on their G4 line [26& overall, but G4 is higher margin than iMac]. That means ~$750 profit on a $2500 system. They can afford to eat the DRAM price rise. The quarterly profit will be severely hampered, but they wouldn't expose themselves to lawsuits and bad PR.
It sounds like the financial guys are in charge, while customer service is at the bottom of the pile. Think Different. The stock price is more important than your customers.
Actually, the Mhz parity is an important issue for marketing Apple computers. People view Mhz as a quick way to judge the processor. Whenever Intel introduces a new product family, they always make sure the Mhz also add up. When PII came out, it started at 233, the highest Pentium speed at the time. When PIII was introduced, it also started at 450mhz, the highest PII speed then. But as far as the "upgrade path" goes, many people don't upgrade their OS. Many people are happily running Win95 right now. Similarly, many people who run Win98 will not lay hands on Win2k or Millenium. To top it all off, piracy is rampant in the Windows world as well. Finally, I think the best processor out there right now is not the G4. It's the Athlon. Running the standard office apps and internet browser software, the PIII is comparable in power to a G4, and the Athlon is 15-30% faster on the same rating. The disparity becomes even greater on games. The G4 gains its performance advantage in applications with lots of data manipulation but little code processing [SETI@home being an excellent example) plus graphical apps [also data intensive]. But overall - Athlon is a clear winner for 80% of the apps out there. Motorola's own benchmarks prove it.
I don't think Apple is quite irrelevant yet, as far as the future of the PPC goes, but maybe in a year or two. . . Face it, Moto needs Apple for the PPC more than Apple needs Moto.
This could not be less true. Motorola sells the vast majority of its PPCs for embedded processor systems. That's also where IBM sells most of their PPC technology. To say that Motorola needs Apple is like saying Microsoft needs Apple to help sell Office.
Both Motorola and Microsoft would like to keep Apple around as a customer, but if Apple folded up today, they'd shrug and go on with their lives.
If Motorola decided to shut down its G4 assembly line today - where would Apple be?
Don't be silly. Do you have idea how many computers fit on that export restriction list?
When I order a laptop, I routinely get asked whether I plan to export it. Athlon chips also fall under the restriction, since its performance is up to 2.4Gflops. Dual processor computers are also on that list, since it's the total computational speed that counts.
First of all, you should have heard the big news that Apple IS reneging on most of the pre-orders. Secondly, Apple is only adding memory to the Ultimate 450Mhz model. If you ordered a G4-450 w/ Sawtooth, you are going to get a G4-400 w/ Sawtooth for the same price. What a bargain?! Also, IBM has signed an manufacturing licensing agreement. No IBM-made G4 chips will appear for a minimum of 3 months. That means the huge backlogs must be filled by Motorola exclusively. The true test is Monday. People will have the opportunity to digest the news, and Apple will face the music.
Slashdot Mirror
Doesn't option 1 basically invalidate the trend toward microkernels and preventing direct hardware writes? The whole point of the NT kernel was to prevent programs from accessing the hardware directly, potentially crashing the entire system.
It's great you enjoy playing Quake3 Arena under 640x480 at 30fps. However, the standard for demanding gamers dictate that 30fps is barely adequate. In order to get really smooth action, you need 40+ fps.
Also, you didn't mentiont the color depth you were running at.
Half-Life is a great game. Fantastic detail as well as good game play. But you lose a lot of the fun if you can't see the game the way it was intended to be seen. Imagine admiring a Monet painting through a telescope or through a TV broadcase. Just not the same thing.
I think Intel gave themselves a lot of time in their original timeline to develop Willliamette. Given the trouble Intel is having with Merced/Itanium, that is probably a good idea.
But AMD's unexpected strong release of the Athlon means Intel can't be complacent any more. Most reviewing sites agree that the PIII-600 was rushed into production as a defensive measure. The fact that AMD promptly added Athlon-650 and -700 neatly demolished that Intel move to claim at least parity.
Let's look at some of Intel's moves lately...
Introducing the 820 chipset in order to sell MBs[diversify]
Dramatic price cuts in PIII and Celeron[price war] which cuts into Intel's profits
Premature release of PIII-600 which resulted in some errata
Hyping up future products like Williamette
Six months has brought Intel and AMD into a whole different situation. Early this year, Intel was also on the defensive as AMD chips flooded the low end market, even making in-roads to Gateway. In response, Intel cut their margins to the bone and flooded the market with cheap Celerons and PIIIs. But now Intel faces a situation that money alone won't solve.
Besides enjoying the fruits of competition, I'd be interested to see whether AMD can finally reach profitability and parity with Intel.
In terms of computational power, I agree that a better architecture counts for more than sheer Mhz ratings. That is where Athlon is beating Pentium-III, after all. At the same clock rating, the Athlon is a handy winner by 10%-30%.
At the same time, speed ratings sell. Motorola recently revised their PPC roadmap to include G4-II. The new development line is meant to push the existing G4 architecture to higher speeds by lengthening the pipeline. In short, trading away architectural advantages to increase the Mhz ratings. The triumph of image over substance.
That's an awfully confident posture to take about the price hike. Michael Dell confidently predicted 2 weeks ago that Dell wouldn't be impacted by the earthquake because of their volume buying power.
Now, Dell is admitting a negative impact from the earthquake because of interruptions in DRAM and LCD production. Apple operates on an even slimmer inventory than Dell, plus outsources more of their manufacturing to Taiwan. The iBook is already shipping in smaller numbers than expected, after all.
Finally, on September 3rd [before the G4 shortage and Taiwan earthquake], the First Call estimate was for $0.89 earning for Q1 '00. So the estimate of $0.86 quoted above is already discounted which is what I have predicted previously.
The breakdown would go like this if you ordered a G4-500 for $3500.
1) Take a G4-450 for $2,500, the price for a G4-450 before Oct. 13th.
2) Take a G4-450 for $3,500, with an extra 128mb of memory and 7 extra GB of HD space.
Now, which choice do you think people will make? What percentage of users would take option #2 and pay the extra $1,000 for 128mb/7GB HD?
1. Apple introduces several new lines of kick-butt products; iBook, iMac DTV, G4 Professional Boxes.
I'd point out the saga of G4 tends to show that a good PR show doesn't always lead to a successful product launch. There's no doubt in my mind that Apple's G4 sales won't meet their internal projections. Not only do they have slimmer margins from DRAM prices, they won't sell many $3500 G4s.
2. Apple announces higher than expected earnings and projections and that because of massive backorders they are going to have a blowout fourth quarter.
Very true. Most analysts who commented did so by focusing on the coming quarter.
3. Apple announces unit shipments are targetted to be 40% higher next year, thus increasing market share.
I would disagree with that, since Apple's sales numbers are actually in decline.
Apple Financial Results show a steady decline in international sales for the last 3 quarters. I'd agree we should see a big bump in unit shipment, much as there was a big jump with the introduction of the iMac. But true to history, Apple's sales just tapers off after a strong start.
It is by no means assured that sales growth will continue.
4. Apple announces G4 supply problems - cannot deliver as many machines as users want to buy.
I guess this pretty much speaks for itself, but let's remember this point as we move to point #6.
5. Apple announces Mac OS X preview due in spring.
I don't see how this would excite investors. OS X is a clear move away from the old MacOS. The selling point of the consumer devices like iMac and iBook is ease of use. It hasn't been proven yet that OS X would offer the same features. The big benefits of OS X would have no impact on the consumer market at all, being more geared toward high power users.
Don't get me wrong, it's an important development. But the market which would benefit from OS X is the one Apple retreated from a long time ago.
6. Apple annonces inventory pipeline averaged TWO DAYS in 1999. Best in industry, better than even Dell.
This folds right back into point #4. When you can't even supply your existing orders, then of course you have almost no inventory in the pipeline.
The point of the build-to-order model is two-fold.
1) Minimize inventory to free up working capital.
2) Minimize delivery time
Apple has achieved objective 1 at the cost of failing objective #2. Not a successful story, I think.
Given a choice, would you prefer to have 2 days inventory and $700 million backlog or 2 weeks inventory and $195 million [$700 million * 28% margin] in your pocket??
First of all, the Kihei iMacs were never slated for delivery in Q4 '99.
Secondly, the iBook was originally scheduled for delivery on Sept. 25-26th. 4 days before books close on Q4. Even had the expected 20,000 iBooks been shipped, it would not have made that much of a difference. The iBook is not a Q4 '99 product. It was meant to impact Q1 '00.
The last major product line, G4, shipped a bit over 64,000 units. That's about 40% of expected shipment, so it's not bad. In short, Apple's supply problems for Q4 '99 wasn't as bad as you'd think.
But to address your point directly, the pent up demand doesn't invalidate these two points:
1) Apple will sell fewer of its highest margin machines.
2) DRAM price hikes will put a crimp in EVERY computer maker's profits
The analyst estimates were made before the G4 repricing thing hit the newswires. Once they've had the chance to digest the fact that Apple's product mix will be even more tilted toward the low-end, I'm sure the earning estimates will be revised down a couple of cents.
Finally, the $0.86 per share earnings is only a 60% increase over the reduced earnings estimates. I remind you that Apple was originally projecting $0.76 per share earnings for Q4 '99. The projection of $0.86 isn't nearly as impressive when viewed in perspective.
Now that the dust has settled, what are the lessons to be learned?
1) Apple did the right thing in the end, neatly avoiding the fiasco of upsetting its customers and inviting a horde of lawsuits.
2) Apple has re-learned a basic economic lesson. You can't renegotiate orders you already accepted.
3) Higher prices = upset customers. I hope other computer OEMs learn this lesson as well, since news snippets indicate Gateway may be considering raising prices in the aftermath of the DRAM price crisis.
4) Don't depend on a single supplier. In 3-9 months, IBM will be ready to sell G4 chips, which would be a great improvement for Apple.
5) Christmas won't be as merry as expected in Cupertino, CA.
One, there won't be any revenues from those $3,500 G4-500's. Sales of $3,500 G4-450s will probably be lower. Add in CFO Andersen's warning about increasing advertising spendings, and you can smell the gross margins falling.
Two, rising DRAM prices will put even more pressure on those profit margins.
Three, Apple is still stuck with supplying the entire backorder of G4s from Motorola alone. For Apple, Q1 of 2000 will depend on how well Motorola delivers.
And the most important lesson to take away from this debacle? The rumor mill isn't always right, but they do have an impact. Apple could have crushed all these rumors with a press release on Thursday or Friday. They didn't. I'm willing to bet Apple was gauging public reaction over the last 4 days to decide on a course of action.
Let me start with a couple of cliches.
"A little knowledge is a dangerous thing"
"There are lies, damn lies, and statistics"
That 12% market share being cited is a classic example of both of the cliches. Apple does have a 12% market share in U.S. Consumer market. But that totally excludes international market, corporate market, etc...
Apple's overall market share in the computer market is estimated to be 6%. That's down from the highs of ~15% in the late 80's and even down from the 7% market share of 1996.
You can put the blame on Motorola for the G4-500 fiasco.
But Apple is solely to blame for 2 actions, which will anger customers far more than not having the 500Mhz processor.
1) Cancelling pre-orders
2) Downgrading processors without reducing the prices
The rationale for #1 is directly connected to #2. In order to protect their profit margin, pre-orders must be cancelled in order to make people order into the 3 price tiers.
The pathetic excuse of DRAM prices is ridiculous. Apple carries a 30-35% margin on their G4 line [26& overall, but G4 is higher margin than iMac]. That means ~$750 profit on a $2500 system. They can afford to eat the DRAM price rise. The quarterly profit will be severely hampered, but they wouldn't expose themselves to lawsuits and bad PR.
It sounds like the financial guys are in charge, while customer service is at the bottom of the pile. Think Different. The stock price is more important than your customers.
Actually, the Mhz parity is an important issue for marketing Apple computers. People view Mhz as a quick way to judge the processor. Whenever Intel introduces a new product family, they always make sure the Mhz also add up. When PII came out, it started at 233, the highest Pentium speed at the time. When PIII was introduced, it also started at 450mhz, the highest PII speed then. But as far as the "upgrade path" goes, many people don't upgrade their OS. Many people are happily running Win95 right now. Similarly, many people who run Win98 will not lay hands on Win2k or Millenium. To top it all off, piracy is rampant in the Windows world as well. Finally, I think the best processor out there right now is not the G4. It's the Athlon. Running the standard office apps and internet browser software, the PIII is comparable in power to a G4, and the Athlon is 15-30% faster on the same rating. The disparity becomes even greater on games. The G4 gains its performance advantage in applications with lots of data manipulation but little code processing [SETI@home being an excellent example) plus graphical apps [also data intensive]. But overall - Athlon is a clear winner for 80% of the apps out there. Motorola's own benchmarks prove it.
How do you feel about the "direct intervention" now?
Unless you are one of the lucky few whose systems are already in production, I bet you don't feel very generous about Apple right now.
I don't think Apple is quite irrelevant yet, as far as the future of the PPC goes, but maybe in a year or two. . . Face it, Moto needs Apple for the PPC more than Apple needs Moto.
This could not be less true. Motorola sells the vast majority of its PPCs for embedded processor systems. That's also where IBM sells most of their PPC technology. To say that Motorola needs Apple is like saying Microsoft needs Apple to help sell Office.
Both Motorola and Microsoft would like to keep Apple around as a customer, but if Apple folded up today, they'd shrug and go on with their lives.
If Motorola decided to shut down its G4 assembly line today - where would Apple be?
Don't be silly. Do you have idea how many computers fit on that export restriction list?
When I order a laptop, I routinely get asked whether I plan to export it. Athlon chips also fall under the restriction, since its performance is up to 2.4Gflops. Dual processor computers are also on that list, since it's the total computational speed that counts.
First of all, you should have heard the big news that Apple IS reneging on most of the pre-orders. Secondly, Apple is only adding memory to the Ultimate 450Mhz model. If you ordered a G4-450 w/ Sawtooth, you are going to get a G4-400 w/ Sawtooth for the same price. What a bargain?! Also, IBM has signed an manufacturing licensing agreement. No IBM-made G4 chips will appear for a minimum of 3 months. That means the huge backlogs must be filled by Motorola exclusively. The true test is Monday. People will have the opportunity to digest the news, and Apple will face the music.