Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
IBM's 90nm PowerPC 970FX
- Thread starter MacRumors
- Start date
- Sort by reaction score
Re: Re: R&D
Not use "Mac" in its name? I can't see them doing that. They sell Macintosh computers, not others. They have a hard enough time growing market share. Releasing a new computer and not calling it a Mac would be branding suicide in my opinion.
Not use "Mac" in its name? I can't see them doing that. They sell Macintosh computers, not others. They have a hard enough time growing market share. Releasing a new computer and not calling it a Mac would be branding suicide in my opinion.
Re: Re: R&D
i think that these numbers account for several reasons. once has to do with the fact that apple was stuck with the g4 powermac, and few had a reason to upgrade until the g5. these numbers might be changing now.
i also believe that apple has a smaller profit margin on the iMachines, as they are in a much more competitive market when it comes to consumers. Pros easily justify spending more for a mac than a pc, but a consumer might not be able to do the same.
i think that these numbers account for several reasons. once has to do with the fact that apple was stuck with the g4 powermac, and few had a reason to upgrade until the g5. these numbers might be changing now.
i also believe that apple has a smaller profit margin on the iMachines, as they are in a much more competitive market when it comes to consumers. Pros easily justify spending more for a mac than a pc, but a consumer might not be able to do the same.
Where's the hook?
Uh, does this mean you haven't noticed that the shift to portables is industry-wide? A lot of companies and home users are buying laptops as desktop replacements. I could never settle for the performance hit, but for the average office drone and the average home user, portables make a lot of sense.
Uh, does this mean you haven't noticed that the shift to portables is industry-wide? A lot of companies and home users are buying laptops as desktop replacements. I could never settle for the performance hit, but for the average office drone and the average home user, portables make a lot of sense.
Re: Base RAM
A 256MB stick for an iMac runs about $50. A 512MB stick about $100. Probably much less a delta at Apple wholesale prices. To stick a paltry 256MB stick as standard and not a 512MB just sucks. A standard 256MB really limits the users expandability.
Same thing they did in G3 iBooks, a paltry 128MB stanard.
Squeeze every last drop of profit. As a user, I don't like it. As an Apple shareholder, it didn't do much, if anything for me.
A 256MB stick for an iMac runs about $50. A 512MB stick about $100. Probably much less a delta at Apple wholesale prices. To stick a paltry 256MB stick as standard and not a 512MB just sucks. A standard 256MB really limits the users expandability.
Same thing they did in G3 iBooks, a paltry 128MB stanard.
Squeeze every last drop of profit. As a user, I don't like it. As an Apple shareholder, it didn't do much, if anything for me.
Re: Where's the hook?
But the iMac is like a portable (takes up little desk space) without the performance hit and less limitations. The PC world has nothing like it, they all sell ugly towers. I think there's way too much trade off in portable as desktop replacements (screen size, keyboard, performance).
I see full page ads every day in the paper for Dell, HP and IBM and they all feature dedktops. Apparently someone is buying these.
Just because I vouldn't resist, I bought a used G3/900 iBook for $601 from compusaauctions (turned out it was new.
No way could I see this as my everyday computer. Quite frankly, it just another of my toys.
OK, I've rambled enough here.
But the iMac is like a portable (takes up little desk space) without the performance hit and less limitations. The PC world has nothing like it, they all sell ugly towers. I think there's way too much trade off in portable as desktop replacements (screen size, keyboard, performance).
I see full page ads every day in the paper for Dell, HP and IBM and they all feature dedktops. Apparently someone is buying these.
Just because I vouldn't resist, I bought a used G3/900 iBook for $601 from compusaauctions (turned out it was new.
No way could I see this as my everyday computer. Quite frankly, it just another of my toys. OK, I've rambled enough here.
Re: Re: Re: Re: R&D
I was wondering when someone would catch on. While there is a lot positive about Mac, it is also perceived as a high-end PC, so many consumers dismiss it out of hand. Whereas professionals are drawn to it.
iPod and iBook are both huge successes. OS X is another huge success. It's AppleWorks and not MacWrite anymore. It's iPhoto and not MacPaint. The original Apple computer was not a Mac it was a II and then IIe, IIc and even a III. The server is not MacServe, it's Xserve. Apple has been careful to give their new products new identities so that customers look at them without any old prejudices.
That is why I think the next Apple consumer computer will not have the name Macintosh in it. Of course it will run OS X, and we will always know it as a Mac computer, but the reality is that Apple needs to give the consumer a message that this is not a new revision but simply new and he needs to give it a fresh look. I think that is why iPod has so many Windows buyers, they don' associate it with Mac myths and prejudices.
I was wondering when someone would catch on. While there is a lot positive about Mac, it is also perceived as a high-end PC, so many consumers dismiss it out of hand. Whereas professionals are drawn to it.
iPod and iBook are both huge successes. OS X is another huge success. It's AppleWorks and not MacWrite anymore. It's iPhoto and not MacPaint. The original Apple computer was not a Mac it was a II and then IIe, IIc and even a III. The server is not MacServe, it's Xserve. Apple has been careful to give their new products new identities so that customers look at them without any old prejudices.
That is why I think the next Apple consumer computer will not have the name Macintosh in it. Of course it will run OS X, and we will always know it as a Mac computer, but the reality is that Apple needs to give the consumer a message that this is not a new revision but simply new and he needs to give it a fresh look. I think that is why iPod has so many Windows buyers, they don' associate it with Mac myths and prejudices.
Re: Re: Re: Re: Re: R&D
Yes, well, my PowerBook doesn't have the name 'Mac' in it, but it's well recognized that it's a Macintosh computer. So is the iBook that my friend has. I think that we're getting into a bit of a nebulous region. What about the Performa? Or the Quadra? Or the Centris? These were all Macs that technically didn't have 'Mac' in the name. I don't think it's quite as simple as you seem to think that it is...
Yes, well, my PowerBook doesn't have the name 'Mac' in it, but it's well recognized that it's a Macintosh computer. So is the iBook that my friend has. I think that we're getting into a bit of a nebulous region. What about the Performa? Or the Quadra? Or the Centris? These were all Macs that technically didn't have 'Mac' in the name. I don't think it's quite as simple as you seem to think that it is...
mini tower
How about my throwing this out?
I was a partner in an audio firm a number of years ago. We produced speakers for professional use, as well as station monitors for broadcast radio, modular amps, etc. I'm pointing this out because I have designed equipment that was of high grade. The G5's are to my mind the best designs Apple has ever produced, and remind me of what my company used to do (though it was analog).
I've looked at the G5 the way a production engineer would, for the purpose of seeing where the "fat" could be removed in order to make it less expensive without cheapening it. I believe that I have succeeded.
Most of us here probably would agree that the iMac, while quite elegant, has probably seen the best of it's sales in the past, and that no matter what Apple does, they will most likely continue to decline.
My feeling is that they don't really have a definite market place. Businesses don't like them because they are too distracting, and not expandable or up-dateable enough, as well as being to expensive. Schools don't like them because they are too maneuverable, and the students will first play with the arm, and then break them. That was why the eMac came out (to Apple's surprise). This leaves the home user. That is most of the sales base. But the wonder of the design has worn off, and sales decline.
I have a proposal. A less expensive tower. This doesn't seem to be so insightful, because everyone has been crying out for this. True. But I think that I know how it can be done in a realistic way that can be produced.
Looking at this as an engineer who has designed product, I have identified areas which would cut costs drastically, and yet leave a unit that would still be a "Mac" in every way.
Here are some of my ideas.
1. Eliminate the handles and feet. These really do add more to the cost than you would believe, and as my proposals would cut the size and weight, would not be needed anyway as the machine could be carried under the arm, or in a case.
2. This would be a single processor machine. Apple now uses the same case and motherboard size for both the single and dual models. This could be smaller.
3. To do this, Apple could cut the size of the motherboard in several ways.
a. Eliminate the almost 3 inches that are unused for the second processor.
b. Cut the memory slots down to two. This would still enable 2 gig's of memory now, and four when the 2 gig sticks come out later this year. Very few businesses or home users, or schools need more than this.
c. Cut the Toslink optical. This was really a mistake on Apple's part, as Toslink is not regarded highly in the audio or recording industry. Most pro digital recording devices use AES/EBU, S/PDIF for semi-pro or home equipment. This could save $50 if they go back to the mini-jack.
This would cut the price of the motherboard by at least 25%, and Apple's motherboards are not cheap!
4. Use a 1.8 processor and PCI 2.1 spec slots. They will certainly do the job. Keep the 8x AGP, as it only adds a few bucks to the total, and still allows the high performance boards that gamer's love.
5. In calculating power needs, I found that Apple was generous. A 325 watt supply is all that is needed on this "new" mini-tower. It would need less fans as well.
6. The case can now be lowered by three inches. When the lack of handles and feet are added in, the difference is more like six inches. We have also lost at least ten pounds.
When we are through, the Mac still looks like a Mac, and performs like a Mac. It still has that industrial but stylish look.
What would this cost? I estimate that it could hit the shelves for $999.
With a new Apple LCD for under $500, this would be a great price point. With a $75 17" CRT, it would be a steal!!
If Apple would wise up, it would understand that, like Microsoft (shudder), its main competitor in the future is Linux. Here, the hardware could compete. But there is one last thing...
Do the oldtimers, such as myself, remember when Apple sold hardware, and gave away the OS? This stopped with System 7. I guess Apple looked around one day and saw that Microsoft was making money on software and decided that they should as well.
The problem is that The OS, for Apple, isn't worth much. They sold about $135,000,000 worth last year. Before you say that that's a lot, figure what that is as a percentage of total sales of (hopefully) at least $7,000,000,000. Almost nothing. A little more than two thirds is profit.
My suggestion is that Apple sell it for $49. They will make a profit on that, but more importantly, as the members of my Mac group here in NYC agree, many more people will upgrade each time. Apple would probably sell a good deal more.
The other advantages are that businesses would not think that Mac's are more expensive, maybe less. Microsoft would have to meet a much lower pricing if Apple gave site licenses based on that one seat price. AND Linux would not have so much of an advantage either. Red Hat and IBM, among others, sell their LINUX support sets for a comparable price. A lot of LINUX users like MAC's and OS X, but they won't pay the price. No more excuses.
What do you think?
How about my throwing this out?
I was a partner in an audio firm a number of years ago. We produced speakers for professional use, as well as station monitors for broadcast radio, modular amps, etc. I'm pointing this out because I have designed equipment that was of high grade. The G5's are to my mind the best designs Apple has ever produced, and remind me of what my company used to do (though it was analog).
I've looked at the G5 the way a production engineer would, for the purpose of seeing where the "fat" could be removed in order to make it less expensive without cheapening it. I believe that I have succeeded.
Most of us here probably would agree that the iMac, while quite elegant, has probably seen the best of it's sales in the past, and that no matter what Apple does, they will most likely continue to decline.
My feeling is that they don't really have a definite market place. Businesses don't like them because they are too distracting, and not expandable or up-dateable enough, as well as being to expensive. Schools don't like them because they are too maneuverable, and the students will first play with the arm, and then break them. That was why the eMac came out (to Apple's surprise). This leaves the home user. That is most of the sales base. But the wonder of the design has worn off, and sales decline.
I have a proposal. A less expensive tower. This doesn't seem to be so insightful, because everyone has been crying out for this. True. But I think that I know how it can be done in a realistic way that can be produced.
Looking at this as an engineer who has designed product, I have identified areas which would cut costs drastically, and yet leave a unit that would still be a "Mac" in every way.
Here are some of my ideas.
1. Eliminate the handles and feet. These really do add more to the cost than you would believe, and as my proposals would cut the size and weight, would not be needed anyway as the machine could be carried under the arm, or in a case.
2. This would be a single processor machine. Apple now uses the same case and motherboard size for both the single and dual models. This could be smaller.
3. To do this, Apple could cut the size of the motherboard in several ways.
a. Eliminate the almost 3 inches that are unused for the second processor.
b. Cut the memory slots down to two. This would still enable 2 gig's of memory now, and four when the 2 gig sticks come out later this year. Very few businesses or home users, or schools need more than this.
c. Cut the Toslink optical. This was really a mistake on Apple's part, as Toslink is not regarded highly in the audio or recording industry. Most pro digital recording devices use AES/EBU, S/PDIF for semi-pro or home equipment. This could save $50 if they go back to the mini-jack.
This would cut the price of the motherboard by at least 25%, and Apple's motherboards are not cheap!
4. Use a 1.8 processor and PCI 2.1 spec slots. They will certainly do the job. Keep the 8x AGP, as it only adds a few bucks to the total, and still allows the high performance boards that gamer's love.
5. In calculating power needs, I found that Apple was generous. A 325 watt supply is all that is needed on this "new" mini-tower. It would need less fans as well.
6. The case can now be lowered by three inches. When the lack of handles and feet are added in, the difference is more like six inches. We have also lost at least ten pounds.
When we are through, the Mac still looks like a Mac, and performs like a Mac. It still has that industrial but stylish look.
What would this cost? I estimate that it could hit the shelves for $999.
With a new Apple LCD for under $500, this would be a great price point. With a $75 17" CRT, it would be a steal!!
If Apple would wise up, it would understand that, like Microsoft (shudder), its main competitor in the future is Linux. Here, the hardware could compete. But there is one last thing...
Do the oldtimers, such as myself, remember when Apple sold hardware, and gave away the OS? This stopped with System 7. I guess Apple looked around one day and saw that Microsoft was making money on software and decided that they should as well.
The problem is that The OS, for Apple, isn't worth much. They sold about $135,000,000 worth last year. Before you say that that's a lot, figure what that is as a percentage of total sales of (hopefully) at least $7,000,000,000. Almost nothing. A little more than two thirds is profit.
My suggestion is that Apple sell it for $49. They will make a profit on that, but more importantly, as the members of my Mac group here in NYC agree, many more people will upgrade each time. Apple would probably sell a good deal more.
The other advantages are that businesses would not think that Mac's are more expensive, maybe less. Microsoft would have to meet a much lower pricing if Apple gave site licenses based on that one seat price. AND Linux would not have so much of an advantage either. Red Hat and IBM, among others, sell their LINUX support sets for a comparable price. A lot of LINUX users like MAC's and OS X, but they won't pay the price. No more excuses.
What do you think?
Re: mini tower
Well it sounds good for me. I guess this is sort of what I am hoping for with the wishful talk about a new cube.
I want G5, good graphics, preferably 4 not 2 slots for ram - its much cheaper to buy 512 sticks, and you need to balance them out with the G5. Smaller form factor would be a big plus.
If you kept everything else you get with the current G5 range, this would be very appealing.
Ability to mount a TFT screen on an arm afixed to the "cube" up to at leats 20" would be very fine as well. I hope the new "metal" screens cater for more than just bezels...
Jason
Jason
Well it sounds good for me. I guess this is sort of what I am hoping for with the wishful talk about a new cube.
I want G5, good graphics, preferably 4 not 2 slots for ram - its much cheaper to buy 512 sticks, and you need to balance them out with the G5. Smaller form factor would be a big plus.
If you kept everything else you get with the current G5 range, this would be very appealing.
Ability to mount a TFT screen on an arm afixed to the "cube" up to at leats 20" would be very fine as well. I hope the new "metal" screens cater for more than just bezels...
Jason
Jason
Re: mini tower
No offense intended -- this is a sincere question: do you really think Apple can't figure out how to make a cheap case & motherboard? You've done a fine job of detailing how they could do it, but your basic hypothesis is so self-evident I'm struggling to see the point of bringing it up. The question was never "Can they?" The question is "Will they?"
I disagree. The current design is absolutely fabulous, especially compared to the competition. Sure, it can and will be updated, but the all-in-one solution is not just elegant -- it's cleaner, simpler, and smaller than separate components. Your argument re: businesses is misguided because Apple doesn't try to sell boxes to the cubical jungle. And businesses don't care about expandability (except RAM, which is easy enough) -- it has long been the case that businesses will buy new cheap boxes instead of hassling with upgrades. Your argument re: schools is misguided, as most schools are going to portables, whether they are Macs or PCs. And I've got anecdotal evidence of my own that at least one school district loves the iMacs because (a) they are more rugged than portables, (b) small footprint, and (c) the kids love them. (My brother-in-law is assistant district superintendent.)
IMHO, the only things wrong with the iMac are cost and the spec sheet. It would sure be nice if Apple could cut a couple hundred off each model. And the megahertz gap is not a functional problem, it's a marketing nightmare.
How many decades in the future?
Um, isn't this all just money-juggling? Apple most certainly has a target gross margin. If they get $135m less from the OS, they'll have to boost prices on hardware to recover it. And it seems to me that your average short-sighted business owner is much more concerned with sticker shock (hardware) than maintenance (OS).
Besides, your argument for lowering the OS cost just doesn't add up. If you are correct that two-thirds of the $129 OS price is profit (that's a big IF -- where the heck did that come from), that means that approx. $44 is the break-even point, with the additional $85 being profit. If you lower the price to $49, profit is cut to $5, and you'd have to sell 17 TIMES the copies to get the same profit. Yeah, that'll happen . . .
OS cost is a tiny part of the Mac's TCO. Tiny! You just got done saying Apple shouldn't care about $135m from their OS sales, and now say that a Linux user will pay $1,500 or more for a Mac and then whine about $129 for a major OS update sometime in the quasi-distant future?
If you're actually saying that the entire system (hardware and software) are too expensive, welcome to reality. Apple has never and will never be able to compete on price, because that would require them to build a commodity box, like Dull. Apple (and SJ) have stated over and over and over that they will never sell a box. Disagree with their strategy all you want -- just don't hold your breath waiting for your cheap Linux box.
That's what I think.
No offense intended -- this is a sincere question: do you really think Apple can't figure out how to make a cheap case & motherboard? You've done a fine job of detailing how they could do it, but your basic hypothesis is so self-evident I'm struggling to see the point of bringing it up. The question was never "Can they?" The question is "Will they?"
I disagree. The current design is absolutely fabulous, especially compared to the competition. Sure, it can and will be updated, but the all-in-one solution is not just elegant -- it's cleaner, simpler, and smaller than separate components. Your argument re: businesses is misguided because Apple doesn't try to sell boxes to the cubical jungle. And businesses don't care about expandability (except RAM, which is easy enough) -- it has long been the case that businesses will buy new cheap boxes instead of hassling with upgrades. Your argument re: schools is misguided, as most schools are going to portables, whether they are Macs or PCs. And I've got anecdotal evidence of my own that at least one school district loves the iMacs because (a) they are more rugged than portables, (b) small footprint, and (c) the kids love them. (My brother-in-law is assistant district superintendent.)
IMHO, the only things wrong with the iMac are cost and the spec sheet. It would sure be nice if Apple could cut a couple hundred off each model. And the megahertz gap is not a functional problem, it's a marketing nightmare.
How many decades in the future?
Um, isn't this all just money-juggling? Apple most certainly has a target gross margin. If they get $135m less from the OS, they'll have to boost prices on hardware to recover it. And it seems to me that your average short-sighted business owner is much more concerned with sticker shock (hardware) than maintenance (OS).
Besides, your argument for lowering the OS cost just doesn't add up. If you are correct that two-thirds of the $129 OS price is profit (that's a big IF -- where the heck did that come from), that means that approx. $44 is the break-even point, with the additional $85 being profit. If you lower the price to $49, profit is cut to $5, and you'd have to sell 17 TIMES the copies to get the same profit. Yeah, that'll happen . . .
OS cost is a tiny part of the Mac's TCO. Tiny! You just got done saying Apple shouldn't care about $135m from their OS sales, and now say that a Linux user will pay $1,500 or more for a Mac and then whine about $129 for a major OS update sometime in the quasi-distant future?
If you're actually saying that the entire system (hardware and software) are too expensive, welcome to reality. Apple has never and will never be able to compete on price, because that would require them to build a commodity box, like Dull. Apple (and SJ) have stated over and over and over that they will never sell a box. Disagree with their strategy all you want -- just don't hold your breath waiting for your cheap Linux box.
That's what I think.
Re: mini tower
First off, we'd love to fill our labs with nice iMacs (graphic design labs excluded of course).
Pros to the iMac:
- screen moves out of the way for lecture style classrooms
- lcd/g4 combo put out little heat, the rooms we uses were never designed for 20+ computers and monitors and they often get too hot (even with retroactive tweaks to HVAC)
- They look pretty so they make the school look good (and for mac guys on a University campus, they draw people into our labs.. over the PC ones ;-)
- They are quiet so they are easy to teach over
Cons to iMacs in schools
- THEY COST TOO MUCH for what you get. We don't look at it being an oh so cool computer.. it's a box that does 'x' for $$. THIS is why the eMac is in existance.. edu demanded a cheaper (sub $1000) machine.
- there are worries about LCD Damage... but I've never found this to be a problem, even in remote 24x7 labs where we find the plugs cut off the mice and keyboards (yes, only the plugs missing).
We've also found that expandability is something we look for, but in practice we typically don't ever expand a machine. We ocassionally upgrade memory, but usually the machine just sits there till the upgrade cycle comes around. I upgrade the memory in a bunch of Blue and White G3s a while back, but only because they had 64 and the budget for replacement was thin.
Working at 2 Universities, I'll comment on what I percieve schools to want.
First off, we'd love to fill our labs with nice iMacs (graphic design labs excluded of course).
Pros to the iMac:
- screen moves out of the way for lecture style classrooms
- lcd/g4 combo put out little heat, the rooms we uses were never designed for 20+ computers and monitors and they often get too hot (even with retroactive tweaks to HVAC)
- They look pretty so they make the school look good (and for mac guys on a University campus, they draw people into our labs.. over the PC ones ;-)
- They are quiet so they are easy to teach over
Cons to iMacs in schools
- THEY COST TOO MUCH for what you get. We don't look at it being an oh so cool computer.. it's a box that does 'x' for $$. THIS is why the eMac is in existance.. edu demanded a cheaper (sub $1000) machine.
- there are worries about LCD Damage... but I've never found this to be a problem, even in remote 24x7 labs where we find the plugs cut off the mice and keyboards (yes, only the plugs missing).
We've also found that expandability is something we look for, but in practice we typically don't ever expand a machine. We ocassionally upgrade memory, but usually the machine just sits there till the upgrade cycle comes around. I upgrade the memory in a bunch of Blue and White G3s a while back, but only because they had 64 and the budget for replacement was thin.
Portability...
I think people are missing some points about portables and G5's. The issues are a lot bigger than just power consumption of the chip. There's also issues like power management (going down to lower consumption modes), there are issues like it is 64 bits so it uses a data path twice as wide; how much more power does that consume (or the fact that two banks of RAM are live at a time). The faster bus consumes more power, and requires that RAM consumes more power as well, not to mention all the I/O chips (which are also faster and thus more power wasteful). There's issues of locality of heat/power and moving that around (hot spots), so cooling can be an issue. And then there's the OS itself (and services) that consumer more and more power.
I would love to see a G5 performance class portable. But the other side is that OSX still sucks on power management. Even though I use OS X exclusively, I get probably half the battery life I used to on OS 9. The scheduler is richer than OS 9, but not really rich in knowing lower power/priority tasks and so on, and some is because UNIX-think and design is about running all these threads that need to be serviced (and burn mips) even if they aren't really doing much. Apple and others spend those MIPs because they have them, which is fine on a desktop, less so on a laptop. But things are getting worse; my PowerBook's fan (15") runs way too much as it is, and eats too much power and has too short a battery life. I can barely (if lucky) watch a DVD. The overheating in my other PowerBook (my wife uses) means she can get abour 20 minutes on her battery; which obviously needs to be replaced. But this is because the machines get uncomfortably hot as it is (which kills your batteries). I can't use my powerbook in clamshell mode as it is, partly because Apple doesn't think I should be able to (and thus design for it), and partly because of ventilation, but the point is that I'd like to. And so on, and so forth.
So I care about balancing the entire system, and I don't think the G5 is going to fix that. There are much more important things to productivity that raw power, and much more to power management than just the average draw of the CPU. So build me a portable that is reliable (not fragile), has a decent battery life, doesn't run the fan all the time, has a dock, works in clamshell mode, doesn't get smudges on the screen, and doesn't get so hot that it cooks its battery, and that'll go a long way.... oh yeah, and I'd also like it to be faster.
I think people are missing some points about portables and G5's. The issues are a lot bigger than just power consumption of the chip. There's also issues like power management (going down to lower consumption modes), there are issues like it is 64 bits so it uses a data path twice as wide; how much more power does that consume (or the fact that two banks of RAM are live at a time). The faster bus consumes more power, and requires that RAM consumes more power as well, not to mention all the I/O chips (which are also faster and thus more power wasteful). There's issues of locality of heat/power and moving that around (hot spots), so cooling can be an issue. And then there's the OS itself (and services) that consumer more and more power.
I would love to see a G5 performance class portable. But the other side is that OSX still sucks on power management. Even though I use OS X exclusively, I get probably half the battery life I used to on OS 9. The scheduler is richer than OS 9, but not really rich in knowing lower power/priority tasks and so on, and some is because UNIX-think and design is about running all these threads that need to be serviced (and burn mips) even if they aren't really doing much. Apple and others spend those MIPs because they have them, which is fine on a desktop, less so on a laptop. But things are getting worse; my PowerBook's fan (15") runs way too much as it is, and eats too much power and has too short a battery life. I can barely (if lucky) watch a DVD. The overheating in my other PowerBook (my wife uses) means she can get abour 20 minutes on her battery; which obviously needs to be replaced. But this is because the machines get uncomfortably hot as it is (which kills your batteries). I can't use my powerbook in clamshell mode as it is, partly because Apple doesn't think I should be able to (and thus design for it), and partly because of ventilation, but the point is that I'd like to. And so on, and so forth.
So I care about balancing the entire system, and I don't think the G5 is going to fix that. There are much more important things to productivity that raw power, and much more to power management than just the average draw of the CPU. So build me a portable that is reliable (not fragile), has a decent battery life, doesn't run the fan all the time, has a dock, works in clamshell mode, doesn't get smudges on the screen, and doesn't get so hot that it cooks its battery, and that'll go a long way.... oh yeah, and I'd also like it to be faster.
Mac OS X issues and the G5's
While melgross felt OS X was not really a money maker for Apple and the future of computing was Linux, there are two other things I feel he forgot about OS X:
First, owning the OS X is not just a chance to get money out of existing users through upgrades. Is the $135 million you quote as OS X revenue adjusted to take Apple's new units sold into account? Did Apple "charge itself" with $119 for every new unit shipped last year? This is a classic "internal transfer fee" problem for a company's books, as the hardware division is a "parasite" of the OS division. Without the OS, the hardware division would (of course!) have a harder time selling those Mac boxes.
And it also saves Apple precious dollars in not paying a license fee to an outside company for every NEW unit sold like the PC world does to MS. (I know that Linux doesn't involve this payment, but you had various arguments, and we'll get to Linux next.)
So while the OS has at times been a drag on it's finances, in most respects it is one of Apple's competitive strengths.
But the second big bonus of Apple's OS X is it's ability to be fine-tuned to both the hardware, and ultimately to the processor (after all the theme of this thread!!) We see this in Apple's push into digital video, wedding the OS, the hardware, and Final Cut Pro.
As has been discussed earlier in this thread by people far more knowledgable of the technical spec's of the 970FX, is it any coincidence that this processor, (and parts of the OS X) are fine tuned to "multimedia like" tasks?
And doesn't this ability to merge the G5 hardware and the customized OS to whatever direction Apple wants (subject to the processor's design) highlight one of the company's overall advantages? Could it do this with Linux? Should it even care about the users who want the cheapest solution, or focus on adding value to those willing to pay for a more elegent solution?
Of course Apple needs to address it's aging lineup - but it seems ironic to argue that Apple should drop OS X now that this new system they've been pushing for years has finally come of age. It's finally getting some traction, and is now being tuned to bring out the power of the G5 chip. Rather than running away from the G5, Apple needs to rush it's entire product line over to IBM and figure how to push that processor to it's limit!
While melgross felt OS X was not really a money maker for Apple and the future of computing was Linux, there are two other things I feel he forgot about OS X:
First, owning the OS X is not just a chance to get money out of existing users through upgrades. Is the $135 million you quote as OS X revenue adjusted to take Apple's new units sold into account? Did Apple "charge itself" with $119 for every new unit shipped last year? This is a classic "internal transfer fee" problem for a company's books, as the hardware division is a "parasite" of the OS division. Without the OS, the hardware division would (of course!) have a harder time selling those Mac boxes.
And it also saves Apple precious dollars in not paying a license fee to an outside company for every NEW unit sold like the PC world does to MS. (I know that Linux doesn't involve this payment, but you had various arguments, and we'll get to Linux next.)
So while the OS has at times been a drag on it's finances, in most respects it is one of Apple's competitive strengths.
But the second big bonus of Apple's OS X is it's ability to be fine-tuned to both the hardware, and ultimately to the processor (after all the theme of this thread!!) We see this in Apple's push into digital video, wedding the OS, the hardware, and Final Cut Pro.
As has been discussed earlier in this thread by people far more knowledgable of the technical spec's of the 970FX, is it any coincidence that this processor, (and parts of the OS X) are fine tuned to "multimedia like" tasks?
And doesn't this ability to merge the G5 hardware and the customized OS to whatever direction Apple wants (subject to the processor's design) highlight one of the company's overall advantages? Could it do this with Linux? Should it even care about the users who want the cheapest solution, or focus on adding value to those willing to pay for a more elegent solution?
Of course Apple needs to address it's aging lineup - but it seems ironic to argue that Apple should drop OS X now that this new system they've been pushing for years has finally come of age. It's finally getting some traction, and is now being tuned to bring out the power of the G5 chip. Rather than running away from the G5, Apple needs to rush it's entire product line over to IBM and figure how to push that processor to it's limit!
Re: Portability...
Apple's not going to take the PowerMac design and just drop it in a laptop and call it a day - and for many practical reasons they can't. Personally of the points you bring up, I think Apple will make the following compromises for a PowerBook:
1) 64-bit data paths will not go everywhere - in fact the boss to the CPU is actually dual 32-bit unidirectional, so while there are more pins in/out of the CPU this isn't really a big deal from a power perspective.
2) The PowerBooks will probably not use Dual Channel. When you have 8 RAM slots that's one thing, but when you have only two you can't really get away with filling them both at the factory. If they decide to do dual channel anyway, then we'll probably see RAM soldered on the motherboard
3) The various system busses don't have to run as fast as they do on the PowerMac. The system controller will be simpler due to not needing to support PCI/PCI-X slots (which removes the 16-bit HT link) and since the chips aren't likely to be 1.8 or 2Ghz the ram will probably be PC2700 instead of the PC3200 in the towers.
4) As for the OS, it's been getting more and more efficient, so I really can't see how it's consuming more power .
I never really spent much time in OS 9 (it's been a LONG time) but what I do remember is that some of the Power management features that gave you that longer battery life was utterly horrid from a usage point of view - OS 9 may have been aggressive about CPU cycling but it was also ignorant of the real amount of CPU you needed - you stop moving the mouse and suddenly that Photoshop operation slows to a crawl...
Apple is aware that they need to do more for Power Management on OS X, but I also think that they are doing what they can for the hardware that they no longer make. The PowerMac G5 can shed gobs of wattage (cranking down to about 1.3Ghz when idle by lowering CPU voltage). I think that a PowerBook G5 would have much the same technology in it.
You wouldn't happen to have an original PowerBook G4 with the original battery would you? It's not a heat issue, but a short-charge cycle issue that occurs when you spent most if not all of your time on the power adapter. Heat does nuke the battery, but I have not seen a PowerBook get so hot so regularly that it is an issue... Of course, if this isn't the case you are probably right .
Someday... someday...
Apple's not going to take the PowerMac design and just drop it in a laptop and call it a day - and for many practical reasons they can't. Personally of the points you bring up, I think Apple will make the following compromises for a PowerBook:
1) 64-bit data paths will not go everywhere - in fact the boss to the CPU is actually dual 32-bit unidirectional, so while there are more pins in/out of the CPU this isn't really a big deal from a power perspective.
2) The PowerBooks will probably not use Dual Channel. When you have 8 RAM slots that's one thing, but when you have only two you can't really get away with filling them both at the factory. If they decide to do dual channel anyway, then we'll probably see RAM soldered on the motherboard
3) The various system busses don't have to run as fast as they do on the PowerMac. The system controller will be simpler due to not needing to support PCI/PCI-X slots (which removes the 16-bit HT link) and since the chips aren't likely to be 1.8 or 2Ghz the ram will probably be PC2700 instead of the PC3200 in the towers.
4) As for the OS, it's been getting more and more efficient, so I really can't see how it's consuming more power
.I never really spent much time in OS 9 (it's been a LONG time) but what I do remember is that some of the Power management features that gave you that longer battery life was utterly horrid from a usage point of view - OS 9 may have been aggressive about CPU cycling but it was also ignorant of the real amount of CPU you needed - you stop moving the mouse and suddenly that Photoshop operation slows to a crawl...
Apple is aware that they need to do more for Power Management on OS X, but I also think that they are doing what they can for the hardware that they no longer make. The PowerMac G5 can shed gobs of wattage (cranking down to about 1.3Ghz when idle by lowering CPU voltage). I think that a PowerBook G5 would have much the same technology in it.
You wouldn't happen to have an original PowerBook G4 with the original battery would you? It's not a heat issue, but a short-charge cycle issue that occurs when you spent most if not all of your time on the power adapter. Heat does nuke the battery, but I have not seen a PowerBook get so hot so regularly that it is an issue... Of course, if this isn't the case you are probably right
.Someday... someday...
Re: mini tower
I don't think your analysis of OS price holds water. If Apple sold the OS for 1/3 it's current price, the revenue from that sale could support, at most, only about 900 engineers (if your being incredibly cheap with their salaries & benefits relative to California standards). There are easily that many paid workers working on Mac OS X, as well as other people whose work is paramount to getting the OS into customer hands. And finally the resellers need to make a profit too or they won't stock the product. If your only selling for $49, and the real cost is $44, then Apple's not going to make ANY profit from reseller sales.
I think a good indicator of the absolute minimum price that Apple could reasonably sell OS X for is the educational price of $69. Much less than that and resellers won't want to stock it, and Apple won't be able to make a reasonable profit.
I don't think your analysis of OS price holds water. If Apple sold the OS for 1/3 it's current price, the revenue from that sale could support, at most, only about 900 engineers (if your being incredibly cheap with their salaries & benefits relative to California standards). There are easily that many paid workers working on Mac OS X, as well as other people whose work is paramount to getting the OS into customer hands. And finally the resellers need to make a profit too or they won't stock the product. If your only selling for $49, and the real cost is $44, then Apple's not going to make ANY profit from reseller sales.
I think a good indicator of the absolute minimum price that Apple could reasonably sell OS X for is the educational price of $69. Much less than that and resellers won't want to stock it, and Apple won't be able to make a reasonable profit.
Originally posted by idkew
you make me laugh.
if only it were that easy. unfortunately, this would never happen. if apple said that in 2 weeks, there would be a new g5 book, then no one would buy a book for those 2 weeks, then apple has overstocks they can't sell........
So the current Books must not be very good? If Intel says there will be a 4 GHz chip in two months or AMD says the AMD 64 goes mobile (which already happened I know) everyone buys the old stuff because it's not bad and it's fast enough for most of us. And the older chips will be cheaper and that's the normal way it goes. I do not understand why Apple makes all this a secret. Stupid game if you ask me. This keynote stuff where everybody is naked and praise the lord.
Sure I wouldn't have buyed the G3 iBook because in my feeling it's not fast enough for OSX. And I knew that before I bought the thing but why waiting there will be no G4 in iBooks.
G5 chips in PB
Hi folks,
IMHO Apple has many alternatives to assemble a G5 PBook:
1) many of the PC manufacturers put batteries in their laptops with a charge greater than 55 Watth.
2) Apples processors did not support big variations in CPU clock. The portable G5 version could have a big range in CPU clock, maybe from 400 MHz to 1.6 GHz. That's standard for mobile pentium boxes.
3) As already mentioned Apple doesn't have to adapt the high memory bandwidth of the desktops fully. The G5 is faster than the G4 not only because of memory and CPU clock speed.
4) Last but not least the 970FX consumes less watts per computing power than the current mobile G4 generation.
5) the new graphics accelerators from ATI could help saving power too if they are properly applied to new pwerbooks.
Doesn't that sound doable?
- regards
Hi folks,
IMHO Apple has many alternatives to assemble a G5 PBook:
1) many of the PC manufacturers put batteries in their laptops with a charge greater than 55 Watth.
2) Apples processors did not support big variations in CPU clock. The portable G5 version could have a big range in CPU clock, maybe from 400 MHz to 1.6 GHz. That's standard for mobile pentium boxes.
3) As already mentioned Apple doesn't have to adapt the high memory bandwidth of the desktops fully. The G5 is faster than the G4 not only because of memory and CPU clock speed.
4) Last but not least the 970FX consumes less watts per computing power than the current mobile G4 generation.
5) the new graphics accelerators from ATI could help saving power too if they are properly applied to new pwerbooks.
Doesn't that sound doable?
- regards
But that doesn't take into account the culture difference. When Apple makes an announcement about something it tends to circulate the community a LOT faster than in the PC world, because there is a higher % of us that actually pay attention to these announcements. In the PC world you have lots more people who couldn't care less about Intel or AMD (or even know what they do) and just buy the box they think they need.
And when there are big days each year where you know that it is virtually guaranteed that something will be announced that can affect your purchases, you tend to hold off until that something emerges or you just must buy what is currently available.
It's not so much a case of no body buying what is available now when Apple announces something, or it is suspected of announcing something soon. But the sales effect is noticeable and affects their bottom-line - something which many many people are always watching and probably to Apple's chagrin.
Re: G5 chips in PB
Apple used to, and then for some reason went with smaller batteries. My TiBook has a 61 W/h battery in it. They may be a little heavier, but the extra time would probably be liked by many...
Most of this variation is needed on P-Ms because at their top rated speeds they use more power than the G4 (current wattage allowances on PC laptops are around 25w or about 2-3 times the G4 typical). Granted it's likely a typical vs peak power comparison, but that still leads one to believe that a 1.6Ghz P-M can consume as much as 20w typical - so being able to ramp down to 1/4 of your top speed makes for good average savings. The current G4s at 1Ghz however use less than 10 watts typical so being able to drop to only 667Mhz isn't as big a deal since you end up using about the same power in both cases (400 P-M vs 667 G4-7447).
Well, we know that it consumes at 1.4Ghz it consumes 12.3 watts. Something has been bugging the heck outta me about that (after rereading the sheet) it says that @ 2Ghz and 1v it is 24.5w, but @ 1.4 Ghz and 1v it is 12.3. either there is something seriously wrong with the 970 at low speeds, or they are using different voltages at both levels. I originally thought that it was a case of 1.3v @ 2Ghz and 1.1v at 1.4Ghz, but perhaps the 1.4 Ghz version is more like .85v...
Either way, it contributes to the possibility of G5 PowerBooks, but just seems like they are missing something in that sheet somewhere =).
Apple used to, and then for some reason went with smaller batteries. My TiBook has a 61 W/h battery in it. They may be a little heavier, but the extra time would probably be liked by many...
Most of this variation is needed on P-Ms because at their top rated speeds they use more power than the G4 (current wattage allowances on PC laptops are around 25w or about 2-3 times the G4 typical). Granted it's likely a typical vs peak power comparison, but that still leads one to believe that a 1.6Ghz P-M can consume as much as 20w typical - so being able to ramp down to 1/4 of your top speed makes for good average savings. The current G4s at 1Ghz however use less than 10 watts typical so being able to drop to only 667Mhz isn't as big a deal since you end up using about the same power in both cases (400 P-M vs 667 G4-7447).
Well, we know that it consumes at 1.4Ghz it consumes 12.3 watts. Something has been bugging the heck outta me about that (after rereading the sheet) it says that @ 2Ghz and 1v it is 24.5w, but @ 1.4 Ghz and 1v it is 12.3. either there is something seriously wrong with the 970 at low speeds, or they are using different voltages at both levels. I originally thought that it was a case of 1.3v @ 2Ghz and 1.1v at 1.4Ghz, but perhaps the 1.4 Ghz version is more like .85v...
Either way, it contributes to the possibility of G5 PowerBooks, but just seems like they are missing something in that sheet somewhere =).
Re: Re: G5 chips in PB
The point is: Many users don't need much power in 80% of their time. I'm mainly doing programming which means: I'm typing 80% of my time and rarely compile or debug something. Only latter tasks need processor power. I could probably get along with 200 MHz in most cases. When you're watching a DVD en-route you might need 400 MHz. What I want to say is: 1.4 GHz is mostly an overkill and dissipating batterie power. That's why I envy people with laptops that provide large scaling in clock frequencies.
Although I do not envy anyone with a non-Apple regardless if he is running Linux or WinniePoo.
The point is: Many users don't need much power in 80% of their time. I'm mainly doing programming which means: I'm typing 80% of my time and rarely compile or debug something. Only latter tasks need processor power. I could probably get along with 200 MHz in most cases. When you're watching a DVD en-route you might need 400 MHz. What I want to say is: 1.4 GHz is mostly an overkill and dissipating batterie power. That's why I envy people with laptops that provide large scaling in clock frequencies.
Although I do not envy anyone with a non-Apple regardless if he is running Linux or WinniePoo.