I had been looking forward to buying a new PowerMac to replace my 9 year old PowerCenter. I even decided that the prior generation G5 2.0 was the best deal for myself with hope that it would last another 9 years. However, given that the MacOS will be running on Intel chips in two years, I believe that it would make more sense to buy the cheapest MacMini to get me through until the new Intel-based machines are available. Any opinions?
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
New Reason to Buy a MacMini - Intel - Opinions?
- Thread starter Greg421
- Start date
- Sort by reaction score
I disagree with those who think it's a mistake to buy PPC Macs. Quite the opposite -- all the aps are written for PPC now and all future aps will be compiled for PPC and x86. None of us know how well Rosetta will really work and God knows how long it will take for the major aps to get the universal binaries ... so buy PPC now.
I think if you want or need a new desktop, it's a great buying opportunity with the last gen PPCs. If you want a new PowerBook, that's another story -- the Intel PBs will probably be the first new products out of the gate and will deliver a huge processing and feature upgrade, enough to make up for any Rosetta performance hit.
What I definitely would NOT do is buy a Mac Mini. Go with a G5 iMac and use it for two to three years, until the transition is complete.
I think if you want or need a new desktop, it's a great buying opportunity with the last gen PPCs. If you want a new PowerBook, that's another story -- the Intel PBs will probably be the first new products out of the gate and will deliver a huge processing and feature upgrade, enough to make up for any Rosetta performance hit.
What I definitely would NOT do is buy a Mac Mini. Go with a G5 iMac and use it for two to three years, until the transition is complete.
I disagree. If you are going to want one of these new Intel equipped Macs, then buy the mini because you are only going to use it a year or two at most. It will be a vast upgrade over what you have now and it won't cost you much cheese. Plus if you buy a iMac then you will be stuck with a monitor all in one setup that you cannot use with you new machine and that would suck, if you know you are going to upgrade, then do a mini instead 
I am perplexed at this time, like so many others. I like the idea of the Intel chips being used. But since I was considering a iMac G5 or a PM G5 soon; I am concerned about the cost to "upgrade" to new software when the MacIntels hit. Surely the software companies won't provide us with a "fat" install to keep us current for the next two to four years.
yoda13 said:I disagree. If you are going to want one of these new Intel equipped Macs, then buy the mini because you are only going to use it a year or two at most. It will be a vast upgrade over what you have now and it won't cost you much cheese. Plus if you buy a iMac then you will be stuck with a monitor all in one setup that you cannot use with you new machine and that would suck, if you know you are going to upgrade, then do a mini instead
Well, first of all, you'd be an idiot to buy an Intel Mac BEFORE the aps came out. Rosetta will (except in the case of PBs and Mac Minis) probably make the new Intels run a bit slower than the last generation PPCs. And like I said above, all new software will be universal code so PPCs will always be compatible. So you can't time now when will be a good time to buy an Intel Mac, we simply don't know. And if you have a lot of aps already, why on earth would you want to run them in Rosetta when you can run them native?
Secondly, a Mac Mini is not necessarily the best buy -- it depends on whether you have a decent monitor already. And as for that point about the iMac, you're assuming either another $1000 for that Mac Mini for a really good monitor or perhaps using an outdated monitor for that Intel Mac. The iMac monitor issue is really a non-issue.
I think a Mac Mini is a great idea for the time being. You could go with an iMac and be just as happy, but if it took you nine years to upgrade I'm thinking you don't have the money to buy a $1500 computer now and then again in a couple of years. From a financial standpoint, I think the Mini is the way to go if it's going to be a temporary thing. And you don't have to spend a thousand bucks on a monitor, there are a LOT of good monitors out there for MUCH less.
I say go Mini for now and go MacIntel a couple of years down the road.
I say go Mini for now and go MacIntel a couple of years down the road.
If you are planning on buying a computer now, I wouldn't worry about the PPC -> x86 transition too much. I am not an expert in every technical area related to this transition, but here are my random thoughts about it categorized into two sections.
What Apple would like consumers to believe.
- OS X will instantly be just as fast on x86 as it is on PPC.
- All ISV's will be shipping fat binaries of their programs by mid-2006.
- For all other issues Rosetta will work its magic and you won't notice the performance hit.
- Consumers don't know anything about optimizing software for certain hardware instruction sets.
What is likely in reality
- OS X will initially run somewhat slower on x86. While Apple has been planning this for years, many multimedia apps like iLife and the Pro apps are currently optimized for Altivec rather than SSE. Overtime this will change, I'm sure Leopard will optimized for x86 rather than PPC, but that's a ways out there.
- Not all ISV's will ship fat binaries right away. Some will be late at offering x86 compatible binaries, some will cut off PPC binariy support early. Fat binaries will run on either architecture but this ignores optmization for a particular platforms hardware instruction set which can make a significant difference in performace.
- For those old apps that are no longer supported/updated or where it was too painful to convert to a universal binary you'll be stuck with Rosetta. No matter what Steve says, anyone who has worked with emulated environments knows there is going to be a performance hit. The emulation may be rather efficient and I'm sure it varies from app to app. Still, converting the instructions on the fly as the app runs is adding another level of complexity and another set of processes that must constantly be done in order for that app to run. There WILL be a performance hit running PPC binaries on x86.
So what does it mean for buying a PPC Mac or waiting for an x86 Mac? No matter what you do you will experience some pain somewhere a long the line - this is the nature of transitions. How much pain you must endure and how long that pain will last can't be known right now as there are too many variables. Here's my opinion.
Buy a PPC Mac now:
First of all, it's the only type of Mac you can get for the next year. You get a good year of solid PPC only or fat binary (still PPC native) and if any optimization has been done it will be for PPC instruction sets. Then x86 Mac hardware comes out and fat binaries are common with some legacy apps running emulated on x86 for a while, still probably better to have a PPC Mac for first 6-12 months that x86 Mac is out. This is two solid years of PPC Mac performance, then I think it'll get grey for a year or two when x86/PPC are about on par as far as performance and transition frustration - this will of course depend a lot upon what apps you need and what state those apps are in at the time. Ultimately, x86 Mac starts to outshine PPC Mac... but by that time you'll probably be considering an upgrade anyway.
Buy an underpowered PPC Mac now and/or hold out for getting an x86 Mac early on:
You'll have less than your ideal hardware now, or none at all if you completely hold out. You do that for 12-18 months, then you buy that x86 Mac that you've so eagerly been waiting for and find that OS X is actually a little more sluggish on it at first, at least until Leopard is out and maybe gone through a maintenance release or two. Plus, any apps you are running will at best be shipped in fat binary so that they will run on x86, but still will probably be optimized for Altivec. If they aren't universal binaries you'll really take a hit by running them with the magic Rosetta instruction translator that doesn't cost any performace degredation - none!
The x86 Mac will eventually out-shine the PPC Mac, but my guess is that this won't happen for 3-4 years at the earliest. For the next three years I think having a PPC Mac will give you the best performance for the money. At some point in 3-5 years x86 Mac will overtake the PPC Mac in terms of performance and you can decide at that point when it is right for your to upgrade from PPC Mac to x86 Mac. Based on this, I would recommend getting a Mac now - one that you are confident will last you 3-4 years. I think this would be an iMac or PowerMac G5.
Good luck with whatever you decide.
What Apple would like consumers to believe.
- OS X will instantly be just as fast on x86 as it is on PPC.
- All ISV's will be shipping fat binaries of their programs by mid-2006.
- For all other issues Rosetta will work its magic and you won't notice the performance hit.
- Consumers don't know anything about optimizing software for certain hardware instruction sets.
What is likely in reality
- OS X will initially run somewhat slower on x86. While Apple has been planning this for years, many multimedia apps like iLife and the Pro apps are currently optimized for Altivec rather than SSE. Overtime this will change, I'm sure Leopard will optimized for x86 rather than PPC, but that's a ways out there.
- Not all ISV's will ship fat binaries right away. Some will be late at offering x86 compatible binaries, some will cut off PPC binariy support early. Fat binaries will run on either architecture but this ignores optmization for a particular platforms hardware instruction set which can make a significant difference in performace.
- For those old apps that are no longer supported/updated or where it was too painful to convert to a universal binary you'll be stuck with Rosetta. No matter what Steve says, anyone who has worked with emulated environments knows there is going to be a performance hit. The emulation may be rather efficient and I'm sure it varies from app to app. Still, converting the instructions on the fly as the app runs is adding another level of complexity and another set of processes that must constantly be done in order for that app to run. There WILL be a performance hit running PPC binaries on x86.
So what does it mean for buying a PPC Mac or waiting for an x86 Mac? No matter what you do you will experience some pain somewhere a long the line - this is the nature of transitions. How much pain you must endure and how long that pain will last can't be known right now as there are too many variables. Here's my opinion.
Buy a PPC Mac now:
First of all, it's the only type of Mac you can get for the next year. You get a good year of solid PPC only or fat binary (still PPC native) and if any optimization has been done it will be for PPC instruction sets. Then x86 Mac hardware comes out and fat binaries are common with some legacy apps running emulated on x86 for a while, still probably better to have a PPC Mac for first 6-12 months that x86 Mac is out. This is two solid years of PPC Mac performance, then I think it'll get grey for a year or two when x86/PPC are about on par as far as performance and transition frustration - this will of course depend a lot upon what apps you need and what state those apps are in at the time. Ultimately, x86 Mac starts to outshine PPC Mac... but by that time you'll probably be considering an upgrade anyway.
Buy an underpowered PPC Mac now and/or hold out for getting an x86 Mac early on:
You'll have less than your ideal hardware now, or none at all if you completely hold out. You do that for 12-18 months, then you buy that x86 Mac that you've so eagerly been waiting for and find that OS X is actually a little more sluggish on it at first, at least until Leopard is out and maybe gone through a maintenance release or two. Plus, any apps you are running will at best be shipped in fat binary so that they will run on x86, but still will probably be optimized for Altivec. If they aren't universal binaries you'll really take a hit by running them with the magic Rosetta instruction translator that doesn't cost any performace degredation - none!
The x86 Mac will eventually out-shine the PPC Mac, but my guess is that this won't happen for 3-4 years at the earliest. For the next three years I think having a PPC Mac will give you the best performance for the money. At some point in 3-5 years x86 Mac will overtake the PPC Mac in terms of performance and you can decide at that point when it is right for your to upgrade from PPC Mac to x86 Mac. Based on this, I would recommend getting a Mac now - one that you are confident will last you 3-4 years. I think this would be an iMac or PowerMac G5.
Good luck with whatever you decide.
chicagdan said:
Speaking as someone who invested some money in an iMac and then decided to upgrade to a PM, the monitor is an issue. And if you are going to run the latest and greatest software, then when the Intel based Macs come out, it will be time to buy one, because that is when the software will begin to be optimized for it. I am not as comfortable as you that all new software will be PPC compatible. Sure I know its possible, but will developers still choose to compile both, which we may view as trivial but isn't necessarily, I am not sure they will. And for the record, I didn't say anything about buying an Intel equipped Mac without apps being available for it, don't know where you got that. As far as the rest of it goes, I was just telling him what I would do if I were him, which is buy a mini now and an Appletel when they come out.
Buying an expensive computer and holding on to it for 5-10 years really doesn't make sense. You will get much better bang for your buck if you buy a $650 computer every 3 years than a $1950 computer every 9 years. No matter how much you spend on a computer, in 9 years it's going to be obsolete. Budget computers, on the other hand may never be cutting edge, but they will always be good enough for nearly any task.
feakbeak said:
Not necessarily. There are already some frameworks/libraries on OS X and with the Apple Apps that 'translate' abstracted multimedia instructions to processor technologies. This way, different PPC processors can have their own optimized libs. Just throw in a MMX/SSE/? library and have all your iLife und Pro apps fly.
feakbeak said:
This is technically not correct. There are actually two binaries made that both sit in the application package. So each of them can be completely optimized by the compiler, and tweaked by hand. The same is true for Framworks, libraries, etc.
Great piece bitfreat.....
I think your analysis is right on the money and should be posted above any thred which contains the text x86....
<end of sucking up>
In honesty... good stuff ....
I think your analysis is right on the money and should be posted above any thred which contains the text x86....
<end of sucking up>
In honesty... good stuff ....
I know that most modern compilers optimize code to some extent during compilation. Although I am a programmer I am not extremely knowledgable about optimization. However, I am doubtful that it is as easy you make it sound. I'm sure there are tools and optimized libraries that will assist developers in converting their apps. However, no matter how much people say this will be a smooth and easy transition, experience has taught me there will be issues, incompatibilities and initially many apps are going to lag on x86 as compared to PPC.madrobby said:
unfaded said:
evilgEEk said:
Speaking for myself, money isn't as large an issue as being practical and frugal. I work hard for a living, and Im choosy about giving up my money when spending it. I got a 128K Mac in 1985 (cost a couple of grand at the time), upgraded to a 512KE, a MacPlus, IIsi, Powerbook 145, SE/30, and then the PowerCenter in 1996. I have been able to upgrade the RAM and the CPU in the PowerCenter a number of times (now, it has 256 MB RAM, a G3 500MHz CPU, 4 internal hard drives, Plextor CD-RW Drive, Radeon Video Card, Firewire, and USB). The G5 will not likely have a processor upgrade, but I would think an Intel machine will have a processor upgrade path. Maybe, an Intel upgrade card will become available to stick in those empty PCI-X slots.
I appreciate all of the replies. Thanks.
Similar dilemma
Thanks
Richard
I am in a similar position. I have a 400mHz G4 sawtooth tower (bus speed 100mHz), now upgraded to 1gHz, an extra HD and a Radeon 9000. I was thinking of the 20inch G5 iMac bit now wonder whether the mac mini (esp after Rev B) would be a better buy to tide me over for 2-3 years. What's your advice? Would I notice much of a difference between the present mac mini and the G4 tower I have now?Greg421 said:
Thanks
Richard
rtjstevens said:
The mac mini would be a hell of a lot faster than your Sawtooth G4. Everything will run more smoothly and things like ripping CD's and other processor intensive tasks will be much faster.
Shadow
I'm in a similar boat as the OP (my details are at https://forums.macrumors.com/posts/1510441/) because I was originally planning on getting a PM G5 dual 2 ghz or so that would last me a few years (since evolutionary improvements on Macs had seemed sluggish). Now, that seems pointless.
Now, I'm probably just going to get a Mini even though I originally ruled out G4s. :/ If Apple suddenly does a big price cut on low end PM G5s, I might go w/that.
Now, I'm probably just going to get a Mini even though I originally ruled out G4s. :/ If Apple suddenly does a big price cut on low end PM G5s, I might go w/that.
Will_reed said:
Well, I finally made up my mind and bought a refurb 2.0 GHz from Apple. By the time I will want to upgrade the computer, the Intels will be established. I guess what tipped the scales was that I have a track record of waiting to buy after others work out the bugs in the hardware and software. I believe the Mac I bought will fit the bill well into the Intel era. Also, I'm damn happy about saving over $800 that I would have spent on the 2.3 GHz machine I was considering.
Thanks for all the help,
Greg
Lacero said:
Why upgrade every 9 years or so? Well, see my previous posts in this thread.
wdlove said:
I got a great deal on a Dell monitor (UltraSharp 1905FP 19-inch Flat Panel Monitor with Height Adjustable Stand) for $350 back in March. I never knew how much room my 19" CRT used till I got the LCD. I am satisfied with my investment. I hear the Apple displays are great, but I can't imagine spending so much money on a monitor.