What's to stop Apple from building a headless desktop and marking it up the same amount as the iMac?
The fact that, by definition, it would allow people to replace some components much more easily than the could with an AIO. This decreases the need for users to replace the entire computer, as single components can be upgraded now. Since those single components aren't actually sold by Apple (and, even if they were, wouldn't be as high-margin as the entire computer), Apple loses out on revenues.
Whereas, for an AIO, people will eventually need (or at least feel the need) to replace the entire machine.
Looks like no-one's 100% sure if it's X2 or just FX.
This 4x4 architecture will require an Athlon with at least 2 Hypertransport links. If in fact the X2 CPUs only have 1 Hypertransport link, than it'll be impossible to put them into a 4x4 system.
AMD could possibly use 1 FX and 1 X2 for the 4x4, but 1 FX (a rebadged, repackaged Opteron essentially) would be required.
Quote:
Anyways, I'd argue that Quad 2.0 is better than a 2.67 GHz Conroe. There are two markets at issue here: high-end "gaming" desktop, and workstation.
There's just too many apps dependent on single threaded performance, and too many multithreaded apps that don't scale beyond 2 cores very well. On top of this, Conroe will be cheaper.
With this sort of situation, Apple has to make sure that its mid-range systems are faster than the lower end systems, otherwise it cannibalizes its higher end systems. So, I think they will have a mid-range system that'll be faster than a low end system for the vast majority of apps.
Quote:
Also, I just want to re-iterate that the custom northbridge was speculation, based on what Apple could be changing in a motherboard and aiming for the "low-hanging fruit" in terms of saving money.
Yup. The odds are low that Apple will use a custom Northbridge.
The fact that, by definition, it would allow people to replace some components much more easily than the could with an AIO. This decreases the need for users to replace the entire computer, as single components can be upgraded now. Since those single components aren't actually sold by Apple (and, even if they were, wouldn't be as high-margin as the entire computer), Apple loses out on revenues.
Whereas, for an AIO, people will eventually need (or at least feel the need) to replace the entire machine.
That's the theory, but I've rarely seen that in practice. People don't generally upgrade all that often, and many people end up buying a new monitor (and any other peripherals) at the same time as a new computer, even they're free to keep the old stuff. If anything, I'd think that having an AIO would make people hold on to computers longer since they know they're stuck with whatever they get and want to make the biggest jump possible. If you have any statistics showing that AIO owners upgrade more often, I'd love to see them. But I'm doubtful.
And for the record, the notion that people will buy more often doesn't mean they have higher profit margins.
I'm not sure if Apple would want to give up the high end like that, they've tried to be as close to the cutting edge of workstation computing as they could in the past, even if only a small portion of their user base would fill up that capacity. Chucker mentioned SAS and I think they'll have to at least offer that sort of thing in order to stay in the league to justify their pricing.
Fully Buffered is more of a server feature. It has higher latency in trade for larger capacity. So, it makes sense for the Xserve to use FB-DIMM. For the Mac Pros, DDR2 should be a win-win. Lower latencies for about the same bandwidth. And it's cheaper.
With the way Intel is placing its products, 2S systems are a "server" feature, and as a results, workstations will be more expensive than a 2S system specifically designed for the workstation market.
By the way, I know that Apple has kept the previous machines in the same enclosure when moving to Intel, but those were mainly consumer machines.
I understand the reasoning there. Don't frighten loyal users with something drastically different. Make the move to Intel as painless as possible.
But we're now talking about the Pro machine. Pros are supposed to know the difference between PCI, PCI-X and PCI-Express. In short, they aren't easily frightened off by technological change.
So putting the Mac Pro in a new dress won't cause major problems here.
So if this is the next-gen machine, it should look like it too.
The reason they would move the power supply to the top of the machine makes complete sense. Think about it -- the power supply generates a lot of heat. Heat rises. Having it at the bottom of the computer this whole time was contributing a lot to the high temperatures of the computer. Putting it at the top of the machine will allow the air to stay at the top and be blown out, leaving the rest of the computer cooler.
I really don't think the power cable will get in the way that much, so it shouldn't be a big deal.
Except that the power supply airflow in G5s is vented front to back and is piped OUT of the machine, rather than hovering within as it does in most PCs. This keeps the majority of the heat from the power supply separate from the rest of the system, In fact, there's an aluminum floor between the current G5 power supply and the main bay. Aluminum is a horrible heat conductor (hence it's use in heatsinks; it cools very quickly).
Most PC designs are just a single box. There's no compartmentalization like in the G5, so yes, having a power supply on top in a system like that makes sense. But then, the power supply is also affected by the heat rising up into IT as well (from the rest of the main bay), which is I'm sure, part of the reason why PC power supplies drop like flies.
Most PC designs are just a single box. There's no compartmentalization like in the G5, so yes, having a power supply on top in a system like that makes sense. But then, the power supply is also affected by the heat rising up into IT as well (from the rest of the main bay), which is I'm sure, part of the reason why PC power supplies drop like flies.
The heat from a PC power supply goes out the back too. The lack of compartmentalization in a typical PC does nothing for or against the power supply because it is enclosed in a box and its heat is ducted out.
Quote:
Aluminum is a horrible heat conductor (hence it's use in heatsinks; it cools very quickly).
That statement is self-contradictory. Aluminum has to be a good conductor of heat to be used as a heat sink material. If it was a bad conductor, that means it's an insulator, which traps heat. Aluminum is a better conductor of heat than many common metals.
The heat from a PC power supply goes out the back too. The lack of compartmentalization in a typical PC does nothing for or against the power supply because it is enclosed in a box and its heat is ducted out.
That statement is self-contradictory. Aluminum has to be a good conductor of heat to be used as a heat sink material. If it was a bad conductor, that means it's an insulator, which traps heat. Aluminum is a better conductor of heat than many common metals.
I agree with Jeff. Many cooking pans are made of aluminum, at least in part.
The heat from a PC power supply goes out the back too. The lack of compartmentalization in a typical PC does nothing for or against the power supply because it is enclosed in a box and its heat is ducted out.
Yes, but there's usually no inlet for cool air in a PC. Some now have side fans, but those don't provide a proper air flow over all the warmed parts; rather just blasting cool air onto a specific area, and not through the system as the G5 does, nor do they have fans specifically designed to pull air into them, and then force it back out the other side. Basically, without an indirect side fan, all most PC power supplies do is just create a vacuum within the case. If it were a perfect vacuum, it would be good, as the thermal conductivity of a vacuum is zero. However, it's not, meaning the small amount of air within the case will heat up. A side fan will help alleviate that, but it's nowhere near efficient.
The faster air moves, the faster it can remove heat. Having a wind-tunnel design will greatly decrease the temperature of a system. Frankly though, I have yet to see a PC as well designed thermodynamically as the G5.
Quote:
That statement is self-contradictory. Aluminum has to be a good conductor of heat to be used as a heat sink material. If it was a bad conductor, that means it's an insulator, which traps heat. Aluminum is a better conductor of heat than many common metals.
Poor choice of words on my part. Yes, Aluminum is a very good conductor of heat, but very poor at thermal retention (hence the reason most professional chefs don't use aluminum cookware, but composites of INOX and or INOX sandwiched with copper). In a forced air system like that in the G5, most of the heat within that aluminum will be sapped away with relative ease.
I agree with Jeff. Many cooking pans are made of aluminum, at least in part.
Yes, but many cooking pans are designed for the average consumer who doesn't like heavy pots and pans. Professional chefs rarely touch the stuff, as it tends to cook things unevenly.
Yes, but many cooking pans are designed for the average consumer who doesn't like heavy pots and pans. Professional chefs rarely touch the stuff, as it tends to cook things unevenly.
OK copper may be better. But aren't we discussing whether aluminum is a good conductor of heat(which it is)?
Yes, but there's usually no inlet for cool air in a PC. Some now have side fans, but those don't provide a proper air flow over all the warmed parts; rather just blasting cool air onto a specific area, and not through the system as the G5 does, nor do they have fans specifically designed to pull air into them, and then force it back out the other side. Basically, without an indirect side fan, all most PC power supplies do is just create a vacuum within the case. If it were a perfect vacuum, it would be good, as the thermal conductivity of a vacuum is zero. However, it's not, meaning the small amount of air within the case will heat up. A side fan will help alleviate that, but it's nowhere near efficient.
A lot of that depends on the specific case, but they all have inlets, many have empty fan mounts on the front, some do have fans. None of that really has anything do do with power supply location. And the vacuum involved in either way is negligible.
Quote:
Frankly though, I have yet to see a PC as well designed thermodynamically as the G5.
My Compaq PWS700 and W8000 are about as good and aren't as accoustically annoying as my dual 2.5 G5, which spinds up to an annoying high pitch whine whenever CPU load gets over 30%.
Originally posted by JeffDM My Compaq PWS700 and W8000 are about as good and aren't as accoustically annoying as my dual 2.5 G5. [/B]
My SP 1.6Ghz G5 is still as quiet as the day I bought it. Though every few months or so, I blow the dust out of the system.
As for placement of the power supply, I'd venture to say it doesn't make much difference in the G5 design, as long as it were to stay isolated and had the same airflow as it currently does. I'm saying that in normal PC designs, it's placement at the top is out of necessity. Why Apple would choose to move the power supply to the top, when it's completely unnecessary and counters the current case design where it's isolated and vented on both sides, makes no sense.
The ATX design doesn't *require* the power supply be in a certain position, it's just put there because that's the way PC cases are designed and have been for the past 20 years or so. You could really put it anywhere in the system you wanted to, it's just a matter of routing power cabling around the inside of the case.
My SP 1.6Ghz G5 is still as quiet as the day I bought it. Though every few months or so, I blow the dust out of the system.
Well, having a low frequency single chip is not much of a heat load. My W8000 has two Netburst 2.8GHz chips and it's noise is a lot more tolerable than my G5.
At any rate, I haven't seen much an argument why moving the power supply for the Mac Pros would be a bad thing.
OK copper may be better. But aren't we discussing whether aluminum is a good conductor of heat(which it is)?
I ceded in another comment that "conductor" was a poor choice of words on my part, when I meant to say something along the lines of thermal retention.
Aluminum will heat up very quickly, but it cools down just as fast. That's why it's used a lot in bakeware rather than cookware -- where most times it's in a situation where there's surrounding or convection heat and it's just used to hold something or form a shape, as opposed to cooking it.
Comments
Originally posted by minderbinder
Who cares if it eats into iMac sales, as long as it's just as profitable and increases mac sales overall?
Because an AIO, by definition, has larger profit margins and revenues than a headless desktop does.
Originally posted by Chucker
Because an AIO, by definition, has larger profit margins and revenues than a headless desktop does.
And they need those margins because far less are sold.
Originally posted by BenRoethig
]And they need those margins because far less are sold.
Yeah, because iMacs are known not to sell well.
Originally posted by Chucker
Because an AIO, by definition, has larger profit margins and revenues than a headless desktop does.
By definition? What's to stop Apple from building a headless desktop and marking it up the same amount as the iMac?
Originally posted by minderbinder
What's to stop Apple from building a headless desktop and marking it up the same amount as the iMac?
The fact that, by definition, it would allow people to replace some components much more easily than the could with an AIO. This decreases the need for users to replace the entire computer, as single components can be upgraded now. Since those single components aren't actually sold by Apple (and, even if they were, wouldn't be as high-margin as the entire computer), Apple loses out on revenues.
Whereas, for an AIO, people will eventually need (or at least feel the need) to replace the entire machine.
Originally posted by ZachPruckowski
Looks like no-one's 100% sure if it's X2 or just FX.
This 4x4 architecture will require an Athlon with at least 2 Hypertransport links. If in fact the X2 CPUs only have 1 Hypertransport link, than it'll be impossible to put them into a 4x4 system.
AMD could possibly use 1 FX and 1 X2 for the 4x4, but 1 FX (a rebadged, repackaged Opteron essentially) would be required.
Anyways, I'd argue that Quad 2.0 is better than a 2.67 GHz Conroe. There are two markets at issue here: high-end "gaming" desktop, and workstation.
There's just too many apps dependent on single threaded performance, and too many multithreaded apps that don't scale beyond 2 cores very well. On top of this, Conroe will be cheaper.
With this sort of situation, Apple has to make sure that its mid-range systems are faster than the lower end systems, otherwise it cannibalizes its higher end systems. So, I think they will have a mid-range system that'll be faster than a low end system for the vast majority of apps.
Also, I just want to re-iterate that the custom northbridge was speculation, based on what Apple could be changing in a motherboard and aiming for the "low-hanging fruit" in terms of saving money.
Yup. The odds are low that Apple will use a custom Northbridge.
Originally posted by Chucker
The fact that, by definition, it would allow people to replace some components much more easily than the could with an AIO. This decreases the need for users to replace the entire computer, as single components can be upgraded now. Since those single components aren't actually sold by Apple (and, even if they were, wouldn't be as high-margin as the entire computer), Apple loses out on revenues.
Whereas, for an AIO, people will eventually need (or at least feel the need) to replace the entire machine.
That's the theory, but I've rarely seen that in practice. People don't generally upgrade all that often, and many people end up buying a new monitor (and any other peripherals) at the same time as a new computer, even they're free to keep the old stuff. If anything, I'd think that having an AIO would make people hold on to computers longer since they know they're stuck with whatever they get and want to make the biggest jump possible. If you have any statistics showing that AIO owners upgrade more often, I'd love to see them. But I'm doubtful.
And for the record, the notion that people will buy more often doesn't mean they have higher profit margins.
Originally posted by JeffDM
I'm not sure if Apple would want to give up the high end like that, they've tried to be as close to the cutting edge of workstation computing as they could in the past, even if only a small portion of their user base would fill up that capacity. Chucker mentioned SAS and I think they'll have to at least offer that sort of thing in order to stay in the league to justify their pricing.
Fully Buffered is more of a server feature. It has higher latency in trade for larger capacity. So, it makes sense for the Xserve to use FB-DIMM. For the Mac Pros, DDR2 should be a win-win. Lower latencies for about the same bandwidth. And it's cheaper.
With the way Intel is placing its products, 2S systems are a "server" feature, and as a results, workstations will be more expensive than a 2S system specifically designed for the workstation market.
Besides, making a custom chipset is not cheap.
Yup. Not good odds.
I understand the reasoning there. Don't frighten loyal users with something drastically different. Make the move to Intel as painless as possible.
But we're now talking about the Pro machine. Pros are supposed to know the difference between PCI, PCI-X and PCI-Express. In short, they aren't easily frightened off by technological change.
So putting the Mac Pro in a new dress won't cause major problems here.
So if this is the next-gen machine, it should look like it too.
Originally posted by West
The reason they would move the power supply to the top of the machine makes complete sense. Think about it -- the power supply generates a lot of heat. Heat rises. Having it at the bottom of the computer this whole time was contributing a lot to the high temperatures of the computer. Putting it at the top of the machine will allow the air to stay at the top and be blown out, leaving the rest of the computer cooler.
I really don't think the power cable will get in the way that much, so it shouldn't be a big deal.
Except that the power supply airflow in G5s is vented front to back and is piped OUT of the machine, rather than hovering within as it does in most PCs. This keeps the majority of the heat from the power supply separate from the rest of the system, In fact, there's an aluminum floor between the current G5 power supply and the main bay. Aluminum is a horrible heat conductor (hence it's use in heatsinks; it cools very quickly).
Most PC designs are just a single box. There's no compartmentalization like in the G5, so yes, having a power supply on top in a system like that makes sense. But then, the power supply is also affected by the heat rising up into IT as well (from the rest of the main bay), which is I'm sure, part of the reason why PC power supplies drop like flies.
Originally posted by zang
Most PC designs are just a single box. There's no compartmentalization like in the G5, so yes, having a power supply on top in a system like that makes sense. But then, the power supply is also affected by the heat rising up into IT as well (from the rest of the main bay), which is I'm sure, part of the reason why PC power supplies drop like flies.
The heat from a PC power supply goes out the back too. The lack of compartmentalization in a typical PC does nothing for or against the power supply because it is enclosed in a box and its heat is ducted out.
Aluminum is a horrible heat conductor (hence it's use in heatsinks; it cools very quickly).
That statement is self-contradictory. Aluminum has to be a good conductor of heat to be used as a heat sink material. If it was a bad conductor, that means it's an insulator, which traps heat. Aluminum is a better conductor of heat than many common metals.
Originally posted by JeffDM
The heat from a PC power supply goes out the back too. The lack of compartmentalization in a typical PC does nothing for or against the power supply because it is enclosed in a box and its heat is ducted out.
That statement is self-contradictory. Aluminum has to be a good conductor of heat to be used as a heat sink material. If it was a bad conductor, that means it's an insulator, which traps heat. Aluminum is a better conductor of heat than many common metals.
I agree with Jeff. Many cooking pans are made of aluminum, at least in part.
Originally posted by JeffDM
The heat from a PC power supply goes out the back too. The lack of compartmentalization in a typical PC does nothing for or against the power supply because it is enclosed in a box and its heat is ducted out.
Yes, but there's usually no inlet for cool air in a PC. Some now have side fans, but those don't provide a proper air flow over all the warmed parts; rather just blasting cool air onto a specific area, and not through the system as the G5 does, nor do they have fans specifically designed to pull air into them, and then force it back out the other side. Basically, without an indirect side fan, all most PC power supplies do is just create a vacuum within the case. If it were a perfect vacuum, it would be good, as the thermal conductivity of a vacuum is zero. However, it's not, meaning the small amount of air within the case will heat up. A side fan will help alleviate that, but it's nowhere near efficient.
The faster air moves, the faster it can remove heat. Having a wind-tunnel design will greatly decrease the temperature of a system. Frankly though, I have yet to see a PC as well designed thermodynamically as the G5.
That statement is self-contradictory. Aluminum has to be a good conductor of heat to be used as a heat sink material. If it was a bad conductor, that means it's an insulator, which traps heat. Aluminum is a better conductor of heat than many common metals.
Poor choice of words on my part. Yes, Aluminum is a very good conductor of heat, but very poor at thermal retention (hence the reason most professional chefs don't use aluminum cookware, but composites of INOX and or INOX sandwiched with copper). In a forced air system like that in the G5, most of the heat within that aluminum will be sapped away with relative ease.
Originally posted by backtomac
I agree with Jeff. Many cooking pans are made of aluminum, at least in part.
Yes, but many cooking pans are designed for the average consumer who doesn't like heavy pots and pans. Professional chefs rarely touch the stuff, as it tends to cook things unevenly.
Originally posted by zang
Yes, but many cooking pans are designed for the average consumer who doesn't like heavy pots and pans. Professional chefs rarely touch the stuff, as it tends to cook things unevenly.
OK copper may be better. But aren't we discussing whether aluminum is a good conductor of heat(which it is)?
Originally posted by zang
Yes, but there's usually no inlet for cool air in a PC. Some now have side fans, but those don't provide a proper air flow over all the warmed parts; rather just blasting cool air onto a specific area, and not through the system as the G5 does, nor do they have fans specifically designed to pull air into them, and then force it back out the other side. Basically, without an indirect side fan, all most PC power supplies do is just create a vacuum within the case. If it were a perfect vacuum, it would be good, as the thermal conductivity of a vacuum is zero. However, it's not, meaning the small amount of air within the case will heat up. A side fan will help alleviate that, but it's nowhere near efficient.
A lot of that depends on the specific case, but they all have inlets, many have empty fan mounts on the front, some do have fans. None of that really has anything do do with power supply location. And the vacuum involved in either way is negligible.
Quote:
Frankly though, I have yet to see a PC as well designed thermodynamically as the G5.
My Compaq PWS700 and W8000 are about as good and aren't as accoustically annoying as my dual 2.5 G5, which spinds up to an annoying high pitch whine whenever CPU load gets over 30%.
Originally posted by JeffDM My Compaq PWS700 and W8000 are about as good and aren't as accoustically annoying as my dual 2.5 G5. [/B]
My SP 1.6Ghz G5 is still as quiet as the day I bought it. Though every few months or so, I blow the dust out of the system.
As for placement of the power supply, I'd venture to say it doesn't make much difference in the G5 design, as long as it were to stay isolated and had the same airflow as it currently does. I'm saying that in normal PC designs, it's placement at the top is out of necessity. Why Apple would choose to move the power supply to the top, when it's completely unnecessary and counters the current case design where it's isolated and vented on both sides, makes no sense.
The ATX design doesn't *require* the power supply be in a certain position, it's just put there because that's the way PC cases are designed and have been for the past 20 years or so. You could really put it anywhere in the system you wanted to, it's just a matter of routing power cabling around the inside of the case.
Originally posted by zang
My SP 1.6Ghz G5 is still as quiet as the day I bought it. Though every few months or so, I blow the dust out of the system.
Well, having a low frequency single chip is not much of a heat load. My W8000 has two Netburst 2.8GHz chips and it's noise is a lot more tolerable than my G5.
At any rate, I haven't seen much an argument why moving the power supply for the Mac Pros would be a bad thing.
Originally posted by backtomac
OK copper may be better. But aren't we discussing whether aluminum is a good conductor of heat(which it is)?
I ceded in another comment that "conductor" was a poor choice of words on my part, when I meant to say something along the lines of thermal retention.
Aluminum will heat up very quickly, but it cools down just as fast. That's why it's used a lot in bakeware rather than cookware -- where most times it's in a situation where there's surrounding or convection heat and it's just used to hold something or form a shape, as opposed to cooking it.