About to buy 3.0 ....should I REALLY go 2.8 instead?

[jconly]

I know, I know...
This has been discussed a million times before. But maybe I should mention my own personal situation.

I'm about to pull the trigger on a much needed new imaging workstation. I spend all my time in Photoshop, Bridge, and Capture One. I shoot 39 megapixel digital backs, and work with high res 4x5 scans. I'm commonly working on files 4GB+

I was going to go with the 3.0, with 10GB of ram. Would this really be a waste? I know that I can't utilize all those cores alone with just PS, but I want to be able to multitask as much as possible. But more importantly, I need this machine to last me AS LONG AS POSSIBLE.

Wether I go with the 2.8 or the 3.0, this is a major investment for me. When all is said and done, I'm looking at a 9k - 10k investment. After that, I can't forsee myself being able to purchasing a new machine for AT LEAST 6 years. Surely within that time frame, a new version of Creative Suite will be released, and I'm sure it will be multi-thread capable. As camera technology evolves, it seems to me that buying bigger now will give me greater capabilities down the road.

But, I'm not doing this to have "the best." If I should really go 2.8, convince me otherwise. Money saved is money saved after all.

[TheThirdMan]

I shoot 39 megapixel digital backs, and work with high res 4x5 scans. I'm commonly working on files 4GB+

You have a hasselblad H3D!??

In answer to ur question, leave it at 2.8. For future proofing, speed will have a negligible impact. 2.8 and 3.0 and 3.2 will all be seriously slow in 6 years.

[ErikAndre]

Additional 200Mhz = $720 + tax

Additional Resell Value = not so much.

Addtional Performance = not so much

Ability to brag that you got a 3.0 instead of a 2.8 = priceless.

UPDATE: I compiled a PDF (by MacWorld) showing the differences between all the Mac's here: http://forums.macrumors.com/showthread.php?t=421580

[jconly]

You have a hasselblad H3D!??

In answer to ur question, leave it at 2.8. For future proofing, speed will have a negligible impact. 2.8 and 3.0 and 3.2 will all be seriously slow in 6 years.

No, I'm not a fan of those Imacon (now Hasselblad branded backs).
I work with a Phase-One P45 on an H2 or 503 body whenever I don't need the movements of a 4x5

But really? I guess I can't lie, there is a part of me that says 3.0 is sweeter, do it.
But am I really not going to see any improvements. Even if I'm running emulation software and a second OS?

[kirkbross]

But, I'm not doing this to have "the best." If I should really go 2.8, convince me otherwise. Money saved is money saved after all.

I was in the same boat, but decided that a 7% speed increase would NOT make my life 7% better. I'm going to get a 2.8 (as soon as Pro Tools is compatible) and then sell it and (probably) get a Nehalem Mac in a year or so.

The common argument of "I want it to last as long as possible" is kind of irrelevant because a slight proc boost in three years is going seem silly compared to the advances in motherboards, RAM, SSD's, etc.. Further, the application upgrades you'll be using in three years will beg for newer "overall" systems, much more so than just a slight proc boost.

Now, if you're income depends on waiting for huge Photoshop filters to render on 50 billion gigabyte image files all day long, then 3.0 will probably make your life slightly easier, but 999 out of 1000 wouldn't notice the time savings -- and saving time is the bulk of this dilema. The other thing is the psychological effect of having a "3.0" box vs. a "2.8" box. If you'll "feel" happier, then factor that in, but keep it low on the totem pole.

[shabba]

Doesnt Photoshop only support 4GB of ram?

[jconly]

I was in the same boat, but decided that a 7% speed increase would NOT make my life 7% better. I'm going to get a 2.8 (as soon as Pro Tools is compatible) and then sell it and (probably) get a Nehalem Mac in a year or so.

The common argument of "I want it to last as long as possible" is kind of irrelevant because a slight proc boost in three years is going seem silly compared to the advances in motherboards, RAM, SSD's, etc.. Further, the application upgrades you'll be using in three years will beg for newer "overall" systems, much more so than just a slight proc boost.

Now, if you're income depends on waiting for huge Photoshop filters to render on 50 billion gigabyte image files all day long, then 3.0 will probably make your life slightly easier, but 999 out of 1000 wouldn't notice the time savings -- and saving time is the bulk of this dilema. The other thing is the psychological effect of having a "3.0" box vs. a "2.8" box. If you'll "feel" happier, then factor that in, but keep it low on the totem pole.

I see what you're saying about the proc diff compared to overall changes. I've been thinking about that though. But what foreseeable technology is going to come out? Faster HD I/O's can be achieved with a PCIe card, and I could care less about USB 3.0. Networking speeds are not so important. Graphics are upgradeable. The major changes are the changes that would come with the newer intel chip.

If I could, I would def wait for Nehalem, but I'm in need of a new machine now. But how much depreciation are we talking here if reselling a year from now is a possibility? It's not a bad thought, but it is a huge hassle.

But anyway, I'm not waiting on huge filter renderings really. In fact, I'm more of a shooter than a "retoucher," although I do all the retouching work since I have trust issues. :rolleyes: I suppose my main issues are based on Disk I/O and RAM.

Perhaps I should go 2.8.

[ErikAndre]

Perhaps I should go 2.8.

It'll be a tough call man. I do HD video editing and I was 100% set on getting the 3.0. Then I saw the article on Bearfeats and then the specs on MacWorld, and instead went with the 2.8. I just couldn't justify spending another $800 on a marginal difference. I got it with the 8800GT, so the order is on hold, but I can't say that I don't feel tempted from time to time to just cancel the thing (being that nothing is charged to my card till it ships) and just get the 3.0.

[jconly]

so stressfull

I suppose I could always find something else to spend the cash on tho.
in fact, some new lighting equip sounds pretty nice

[drrich2]

If the hopes of the performance jump with the advent of Nehalem are even half-way accurate, even 4 years from now I suspect the performance standard for high-end desktop workstation PCs (Macs, Windows or what-have-you) will have advanced to far that you'll want a new system.

Perhaps it might make more sense to 'bank' the money a 3.0 or 3.2 would cost, get a 2.8, and simply get used to the idea of upgrading in 3 or 4 years?

I know you're looking at a large outlay (maybe 10 grand), so easier said than done. I also know that when the standard is for computers in your niche to be 3x's faster, have 3x's the storage, triple the RAM & much faster hard drives, well, it's hard to stay with the old product.

I don't just just what all is running the price of your system to 10 grand.

For sake of argument, if you bought a 5 grand Penryn 2.8 setup NOW, and then sold it & bought a 5 grand Nehalem system in first quarter '09, would the sacrifices you make this year really have enough practical impact to offset the gains you'd have the next 5 years (of your proposed 6+ year cycle)?

Just my 2 cents; hope it's somehow helpful. Good luck with your system.

Richard.

[jconly]

If the hopes of the performance jump with the advent of Nehalem are even half-way accurate, even 4 years from now I suspect the performance standard for high-end desktop workstation PCs (Macs, Windows or what-have-you) will have advanced to far that you'll want a new system.

Perhaps it might make more sense to 'bank' the money a 3.0 or 3.2 would cost, get a 2.8, and simply get used to the idea of upgrading in 3 or 4 years?

I know you're looking at a large outlay (maybe 10 grand), so easier said than done. I also know that when the standard is for computers in your niche to be 3x's faster, have 3x's the storage, triple the RAM & much faster hard drives, well, it's hard to stay with the old product.

I don't just just what all is running the price of your system to 10 grand.

For sake of argument, if you bought a 5 grand Penryn 2.8 setup NOW, and then sold it & bought a 5 grand Nehalem system in first quarter '09, would the sacrifices you make this year really have enough practical impact to offset the gains you'd have the next 5 years (of your proposed 6+ year cycle)?

Just my 2 cents; hope it's somehow helpful. Good luck with your system.

Richard.


Its not so much the cost of the initial system that is running up the cost as it is a new monitor and even more so the cost of building a proper image archive.

I apprecite the post though. You make some very valid points.

The more I think about it, I think I will indeed go with a 2.8 and just hop on the "technology train"
I guess I need to embrace the fact that things are always changing. No sense in buying more than I need at the moment.

[drrich2]

The monitor will carry forward to the new system. I've got a 2.8 GHz MacPro on order, & the old 24" Dell monitor currently hooking to a Dell XPS is what I plan to use with it. Sure I'd like a new monitor; just don't want to pay for it.

Ditto the image archive. I enjoy digital photography & me, my wife & buddy have well over 50 gigs in photos; it's due to migrate to the new MacPro, too.

Richard.

[jconly]

The monitor will carry forward to the new system. I've got a 2.8 GHz MacPro on order, & the old 24" Dell monitor currently hooking to a Dell XPS is what I plan to use with it. Sure I'd like a new monitor; just don't want to pay for it.

Ditto the image archive. I enjoy digital photography & me, my wife & buddy have well over 50 gigs in photos; it's due to migrate to the new MacPro, too.

Richard.

Ya, I have about 2tb to deal with, and its always growing

But yes, the monitor is more of a long term investment. Which is also why I'm putting that off a little longer. Before I buy a samsung xl24, I want to see what other led displays hit the market in the next 6 months. Eizo is also releasing a new 30 in soon. Should be interesting to see the specs on that

[deathshrub]

Dude, buy whatever the hell you want to.

[ErikAndre]

Dude, buy whatever the hell you want to.

Good call.

[jconly]

Dude, buy whatever the hell you want to.

and I will, dude

By why not get the free opinion of others first?
plus some good points worth considering have been raised.

[choreo]

I was all set to buy the 3.2 and decided to go with the 2.8. It came in two weeks ago and it is much faster than the Dual 2-Gig G5 that I just retired. I added 16-gig of RAM and 4 super-fast 750-gig hard drives and I am totally pleased with the speed I am getting out of the hardware (at least on what works with Leopard so far).

I have been doing Macs for 18 years and usually swap out about every 2 years and manage about 10-terabytes of graphics on my current system for clients. This last G5 tower is the longest I have ever kept a computer (4 years). But what you may find is that it will not be "speed" of the computer that forces you out of what ever you have. The reason I get "forced" to "upgrade" (really more of a cross-grade) is not usually due to the computer being too slow, but rather they keep switching industry standards on storage, peripherals, architectures, etc. and the software writing just keeps getting sloppier and heavier - throwing the load onto the machines to do more work to make up the difference.

The computer is also the cheapest part of every upgrade I have done. So far my switch to MacPro/Leopard has cost me over $8000 the past month (not including the computer) - Why? Because the new price for processor speed is all new associated expenses. Scanners not recognized - my choice, purchase Silverfast software at a cost that is more than a new scanner and hope the 2 year old scanner doesn't break (which cannot be replaced since it is out of production) or purchase new scanners which probably are not ready for prime time and debug that issue. Sony Artisan monitors - in perfect condition with proprietary software (now deprecated) - $750 calibrators just turned into door stops thanks to Leopard. People say "stay with Tiger", but the new Mac Pros will not even boot with Tiger. CS3 Leopard upgrades were - no extra charge this time, but I have spent a few thousand dollars to get all my Photoshop and After Effects plugins updated to run. anyway you get the idea. "My" experience has been that at no point will everything just "work" for any length of time. Last night I spent 8 hours straight just trying to work around a Leopard Users Permissions issue... That is why I never get "excited" about so called "speed" tests. One incompatibility can cost you WAY MORE time than ANY hardware speed gain can save you!!! ... and there will be many if your experience is like mine. This is one reason I did not go with the 8800 Nvidia card. I don't play "games", I make a living on my Mac and I must have 3 monitors to do the work I have to do. (2) 8800's "MIGHT" just work? (1) 8800 and (1) 2600 ATI "MIGHT" work? ... but if they don't you could spend months trying to get a fix as NOBODY will claim responsibility for that configuration. I bought (2) ATI 2600's installed - then if it doesn't work I can at least yell at Apple until something maybe gets done!

One thing for sure... anyone who has time to run theoretical speed tests is majoring in the minors. I have watched many people work on their computers and taught many students and for most, the "human" element will be the biggest slow down in productivity - not the machine (Game players excluded - if you are not producing anything, then speed is everything).

I guess what I am saying is at the end of every year, truth be told, the machine spends more time waiting on me (to get upgrades installed, type in serial numbers - repeatedly, activating software, deactivating software, getting technical assistance, etc.) than I spend waiting on my machine to render.

Oh, 2.8.

[Grimace]

I once bought the top of the line system, just because -- even though I knew that the next notch (or two) down would have the same real-world effect. The thrill of having the fastest beast on the block was awesome, but in a month, I was over it.

The cost savings compared to performance increase is important. It sure as heck isn't worth the extra $700+, save that for the sell-and-buy-what's-newer time in 1.5 years. Seriously, you can take a $700 hit in a year and get a better machine. If you buy/sell/re-buy - your only "new" investment is around $700.

2.8

[zdobson]

Take that $700 and invest it. Add $20 a month to that account and in 4 years you'll have almost enough for a new system:

Your monthly deposit for 4 years of $ 20.00 for an interest rate of 9.000 % compounded monthly with an initial amount of $ 700.00:

Year Balance
0 $ 1,015.82
1 $ 1,361.26
2 $ 1,739.11
3 $ 2,152.40

Final Savings Balance: $ 2,152.40

[GoKyu]

I don't (currently) use my computer to make money, but for my first Mac, I wanted something that would LAST a long time (as you said you wanted too) - I ended up getting an octo-core 2.8, 8 extra gigs of RAM (10 total), because I use Lightroom and Photoshop a lot (true, Photoshop doesn't currently use more than 3 gigs of RAM, but I've already heard that CS4 is due out near the end of the year, or slightly beyond that...)

Once they make a 64 bit version of Photoshop, then you can really start loading up on your memory.

For me, this machine will last at least 6 or 7 years, and by the time I'm ready for something else, I should still be able to sell this one at a fairly good price.

You definitely don't need more than 2.8 unless you do video editing for a living with your Mac, where time really IS money, and the faster you go, the sooner you finish.

-Bryan

[macenforcer]

Look at the Powermac G5 resale values. The dual 2.3ghz is the highest valued computer of all of the except the quad and that is becuase it runs cooler and is almost just as fast as the dual 2.7ghz. Nobody wants all that heat. The liquid cooling killed the resale.

2.8 is the best system out there. Way cheaper, way cooler and only milliseconds slower. You will never notice the difference.

[Reach]

Look at the Powermac G5 resale values. The dual 2.3ghz is the highest valued computer of all of the except the quad and that is becuase it runs cooler and is almost just as fast as the dual 2.7ghz. Nobody wants all that heat. The liquid cooling killed the resale.

2.8 is the best system out there. Way cheaper, way cooler and only milliseconds slower. You will never notice the difference.

Not in my market..

The top-of-the line machines from their days has always commanded better resale-value when I have sold them. (And I just sold a 2x2,5ghz liquid cooled PM)

BUT, with the old machines, there was more than the proc that was diferent too usually, now it's basically the same machine, so it doesn't matter so much anymore.

Anyway, I'm going with the 3ghz. When working with Photoshop and those applications, it's the ghz that will help you, NOT the 8 cores. And I am fully aware that it will not be an insane increase in performance, but it will be some. That means it will last me a little longer, AND be differentiated from the horde of 2.8ghz Mac Pros that will flood the market when we start selling these things. I personally think that will make some difference when selling, but of course not the whole cost of the upgrade. Still though, slightly higher price, probably slightly easier to sell, and will last you a little longer, while you'll get your work done slightly faster while you've got it. Costs a little and gives a little, this is a machine I make money on, so I could take that hit, considering I'll have the machine as my main source of income for the next years.

[fernmeister]

If we drew a picture of where the high end of photo processing will be in 4-5 years, would 200Mhz per processor be that big a difference? I doubt it.

[MacProUser]

Not in my market..

The top-of-the line machines from their days has always commanded better resale-value when I have sold them. (And I just sold a 2x2,5ghz liquid cooled PM)

BUT, with the old machines, there was more than the proc that was diferent too usually, now it's basically the same machine, so it doesn't matter so much anymore.

Anyway, I'm going with the 3ghz. When working with Photoshop and those applications, it's the ghz that will help you, NOT the 8 cores. And I am fully aware that it will not be an insane increase in performance, but it will be some. That means it will last me a little longer, AND be differentiated from the horde of 2.8ghz Mac Pros that will flood the market when we start selling these things. I personally think that will make some difference when selling, but of course not the whole cost of the upgrade. Still though, slightly higher price, probably slightly easier to sell, and will last you a little longer, while you'll get your work done slightly faster while you've got it. Costs a little and gives a little, this is a machine I make money on, so I could take that hit, considering I'll have the machine as my main source of income for the next years.


I totally agree with you - as long 8 core isn´t supported by most of the current software versions you benefit from the absolut more processor power of a 3,0 or 3,2 system !

[JesterJJZ]

I would get the 2.8 and buy more ram and/or drives.

[macz1]

You can see it this way:

You buy a 2.8 GHz Mac Pro, but it's as fast as an 3.0 GHz Mac Pro of the last generation. Performance wise, you get an 3.0 GHz Clovertown witch consumes much less energy.

an 8core 3GHz Clovertown was the ultra top of the line a couple of weeks ago...

[jconly]

Not in my market..

The top-of-the line machines from their days has always commanded better resale-value when I have sold them. (And I just sold a 2x2,5ghz liquid cooled PM)

BUT, with the old machines, there was more than the proc that was diferent too usually, now it's basically the same machine, so it doesn't matter so much anymore.

Anyway, I'm going with the 3ghz. When working with Photoshop and those applications, it's the ghz that will help you, NOT the 8 cores. And I am fully aware that it will not be an insane increase in performance, but it will be some. That means it will last me a little longer, AND be differentiated from the horde of 2.8ghz Mac Pros that will flood the market when we start selling these things. I personally think that will make some difference when selling, but of course not the whole cost of the upgrade. Still though, slightly higher price, probably slightly easier to sell, and will last you a little longer, while you'll get your work done slightly faster while you've got it. Costs a little and gives a little, this is a machine I make money on, so I could take that hit, considering I'll have the machine as my main source of income for the next years.

Def some food for thought, and in contrast to the opinion of most as well.
This was my initial thinking, but what I need to figure out is just how processor dependent Photoshop actually is. From what I've been observing, not much. It's mainly come down to RAM and HD I/O. So, with only minimal speed increases from 2.8--->3.0, is it even all that likely I would be noticing the differences in a Program that basically relies on RAM and Scratch disk to do its job?

[krye]

If you're concerned about resale, then look at eBay. The last generation of Mac Pros still seem to be fetching pretty close to retail prices.

[dblester]

i am in the same boat here i am unsure of what processor to choose i was originally dead set on the 3.2 but then after looking over benchmarks and looking over the forums i have become unsure.. i am going to be using my machine for photoshop / illustrator/ some after effects/ indesign apps like that... will there honestly be any actual noticeable speed difference between the different processors i already ordered 16gb of ram from owc, so going with the 3.0 or 2.8 will really only allowing me to save some money on this but i want to have a machine that will last me about 4-5 years.. anyone here have the 3.0 or 3.2 yet? and what are your thoughts on it? thanks in advance

[MacProUser]

i am in the same boat here i am unsure of what processor to choose i was originally dead set on the 3.2 but then after looking over benchmarks and looking over the forums i have become unsure.. i am going to be using my machine for photoshop / illustrator/ some after effects/ indesign apps like that... will there honestly be any actual noticeable speed difference between the different processors i already ordered 16gb of ram from owc, so going with the 3.0 or 2.8 will really only allowing me to save some money on this but i want to have a machine that will last me about 4-5 years.. anyone here have the 3.0 or 3.2 yet? and what are your thoughts on it? thanks in advance


With the 3GHz model you´ll get about 10% more speed compared to the previous model (clovertown) - for a significant reduced price ! Can´t get wrong with that ...

[dogbait]

If you're even asking the question 'Do I need it', chances are you don't. That extra 7-15% increase in speed is of less benefit than spending the money you save on more memory or a fast drive array (http://www.g-technology.com/); because the majority of the time your Mac's processors will be sitting around waiting on the hard drives.

[kingtj]

I've been in the computer field for close to 17 years now, and I have *yet* to see almost anyone hang onto a system for 6 years and still find that more "sensible" than replacing it with a newer one.

(Ok, I met *one* guy who was still happily chugging along with an old IBM 286 class PC running MS-DOS, a good 10 years after he originally purchased the system. The only things he ever did with the computer were spreadsheets in Lotus 1-2-3, and printed out address labels with "My Maillist Pro". In his case, it made no sense to buy anything newer, as long as his old machine didn't totally die on him.)

But usually, upgrading a computer every 3-4 years is about the smartest time-frame. Your old machine still holds a little bit of resale value when it's only 3 or 3.5 years old, so you can use that to offset your new system's cost. (At 5-6 years out, your old machine becomes MUCH harder to resell, and you're often lucky to get pennies on the dollar of its initial cost.)

It's true that you get a little more "useful life" out of your machine if you spend more, up front, going with a bigger/better configuration. (That's similar to the theory that it's wise to buy a new car "fully loaded" with options, because they have better resale value.) But you still have to find the "sweet spot" with computer upgrades.... EG. When upgrading a hard drive today, the largest you can go and still stay in the optimal "price vs. storage space" area is a 750GB drive. The 1TB drives generally cost you about a $50 premium on "gigabytes per dollar". (Of course, certain people will pay the premium, like if you're an iMac owner who can only fit ONE drive in the machine, and you want as much internal storage space as possible.)

Same thing with CPUs. Intel always releases a range of speeds of any new CPU of theirs, with the top couple fastest models commanding big price premiums. That's part of how they recoup their R&D costs. The people having to have "the best, at any cost" will pay the big markup, subsidizing the rest of the product line for them. The "sweet spot" for the new Xeons happens to be the 2.8Ghz model right now.


If the hopes of the performance jump with the advent of Nehalem are even half-way accurate, even 4 years from now I suspect the performance standard for high-end desktop workstation PCs (Macs, Windows or what-have-you) will have advanced to far that you'll want a new system.

Perhaps it might make more sense to 'bank' the money a 3.0 or 3.2 would cost, get a 2.8, and simply get used to the idea of upgrading in 3 or 4 years?

I know you're looking at a large outlay (maybe 10 grand), so easier said than done. I also know that when the standard is for computers in your niche to be 3x's faster, have 3x's the storage, triple the RAM & much faster hard drives, well, it's hard to stay with the old product.

I don't just just what all is running the price of your system to 10 grand.

For sake of argument, if you bought a 5 grand Penryn 2.8 setup NOW, and then sold it & bought a 5 grand Nehalem system in first quarter '09, would the sacrifices you make this year really have enough practical impact to offset the gains you'd have the next 5 years (of your proposed 6+ year cycle)?

Just my 2 cents; hope it's somehow helpful. Good luck with your system.

Richard.

[jconly]

If you're even asking the question 'Do I need it', chances are you don't. That extra 7-15% increase in speed is of less benefit than spending the money you save on more memory or a fast drive array (http://www.g-technology.com/); because the majority of the time your Mac's processors will be sitting around waiting on the hard drives.

I guess I neglected to respond earlier to the other individual who made the same comment on more memory and better storage.

The 8 GB's that I already have budgeted are more than sufficient, as is the Raid 10 "working images" array that will be built to accompany my JBOD image archive. Really, that argument is irrelevant.

And perhaps you misread the title of my thread. I didn't ask if I needed it. I asked wether everybody's advice to go with 2.8 is valid in my specific situation. The question isn't whether or not I'll be able to utilize 3.0, as I know I will at some point in time. The question is whether it's a financially sensible investment at THIS moment in time.

Perhaps I should have been more clear.

[Reach]

Def some food for thought, and in contrast to the opinion of most as well.
This was my initial thinking, but what I need to figure out is just how processor dependent Photoshop actually is. From what I've been observing, not much. It's mainly come down to RAM and HD I/O. So, with only minimal speed increases from 2.8--->3.0, is it even all that likely I would be noticing the differences in a Program that basically relies on RAM and Scratch disk to do its job?

The difference is felt towards the end of the computers life, not much now. I usually keep my stationary Mac about three years, and towards the end I really notice whatever extra I got when I bought the machine.

Don't know if that makes sense, but if you don't do überheavy work in PS it won't matter that much now. But it will matter once time passes, CS4-5 arrives etc etc.

[MacProUser]

The difference is felt towards the end of the computers life, not much now. I usually keep my stationary Mac about three years, and towards the end I really notice whatever extra I got when I bought the machine.

Don't know if that makes sense, but if you don't do überheavy work in PS it won't matter that much now. But it will matter once time passes, CS4-5 arrives etc etc.



I totally agree with that !

[jconly]

The difference is felt towards the end of the computers life, not much now. I usually keep my stationary Mac about three years, and towards the end I really notice whatever extra I got when I bought the machine.

Don't know if that makes sense, but if you don't do überheavy work in PS it won't matter that much now. But it will matter once time passes, CS4-5 arrives etc etc.

This is what I was trying to get at in my last post.
I'm well aware I won't be fully utilizing 8 cores of 3.0 NOW..

But down the road, thats a different story.
And alright, maybe I'll be upgrading before the 6 year mark, so let's say I do go on a 4 year cycle. I don't doubt that in 4 years I'll be using the processor to it's potential.

[kem]

Buy the fastest you can afford. If total cost is an issue, getting more RAM over processor speed will be more important. The $800 or whatever you save by not getting a 3.0 buys you 16GB of ram.

Photoshop will make use of more then 4gb of ram in CS3: (quoted from an forum post I found)
"VM BUFFERING IN PHOTOSHOP CS3
On Macintosh computers, Photoshop can directly access up to about 3.5GB. When there is more than 3.5GB of document data, Photoshop writes data to its scratch files as necessary. On a computer with 4GB or less of RAM, the data is transferred directly between the scratch files on disk and the Photoshop RAM. On a computer with more than 4GB of RAM, Photoshop tells the operating system to use the extra RAM as a buffer for the Photoshop scratch file. In this case, when document data no longer fits in the 3.5GB of Photoshop RAM and is written to the scratch file, the operating system stores it in the extra RAM and can retrieve it from there much faster than it could be read from disk. This lets Photoshop take advantage of more than 4GB of RAM to significantly increase performance with very large documents."

And an article i read on barefeats: while processing a 300mb file CS3 was using about 13gb of memory on a 16gb system.

[relbbircs]

This is a helpful thread.

It got me thinking: I'd had on order the 3.0, with 4 GB of ram and the 8800, but to be honest, the only thing I really needed of those three enhancements to the stock machine was ram. Mostly I'm using the machine for audio editing (large files that are usually 2 GB or larger, using Sound Track Pro) and routine web tasks. Never or almost never use the thing to play games.

So this morning I canceled the order, placed an order at OWC for 8 GB of ram (4x2), for delivery tomorrow, and went to a local Apple store and bought the stock machine.

I'm too exhausted from all this to have opened it up yet or started to transfer everything from my Power Mac (dual 2.3 G5), but I saved net about $1,050, and have a machine much more suited to my purposes with twice as much ram.

I figure I'll upgrade again in 2-3 years.

[jconly]

Buy the fastest you can afford. If total cost is an issue, getting more RAM over processor speed will be more important. The $800 or whatever you save by not getting a 3.0 buys you 16GB of ram.

Photoshop will make use of more then 4gb of ram in CS3: (quoted from an forum post I found)
"VM BUFFERING IN PHOTOSHOP CS3
On Macintosh computers, Photoshop can directly access up to about 3.5GB. When there is more than 3.5GB of document data, Photoshop writes data to its scratch files as necessary. On a computer with 4GB or less of RAM, the data is transferred directly between the scratch files on disk and the Photoshop RAM. On a computer with more than 4GB of RAM, Photoshop tells the operating system to use the extra RAM as a buffer for the Photoshop scratch file. In this case, when document data no longer fits in the 3.5GB of Photoshop RAM and is written to the scratch file, the operating system stores it in the extra RAM and can retrieve it from there much faster than it could be read from disk. This lets Photoshop take advantage of more than 4GB of RAM to significantly increase performance with very large documents."

And an article i read on barefeats: while processing a 300mb file CS3 was using about 13gb of memory on a 16gb system.

Wow, a 300MB file using 13GB of RAM.
Right before I read this, I had decided to just go with the 3.0.

But, perhaps I might be better off with 16gb of ram instead of the 10.

[drag0nreb0rn]

Crazy talk, all Crazy talk.. Most of you buy new machines every few years. Those of you that don't, this post is not for you.

As I buy a new machine every 1-3 years, getting 3.2 makes no difference whatsover. so what shave a couple of seconds off your rendering? There are upgrades that add a lot of value for low cost and upgrades that add little value for high cost. This is little value, high-cost.

If you are a gamer, the extra mhz are good, but are you really playing Crysis on this thing? And if so, get the 8800gt instead of ATI.

If you are an Adobe /final cut type of person, 8 cores does a hell of a lot of good. the extra mhz don't buy you enough to pay the inflated price. Take the extra cash and buy more ram or a blu-ray burner.

If you were already doing that, then get RAID 5 going so your data is secure.

if you are rich, then don't bother checking to see if it makes sense, you will buy the best anyway. These puppies are crazy fast for $2799. I would save your money for your next machine in 2 years.

Just an opinion.. I wouldn't waste 3 seconds considering 3.0ghz or 3.2ghz.. Now if the FSB was slower at 2.8 like it was originally rumored, then we could talk.

[kem]

To clarify more, they were running the Retouch Artists - Photoshop Speed Test (as seen in the stickied thread in this forum), but instead of using the 3.5mbjpg that comes with the test they substituted it with a 300mb file. So it a pretty intense operation that was occuring to use up that much ram.

Wow, a 300MB file using 13GB of RAM.
Right before I read this, I had decided to just go with the 3.0.

But, perhaps I might be better off with 16gb of ram instead of the 10.

[jconly]

To clarify more, they were running the Retouch Artists - Photoshop Speed Test (as seen in the stickied thread in this forum), but instead of using the 3.5mbjpg that comes with the test they substituted it with a 300mb file. So it a pretty intense operation that was occuring to use up that much ram.

I'm going to read more about the specifics of this test to see exactly what is being done to the file.

But regardless....
It's a really tough call for me. My future (3+ years) is very difficult to predict at the moment, as I'll be entering a new geographic market.

I can afford either system at the moment, with sufficient funds remaining for the proper amount of RAM and storage (I don't care THAT MUCH about the test, I'll be more then happy with 10GB.) So the decision comes down to when I think I'll be able to upgrade the system.

I've always updated on a 3/4-year cycle previous to this. But then again, this system is much more expensive than any other I've ever purchased.

But, if I won't be upgrading for 5 or 6 years, I'm going the 3.0 route. If I think I will be able to upgrade within the 3/4 year cycle, or even if I want to go through the hassle of selling and buying Nehalem, I'll buy the 2.8.

Sure I could invest the money saved and then sell of the system to upgrade on a shorter cycle.....

But honestly, the more I think about it, reliability is a necessity with the short deadlines I encounter. More frequent computer updates is a bittersweet thing. The downtime of dealing with upgrades is not an additional burden I wish to inflict upon myself, especially while managing a freelance business. In fact, after I make this purchase, I'll be doing what I can to break my ties with the time consuming evils of Apple-RSS feeds and forum trolling. Ive gotten somewhat addicted as I've been impatiently awaiting the release of these new systems. I need to drop that and put the focus back on my work.

It's still up for debate till the end of the night (as I NEED to purchase the machine tomorrow morning), but right now I think I'm leaning to the 3.0.

[MacProUser]

By the way - nobody knows what a Nehalem system exactly looks like in future. It´s a completely new architecture and I bet there will be some extra trouble when changing to one of those early models (remember when the first MacPro came out). So why not buying a proofed running system as the current Mac Pro in a better equipped version - as for example the 3 GHz model. You win at least some extra time to see how this upcoming technology (hardware AND software) will work. A Mac Pro should be useful even running for more than 3-4 years ...
Some people mentioned the economical side of the story - that I understand. They also mentioned, more RAM would boost performance more than processor speed - and I agree, too ! So in my eyes, the advice has to be:


Choose the 3.0 instead of the 3.2, get some extra RAM for the money you´ve saved - and use your system for a minimum of 5 years !


That´s what I´m going to do ;-)

[SCLlama]

I personally went for the 3.0. I contemplated the 2.8 just like you, but I like the idea that in the next "performance bump", the 3.0 will probably still be around. Though that is really only my own mind games :D. Not to mention I went for the middle of the road when I got my G5 about 5 years ago, and I feel it gave me an extra year or so of use for what I am doing.

[irishgrizzly]

I see it like a sports car – you can spend more on the model that does an extra 20kph but I'll rarely get up there to use it.

I do take the earlier point about distinguishing your model at reselling time – but in 3 years that extra .2ghz will seem like weak sauce.