Idea why Apple doesn't want a CPU expandable Mid-tower
Apple currently targets the Mac Pro with Xeons. If they are upgradeable to the upcoming Penryn, Apple gets nothing out of it, unless someone is truly dense and purchases them from Apple.
With the iMac, MacBook, MacBook Pro, Mac mini they don't have CPU upgrade paths.
A mid-tower system people have been clammering for Apple to release opens up most of the upgrade costs to 3rd parties and not Apple.
Outside of having a marginal increase in marketshare, what financial benefit does Apple make, compared to Intel or others by allowing a completely upgradeable system?
With the iMac, MacBook, MacBook Pro, Mac mini they don't have CPU upgrade paths.
A mid-tower system people have been clammering for Apple to release opens up most of the upgrade costs to 3rd parties and not Apple.
Outside of having a marginal increase in marketshare, what financial benefit does Apple make, compared to Intel or others by allowing a completely upgradeable system?
Comments
Apple currently targets the Mac Pro with Xeons. If they are upgradeable to the upcoming Penryn, Apple gets nothing out of it, unless someone is truly dense and purchases them from Apple.
With the iMac, MacBook, MacBook Pro, Mac mini they don't have CPU upgrade paths.
A mid-tower system people have been clammering for Apple to release opens up most of the upgrade costs to 3rd parties and not Apple.
Outside of having a marginal increase in marketshare, what financial benefit does Apple make, compared to Intel or others by allowing a completely upgradeable system?
For Apple there would be no benefit, as it would hurt their high profit margins...
But part of the problem, is that the current Mac Pro is a poor value - a 7300GT standard is weak, same with 1 GB. Not all pros/semi-pros, need or want a server-class workstation, and the 20" iMac has a cheap LCD panel, compounded that it's covered by glass. And it uses slower, more expensive laptop parts. The 24" is nice, but again, it's just a laptop in different case - and I personally dislike ATI; weaker cards in comparison to Nvidia, and their poor Vista/Linux drivers make want to write them off for good.
If the Mac Pro was $1500, it wouldn't be a huge deal - ECC DDR2 and Socket 771 processors are fairly cheap now, but $2500, no.
Intel Xeon E5410 Harpertown 2.33GHz 12MB L2 Cache LGA 771 80W Quad-Core Processor - Retail -$310
http://www.newegg.com/Product/Produc...82E16819117149
Kingston 4GB(2 x 2GB) 240-Pin DDR2 SDRAM ECC Registered DDR2 667 (PC2 5300) Dual Channel Kit Server Memory - Retail - $150
http://www.newegg.com/Product/Produc...82E16820134329
The simple answer to why Apple isn't selling a mid-tower is probably that Steve just doesn't want to be in that market. We'll see.
The percentage of PC users (outside the home-builder segment) who ever upgrade the processor in their machines -or upgrade anything in their machines, for that matter- is incredibly low. It might look different on forums like this, but I doubt Apple users are much more active upgraders.
The simple answer to why Apple isn't selling a mid-tower is probably that Steve just doesn't want to be in that market. We'll see.
I know a lot of people that just buy the newer machine rather than upgrade. I'm one of the only people I know that upgraded their computer. And my only upgrades were straight from the factory. I have not continued to u/g and I could. But I would rather go with newer technology, and newer parts than build a "hot rod" out of a technology item.
What apple should do is to give a better price for Xeon based workstation coz they choose manufacturers, place big orders so it should be cheaper for a customer to buy all ready machine with Xeons, than to buy the same thing part by part and assemble it at home, like I did all that time with PC.
When looking at the economics of upgrade vs. replace you should also look at resale value. The resale on Macs is pretty good (partly because they generally aren't upgradeable), which makes upgrading less attractive. I took one Mac down the upgrade path a decade ago... it really wasn't worth it.
In a year I put in a Pentium D 940 Dual-core and an additional 2 Gigs of Ram. The power supply is next to be upgraded to 620W and SLI ready GPU.
This machine runs Linux. By spreading the cost over time it's not much of a dent.
On OS X hardware, I'll just buy the Mac Pro and depending on the motherboard LGA 775 socket capabilities, upgrade a newer CPU if it is enabled.
With PCIe 2.0 out, I'd expect the new Mac Pros to be PCIe 2.0 ready and that opens up a lot of GPU options recently released by ATi and nVidia for PCIe 2.0.
As I buy another system for Linux I'll build a Grid Distributed System with the prior system and use it for a few more years before moving it to File server/mail service/Web Dav/etc.
I know a lot of people that just buy the newer machine rather than upgrade. I'm one of the only people I know that upgraded their computer. And my only upgrades were straight from the factory. I have not continued to u/g and I could. But I would rather go with newer technology, and newer parts than build a "hot rod" out of a technology item.
This is a very important consideration, upgrading rarely gives you an opportunity to keep your machine up to date. It may or may not result in a faster machine or other capabilities.
As to recent considerations, upgrading is really stupid if you want to take advantage of recent OS technologies such as virtualization. Only some of the most recent chip sets from Intel are heavily virtualized. For at least the last couple of years it has made more sense to buy a new motherboard, processor and memory instead of trying to upgrade.
The days of slapping a new 486 into your motherboard and doubling your performance is gone.
Dave
Apple currently targets the Mac Pro with Xeons. If they are upgradeable to the upcoming Penryn, Apple gets nothing out of it, unless someone is truly dense and purchases them from Apple.
With the iMac, MacBook, MacBook Pro, Mac mini they don't have CPU upgrade paths.
A mid-tower system people have been clammering for Apple to release opens up most of the upgrade costs to 3rd parties and not Apple.
Outside of having a marginal increase in marketshare, what financial benefit does Apple make, compared to Intel or others by allowing a completely upgradeable system?
If by "financial benefit" you mean an immediate spike in profits, then my answer is 'none'; however, the market share gain will offset the "lost" profits from increased costs for producing/supporting a mid-tower, cannibalization of the iMac, Mac Mini, and Mac Pro, as well as, people upgrading components instead of purchasing new hardware. Apple should care less about losing sales of the iMac, Mac mini, and Mac Pro to the mid-tower and care more about the people that will buy a mid-tower but not an iMac, Mac mini, or Mac Pro.
Apple currently targets the Mac Pro with Xeons. If they are upgradeable to the upcoming Penryn, Apple gets nothing out of it, unless someone is truly dense and purchases them from Apple.
We the Mac Pros that has nothing to do with it. Apple can not move forward without the use of modern chip sets. That is what it comes down to.
As for Apple not selling a machine with a CPU slot for you to upgrade that make no sense whatever when they are being marketed to businesses as professional machines.
With the iMac, MacBook, MacBook Pro, Mac mini they don't have CPU upgrade paths.
Nothing supported by Apple that is. The Mini can be upgraded by CPU replacement for example. The fact that it is done infrequently suggest that many people have not found value in the upgrades. Futher the people who do find value probably do so because Apple takes so long to update the hardware to modern chipsets. For example if a Mini came out with an X3100 GPU or even a discrete GPU from NVidia, there would be value and a lot of incentive to go out and simply buy a new machine.
A mid-tower system people have been clammering for Apple to release opens up most of the upgrade costs to 3rd parties and not Apple.
Almost everyone I know isn't looking for a mid tower so they can upgrade the CPU. The goal is different, that is to provide a low cost platform for people with needs that can't easily be molded into a standard Apple product. In fact I suspect that most people wouldn't be bothered by a soldered in CPU on a Apple mid tower.
The goal with a mid tower is the flexibility to add non standard I/O and storage on a low cost platform.
Outside of having a marginal increase in marketshare, what financial benefit does Apple make, compared to Intel or others by allowing a completely upgradeable system?
Well first I don't think they would ever offer a completely upgradeable system! At least not in the way I'm thinking your are thinking! The biggest advantage for Apple is they would capture some of that portion of the market that has need for a low cost expandable machine. For example I have a few Linux machines around the house and one of the most important features on these machines is the PCI bus. That to allow me to use the machines in alternative ways such as a CNC controller.
Is that a lot of machines, that is people that see a need for a PCI bus or a couple of 3.5" drive bays. Well at the moment it is obvious that Apple doesn't think so. Personally I think it is much more than the zero Apple seems to attribute.
Basically Apple misses a lot of sales simply because they don't have the right hardware mix. Surprisingly this is important to business too, as a $400 computer that can have a USB or serial card installed is much more useful than a $1400 computer.
Dave
I know a lot of people that just buy the newer machine rather than upgrade. I'm one of the only people I know that upgraded their computer. And my only upgrades were straight from the factory. I have not continued to u/g and I could. But I would rather go with newer technology, and newer parts than build a "hot rod" out of a technology item.
I think it depends on a person's needs - with AMD moreso than Intel, it makes sense to upgrade the CPU. For those users that have an AMD AM2 socket motherboard, you can put the latest Phenom quad core processor in it, and still use all your other parts (DDR2 memory and PCI-X graphics), and for $200 dollars, that's a cheap boost in performance.
For example, my friend has a single-core Athlon 64 in his AM2 board, but doesn't have the money to spend on an a C2D rig. It would benefit someone like him the most.
And it also depends on the expected lifetime and cost of upgrading - I'm still using AGP, DDR, and Socket 939 for my PC tower, it's way beyond its EOL, but I have no reason to upgrade, as it runs Vista perfectly, and all I really use it for is storage. But I've also updated the graphics card and CPU once, and that made sense, as it speed up the system for little money, especially with the CPU - single to dual core.
As for most of the current Apple's, It doesn't made as much sense to upgrade the CPU's(the MB, Mini, or iMac), as again, those all use mobile CPU's and are much more expensive to upgrade (double to triple desktop CPU prices).
For what it would cost to buy a T7200 2.0 GHz mobile CPU, you could buy the low-end Intel quad-core desktop chip.
The percentage of PC users (outside the home-builder segment) who ever upgrade the processor in their machines -or upgrade anything in their machines, for that matter- is incredibly low.
So I have to wonder, why is it always claimed on these threads that Apple would lose a ton of new machine sales from selling an upgradable machine, when upgrading is the exception in the rest of the market?
Almost everyone I know isn't looking for a mid tower so they can upgrade the CPU. The goal is different, that is to provide a low cost platform for people with needs that can't easily be molded into a standard Apple product.
Yep. Apple cannot lose business it never made an attempt to grab in the first place. If I want a machine with a quick GPU, either Apple sells me one, or they sell me a consumer class machine with a PCI-Express slot (ATi and nVidia are guaranteed to handle the rest), or I deal with someone else entirely. Currently they're set on the last option.
with AMD moreso than Intel, it makes sense to upgrade the CPU
Is AM2 much better than the 775 on upgrades, relatively speaking? I'd think both are fine. P4->C2Q or Pentium D->C2Q is huge. Slow C2D->fast C2Q is a still a big upgrade.
Upgradeability shouldn’t be mistaken with flexibility
a system with slots and desktop hardware give you more BTO flexibility and why can't you get a desktop mac for $600-$1900 with a good video card with out a build in screen.
Why must I pay $2200 base + $250 for a ATI Radeon X1900 XT 512MB + the high cost of FB-DIMMS around $100 a gig to get a good system with 2gb of ram and a ok video card.
Apple could easily sell a mid-tower but lock out upgrades in the firmware.
The point is that they simply don't have to do this at all. All they have to do is to solder the processor to the mother board. For the vast majority of people looking for an expandable PC, that is low cost, this would do just fine. I think that is key, Apple could enter the low cost expandable platform, with custom mother board, without issue. As all they have to do is to provide the slots and bays the vast majority of people desire.
Every PC I own has a multicard reader stuck into an external 3.5 in bay. Can't do that on any Apple machine at all. Not to mention installing a low cost parallel port card, or any other card that might fit into xyz variant of the PCI standard.
The biggest problem with all of this is that Apple customers have been spoon feed the idea that external devices are always better, They aren't always and its a shame that Apple doesn't recognize that.
Dave
Another issue with upgradability that should be mentioned: Apple loses control of the machine's innards. Any old crappy parts maker can come along and sell you junk to swap into your nice, stable Apple-built computer... and the result is hardware instability and software incompatibilities. It spoils the Mac experience. It spoils the Windows experience too, but its not as noticeable.
This may be true to some extent, but in my experience I had absolutely no trouble when I upgraded a 7500 I paid only $89 for. I am not a geek, I have only a very rudimentary understanding of computers. To the 7500 I added, an ATI Rage Pro card, a Sonnet G4 upgrade card, a USB/Firewire upgrade card and a 9 gig SCSI hard drive. I had no issues, no real problems and this Frankentein monster lasted about 4 years of use by me, my wife and children. I will say it wasn't trouble free, but all issues had to do with OS 9 and its' legendary propensity to crash.
When looking at the economics of upgrade vs. replace you should also look at resale value. The resale on Macs is pretty good (partly because they generally aren't upgradeable), which makes upgrading less attractive. I took one Mac down the upgrade path a decade ago... it really wasn't worth it.
Macs in general have good resale value and this may be due to the lack of models. Some buyers would rather buy a used higher end model or, for those buyers with entry price sticker shock(ie: Apple doesn't sell econo models) they can afford used models.
For Apple there would be no benefit, as it would hurt their high profit margins...
This can not be proved. The iMac and Mac mini both use more expensive laptop parts. A reasonable argument may be made that Apple's profit margins may actually go up with a mid tower that used less expensive desktop parts. This argument has gone on for years and likely will never be resolved. After Steve Jobs statement concerning it is an AIO world, I don't expect Apple to ever offer a reasonably priced mid tower. So the eternal questions about why Apple doesn't offer a mid tower will continue for years to come, spawning more of these threads and perpetuating these very arguments.
This is a very important consideration, upgrading rarely gives you an opportunity to keep your machine up to date. It may or may not result in a faster machine or other capabilities.
As to recent considerations, upgrading is really stupid if you want to take advantage of recent OS technologies such as virtualization. Only some of the most recent chip sets from Intel are heavily virtualized. For at least the last couple of years it has made more sense to buy a new motherboard, processor and memory instead of trying to upgrade.
The days of slapping a new 486 into your motherboard and doubling your performance is gone.
Dave
There is no heavy or light virtualization by Intel. The chip either has Virtualization support or it doesn't.
My Pentium D 940 Dual-Core has it.
The Pentium D 945 Dual-Core doesn't.
There are tradeoffs for the various chip offerings:
http://compare.intel.com/pcc/default...&culture=en-US
By eliminating this and not having a Mid-Tower BTO system that can continue to be upgraded for at least 18 months, Apple doesn't lose all that extra revenue stream.
I think it depends on a person's needs - with AMD moreso than Intel, it makes sense to upgrade the CPU. For those users that have an AMD AM2 socket motherboard, you can put the latest Phenom quad core processor in it, and still use all your other parts (DDR2 memory and PCI-X graphics), and for $200 dollars, that's a cheap boost in performance.
For example, my friend has a single-core Athlon 64 in his AM2 board, but doesn't have the money to spend on an a C2D rig. It would benefit someone like him the most.
And it also depends on the expected lifetime and cost of upgrading - I'm still using AGP, DDR, and Socket 939 for my PC tower, it's way beyond its EOL, but I have no reason to upgrade, as it runs Vista perfectly, and all I really use it for is storage. But I've also updated the graphics card and CPU once, and that made sense, as it speed up the system for little money, especially with the CPU - single to dual core.
As for most of the current Apple's, It doesn't made as much sense to upgrade the CPU's(the MB, Mini, or iMac), as again, those all use mobile CPU's and are much more expensive to upgrade (double to triple desktop CPU prices).
For what it would cost to buy a T7200 2.0 GHz mobile CPU, you could buy the low-end Intel quad-core desktop chip.
Same goes for the LGA-775 Socket boards for Intel that go from Pentium D to Core2Duo and now the CoreDuo to QuadCore to these:
http://www.asus.com/products.aspx?l1=3&l2=11&l3=572
I buy one of these motherboards and it's easily upgradeable even when Intel switches to Nahalem and 32nm fabs. They are pin compatible chipsets and part of the LGA 775 family.
For Apple to expand into the BTO market they'd have to provide some advanced Bundling options that would be a combo CPU-GPU-RAM combo that virtually makes it less appealing for one to go to some other company.
If they added a BTO Service Insurance Plan to provide a guaranteed upgrade path for a nominal fee I can see people buying this under the extended AppleCare (AppleCare Pro) Plan.
This may be true to some extent, but in my experience I had absolutely no trouble when I upgraded a 7500 I paid only $89 for. I am not a geek, I have only a very rudimentary understanding of computers. To the 7500 I added, an ATI Rage Pro card, a Sonnet G4 upgrade card, a USB/Firewire upgrade card and a 9 gig SCSI hard drive. I had no issues, no real problems and this Frankentein monster lasted about 4 years of use by me, my wife and children. I will say it wasn't trouble free, but all issues had to do with OS 9 and its' legendary propensity to crash.
It only has to be true to a small extent in order to significantly increase the burden on Apple's support department.