1999 entry price. ... The minimum config costs, and that's what you'd get, moreover the costs would be exacerbated by the miniturization penalty you have to pay to get the thing to fit into an iMac (with the better PS) or PB.
The 1.6 GHz PowerMac G5 is a start yes. So Apple can cheapen it by eliminating the PCI slots, eliminate the Hypertransport bridge chip, eliminate the another 2 DIMM slots (to make 2). They can also use previous sound systems, Firewire/Ethernet physical layers, and ATA/100 disks. So they have eliminated around 80% of the board. I don't think it is as hard as it looks.
If Apple did it right, they should have designed the system ASIC to support single channel DDR SDRAM too.
Quote:
Look at the case, yes the size is for SILENT cooling, but it's big, it's got a lot of fans, and it doesn't have a ton of room. Is it cool? Yes, compared to desktop X86, is it cooler than 7447/57? No, and the Mobo heat/cost numbers probably stack up not so well at all against plain old 7445/55.
You should really do the power consumption numbers. A 1.3 GHz 7457 runs around 17 watts. A 1.3 GHz 970 will run around 25 watts. A little bit of circuit tuning, floor plan changes, and some voltage/frequency cycling, a 130 nm 1.3 GHz 970 can go into a Powerbook.
Quote:
IBM mentioned 1.2Ghz, but with a debut at 1.6 and looking to 3Ghz in 12 months, it's a safe bet that 1.2 is just a guestimation for the purposes of Microprocessor forum presentations and not a real product. And since you can't see one, it's a much safer guess that dropping the speed to 1.2 doesn't get a chip as cool as IBM originally extrapolated from non-shipping samples.
The current PowerMac G5 machines all run the PPC 970 chips at 2/3 the advertised clock in non-demanding situations, ie, most of the time. So the 1.8 GHz PowerMac G5 runs at 1.2 GHz for most of its life. It'll upclock to 1.8 GHz in a millisecond or 2 when the situation calls for it.
Quote:
As for the cube. It's funny to me that the same people who insist that a G5 should be crammed into an iMac reject the notion of such a cube.
The idea of the cube is silent running. A wonderful idea. However it was an ergonomic nightmare of a machine. I hope Apple would use the current PowerMac G5 industrial design for a single processor, 1 AGP, 1 PCI, 1 disk, 1 optical, and 2 DIMM slot box.
I can buy 3.06HT P4 towers, WITH 17" LCD's included and seperate DVD-multi(+/-) burners and FAST CDrw with 128MB AGP video and 160GB Hdd's for 2399 CANADIAN!!!. That's 1700USD (give or take) all further deductions in USD in USD
Drop one drive, that's 50 bucks.
Cut the HDD down to 80GB, that's another 50-60 bucks
Drop the video down to the iMac's 64MB MX solution, at least 150, but closer to 200 bucks.
HT P4 and 800MHz FSB versus Ghz G4, let's say 100 to be extremely kind to Apple.
Are you seeing the picture yet?
because there's more,
The 17" LCD that come bundled with the tower costs MORE than a 17" widescreen, more surface area. iMac's 17 neatly stradles the difference between a 4:3 15" and a 5:4 17" and does so in price as well, no where near 300, guaranateed. I'm buying RETAIL 17" LCD's for 550 Canadian, and they include seperate case, PS, and analogue circuitry. iMac's LCD needs none of that, and it's a cheaper panel to begin with.
RAM, the PC comes with 512MB of DDR400, the iMac comes with half that, at 266.
Just adding up component costs shows you that the iMacs are 300-500USD too expensive, you don't have to look hard to find that a SUBSTANTIALLY CHEAPER iMac is not just doable NOW, it's long overdue. If for some reason Apple cannot lower the price because of component costs or assembly, even while getting a long life out of the same tired integrated Mobo tech, then they have a problem with their suppliers that they need to adress.
Frankly, that's none of my concern, I'm not in business to prop up Apple from their own failings. I wouldn't buy an iMac and I cannot recommend that any other savvy consumer buy one either.
That doesn't mean you have to get a PC, though, there are plenty of better deals right ther within Apple's own line-up.
What I've played with is just an offereing from the local costco, the white box builders will do considerable better.
To that end, 2.8Ghz (533fsb) machines with 64MB graphics and 120GB hdd's but also with 17" LCD's included AND SUperdrives, can be had for 1999 Canadian. Still trouncing an iMac, and still in reach of 1399. That's already 1449USD with a comparable display and superdrive from a name brand, (Gateway or HP)
In a couple of months 1399 will be more than possible. In fact, by september, such a machine will be under 1299!!!
I can buy 3.06HT P4 towers, WITH 17" LCD's included and seperate DVD-multi(+/-) burners and FAST CDrw with 128MB AGP video and 160GB Hdd's for 2399 CANADIAN!!!. That's 1700USD (give or take) all further deductions in USD in USD
Drop one drive, that's 50 bucks.
Cut the HDD down to 80GB, that's another 50-60 bucks
Drop the video down to the iMac's 64MB MX solution, at least 150, but closer to 200 bucks.
HT P4 and 800MHz FSB versus Ghz G4, let's say 100 to be extremely kind to Apple.
Are you seeing the picture yet?
What's the picture? Apple has to price its products according to what it pays for the parts it uses. You're tapping into massive economies of scale that Apple doesn't have, and razor-thin profit margins that Apple can't sustain.
Besides, you're comparing Apples and oranges. How much for an all-digital, nearly color-correct 17" LCD? Sure, you can eliminate the analog circuitry, but all the parts have to be higher quality because they no longer have anything to hide behind. The LCD panel itself is only one variable in that mix. And oh, yes, what about that lovely little arm the iMac has? How quiet does that box of yours run? How easily and reliably can you leverage the tech on board? If you can't use it reliably, effortlessly and intuitively, including it as a feature is tantamount to false advertizing.
Griping that Apple has to do better accomplishes nothing, especially when you make a point of ignoring all their advantages.
Which is not to say that I think the iMac is all hunky dory as it currently stands.
Just making the case, Apple is doing something wrong if it cant get prices to beat what I pay retail in Canadian dollars.
And as for the panels, some were digital and others were analogue. For 2399, you got a DVI panel.
If an arm and dome are costing them that much, then they desperately need to rethink their pure design focus, and start working to a happier medium between design and costs.
The PC's in question all have good reliable hardware, and work well out of the box. Some of their software isn't the best, but a quick trip to Kazaa fixes that up in an afternoon.
Just adding up component costs shows you that the iMacs are 300-500USD too expensive, you don't have to look hard to find that a SUBSTANTIALLY CHEAPER iMac is not just doable NOW, it's long overdue.
Oh, I agree with you. With the current configuration, the iMac needs to drop around $300. But your statement was to drop the cost of the iMac to eMac levels. That's $800 to $1300, which is rather ambitious. Most machines at this cost, sub $1000, usually have integrated video, at least that's the demarcation at the Dell site. For non-integrated video, the price seems to jump $200 from the integrated one and moves the machine to the $1000 range.
I also think the eMac needs to drop another $200 too, but that has a lot of assumptions in it, such as a 1 GHz G4 costing $50.
Quote:
To that end, 2.8Ghz (533fsb) machines with 64MB graphics and 120GB hdd's but also with 17" LCD's included AND SUperdrives, can be had for 1999 Canadian. Still trouncing an iMac, and still in reach of 1399. That's already 1449USD with a comparable display and superdrive from a name brand, (Gateway or HP)
That's why the iMac needs a PPC 970 chip. Powerbooks too.
You know, it's a pretty safe bet that sub 1.4 G5's don't even exist.
I wouldn't take that bet. It's all a matter of a little process and circuit tuning.
Good choice -- IBM is trying to push the 970 into the embedded space as well and their presentation a couple of weeks ago discussed 1 GHz and 1.2 GHz 970s.
Also, somebody above said "we now know that the power consumption and cost are higher than expected". I haven't seen any power consumption figures except for the 2 GHz processor, whereas the estimates from last October were based on the 1.8 GHz version. We've heard nothing at all about pricing, although since they are going to ship dual 2 GHz machines the yields must be reasonably good. Please support your statement with evidence.
Kickaha and Amorph couldn't moderate themselves out of a paper bag. Abdicate responsibility and succumb to idiocy. Two years of letting a member make personal attacks against others, then stepping aside when someone won't put up with it. Not only that but go ahead and shut down my posting priviledges but not the one making the attacks. Not even the common decency to abide by their warning (afer three days of absorbing personal attacks with no mods in sight), just shut my posting down and then say it might happen later if a certian line is crossed. Bullshit flag is flying, I won't abide by lying and coddling of liars who go off-site, create accounts differing in a single letter from my handle with the express purpose to decieve and then claim here that I did it. Everyone be warned, kim kap sol is a lying, deceitful poster.
Now I guess they should have banned me rather than just shut off posting priviledges, because kickaha and Amorph definitely aren't going to like being called to task when they thought they had it all ignored *cough* *cough* I mean under control. Just a couple o' tools.
Don't worry, as soon as my work resetting my posts is done I'll disappear forever.
Isn't shrinking Apple's chipset design (and moving the DDR controller on-chip) as big of a barrier to the iMac getting a G5 than any theoretical lower-clocked G5s?
Designing a new, smaller motherboard for the iMac and Powerbook is gonna take both time and money...
What's the picture? Apple has to price its products according to what it pays for the parts it uses. You're tapping into massive economies of scale that Apple doesn't have, and razor-thin profit margins that Apple can't sustain.
Matsu mentioned Gateway as one of the PC manufactures with way lower prices. I may be wrong here but I think Apple has a larger market share then Gateway and therefore larger economies of scale. The argument is constantly made for lowering prices to drive market share and the economies of scale.
Obviously though, Gateway's not doing all that good right now (with their razor-thin profit margins) and is a poor example of profitability/ sustainability. Apple does have nice margins on it's hardware and it seems as though they can have some flexibility in dropping prices.
Isn't shrinking Apple's chipset design (and moving the DDR controller on-chip) as big of a barrier to the iMac getting a G5 than any theoretical lower-clocked G5s?
No.
Leave the memory controller off the CPU for now. It only adds complexity to the G5 daughtercard socket. The U3 memory controller is overkill because it supports 2 PPC 970 buses and dual channel DDR SDRAM. A theoritical ASIC that only supports 1 970 bus and 1 DDR SDRAM channel would be cheaper, but Apple can maybe get the same cost reduction by doubling the number of U3 controllers it pays IBM to manufacture. One would hope that the U3 controller also supports 8 bit Hypertransport. If so, it simply uses the same K2 I/O chip the PowerMac G5 board has.
And for the 3rd time, 1 to 1.4 GHz PPC 970 CPUs may be cheaper than 1 to 1.4 GHz Moto 7457 chips. If that is the situation, Apple would be fools to use the 7457 on any of their hardware. Maybe this will force Moto to sell 7457 for less $100, I don't know, but their advertized prices are non-starters for Apple low end machines.
Quote:
Designing a new, smaller motherboard for the iMac and Powerbook is gonna take both time and money...
Time and money yes. But it should be less complicated than the PowerMac G5 motherboard and less costly. There will be a considerable reduction in traces alone: no PCI-X/PCI slots and accompanying traces, no traces for second CPU, no traces used for 4 to 6 DIMMs, no Hypertransport PCI bridge, and assuming Apple uses PC2100/PC2700/AGP4x, those traces become less expensive board-wise (lower Hz, less tightness in the board to pay for). And of course, there will be lower performance components: ATA/100 instead of SATA, AGP4x instead of AGP8x, PC2100/2700 instead of PC2700/3200, slower opticals, et al.
Matsu mentioned Gateway as one of the PC manufactures with way lower prices. I may be wrong here but I think Apple has a larger market share then Gateway and therefore larger economies of scale. The argument is constantly made for lowering prices to drive market share and the economies of scale.
Obviously though, Gateway's not doing all that good right now (with their razor-thin profit margins) and is a poor example of profitability/ sustainability. Apple does have nice margins on it's hardware and it seems as though they can have some flexibility in dropping prices.
-tink
Gateway is a good example of what happens when you are trying to live on razor thin margins and yet are selling boring commodity hardware with very little value added. They get crushed by the larger hardware provider (Dell, Compaq). Gateway is the exact model that Apple must avoid- they must provide value that justifies the extra $$. I can say that they are doing a very good job at this.
I fully realize that the longer we wait the less true any of my arguments can ring. IBM's fab will only pick up steam, .09u parts will be both cheaper and faster when they come, to say nothing of being cooler, and Apple will have time to get the controller ready to rumble for PB and eventually iMac use.
However, I put the time frame for G5 nirvana at about 12+ months for a Powerbook, and a few months afetr that for an iMac.
Lots of stuff can change betwen now and then. That may be, Apple decides to back the iMac off the high-end a bit, since they'll always be chasing their tails to get margins if they insist on stuffing higher end stuff in it. From their claims they seem unable to eek margin from it even with relatively sedate components.
Backing the iMac down into eMac territory 899-1399 (a little higher than eMac, but close enough for me) and sticking a "cube" tier between the PM and the iMac seems like a good idea to me. Especially if the cube tier offers something the iMac doesn't: 64bit "workstation" performance.
Want easy to use, non-threatening AIO? Get iMac. Want a "pro"ish power and a modicum of expansion? Get "cube"
Eventually it has to come that iMacs too will get a G5, but I think it will take some time and a little waiting by Apple to see where Moto stands with it's own PPC evolution.
I do not think that Apple wants to rely on one CPU supplier, however enticing their offerings may be, and however painful their past partnerships have been. It's just wise to see if they can keep both players in the loop.
Seeing as how IBM is also farming out fab expertise at this new plant, it may not be out of the question for a little more co-operation between AIM over the coming months.
One more time, I went through this so many times, jeez. It isn't the margins that hurt gateway. Possibly the WORST retail store strategy ever, agressively pursued at that, hurt them badly. They're still reeling. If they'd never left the barn and just kept plugging away at the direct to order model, they might not be Dell, but they'd be doing quite well. They're still a big outfit, they still get good volume prices and they still make money on their computers. They just spent that money badly, and now they're paying for it.
One more time, I went through this so many times, jeez. It isn't the margins that hurt gateway. Possibly the WORST retail store strategy ever, agressively pursued at that, hurt them badly. They're still reeling.
They are still reeling, and that's largely because they are unable to get the margins they had when they started building the stores. Dell is keeping prices at a level where Gateway's PC margins get squeezed. Even after bailing out of the stores, Gateway hasn't been able to get costs under control and with the pricing pressure from Dell and HP they can't increase margins. Which is why their new strategy is to move to products where the pricing pressure isn't as severe. Gateway may eventually get out of the PC business entirely.
Hmmm, speaking of Microprocessor Forum...I wonder what IBM will present this year?
Me thinks, as in has a strong hunch , they may talk about the fabled 980 processor.
Quote:
After all, by the time the G5 ships (let's face it) in September in volume, the MPF will be due the month after.
But Apple's MB manufacturers will have been receiving 970s since late May '03 to beginning June '03. The shipping 970s will develop more hype for the 980s.
Quote:
GPUL 2? And...what is a 'swift' transition to 0.09 anyhow? Would that be Jan' 04 San Fran'? Or Boston 04?
Some call the transition to 90nm the 970+. I've heard speculation that Sept '03 was a "target ship date", but haven't heard/seen anything recent . . . except where the lead IBM 970 engineer mentioned in the WWDC keynote that IBM already had versions of the "follow-on" processor to the 970. If you do the release math . . . the 970 production was scheduled for 2H 2003 and Apple was supposed to ship Sept 2nd, 2003, but moved it up to Aug 29th (not much I know, but essentially 2 monthes after). Now, If IBM starts 90nm 970s, in say the end of Sept . . . they can have 970+'s available for almost immediate delivery at MWSF '04. Which also happens to fit in with Apple's recent history of releasing the consumer Macs at MacWorlds. Just a thought. . .
Quote:
Looks to me that G5 case could take a 2.5 970 on .13?
. . . And for the 3rd time, 1 to 1.4 GHz PPC 970 CPUs may be cheaper than 1 to 1.4 GHz Moto 7457 chips. If that is the situation, Apple would be fools to use the 7457 on any of their hardware. . .
When discussing slower chips we could be talking about two different things. We might be running higher rated parts at minimum supply voltage, or we might be running lower clock rate parts at normal supply voltage. Let's just take IBM's early remarks about power as an example, knowing that the real numbers today will be a little different.
Back then, a 1.8 GHz 970 was said to dissipate about 42 Watts running at the normal supply voltage. Now if we use the minimum supply voltage, we can run this same part at only 1.2 GHz and it dissipates 19 Watts. On the other hand, if we use a 1.2 GHz rated part, we must use normal supply voltage at this clock rate and it dissipates 28 Watts.
So, the way we get to the lower clock rate makes a significant difference in power dissipation, 19 Watt or 28 Watts in this example. It might be helpful to think of each 970 chip having two different ratings for clock speed, one at normal supply voltage and one at minimum supply voltage.
Right. So a low power 1.2 G5 would have to be able to switch at lower voltages, which would make it expensive, not cheap. The cheaper part would be hotter, assuming that it couldn't run at lower voltages (mebbe assuming a lot)
As I said before, there is no evidence of 1.2Ghz G5 parts, that number could merely have been extrapolated from then sampling parts tested at different clock rates. It might be the case that IBM can do a 1.2 for Apple, but that it can't, as many have assumed, do it cheaper than a 1.6-2Ghz stuff. Not yet.
The 7457 looks underwhelming, according to the PDF floating around, but it may actually be the only option for many of Apple's products for the next 12 months.
As time passes and prices drop and Apple gets time to change things up, then G5's are a natural evolution for the PB's first, and some time later, mebbe the iMac's (depending on whether they intend to keep them relatively upmarket, or to drop the prices.
Comments
Originally posted by I, Fred
I never claimed to have any evidence, only an idea to get IBM more money so they would see the profitability of Fishkill, 970's and so forth.
Well, you started out with this gem:
"IBM is disappointed with 970 demand"
And we're still waiting for a link.
If 970 production is the 'savior' of Fishkill in your mind, why not (instead of a G5 iMac) say something like:
"Dual processors across the line at current price points!"
That's something we could all get behind. Even Matsu.
Originally posted by Matsu
1999 entry price. ... The minimum config costs, and that's what you'd get, moreover the costs would be exacerbated by the miniturization penalty you have to pay to get the thing to fit into an iMac (with the better PS) or PB.
The 1.6 GHz PowerMac G5 is a start yes. So Apple can cheapen it by eliminating the PCI slots, eliminate the Hypertransport bridge chip, eliminate the another 2 DIMM slots (to make 2). They can also use previous sound systems, Firewire/Ethernet physical layers, and ATA/100 disks. So they have eliminated around 80% of the board. I don't think it is as hard as it looks.
If Apple did it right, they should have designed the system ASIC to support single channel DDR SDRAM too.
Look at the case, yes the size is for SILENT cooling, but it's big, it's got a lot of fans, and it doesn't have a ton of room. Is it cool? Yes, compared to desktop X86, is it cooler than 7447/57? No, and the Mobo heat/cost numbers probably stack up not so well at all against plain old 7445/55.
You should really do the power consumption numbers. A 1.3 GHz 7457 runs around 17 watts. A 1.3 GHz 970 will run around 25 watts. A little bit of circuit tuning, floor plan changes, and some voltage/frequency cycling, a 130 nm 1.3 GHz 970 can go into a Powerbook.
IBM mentioned 1.2Ghz, but with a debut at 1.6 and looking to 3Ghz in 12 months, it's a safe bet that 1.2 is just a guestimation for the purposes of Microprocessor forum presentations and not a real product. And since you can't see one, it's a much safer guess that dropping the speed to 1.2 doesn't get a chip as cool as IBM originally extrapolated from non-shipping samples.
The current PowerMac G5 machines all run the PPC 970 chips at 2/3 the advertised clock in non-demanding situations, ie, most of the time. So the 1.8 GHz PowerMac G5 runs at 1.2 GHz for most of its life. It'll upclock to 1.8 GHz in a millisecond or 2 when the situation calls for it.
As for the cube. It's funny to me that the same people who insist that a G5 should be crammed into an iMac reject the notion of such a cube.
The idea of the cube is silent running. A wonderful idea. However it was an ergonomic nightmare of a machine. I hope Apple would use the current PowerMac G5 industrial design for a single processor, 1 AGP, 1 PCI, 1 disk, 1 optical, and 2 DIMM slot box.
I can buy 3.06HT P4 towers, WITH 17" LCD's included and seperate DVD-multi(+/-) burners and FAST CDrw with 128MB AGP video and 160GB Hdd's for 2399 CANADIAN!!!. That's 1700USD (give or take) all further deductions in USD in USD
Drop one drive, that's 50 bucks.
Cut the HDD down to 80GB, that's another 50-60 bucks
Drop the video down to the iMac's 64MB MX solution, at least 150, but closer to 200 bucks.
HT P4 and 800MHz FSB versus Ghz G4, let's say 100 to be extremely kind to Apple.
Are you seeing the picture yet?
because there's more,
The 17" LCD that come bundled with the tower costs MORE than a 17" widescreen, more surface area. iMac's 17 neatly stradles the difference between a 4:3 15" and a 5:4 17" and does so in price as well, no where near 300, guaranateed. I'm buying RETAIL 17" LCD's for 550 Canadian, and they include seperate case, PS, and analogue circuitry. iMac's LCD needs none of that, and it's a cheaper panel to begin with.
RAM, the PC comes with 512MB of DDR400, the iMac comes with half that, at 266.
Just adding up component costs shows you that the iMacs are 300-500USD too expensive, you don't have to look hard to find that a SUBSTANTIALLY CHEAPER iMac is not just doable NOW, it's long overdue. If for some reason Apple cannot lower the price because of component costs or assembly, even while getting a long life out of the same tired integrated Mobo tech, then they have a problem with their suppliers that they need to adress.
Frankly, that's none of my concern, I'm not in business to prop up Apple from their own failings. I wouldn't buy an iMac and I cannot recommend that any other savvy consumer buy one either.
That doesn't mean you have to get a PC, though, there are plenty of better deals right ther within Apple's own line-up.
What I've played with is just an offereing from the local costco, the white box builders will do considerable better.
To that end, 2.8Ghz (533fsb) machines with 64MB graphics and 120GB hdd's but also with 17" LCD's included AND SUperdrives, can be had for 1999 Canadian. Still trouncing an iMac, and still in reach of 1399. That's already 1449USD with a comparable display and superdrive from a name brand, (Gateway or HP)
In a couple of months 1399 will be more than possible. In fact, by september, such a machine will be under 1299!!!
Apple needs to get off their asses, NOW!!!
Originally posted by Matsu
No it is possible and then some.
I can buy 3.06HT P4 towers, WITH 17" LCD's included and seperate DVD-multi(+/-) burners and FAST CDrw with 128MB AGP video and 160GB Hdd's for 2399 CANADIAN!!!. That's 1700USD (give or take) all further deductions in USD in USD
Drop one drive, that's 50 bucks.
Cut the HDD down to 80GB, that's another 50-60 bucks
Drop the video down to the iMac's 64MB MX solution, at least 150, but closer to 200 bucks.
HT P4 and 800MHz FSB versus Ghz G4, let's say 100 to be extremely kind to Apple.
Are you seeing the picture yet?
What's the picture? Apple has to price its products according to what it pays for the parts it uses. You're tapping into massive economies of scale that Apple doesn't have, and razor-thin profit margins that Apple can't sustain.
Besides, you're comparing Apples and oranges. How much for an all-digital, nearly color-correct 17" LCD? Sure, you can eliminate the analog circuitry, but all the parts have to be higher quality because they no longer have anything to hide behind. The LCD panel itself is only one variable in that mix. And oh, yes, what about that lovely little arm the iMac has? How quiet does that box of yours run? How easily and reliably can you leverage the tech on board? If you can't use it reliably, effortlessly and intuitively, including it as a feature is tantamount to false advertizing.
Griping that Apple has to do better accomplishes nothing, especially when you make a point of ignoring all their advantages.
Which is not to say that I think the iMac is all hunky dory as it currently stands.
And as for the panels, some were digital and others were analogue. For 2399, you got a DVI panel.
If an arm and dome are costing them that much, then they desperately need to rethink their pure design focus, and start working to a happier medium between design and costs.
The PC's in question all have good reliable hardware, and work well out of the box. Some of their software isn't the best, but a quick trip to Kazaa fixes that up in an afternoon.
Originally posted by Matsu
Just adding up component costs shows you that the iMacs are 300-500USD too expensive, you don't have to look hard to find that a SUBSTANTIALLY CHEAPER iMac is not just doable NOW, it's long overdue.
Oh, I agree with you. With the current configuration, the iMac needs to drop around $300. But your statement was to drop the cost of the iMac to eMac levels. That's $800 to $1300, which is rather ambitious. Most machines at this cost, sub $1000, usually have integrated video, at least that's the demarcation at the Dell site. For non-integrated video, the price seems to jump $200 from the integrated one and moves the machine to the $1000 range.
I also think the eMac needs to drop another $200 too, but that has a lot of assumptions in it, such as a 1 GHz G4 costing $50.
To that end, 2.8Ghz (533fsb) machines with 64MB graphics and 120GB hdd's but also with 17" LCD's included AND SUperdrives, can be had for 1999 Canadian. Still trouncing an iMac, and still in reach of 1399. That's already 1449USD with a comparable display and superdrive from a name brand, (Gateway or HP)
That's why the iMac needs a PPC 970 chip.
Originally posted by THT
Mastu:
Quote:
You know, it's a pretty safe bet that sub 1.4 G5's don't even exist.
I wouldn't take that bet. It's all a matter of a little process and circuit tuning.
Good choice -- IBM is trying to push the 970 into the embedded space as well and their presentation a couple of weeks ago discussed 1 GHz and 1.2 GHz 970s.
Also, somebody above said "we now know that the power consumption and cost are higher than expected". I haven't seen any power consumption figures except for the 2 GHz processor, whereas the estimates from last October were based on the 1.8 GHz version. We've heard nothing at all about pricing, although since they are going to ship dual 2 GHz machines the yields must be reasonably good. Please support your statement with evidence.
Now I guess they should have banned me rather than just shut off posting priviledges, because kickaha and Amorph definitely aren't going to like being called to task when they thought they had it all ignored *cough* *cough* I mean under control. Just a couple o' tools.
Don't worry, as soon as my work resetting my posts is done I'll disappear forever.
Originally posted by Programmer
We've heard nothing at all about pricing, although since they are going to ship dual 2 GHz machines the yields must be reasonably good.
Well, they've got to be cheaper than Motorola's 7457. Now that's an expensive chip!
After all, by the time the G5 ships (let's face it) in September in volume, the MPF will be due the month after.
GPUL 2? And...what is a 'swift' transition to 0.09 anyhow? Would that be Jan' 04 San Fran'? Or Boston 04?
Looks to me that G5 case could take a 2.5 970 on .13?
Lemon Bon Bon
Designing a new, smaller motherboard for the iMac and Powerbook is gonna take both time and money...
Originally posted by Amorph
What's the picture? Apple has to price its products according to what it pays for the parts it uses. You're tapping into massive economies of scale that Apple doesn't have, and razor-thin profit margins that Apple can't sustain.
Matsu mentioned Gateway as one of the PC manufactures with way lower prices. I may be wrong here but I think Apple has a larger market share then Gateway and therefore larger economies of scale. The argument is constantly made for lowering prices to drive market share and the economies of scale.
Obviously though, Gateway's not doing all that good right now (with their razor-thin profit margins) and is a poor example of profitability/ sustainability. Apple does have nice margins on it's hardware and it seems as though they can have some flexibility in dropping prices.
-tink
Originally posted by Gizzmonic
Isn't shrinking Apple's chipset design (and moving the DDR controller on-chip) as big of a barrier to the iMac getting a G5 than any theoretical lower-clocked G5s?
No.
Leave the memory controller off the CPU for now. It only adds complexity to the G5 daughtercard socket. The U3 memory controller is overkill because it supports 2 PPC 970 buses and dual channel DDR SDRAM. A theoritical ASIC that only supports 1 970 bus and 1 DDR SDRAM channel would be cheaper, but Apple can maybe get the same cost reduction by doubling the number of U3 controllers it pays IBM to manufacture. One would hope that the U3 controller also supports 8 bit Hypertransport. If so, it simply uses the same K2 I/O chip the PowerMac G5 board has.
And for the 3rd time, 1 to 1.4 GHz PPC 970 CPUs may be cheaper than 1 to 1.4 GHz Moto 7457 chips. If that is the situation, Apple would be fools to use the 7457 on any of their hardware. Maybe this will force Moto to sell 7457 for less $100, I don't know, but their advertized prices are non-starters for Apple low end machines.
Designing a new, smaller motherboard for the iMac and Powerbook is gonna take both time and money...
Time and money yes. But it should be less complicated than the PowerMac G5 motherboard and less costly. There will be a considerable reduction in traces alone: no PCI-X/PCI slots and accompanying traces, no traces for second CPU, no traces used for 4 to 6 DIMMs, no Hypertransport PCI bridge, and assuming Apple uses PC2100/PC2700/AGP4x, those traces become less expensive board-wise (lower Hz, less tightness in the board to pay for). And of course, there will be lower performance components: ATA/100 instead of SATA, AGP4x instead of AGP8x, PC2100/2700 instead of PC2700/3200, slower opticals, et al.
Originally posted by tink
Matsu mentioned Gateway as one of the PC manufactures with way lower prices. I may be wrong here but I think Apple has a larger market share then Gateway and therefore larger economies of scale. The argument is constantly made for lowering prices to drive market share and the economies of scale.
Obviously though, Gateway's not doing all that good right now (with their razor-thin profit margins) and is a poor example of profitability/ sustainability. Apple does have nice margins on it's hardware and it seems as though they can have some flexibility in dropping prices.
-tink
Gateway is a good example of what happens when you are trying to live on razor thin margins and yet are selling boring commodity hardware with very little value added. They get crushed by the larger hardware provider (Dell, Compaq). Gateway is the exact model that Apple must avoid- they must provide value that justifies the extra $$. I can say that they are doing a very good job at this.
However, I put the time frame for G5 nirvana at about 12+ months for a Powerbook, and a few months afetr that for an iMac.
Lots of stuff can change betwen now and then. That may be, Apple decides to back the iMac off the high-end a bit, since they'll always be chasing their tails to get margins if they insist on stuffing higher end stuff in it. From their claims they seem unable to eek margin from it even with relatively sedate components.
Backing the iMac down into eMac territory 899-1399 (a little higher than eMac, but close enough for me) and sticking a "cube" tier between the PM and the iMac seems like a good idea to me. Especially if the cube tier offers something the iMac doesn't: 64bit "workstation" performance.
Want easy to use, non-threatening AIO? Get iMac. Want a "pro"ish power and a modicum of expansion? Get "cube"
Eventually it has to come that iMacs too will get a G5, but I think it will take some time and a little waiting by Apple to see where Moto stands with it's own PPC evolution.
I do not think that Apple wants to rely on one CPU supplier, however enticing their offerings may be, and however painful their past partnerships have been. It's just wise to see if they can keep both players in the loop.
Seeing as how IBM is also farming out fab expertise at this new plant, it may not be out of the question for a little more co-operation between AIM over the coming months.
Originally posted by Matsu
One more time, I went through this so many times, jeez. It isn't the margins that hurt gateway. Possibly the WORST retail store strategy ever, agressively pursued at that, hurt them badly. They're still reeling.
They are still reeling, and that's largely because they are unable to get the margins they had when they started building the stores. Dell is keeping prices at a level where Gateway's PC margins get squeezed. Even after bailing out of the stores, Gateway hasn't been able to get costs under control and with the pricing pressure from Dell and HP they can't increase margins. Which is why their new strategy is to move to products where the pricing pressure isn't as severe. Gateway may eventually get out of the PC business entirely.
Originally posted by Lemon Bon Bon
Hmmm, speaking of Microprocessor Forum...I wonder what IBM will present this year?
Me thinks, as in has a strong hunch
After all, by the time the G5 ships (let's face it) in September in volume, the MPF will be due the month after.
But Apple's MB manufacturers will have been receiving 970s since late May '03 to beginning June '03. The shipping 970s will develop more hype for the 980s.
GPUL 2? And...what is a 'swift' transition to 0.09 anyhow? Would that be Jan' 04 San Fran'? Or Boston 04?
Some call the transition to 90nm the 970+. I've heard speculation that Sept '03 was a "target ship date", but haven't heard/seen anything recent . . . except where the lead IBM 970 engineer mentioned in the WWDC keynote that IBM already had versions of the "follow-on" processor to the 970. If you do the release math . . . the 970 production was scheduled for 2H 2003 and Apple was supposed to ship Sept 2nd, 2003, but moved it up to Aug 29th (not much I know, but essentially 2 monthes after). Now, If IBM starts 90nm 970s, in say the end of Sept . . . they can have 970+'s available for almost immediate delivery at MWSF '04. Which also happens to fit in with Apple's recent history of releasing the consumer Macs at MacWorlds. Just a thought. . .
Looks to me that G5 case could take a 2.5 970 on .13?
Lemon Bon Bon
That's probably as high as it will go, for now.
Originally posted by THT
. . . And for the 3rd time, 1 to 1.4 GHz PPC 970 CPUs may be cheaper than 1 to 1.4 GHz Moto 7457 chips. If that is the situation, Apple would be fools to use the 7457 on any of their hardware. . .
When discussing slower chips we could be talking about two different things. We might be running higher rated parts at minimum supply voltage, or we might be running lower clock rate parts at normal supply voltage. Let's just take IBM's early remarks about power as an example, knowing that the real numbers today will be a little different.
Back then, a 1.8 GHz 970 was said to dissipate about 42 Watts running at the normal supply voltage. Now if we use the minimum supply voltage, we can run this same part at only 1.2 GHz and it dissipates 19 Watts. On the other hand, if we use a 1.2 GHz rated part, we must use normal supply voltage at this clock rate and it dissipates 28 Watts.
So, the way we get to the lower clock rate makes a significant difference in power dissipation, 19 Watt or 28 Watts in this example. It might be helpful to think of each 970 chip having two different ratings for clock speed, one at normal supply voltage and one at minimum supply voltage.
As I said before, there is no evidence of 1.2Ghz G5 parts, that number could merely have been extrapolated from then sampling parts tested at different clock rates. It might be the case that IBM can do a 1.2 for Apple, but that it can't, as many have assumed, do it cheaper than a 1.6-2Ghz stuff. Not yet.
The 7457 looks underwhelming, according to the PDF floating around, but it may actually be the only option for many of Apple's products for the next 12 months.
As time passes and prices drop and Apple gets time to change things up, then G5's are a natural evolution for the PB's first, and some time later, mebbe the iMac's (depending on whether they intend to keep them relatively upmarket, or to drop the prices.