I don't think there need to be a separate line for a 'mini' G5. Take the current Powermac line, and for the next revision simply make the low end version a cut down one.
Leave just 1 AGP/1PCI-X for expansion. Trim the processor bay down to a single processor /HS unit in height, 2 memory slots (DDR400). Stick with 2 HD slots/ 1 optical. Stick a single G5 2.0 in it.
For the standard machine, simply turn the HD's through 90 degrees and shuffle slightly to make space for 4 HDD instead of 2, and most of the cmplints about the box would disappear. Alternatively, add the second optical drive and two more hdd to the stack. Go all duals, lets say 2x2.0(DDR 400), 2x2.5(DDR500).
End up with:
2x2.5 G5 @ 2999
2x2.0 G5 @ 2399
1x2.0 G5L @ 1799
What this would do is provide a differentiated lower end that has more of a USP that the current one, and a more capable normal model for those who want that.
It sure does, but a faster CPU is way more expensive! For the AMD Barton XP CPU the 3 GHz cost more two and half time as much as the 2.5 GHz one. So while having a more complex motherboard dual sockets, two CPUs etc do cost more the skyrocketing cost of the leading edge CPUs off set a large part of that cost.
If Apple now let the eMac linger in the G4 domain for the next year as low cost alternative and have G5 iMacs then the option of having the towers all dual again looks quite natrual to Apple and the currently steep price would look better for us buyers
So my suggestion is:
DP 1.6 (a substantial boost aviable right away)
DP 2.0 (a substantial boost aviable right away)
DP 2.x (as fast as IBM can crank them up, aviable within a resonable time frame)
Good point, but we dont know really the real price of CPU. AMD and INTEL sell at very high price their high clocked CPU. But i suspect they have a far better margin on the high clock CPU than the low ones. I suspect that this way of make the price is exagerated. IBM and Apple have a different agrement. We don't know it, but i am sure that the increase in price is more linear than the exponential one in the PC market.
I really don't see Apple bringing back the cube or coming out with a mini. The only reason to add to the midrange of the line up is to sell significantly more computers. It bad for Apple to sell fewer powermacs or imacs by replacing them with cubes/minis. This makes it hard for them to manage supplies and reduces their volumn purchases of components. It could work if they made a killing on the new product, but they tried that with the cube didn't they.
I see the midrange as having many options right now: The imac for AIO buyers-its really an improvement of the original cube IMHO. The imac becomes even more of an attractive buy if they stuff a G5 in it. The low end Powermac if you want expansion. Also, laptops are being used by many as a small footprint desktop. Again, if they get a G5 in a Powerbook than you have a "mini."
The only place I can see the need for a cube is at the bottom end. A G4 cube, i.e. headless eMac, that could expand the number of sales to those looking for a cheap computer. This would work well for those with a monitor already and wouldn't buy another computer unless they could use their old monitor (buyers on a tight budget). However, I doubt Apple could make much profit on such a cube and keep it price well below an eMac. It would also look a lot like a simple update to the original cube-not very ground breaking and could result in bad PR. I doubt it will ever happen.
As for staying on topic:
I guess in the first quarter of next year we'll see SP 1.8, DP 2.0 and DP 2.5 at the current price structure.
True, but having more transistors and having them more densly packed has to affect production costs. If I remember correctly, it has been mentioned that IBM utilizes some automated design tools to speed up the actual design. This might explain the less densly packed transistors??
No comment here, I've not a clue, well maybe, but no point in appearing to bash Motorola. More densly packed transistors/traces may affect yields??
Kind of my point. The MPC7455 is still on the 0.18µm process and unless Motorola is having a fire sale or their yields are tremedous, the IBM 970 should be price competative with the MPC7455/7445 and the MPC7457/7447.
I just can't see why using the G4's, in and of itself, would allow Apple to produce computers cheaper than using the G5. Unless of course all the other stuff, companion chip, Hypertransport, etc. disproportionately increases the cost over the G4's equivalents.
The cost to manufacture something isn't the only thing that determines the price it sells for. For example the 3.2 Ghz P4 costs Intel the same exact amout to make as does the 2.6 Ghz P4 but Intel sells the 3.2Ghz chip at a much higher price. Actually the 2.6 Ghz chip costs more to make since it has to go through the testers a couple more times.
The probablility that the 7455 is cheaper than the 970 isn't really anything new. Every time a new generation of chips come out the older chips drop in price, often by a substantial amount. There are some factors that could possibly change all this. IBM could be offering Apple some exceptionally good prices on the 970, which I doubt. The other is that Motorola is somewhat screwed up at the moment.
The cost to manufacture something isn't the only thing that determines the price it sells for. For example the 3.2 Ghz P4 costs Intel the same exact amout to make as does the 2.6 Ghz P4 but Intel sells the 3.2Ghz chip at a much higher price. Actually the 2.6 Ghz chip costs more to make since it has to go through the testers a couple more times.
The probablility that the 7455 is cheaper than the 970 isn't really anything new. Every time a new generation of chips come out the older chips drop in price, often by a substantial amount. There are some factors that could possibly change all this. IBM could be offering Apple some exceptionally good prices on the 970, which I doubt. The other is that Motorola is somewhat screwed up at the moment.
This is just wrong. There are only so many chips per manufacturing 'unit' that are able to run at 3.2 GHz while bein operational. Similarily, there are a lot more cpus that run stable at 2.6 GHz, making its yield better, and hence cheaper.
True that the material might be cheaper, but the fastest chip is still more expensive...
The price of CPUs are really just arbitary marketing prices. By having really high prices on the fastert CPUs IBM/AMD/Intel stear the consumer to the midrange were they have the highest yield and have the highest margins on the fastest ones.
If you get a batch of 100 CPUs for 10 000 dollars the cost really is 100 dollar/ CPU. If they all would run a 2 GHz they would have the same price. Now if they do not you try to jack up the price of the fastest CPU as much as the market can take. The lower yield the smaller segment of the market you are aiming at and the higher price you can extract
The "science" part of the price is not that difficult to calculate in terms of R&D yields, die size and so on. But that only set the basal level of the price (no not sell at a loss). The real price is more of making as good ROI as the consumer can be billed. If IBM could make CPUs at 1% of current cost would they sell them at 1% the price? Hell no, they would just undercut the competition as little as possible to gain market share for maximum profits.
A while back there was a development conference in China regarding there future investments in computer technology, and of course IBM was there and showed a presentation. Read it [page 23] and you you'll see that he list the (?current?) IBM PPC 970 to clock up to 2.5Ghz.
The only place I can see the need for a cube is at the bottom end. A G4 cube, i.e. headless eMac, that could expand the number of sales to those looking for a cheap computer.[/B]
One of the things I've been trying to figure out recently is how possible it is to do a case-mod on an eMac. It looks, from everything I've seen, not terribly practical - circuit boards and key parts being difficult to get to.
Maybe Apple's best-of-both-worlds option would be to have the non-monitor part of the next generation eMac easily seperatable. That way they can sell the thing as one unit (appealing to the AIO market), but people who do not want the bulk of a monitor they have no intention of using can easily discard it.
This doesn't decrease the price any, but I suspect those calling for a monitor-less eMac in order to simply decrease prices are not looking at as much of a cost saving as they realise.
A good 17" monitor can be bought for just above 100 dollars and that include powercable, onof button and a case all that stuff is shared in the eMac so the price reduction of a headless eMac would be even less than that.
Both the headless eMac and the small tower design sound like a good idea but there is a risk of leaving the clean pro/home portable/stationary quadrants and stray into a product jungle as bad as the LC, performa Quadra Centrino days...
The over priced underpowered iMac is a greater problem than the lack of a headless eMac. The current top of the line iMac is not that mcuh faster than the budget portable iBook... Apple have to first get the current lines in good shape before they add others to the mix.\
With the pro towers are all duals, the servers using G5 and the iMac having fast G5 substantially outperforming the portable and the eMac then they can add a Minitower for gaming students and other that want more than the iMac but not the big tower. They can have a Single CPU tower (quadra 700 size or less but G5 cheese grater design) AGP slot, one PCI slot one optical and one HD. It would enable users to upgrade the graphcial card and also get various funcionalities from the PCI card.
I don't think there need to be a separate line for a 'mini' G5. Take the current Powermac line, and for the next revision simply make the low end version a cut down one.
Leave just 1 AGP/1PCI-X for expansion. Trim the processor bay down to a single processor /HS unit in height, 2 memory slots (DDR400). Stick with 2 HD slots/ 1 optical. Stick a single G5 2.0 in it.
This is precisely what I had in mine for the G5 mini. Essentially, take the 1.6 GHz PowerMac G5 board, remove 2 PCI slots, and shrink the board down to half the size. Then wrap a 12x12x8 inch box of the same design around it.
Same deal with a G4 mini. Take the current El Capitan case and board, remove 3 PCI slots, 1 external bay, and shrink the El Capitan case down to 12x12x9.
They are however separate lines of "computer" even though though use the same components.
Quote:
End up with:
2x2.5 G5 @ 2999
2x2.0 G5 @ 2399
1x2.0 G5L @ 1799
What this would do is provide a differentiated lower end that has more of a USP that the current one, and a more capable normal model for those who want that.
A lot of people have a single processor model for the PowerMac G5. Given the existence of a G5 mini, most assuredly with 1 processor, that'll make a single processor PowerMac G5 and its cost rather undesireable. If Apple doesn't have a G5 mini line, well, forget what I just said.
It hardly becomes obsolete. Not leading-edge, yes... but not obsolete and certainly not any less capable than when you bought it.
True. I guess obsolete is the wrong word. "Dated" maybe? Anyway, I'm sure the machine I'm going to get will more than adequately meet my needs for at least 2.5 years. Probably 3+ unless something drastic happens to software requirements in the interim.
With what I'm seeing from Adobe, I can't see myself upgrading next time around (CS part deux) unless there are some *really* compelling reasons to do so. Feature-wise I think the CS Suite has reached a point where, for my particular work, I won't need anything more for a good while.
Not unless Photoshop adds in some professional grain removal tools for example. That's if the 3rd party tools out there don't improve in the meantime. There are some pretty effective ones out there, but none that I've found to be particularly intuitive or efficient.
A while back there was a development conference in China regarding there future investments in computer technology, and of course IBM was there and showed a presentation. Read it [page 23] and you you'll see that he list the (?current?) IBM PPC 970 to clock up to 2.5Ghz.
Interesting find if true, Eric_Z. That particular page is gone now; would you have a copy of it?
DrBoar wrote:
Quote:
Both the headless eMac and the small tower design sound like a good idea but there is a risk of leaving the clean pro/home portable/stationary quadrants and stray into a product jungle as bad as the LC, performa Quadra Centrino days...
You meant Centris, not Centrino, just for the record.
I really hate to break the news to you but if you cut up the mother board, offer different options and a new case you pretty much have a new line. Sure this product may just offer the essentials but in a sense so does the IMac line.
What I was hoping for was a minimal system that would almost be considered a transportable that would use an external display device. The idea is zero impact on the desktop and reasonablly close to state of the art pefromance.
I still think this ( a mini) is extremely doable if a couple of things come to fruition. Key here is a 970 based processor with a built in memory interface, and a low cost high performance interface to a I/O chip. The I/O chip should support nothing more than is required to deliver a computational machine to the user. I'm even thinking that an integrated GPU on this I/O chip may be doable.
So in the end you have two or three major chips, memory and the associated buffering logic on the motherboard. Very low costs and with a enough modern ports (Firewire and USB) very flexible.
Thanks
Dave
Quote:
Originally posted by THT
This is precisely what I had in mine for the G5 mini. Essentially, take the 1.6 GHz PowerMac G5 board, remove 2 PCI slots, and shrink the board down to half the size. Then wrap a 12x12x8 inch box of the same design around it.
Same deal with a G4 mini. Take the current El Capitan case and board, remove 3 PCI slots, 1 external bay, and shrink the El Capitan case down to 12x12x9.
They are however separate lines of "computer" even though though use the same components.
A lot of people have a single processor model for the PowerMac G5. Given the existence of a G5 mini, most assuredly with 1 processor, that'll make a single processor PowerMac G5 and its cost rather undesireable. If Apple doesn't have a G5 mini line, well, forget what I just said.
For me, the iMac line replaced the desktop models of long ago (powermac 7200-7600). At that time they had desktops and towers - consumer and pro models. I purchased a 7500 because it was the first consumer mac to have PCI slots, nice for upgrading.
I purchase 2 PCI cards and considered a third. The first was for USB and Firewire. The second was for a good graphics card. The other slot, which would have been for a video or audio converter didn't get filled because I realized my system would have been to slow to do me any good with it.
I also upgraded the processor to the best I could, which only made the system viable for 1 more year. That happened over about 4 years.
My point with all of this is, a computer, pro or consumer will only last 5 years max in my opinion before software advances make it unusable. A consumer grade computer, like the iMac, wouldn't really need PCI slots, only pro users need audio/video PCI cards, and all graphics cards are now AGP based. New formats of I/O don't work as well as they should through a PCI bus anyway, so they can wait. So I don't believe PCI slots are necessary on a consumer grade model, but an AGP slot is essential. I realize an upgradible processessor is becoming more difficult with the new systems, but it would help consumers believe their system had more longevity to it even if it's no more than 1 year.
For the most part I see very few problems with the current iMac, except for the price/performance ratio, but that's another story. As far as it being headless, you know I'd love that. I think everyone would. Being able to choose your own display - size and price - is a very personal decision when buying a computer. It's what you stare at for hours each day, not the case. Apple would be wise to return this option to the consumer if they want consumer sales. I'm sure Ives could figure out a way to make the current LCD's on the swivel arm swappable. I really like being able to adjust the screen on my LCD iMac with just a touch, and do so very often.
I really hate to break the news to you but if you cut up the mother board, offer different options and a new case you pretty much have a new line. Sure this product may just offer the essentials but in a sense so does the IMac line.
That's pretty much what I would to see. Actually I would like to see is a 1.8 GHz 17" Powerbook G5, I'm fine with it being 1.5+ inch thick, with a 1680x1120 resolution screen.
Quote:
What I was hoping for was a minimal system that would almost be considered a transportable that would use an external display device. The idea is zero impact on the desktop and reasonablly close to state of the art pefromance.
That's pretty much what I would to see. Actually I would like to see is a 1.8 GHz 17" Powerbook G5, I'm fine with it being 1.5+ inch thick, with a 1680x1120 resolution screen.
You want a laptop without a screen?
Well not exactly, more of an IMac without the screen. Also without the cpu limitations that seem to crop up in the IMac line. In other words a 2.5 GHz 970 in a small box with just the ports it needs to run. Most likely that would be SATA, Firewire and USB. The unit could potentially also have Blutooth or other wireless technology. The idea is to have plenty of Both Firewire and USB ports available. The unit should take either standard AGP GPU's or possibly a Hypertransport based GPU.
The whole idea is to get really good CPU peformance and GPU peformance and drop the other crud. Mind you I believe the IMac is close to providing what is needed, but there are a couple of problems. One as already mentioned is the ability to upgrade the GPU or atleast have options. Another problem is the limited performance of the IMac, the lack of a screen selection and memory expansion.
I look at it this way the G5 is too limited in its expansion capabilities for its price and size. Eventually the G5 will have to be addressed or Apple is likely to see a huge slow down in sales as the early adopters satisfy themselves. Once the G5 becomes a proper tower then a limited expansion potential machine will fill that void.
Getting a larger tower and still lose one optical bay and two hard disk bays
And I do not care for getting a stack of FW boxes. The whole point of having a big tower is to get rid of the clutter. Otherwise I could have a iMac and then daisychain what ever I wish for
Comments
Leave just 1 AGP/1PCI-X for expansion. Trim the processor bay down to a single processor /HS unit in height, 2 memory slots (DDR400). Stick with 2 HD slots/ 1 optical. Stick a single G5 2.0 in it.
For the standard machine, simply turn the HD's through 90 degrees and shuffle slightly to make space for 4 HDD instead of 2, and most of the cmplints about the box would disappear. Alternatively, add the second optical drive and two more hdd to the stack. Go all duals, lets say 2x2.0(DDR 400), 2x2.5(DDR500).
End up with:
2x2.5 G5 @ 2999
2x2.0 G5 @ 2399
1x2.0 G5L @ 1799
What this would do is provide a differentiated lower end that has more of a USP that the current one, and a more capable normal model for those who want that.
Originally posted by DrBoar
It sure does, but a faster CPU is way more expensive! For the AMD Barton XP CPU the 3 GHz cost more two and half time as much as the 2.5 GHz one. So while having a more complex motherboard dual sockets, two CPUs etc do cost more the skyrocketing cost of the leading edge CPUs off set a large part of that cost.
If Apple now let the eMac linger in the G4 domain for the next year as low cost alternative and have G5 iMacs then the option of having the towers all dual again looks quite natrual to Apple and the currently steep price would look better for us buyers
So my suggestion is:
DP 1.6 (a substantial boost aviable right away)
DP 2.0 (a substantial boost aviable right away)
DP 2.x (as fast as IBM can crank them up, aviable within a resonable time frame)
Good point, but we dont know really the real price of CPU. AMD and INTEL sell at very high price their high clocked CPU. But i suspect they have a far better margin on the high clock CPU than the low ones. I suspect that this way of make the price is exagerated. IBM and Apple have a different agrement. We don't know it, but i am sure that the increase in price is more linear than the exponential one in the PC market.
I see the midrange as having many options right now: The imac for AIO buyers-its really an improvement of the original cube IMHO. The imac becomes even more of an attractive buy if they stuff a G5 in it. The low end Powermac if you want expansion. Also, laptops are being used by many as a small footprint desktop. Again, if they get a G5 in a Powerbook than you have a "mini."
The only place I can see the need for a cube is at the bottom end. A G4 cube, i.e. headless eMac, that could expand the number of sales to those looking for a cheap computer. This would work well for those with a monitor already and wouldn't buy another computer unless they could use their old monitor (buyers on a tight budget). However, I doubt Apple could make much profit on such a cube and keep it price well below an eMac. It would also look a lot like a simple update to the original cube-not very ground breaking and could result in bad PR. I doubt it will ever happen.
As for staying on topic:
I guess in the first quarter of next year we'll see SP 1.8, DP 2.0 and DP 2.5 at the current price structure.
Originally posted by rickag
True, but having more transistors and having them more densly packed has to affect production costs. If I remember correctly, it has been mentioned that IBM utilizes some automated design tools to speed up the actual design. This might explain the less densly packed transistors??
No comment here, I've not a clue, well maybe, but no point in appearing to bash Motorola. More densly packed transistors/traces may affect yields??
Kind of my point. The MPC7455 is still on the 0.18µm process and unless Motorola is having a fire sale or their yields are tremedous, the IBM 970 should be price competative with the MPC7455/7445 and the MPC7457/7447.
I just can't see why using the G4's, in and of itself, would allow Apple to produce computers cheaper than using the G5. Unless of course all the other stuff, companion chip, Hypertransport, etc. disproportionately increases the cost over the G4's equivalents.
The cost to manufacture something isn't the only thing that determines the price it sells for. For example the 3.2 Ghz P4 costs Intel the same exact amout to make as does the 2.6 Ghz P4 but Intel sells the 3.2Ghz chip at a much higher price. Actually the 2.6 Ghz chip costs more to make since it has to go through the testers a couple more times.
The probablility that the 7455 is cheaper than the 970 isn't really anything new. Every time a new generation of chips come out the older chips drop in price, often by a substantial amount. There are some factors that could possibly change all this. IBM could be offering Apple some exceptionally good prices on the 970, which I doubt. The other is that Motorola is somewhat screwed up at the moment.
Originally posted by Geddoe
The cost to manufacture something isn't the only thing that determines the price it sells for. For example the 3.2 Ghz P4 costs Intel the same exact amout to make as does the 2.6 Ghz P4 but Intel sells the 3.2Ghz chip at a much higher price. Actually the 2.6 Ghz chip costs more to make since it has to go through the testers a couple more times.
The probablility that the 7455 is cheaper than the 970 isn't really anything new. Every time a new generation of chips come out the older chips drop in price, often by a substantial amount. There are some factors that could possibly change all this. IBM could be offering Apple some exceptionally good prices on the 970, which I doubt. The other is that Motorola is somewhat screwed up at the moment.
This is just wrong. There are only so many chips per manufacturing 'unit' that are able to run at 3.2 GHz while bein operational. Similarily, there are a lot more cpus that run stable at 2.6 GHz, making its yield better, and hence cheaper.
True that the material might be cheaper, but the fastest chip is still more expensive...
.:BoeManE:.
If you get a batch of 100 CPUs for 10 000 dollars the cost really is 100 dollar/ CPU. If they all would run a 2 GHz they would have the same price. Now if they do not you try to jack up the price of the fastest CPU as much as the market can take. The lower yield the smaller segment of the market you are aiming at and the higher price you can extract
The "science" part of the price is not that difficult to calculate in terms of R&D yields, die size and so on. But that only set the basal level of the price (no not sell at a loss). The real price is more of making as good ROI as the consumer can be billed. If IBM could make CPUs at 1% of current cost would they sell them at 1% the price? Hell no, they would just undercut the competition as little as possible to gain market share for maximum profits.
Originally posted by Carson O'Genic
The only place I can see the need for a cube is at the bottom end. A G4 cube, i.e. headless eMac, that could expand the number of sales to those looking for a cheap computer.[/B]
One of the things I've been trying to figure out recently is how possible it is to do a case-mod on an eMac. It looks, from everything I've seen, not terribly practical - circuit boards and key parts being difficult to get to.
Maybe Apple's best-of-both-worlds option would be to have the non-monitor part of the next generation eMac easily seperatable. That way they can sell the thing as one unit (appealing to the AIO market), but people who do not want the bulk of a monitor they have no intention of using can easily discard it.
This doesn't decrease the price any, but I suspect those calling for a monitor-less eMac in order to simply decrease prices are not looking at as much of a cost saving as they realise.
Both the headless eMac and the small tower design sound like a good idea but there is a risk of leaving the clean pro/home portable/stationary quadrants and stray into a product jungle as bad as the LC, performa Quadra Centrino days...
The over priced underpowered iMac is a greater problem than the lack of a headless eMac. The current top of the line iMac is not that mcuh faster than the budget portable iBook... Apple have to first get the current lines in good shape before they add others to the mix.
With the pro towers are all duals, the servers using G5 and the iMac having fast G5 substantially outperforming the portable and the eMac then they can add a Minitower for gaming students and other that want more than the iMac but not the big tower. They can have a Single CPU tower (quadra 700 size or less but G5 cheese grater design) AGP slot, one PCI slot one optical and one HD. It would enable users to upgrade the graphcial card and also get various funcionalities from the PCI card.
Originally posted by Stecs
I don't think there need to be a separate line for a 'mini' G5. Take the current Powermac line, and for the next revision simply make the low end version a cut down one.
Leave just 1 AGP/1PCI-X for expansion. Trim the processor bay down to a single processor /HS unit in height, 2 memory slots (DDR400). Stick with 2 HD slots/ 1 optical. Stick a single G5 2.0 in it.
This is precisely what I had in mine for the G5 mini. Essentially, take the 1.6 GHz PowerMac G5 board, remove 2 PCI slots, and shrink the board down to half the size. Then wrap a 12x12x8 inch box of the same design around it.
Same deal with a G4 mini. Take the current El Capitan case and board, remove 3 PCI slots, 1 external bay, and shrink the El Capitan case down to 12x12x9.
They are however separate lines of "computer" even though though use the same components.
End up with:
2x2.5 G5 @ 2999
2x2.0 G5 @ 2399
1x2.0 G5L @ 1799
What this would do is provide a differentiated lower end that has more of a USP that the current one, and a more capable normal model for those who want that.
A lot of people have a single processor model for the PowerMac G5. Given the existence of a G5 mini, most assuredly with 1 processor, that'll make a single processor PowerMac G5 and its cost rather undesireable. If Apple doesn't have a G5 mini line, well, forget what I just said.
Originally posted by Programmer
It hardly becomes obsolete. Not leading-edge, yes... but not obsolete and certainly not any less capable than when you bought it.
True. I guess obsolete is the wrong word. "Dated" maybe? Anyway, I'm sure the machine I'm going to get will more than adequately meet my needs for at least 2.5 years. Probably 3+ unless something drastic happens to software requirements in the interim.
With what I'm seeing from Adobe, I can't see myself upgrading next time around (CS part deux) unless there are some *really* compelling reasons to do so. Feature-wise I think the CS Suite has reached a point where, for my particular work, I won't need anything more for a good while.
Not unless Photoshop adds in some professional grain removal tools for example. That's if the 3rd party tools out there don't improve in the meantime. There are some pretty effective ones out there, but none that I've found to be particularly intuitive or efficient.
Originally posted by Eric_Z
A while back there was a development conference in China regarding there future investments in computer technology, and of course IBM was there and showed a presentation. Read it [page 23] and you you'll see that he list the (?current?) IBM PPC 970 to clock up to 2.5Ghz.
Interesting find if true, Eric_Z. That particular page is gone now; would you have a copy of it?
DrBoar wrote:
Both the headless eMac and the small tower design sound like a good idea but there is a risk of leaving the clean pro/home portable/stationary quadrants and stray into a product jungle as bad as the LC, performa Quadra Centrino days...
You meant Centris, not Centrino, just for the record.
I really hate to break the news to you but if you cut up the mother board, offer different options and a new case you pretty much have a new line. Sure this product may just offer the essentials but in a sense so does the IMac line.
What I was hoping for was a minimal system that would almost be considered a transportable that would use an external display device. The idea is zero impact on the desktop and reasonablly close to state of the art pefromance.
I still think this ( a mini) is extremely doable if a couple of things come to fruition. Key here is a 970 based processor with a built in memory interface, and a low cost high performance interface to a I/O chip. The I/O chip should support nothing more than is required to deliver a computational machine to the user. I'm even thinking that an integrated GPU on this I/O chip may be doable.
So in the end you have two or three major chips, memory and the associated buffering logic on the motherboard. Very low costs and with a enough modern ports (Firewire and USB) very flexible.
Thanks
Dave
Originally posted by THT
This is precisely what I had in mine for the G5 mini. Essentially, take the 1.6 GHz PowerMac G5 board, remove 2 PCI slots, and shrink the board down to half the size. Then wrap a 12x12x8 inch box of the same design around it.
Same deal with a G4 mini. Take the current El Capitan case and board, remove 3 PCI slots, 1 external bay, and shrink the El Capitan case down to 12x12x9.
They are however separate lines of "computer" even though though use the same components.
A lot of people have a single processor model for the PowerMac G5. Given the existence of a G5 mini, most assuredly with 1 processor, that'll make a single processor PowerMac G5 and its cost rather undesireable. If Apple doesn't have a G5 mini line, well, forget what I just said.
I purchase 2 PCI cards and considered a third. The first was for USB and Firewire. The second was for a good graphics card. The other slot, which would have been for a video or audio converter didn't get filled because I realized my system would have been to slow to do me any good with it.
I also upgraded the processor to the best I could, which only made the system viable for 1 more year. That happened over about 4 years.
My point with all of this is, a computer, pro or consumer will only last 5 years max in my opinion before software advances make it unusable. A consumer grade computer, like the iMac, wouldn't really need PCI slots, only pro users need audio/video PCI cards, and all graphics cards are now AGP based. New formats of I/O don't work as well as they should through a PCI bus anyway, so they can wait. So I don't believe PCI slots are necessary on a consumer grade model, but an AGP slot is essential. I realize an upgradible processessor is becoming more difficult with the new systems, but it would help consumers believe their system had more longevity to it even if it's no more than 1 year.
For the most part I see very few problems with the current iMac, except for the price/performance ratio, but that's another story. As far as it being headless, you know I'd love that. I think everyone would. Being able to choose your own display - size and price - is a very personal decision when buying a computer. It's what you stare at for hours each day, not the case. Apple would be wise to return this option to the consumer if they want consumer sales. I'm sure Ives could figure out a way to make the current LCD's on the swivel arm swappable. I really like being able to adjust the screen on my LCD iMac with just a touch, and do so very often.
Anyway, just a few thoughts,
Zhar
Originally posted by Big Mac
Interesting find if true, Eric_Z. That particular page is gone now; would you have a copy of it?
Yes, can I mail it to you so that you can put it upp for all to see?
[Ed]You should be able to see it now, try the link again.[/Ed]
Originally posted by wizard69
I really hate to break the news to you but if you cut up the mother board, offer different options and a new case you pretty much have a new line. Sure this product may just offer the essentials but in a sense so does the IMac line.
That's pretty much what I would to see. Actually I would like to see is a 1.8 GHz 17" Powerbook G5, I'm fine with it being 1.5+ inch thick, with a 1680x1120 resolution screen.
What I was hoping for was a minimal system that would almost be considered a transportable that would use an external display device. The idea is zero impact on the desktop and reasonablly close to state of the art pefromance.
You want a laptop without a screen?
Originally posted by THT
That's pretty much what I would to see. Actually I would like to see is a 1.8 GHz 17" Powerbook G5, I'm fine with it being 1.5+ inch thick, with a 1680x1120 resolution screen.
You want a laptop without a screen?
Well not exactly, more of an IMac without the screen. Also without the cpu limitations that seem to crop up in the IMac line. In other words a 2.5 GHz 970 in a small box with just the ports it needs to run. Most likely that would be SATA, Firewire and USB. The unit could potentially also have Blutooth or other wireless technology. The idea is to have plenty of Both Firewire and USB ports available. The unit should take either standard AGP GPU's or possibly a Hypertransport based GPU.
The whole idea is to get really good CPU peformance and GPU peformance and drop the other crud. Mind you I believe the IMac is close to providing what is needed, but there are a couple of problems. One as already mentioned is the ability to upgrade the GPU or atleast have options. Another problem is the limited performance of the IMac, the lack of a screen selection and memory expansion.
I look at it this way the G5 is too limited in its expansion capabilities for its price and size. Eventually the G5 will have to be addressed or Apple is likely to see a huge slow down in sales as the early adopters satisfy themselves. Once the G5 becomes a proper tower then a limited expansion potential machine will fill that void.
Thanks
Dave
And I do not care for getting a stack of FW boxes. The whole point of having a big tower is to get rid of the clutter. Otherwise I could have a iMac and then daisychain what ever I wish for