well, i'm not under that impression.. the point i was trying to make with showing some of my high dollar clients was that these people are at the top of their fields and they're mainly using midrange macs in their operations/productions.
Well maybe you should consider why that might be the case, instead of them spending $5,000-10,000 on each machine.
the question meant to be posed by those examples was how come these people doing topend work are fine with midrange offerings with a few mac pros sprinkled in while people on the internet are saying these computers are too underpowered for their work
When the Mac Pro is overpriced, and the iMac is not sufficient, there's a gap in the market which is only filled by PC's or Hackintoshes right now.
Just because you
can use a slower machine, does not mean it is a smart thing to do, or a good use of your money.
There are very few times where the speed of a computer actually
prevents you from doing anything - it just makes it too slow to be worthwhile.
while conveniently leaving out any sort of details of what their needs actually are
People have clearly gone into what their requirements are several times. You only seem to be concerned with what people are doing for a living - those of us who have a need for some sort of "xMac" already
know what kind of performance we need. We don't need you to tell us.
That's why we mention specs, which you don't seem to care about.
me seeing high end production on midrange computers or me seeing zero production from someone complaining about lack of specs?
Just because something may
work does not mean it's efficient or cost-effective. You can obviously get work done on a Mac Mini or an iMac. It doesn't change the fact that you could be doing that work much faster with a high-end desktop. (and that does
not mean stepping up to a Mac Pro which is built for a different target market)
i can afford a mac pro- sure.. or maybe more importantly, i can use a mac pro to generate money in which it pays for itself
By the time new models are out, a Mac Pro will not have paid for itself for me (I definitely won't be 2-3x more productive with it compared to a $2000 PC - if anything it will be slower) and there's no clear upgrade path for current Mac Pro owners, unlike PC's, as they're using custom hardware throughout the machine.
In all likelihood, to upgrade to new hardware, you simply throw out the old machine and replace it with another $5000 purchase.
Now do you begin to understand the problem with the it?
i'm not the type of person who feels comfortable with ripping off my clients for my own luxuries
I would sincerely hope you're not in any position to be recommending hardware purchases if you're having such difficulty understanding why people might require something between an iMac and a Mac Pro.
it would be neat if for one minute you could pretend like you're perfectly content with the computers you own and see no need for an xmac.
Or maybe I'm in the market for a new system and would like to purchase something from Apple that runs OS X, but they don't offer a system that meets my requirements - requirements which are easily met with a PC, but then you have to run Windows or turn it into a Hackintosh. (which I would not count on for
work)
For
most professional tasks, OS X is completely optional. When a $2000 PC gets the job done, that a $3000 Mac Pro won't, it's difficult to not end up with a PC.
but in this thread alone, we have one guy saying an xmac is an imac without a monitor, another guy saying imac graphics suck, another guy saying an xmac requires the ability to use any gpu on the market
The two are not mutually exclusive. If the GPU is easily replaceable, there's nothing to say the base spec can't come without a GPU, or only with a low-end GPU, offering options up to the highest-end GPUs.
another guy saying it needs fast hex-i7s
If you didn't know that those "i7's" are also LGA2011 CPUs, it's an understandable request.
ok.. for the billionth time.. how about you explain your performance requirements.. not his not hers not them not they - but you exactly.
I have - multiple times in fact. I want the fastest Haswell (LGA1150) CPU that there is, paired with the fastest single GPU that offers CUDA support.
Many of us want a machine that costs ~$2000 where it's all spent on performance, and not a built-in display.
then go buy one.. seriously man, what's the hold up?
Perhaps you conveniently forgot that Apple does not make such a computer.
just to put some things into perspective as to the position i personally am talking from.. two of my apps are already leveraging openCL and have been for over a year now.. one of them, my rendering app, the head dev has told me 'the d300s should be just fine for our current implementation but before the year is up, you're going to see huge performance boosts in which the d700s are going to give you a great advantage".. so while i can understand that everybody's software won't be able to utilize the dual gpus, mine will so it's exciting and i speak from that perspective..
i mean, i'm going to be getting upwards of 30x performance gains at much much cheaper costs than if i wanted to get those types of gains via dual socket mega-core machines.. for instance- getting only 2x performance gains over a 12 core would require a 24 core system.. imagine what the cost would be if i wanted to get 30x increase via cpu.. it's unfathomable (and thankfully, apple has seen this and designed the nmp in the way they have.. whether or not people on the internet realize this is irrelevant to me but i guess it's fun to argue about .)
So because
your requirements are met by the Mac Pro's focus on OpenCL performance,
everyone's performance needs are met or should be met some time in the future when every application updates itself to use OpenCL and automatically see a 30x improvement in performance?
Or perhaps the rest of us are here in reality where our current tools use the CPU or CUDA for calculations.
These tools
may move from CUDA to OpenCL - but maybe they won't. Or maybe it will be three years from now. Or perhaps CUDA performance will be twice that of OpenCL's performance for that task.
And some of the tools we're using today likely won't move off the CPU to the GPU, or will only see very minor performance increases when moving to the GPU.
Just because your workload happens to be OpenCL accelerated (which mostly seems to be video-related applications right now) does not mean everyone else's is, or ever will be.
i'm on a 1,1 right now and it still works well.. if i had purchased another company's product 7 years ago, would it still be going strong in the same way my mac pro is or would i have had to replace it by now?
And here we see the real reason why you are happy to spend $5,000 on the system, and don't want to talk specs.
A seven year old computer is not close to offering the performance I need.
You would not have been
required to replace a PC, but it would have been a lot easier to keep it up to date as you only need to swap out the relevant parts. Want a new CPU? Drop it in - though you might have to replace the motherboard at the same time. Can't do that easily with a Mac. Same goes for the GPU.
If you see the computer as an appliance you only upgrade once every five years or more, then I guess spending $5,000 on it makes some amount of sense.
We upgrade our systems every two years on average - though it has been three this time with Intel's push for power efficiency rather than performance lately.
If you were to spend $1000 on upgrades each year rather than $5000 every five years, you would be far more productive over the course of those five years, rather than waiting and making a big jump in performance.
)
