remember, most os x programs are cocoa which just need to be compiled for both at the same time..it really doesnt seem like hard work to me
Spoken like a true non-programmer. Anyone who has worked to support software on two different platforms (with different endians) would completely disagree.
Aren't there relatively few applications that make good use of Altivec? I've never seen a list but my impression was that Photoshop and the Apple iLife and Pro apps were the major examples.
Add OS X to the list. It uses Altivec extensively.
The issue is Rosetta and which applications will need it and be able to use it. Quite obviously, OS X won't need Rosetta. Applications that make significant use of Altivec but won't have an x86 Mac binary are the ones in question. It's hard to think of any that would fall into that box.
Well the Intel 10.4 doesn't use Altivec. It can't.
Right. You can expect a performance hit from that. No wonder why Apple will leave the more powerful machines for the end (2.5 years from now). They certainly hope that the x86 clock speed (and eventually the multi-core CPUs) at this time will make up for that.
Right. You can expect a performance hit from that. No wonder why Apple will leave the more powerful machines for the end (2.5 years from now). They certainly hope that the x86 clock speed (and eventually the multi-core CPUs) at this time will make up for that.
It should be just as fast using SSE2 (which I'd guess will be rebranded Velocity Engine 2). Plus it will be running at a minimum of 3.2GHz with decent bandwidth to fast RAM.
Don't worry about the OS, worry about the apps that need Altivec.
I've been waiting to get a powermac for a while now. I guess my main problem isnt compatability or anything.
The thing I'm concerned about is, I will eventually want to upgrade. Lets say I bought a dual g4 2-3 years ago. And 6 months ago I wanted to upgrade to a dual g5. I could have hopped on e-bay and gotten a good amount of money for my g4.
If I buy a g5 now, in 2-3 years who would want to buy it? There is no longer any resale value for these older systems.
Maybe I'm wrong, but it seems it will head this way.
I've been waiting to get a powermac for a while now. I guess my main problem isnt compatability or anything.
The thing I'm concerned about is, I will eventually want to upgrade. Lets say I bought a dual g4 2-3 years ago. And 6 months ago I wanted to upgrade to a dual g5. I could have hopped on e-bay and gotten a good amount of money for my g4.
If I buy a g5 now, in 2-3 years who would want to buy it? There is no longer any resale value for these older systems.
Maybe I'm wrong, but it seems it will head this way.
These are the things that depress me to think about.
It should be just as fast using SSE2 (which I'd guess will be rebranded Velocity Engine 2).
No. SSE2 has much lower performance than Altivec, even on much higher clocked P4. Apple will wait the clock speed and the cores to catch up and hide the performance hit.
I have decided to count my blessings and ignore the fact that my machine could have been so much better if IBM would have stayed up to par with Intel. I will buy my PB and accept that I could be screwed (software wise) in a year+.
I think you are making a good decision and I do not think you will be screwed with software in a year. I was worried about it too after just buying a Power Mac a month ago, but after reading the highlights from the keynote, I am not worried at all. I am excited for Apple.
I just left my local CompUSA store. There was gentlmen in there buying the cheapest iMac he could. He commented that he had planned to walk in and purchase several power systems but said he was going to minimize his investment at his company until the switch is over.
I then walked out with my DC power adapter for my G4 iBook thinking, man, can this $80 DC adapter wait a year.
Regarding the presumed hit to resale values of machines purchased now, I think people are making a mountain out of a molehill.
An earlier poster suggested that he may be concerned about the resale value of a powermac purchased now and sold in 3 years. Well, three years ago, people were buy Quicksilvers. I just went on ebay and had a looksie. It appears that quicksilvers are selling for in the 450-650 range today. It is not true that a DP G5 bought today will be worth zero dollars in three years. At worst the delta from the resale value of machines purchased three years ago, and sold today will be $100-$200. And in fact, there may be significant aftermarket demand for these machines, as they will run legacy software with no performance hit. I suspect there will be a tiny marginal deprecation hit taken by buyers today. But it won't be anything to sweat over. Something along the lines of a stick of RAM in value lost.
I'd even go so far as to suggest that if you are thinking about buying a powermac, now would be the best time to jump in! You will be insulated from the turmoil over the next three years. Get your machine! Get your software! And compute in peace for the next 36-48 months. When you are ready to buy again, you will be able to buy a mature Rev2 Macintel box.
Well, I don't know how fast Photoshop is going to be on the new Macs now that Altivec has been pulled from underneath it.
Isn't it the case that only certain FILTERS in Photoshop utliize Altivec? And that's why Steve always chose those filters for those nutty G4 bakeoffs he used to do at keynotes?
I've been dithering about a new computer for home, and after yesterday I thought I'd just sit tight for a while, but I am now seriously thinking about jumping on a 20-inch iMac. They're a good value, and it will last me three years, by which time the transition should be sorted out. I'd be in a different frame of mind if new machines were coming in six months, but I think the G5 models will be replaced last, so it might well be two years before we see a next-generation iMac.
I just spent $2200 CAD (plus taxes, and an extra 512 megs of ram) on a new iMac G5 20". I did this two weeks ago, before the Intel rumors.
My parents were all set to also pick up an iBook and iMac. After the announcement, my dad said "I guess I'll wait a year."
There's no need to wait. If the computer does now what you want and need it to do now, that's what matters. In a year, my iMac will still be playing WoW, still run iMovie and all the other iApps, and still store all my music in iTunes and stream it to the home stereo. In fact, I'd wager that even in 5 years, my iMac will still be doing all of this.
Yes, your Apple may have a slightly lower resale value after 3 years than you've been used to - but at least it will still have a resale value. Anyone try selling a Pentium III lately? Be happy that you can still retain value after an extended period of time.
Businesses who lease new products (and you should) get to amortize hardware over 2 years. After 2 years, it's worth nothing to the business in terms of tax breaks, so feel free to switch to Mac-Intel in two years with no regrets, and rest assured that you can pass the "old" G5's down to junior members of the firm who can still run Office and Photoshop and any other current-gen app just great.
I will admit that I reacted with shock and disappointment when I first discovered there'd be a transition after a 2-week old purchase. Take a step back, breathe, and think about the future. 2 years is an eternity in computer years (think of computer years as dogyears x 2 ) and we have no idea what the world will look like then.
I will say, however, that it looked like the P4 3.6 on stage with Steve sure opened apps faster than my 2.0GHz G5.
I think I'll start saving my pennies for my NEXT Mac, and reassure my parents that whatever they wanted their new Macs to do now, they will still be doing in 2 years. My mom's old Toshiba notebook still runs windows 98 and still gets email for her, so a new iBook will be no different.
Think less about the insides, and more about your real-world usage for a computer.
I'm not buying a PPC based mac (unless Apple lowers prices by 40 to 60%)...
No way Apple will cut prices that much. If your need for a new Mac isn't urgent, however, I suppose a little waiting might pay off if Apple sees that products are moving slowly due to PPC/Intel transition concerns.
This wouldn't be a bad time for Apple to sacrifice some of their famously high margins to see if they can actually grow their market share during this transition.
Quote:
...because as of today, PPC based macs are marked for obsoletion. .
All computers are destined to become obsolete. How long do you expect to hold on to any computer and still have the latest and greatest of every single new piece of software work well, if at all, during your computer's senior years?
If you want to hold onto to any computer for a long time, you have to be prepared to hold onto the software you buy for it too, and be happy with whatever service that software gives you up until the point where updates are no longer available.
PPC Macs will easily be supported for the next 4-5 years. How much more life can you expect out of any computer before it becomes starts becoming difficult to get updates that are optimized, or even work at all, for that computer?
Quote:
And I can't imagine most developers writing two sets of code knowing this.
Developers don't have to write two sets of code. All they have to do is move to Xcode if they haven't already, and clean up any bad coding practices which might stand in the way of processor neutrality. After that, one set of code will automatically generate universal binaries. No need for two entirely different sets of code.
There might be, in a few apps, some assembly language or hand-tweaked "#ifdef OSX_X86" logic needed by developers who are trying to optimize performance. I suspect that this kind of tweaking is the one thing that might be done for only one of two platforms, or sacrificed altogether in the name code maintenance simplicity.
At first, developers will remain most interested in tweaking for PPC performance. As time goes on and the number of x86 Macs grows, the emphasis will shift to x86 performance.
Somehow I don't beleive that an old iBook will be your best choice for running new software in a few years from now, Intel or no Intel.
I use most of my Macs for about 5 to 7 years, but I really don't expect the latest and greatest software to run on my Mac "WELL" for more than 3 or 4 years.
Then when that happens switch to a stable version of Linux.
Comments
Originally posted by cj171
remember, most os x programs are cocoa which just need to be compiled for both at the same time..it really doesnt seem like hard work to me
Spoken like a true non-programmer. Anyone who has worked to support software on two different platforms (with different endians) would completely disagree.
Originally posted by Hudson1
Aren't there relatively few applications that make good use of Altivec? I've never seen a list but my impression was that Photoshop and the Apple iLife and Pro apps were the major examples.
Add OS X to the list. It uses Altivec extensively.
Originally posted by PB
Add OS X to the list. It uses Altivec extensively.
But at least Apple has already fixed that.
Originally posted by Blackcat
But at least Apple has already fixed that.
What do you mean?
Originally posted by PB
What do you mean?
Well the Intel 10.4 doesn't use Altivec. It can't.
Originally posted by PB
What do you mean?
The issue is Rosetta and which applications will need it and be able to use it. Quite obviously, OS X won't need Rosetta. Applications that make significant use of Altivec but won't have an x86 Mac binary are the ones in question. It's hard to think of any that would fall into that box.
Originally posted by Blackcat
Well the Intel 10.4 doesn't use Altivec. It can't.
Right. You can expect a performance hit from that. No wonder why Apple will leave the more powerful machines for the end (2.5 years from now). They certainly hope that the x86 clock speed (and eventually the multi-core CPUs) at this time will make up for that.
Originally posted by zaz You'd have to have a pretty huge set of dual processors to be buying a Mac with intel day 1.
giggling.
Well put.
Originally posted by PB
Right. You can expect a performance hit from that. No wonder why Apple will leave the more powerful machines for the end (2.5 years from now). They certainly hope that the x86 clock speed (and eventually the multi-core CPUs) at this time will make up for that.
It should be just as fast using SSE2 (which I'd guess will be rebranded Velocity Engine 2). Plus it will be running at a minimum of 3.2GHz with decent bandwidth to fast RAM.
Don't worry about the OS, worry about the apps that need Altivec.
The thing I'm concerned about is, I will eventually want to upgrade. Lets say I bought a dual g4 2-3 years ago. And 6 months ago I wanted to upgrade to a dual g5. I could have hopped on e-bay and gotten a good amount of money for my g4.
If I buy a g5 now, in 2-3 years who would want to buy it? There is no longer any resale value for these older systems.
Maybe I'm wrong, but it seems it will head this way.
Originally posted by AlmightyMac
I've been waiting to get a powermac for a while now. I guess my main problem isnt compatability or anything.
The thing I'm concerned about is, I will eventually want to upgrade. Lets say I bought a dual g4 2-3 years ago. And 6 months ago I wanted to upgrade to a dual g5. I could have hopped on e-bay and gotten a good amount of money for my g4.
If I buy a g5 now, in 2-3 years who would want to buy it? There is no longer any resale value for these older systems.
Maybe I'm wrong, but it seems it will head this way.
These are the things that depress me to think about.
Originally posted by ryanschmidt
These are the things that depress me to think about.
Well, I don't know how fast Photoshop is going to be on the new Macs now that Altivec has been pulled from underneath it.
Originally posted by Blackcat
It should be just as fast using SSE2 (which I'd guess will be rebranded Velocity Engine 2).
No. SSE2 has much lower performance than Altivec, even on much higher clocked P4. Apple will wait the clock speed and the cores to catch up and hide the performance hit.
Originally posted by ryanschmidt
I have decided to count my blessings and ignore the fact that my machine could have been so much better if IBM would have stayed up to par with Intel. I will buy my PB and accept that I could be screwed (software wise) in a year+.
I think you are making a good decision and I do not think you will be screwed with software in a year. I was worried about it too after just buying a Power Mac a month ago, but after reading the highlights from the keynote, I am not worried at all. I am excited for Apple.
I then walked out with my DC power adapter for my G4 iBook thinking, man, can this $80 DC adapter wait a year.
An earlier poster suggested that he may be concerned about the resale value of a powermac purchased now and sold in 3 years. Well, three years ago, people were buy Quicksilvers. I just went on ebay and had a looksie. It appears that quicksilvers are selling for in the 450-650 range today. It is not true that a DP G5 bought today will be worth zero dollars in three years. At worst the delta from the resale value of machines purchased three years ago, and sold today will be $100-$200. And in fact, there may be significant aftermarket demand for these machines, as they will run legacy software with no performance hit. I suspect there will be a tiny marginal deprecation hit taken by buyers today. But it won't be anything to sweat over. Something along the lines of a stick of RAM in value lost.
I'd even go so far as to suggest that if you are thinking about buying a powermac, now would be the best time to jump in! You will be insulated from the turmoil over the next three years. Get your machine! Get your software! And compute in peace for the next 36-48 months. When you are ready to buy again, you will be able to buy a mature Rev2 Macintel box.
Originally posted by the cool gut
Well, I don't know how fast Photoshop is going to be on the new Macs now that Altivec has been pulled from underneath it.
Isn't it the case that only certain FILTERS in Photoshop utliize Altivec? And that's why Steve always chose those filters for those nutty G4 bakeoffs he used to do at keynotes?
I've been dithering about a new computer for home, and after yesterday I thought I'd just sit tight for a while, but I am now seriously thinking about jumping on a 20-inch iMac. They're a good value, and it will last me three years, by which time the transition should be sorted out. I'd be in a different frame of mind if new machines were coming in six months, but I think the G5 models will be replaced last, so it might well be two years before we see a next-generation iMac.
My parents were all set to also pick up an iBook and iMac. After the announcement, my dad said "I guess I'll wait a year."
There's no need to wait. If the computer does now what you want and need it to do now, that's what matters. In a year, my iMac will still be playing WoW, still run iMovie and all the other iApps, and still store all my music in iTunes and stream it to the home stereo. In fact, I'd wager that even in 5 years, my iMac will still be doing all of this.
Yes, your Apple may have a slightly lower resale value after 3 years than you've been used to - but at least it will still have a resale value. Anyone try selling a Pentium III lately? Be happy that you can still retain value after an extended period of time.
Businesses who lease new products (and you should) get to amortize hardware over 2 years. After 2 years, it's worth nothing to the business in terms of tax breaks, so feel free to switch to Mac-Intel in two years with no regrets, and rest assured that you can pass the "old" G5's down to junior members of the firm who can still run Office and Photoshop and any other current-gen app just great.
I will admit that I reacted with shock and disappointment when I first discovered there'd be a transition after a 2-week old purchase. Take a step back, breathe, and think about the future. 2 years is an eternity in computer years (think of computer years as dogyears x 2
I will say, however, that it looked like the P4 3.6 on stage with Steve sure opened apps faster than my 2.0GHz G5.
I think I'll start saving my pennies for my NEXT Mac, and reassure my parents that whatever they wanted their new Macs to do now, they will still be doing in 2 years. My mom's old Toshiba notebook still runs windows 98 and still gets email for her, so a new iBook will be no different.
Think less about the insides, and more about your real-world usage for a computer.
Originally posted by sc_markt
I'm not buying a PPC based mac (unless Apple lowers prices by 40 to 60%)...
No way Apple will cut prices that much. If your need for a new Mac isn't urgent, however, I suppose a little waiting might pay off if Apple sees that products are moving slowly due to PPC/Intel transition concerns.
This wouldn't be a bad time for Apple to sacrifice some of their famously high margins to see if they can actually grow their market share during this transition.
...because as of today, PPC based macs are marked for obsoletion. .
All computers are destined to become obsolete. How long do you expect to hold on to any computer and still have the latest and greatest of every single new piece of software work well, if at all, during your computer's senior years?
If you want to hold onto to any computer for a long time, you have to be prepared to hold onto the software you buy for it too, and be happy with whatever service that software gives you up until the point where updates are no longer available.
PPC Macs will easily be supported for the next 4-5 years. How much more life can you expect out of any computer before it becomes starts becoming difficult to get updates that are optimized, or even work at all, for that computer?
And I can't imagine most developers writing two sets of code knowing this.
Developers don't have to write two sets of code. All they have to do is move to Xcode if they haven't already, and clean up any bad coding practices which might stand in the way of processor neutrality. After that, one set of code will automatically generate universal binaries. No need for two entirely different sets of code.
There might be, in a few apps, some assembly language or hand-tweaked "#ifdef OSX_X86" logic needed by developers who are trying to optimize performance. I suspect that this kind of tweaking is the one thing that might be done for only one of two platforms, or sacrificed altogether in the name code maintenance simplicity.
At first, developers will remain most interested in tweaking for PPC performance. As time goes on and the number of x86 Macs grows, the emphasis will shift to x86 performance.
Originally posted by Carson O'Genic
Somehow I don't beleive that an old iBook will be your best choice for running new software in a few years from now, Intel or no Intel.
I use most of my Macs for about 5 to 7 years, but I really don't expect the latest and greatest software to run on my Mac "WELL" for more than 3 or 4 years.
Then when that happens switch to a stable version of Linux.