Will there be a new Mac Pro in 2015?

[deconstruct60]

Just to get back on this thread's subject line.... I feel like the new retina iMac has opened a path for a cheaper cylindrical mac pro?

Just the opposite. The retina iMac took the old $2,499 price point that the entry Mac Pro sat in for 3-4 years. Apple filled the "hole" they created when bumped the minimal entry point to the Mac Pro up to $2,999

As the retina iMac goes mainstream over time ( retina 21" and then dropping of non retina models all together .... except perhaps lowest, gutted edu focused versions ) then maybe there will be a price point for the Mac Pro to drop down into.

Not likely over next couple of iterations. For several factors.

1) Single storage drive ( hence pressure to crank the SSD capacity up)
[Apple could put in a m.2 SATA socket on Compute GPU and use the chipset SATA lines but I doubt it. ]

2) "Fire Pro". .... as long as using that to mark up the GPUs , I doubt going to sink low enough to do two and still hit old lower entry price points.

3) For next couple of years DDR4 isn't at the tail end of the maturity curve.

4) Intel prices aren't coming down in the Xeon E5 space.

Since a mac pro doesn't have a screen, Apple could potentially build the same innards that is in the retina Imac and put it in a cylindrical shape and make it cheaper.

That makes no sense at all. You are simply just trying to reuse the case "just because". There is no need for a huge three sided heat sink with iMac innards. Folks would likely have their underwear in even bigger twist if coupled the iMac GPU to a container that had a much bigger thermal envelope.

The Retina iMac uses the following $213 part

http://ark.intel.com/products/80810

The nMP uses the following $294 part

http://ark.intel.com/products/75779/Intel-Xeon-Processor-E5-1620-v2-10M-Cache-3_70-GHz


An $81 swing in CPU package cost isn't going to cause a huge drop in system costs or pricing.

It is the same farce "Core i7" is radically cheaper claim that has gone no where for years .... still going no where in making huge changes to Apple system pricing.

 

[lowendlinux]

Yes they are. In terms of getting applications to certify on top of them they will pull out the "my list is longer than your list" . The OS doesn't matter as much. Broad spectrum users deal with Apps, not OS. Linux's "compile it from scratch open source " crowd often just masks the disconnect more than anything else.



They are asking different subsets of people and doing what the different subsets want. Kind of like a restaurant district where they all serve food but there are Chinese , Indian , Italian , etc. focused restaurants. Too many chiefs with too many different priority orderings.

If it was the same set of folks all asking for the exact same things there would be tons more convergence in the Linux space. There isn't.

I apologize after I made the comment I realized he was probably talking about distos where I was talking about desktops i.e. Gnome, KDE, FVWM, Mate, e etc. The distros are in some sense, competing but the reality is that if something is made for Linux you will be able to use it on your flavor of Linux. There has been a great deal of push recently to have a universal installer for Linux but I don't know if Poettering and Linus himself will be the distros to adopt something like that. IMHO Linux enjoys being the outsider or rebel and to remove that removes the identity of Linux and it's distros.

 

[Nosferax]

Yes they are. In terms of getting applications to certify on top of them they will pull out the "my list is longer than your list" . The OS doesn't matter as much. Broad spectrum users deal with Apps, not OS. Linux's "compile it from scratch open source " crowd often just masks the disconnect more than anything else.



They are asking different subsets of people and doing what the different subsets want. Kind of like a restaurant district where they all serve food but there are Chinese , Indian , Italian , etc. focused restaurants. Too many chiefs with too many different priority orderings.

If it was the same set of folks all asking for the exact same things there would be tons more convergence in the Linux space. There isn't.

There not competing. You can run KDE/Qt apps on a gnome based distro just like you can run Gnome apps on a KDE/Qt based distro. People develop with what they are confortable with. Linux is about choice and the absence of an overseer.

You can also install apps from different package managers on just about any distro.

 

[burnsranch]

Lets look at a new generation real world application.

Take a small police force with 30 8 hours shifts and they video the entire shifts. Everyday they have 240 hours of video to store. Then they have to have a video data base to search and recover data.

If you take this and move it to an educational class, where a random real world situation comes up and now the instructor has the instant ability to search the online video database for multiple real world videos and can use the real world examples to search and teach from. Yes you have to edit the dead time, that is probably pretty easy to do automatically, maybe by measuring heart rate.

I am curious what system out there would be better than a nMP and some thunderbolt raid drives? It is portable and could drive a big TV, search and edit on the fly, the possibilities are endless.

 

[handsome pete]

Lets look at a new generation real world application.

Take a small police force with 30 8 hours shifts and they video the entire shifts. Everyday they have 240 hours of video to store. Then they have to have a video data base to search and recover data.

If you take this and move it to an educational class, where a random real world situation comes up and now the instructor has the instant ability to search the online video database for multiple real world videos and can use the real world examples to search and teach from. Yes you have to edit the dead time, that is probably pretty easy to do automatically, maybe by measuring heart rate.

I am curious what system out there would be better than a nMP and some thunderbolt raid drives? It is portable and could drive a big TV, search and edit on the fly, the possibilities are endless.

Maybe I'm misreading you're whole hypothetical, but why does any of that require a Mac Pro? Why does it need to be portable? If anything, wouldn't such technology move in the direction of direct streaming from the cameras to the server?

 

[jmcube]

I bought a MP last August, with work funds.

I used it in Hamburg, Lecce, Santa Barbara, and a limited amount in Chicago as well as Valparaiso where I am based.

I primarily used the increased computational power although I used the GPUs (computationally) a bit.

Overall it is worth it, although not as much as the 2012 MBP (also purchased with work funds). That machine I have dropped more times than I can count, and have spilled water and coffee on (I think water twice?). I really hope they keep at least the 13inch MBP going as I doubt the rMBP is as robust (I know the MBA is not as robust).

I would have been stuck with the MBP if I had not had the MP, so the form factor is a lot of the reason I am satisfied with it.

JM

 

[goMac]

I am curious what system out there would be better than a nMP and some thunderbolt raid drives? It is portable and could drive a big TV, search and edit on the fly, the possibilities are endless.

A Mac Mini with some Thunderbolt drives would be much cheaper, and do the same thing. Or even a Macbook Air or a Macbook Pro.

The thing about Thunderbolt is it's kind of equalized expansion possibilities over all the different Macs. A Mac Pro is no longer required for expansion. Needing a lot of storage of a lot of custom hardware no longer automatically means a Mac Pro.

The biggest reason to go for a Mac Pro at this point is a lot of CPU power, which you don't need for that situation. I'd say GPU power, but even the home grown Thunderbolt GPU solutions are pulling ahead on that.

 

[walterfog]

hello at all... i just need to have more information about a probably upgrade of the new Mac Pro in the 2015.
I decide to change my old 2008 Mac Pro 8 cores... he was a great machine nut now in the time to change.. so i was wondering to buy a new mac pro.. but i hear about the new xeon e5 v3 upcoming... and now i don't know if buy now the new mac pro or wait till the new upgrade...
there are more exact information about the next upgrade?
does the performance will greatly boost or change from the actually configuration?
the support for ddr4 ram will increase the performance considerably?
thanks a lot for the information
bye
walter

 

[Nosferax]

hello at all... i just need to have more information about a probably upgrade of the new Mac Pro in the 2015.
I decide to change my old 2008 Mac Pro 8 cores... he was a great machine nut now in the time to change.. so i was wondering to buy a new mac pro.. but i hear about the new xeon e5 v3 upcoming... and now i don't know if buy now the new mac pro or wait till the new upgrade...
there are more exact information about the next upgrade?
does the performance will greatly boost or change from the actually configuration?
the support for ddr4 ram will increase the performance considerably?
thanks a lot for the information
bye
walter

The v3 probably won't bring that much of a speed boost over the v2. DDR4 might but even then the difference is measured in ms... The greatest boost would come from up to date gpu.

 

[burnsranch]

"Maybe I'm misreading you're whole hypothetical, but why does any of that require a Mac Pro? Why does it need to be portable? If anything, wouldn't such technology move in the direction of direct streaming from the cameras to the server?"

It is the video data base that is the next technology wave. My 500MBs per second raid drive is almost fast enough for seamless searches through 12TB of videos. Videos has now become cheap.
Let say you are monitoring a pond for wild life. you could set up cameras monitor realtime in HD videos and build searchable data base for a years activity a that pond. That would give you a searchable realtime resource. You might be researching deer, someone else might be researching birds, It does not matter what you are looking for, you have a one year searchable database you can use forever.

My data base searches work OK at 500MBs, my 4k downwards are 10 minutes per hour of video at 40mbs. It take about 10 minutes to transcode an hour of 4k videos into the database so it is searchable.
I can quick scan a 4k video with the cursor almost as quick as I can move the cursor.

Now I have a working data base on a thunderbolt network, there is no reason I cannot hang unlimited CPU units off of that database. One to view and edit, on to import and transcode raw video, another one to render edited projects.


The people who built all of this technology look at from the inside, not from as a tool from the outside. If I have 5 years of a raw video database and technology creates a new wizbang, I just plug it into my database. New editor, just plug it into the database.
I can run my database on TB2 TB1 Usb3 usb2. it sucks at the lower speeds but it is usable.

If I stress test my drives, I and run them all at full speed at the same time on my nMP. So I have almost unlimited capability to build a huge video database off of a $3000 CPU. If you take that wildlife video database and let every school use the video database a $3000 cpu and $2000 raid drive, they can search and research a years worth of real time natural activity video from a laptop in real time.

If I look at the value of one years worth of realtime data for a natural event and the ability to search that videos data base, do I really care it take 1 hour or 2 hours to render my edited project? The value is now in the video database, not the rendered end product. That is a paradigm shift in how we deal with video.

While it might make sense to use the cloud to render a final product, there is now you can get the 500+mbs needed for a video data base to work over the cloud today. We were at 10mbs at best on the cloud?

I really do not know if the nMP is the best tool for this job. I really did not expect this functionality, this soon. I sure was surprise to see how well it all works and now I need to restructure my project from a file based video library to a searchable video database. I am lucky because I started my project 5 years ago, knowing we would have the technology to build video databases.

I have a real world application for my nMP, I am in the process of building a realtime video database of a professional 12 year horsemanship apprenticeship. Once I have the raw video in a searchable database, the output possibilities are endless. This is the exact opposite from today's video professionals who are used to working with expensive video and only shoot what they need for their finished project.

 

[beaker7]

"Maybe I'm misreading you're whole hypothetical, but why does any of that require a Mac Pro? Why does it need to be portable? If anything, wouldn't such technology move in the direction of direct streaming from the cameras to the server?"

It is the video data base that is the next technology wave. My 500MBs per second raid drive is almost fast enough for seamless searches through 12TB of videos. Videos has now become cheap.
Let say you are monitoring a pond for wild life. you could set up cameras monitor realtime in HD videos and build searchable data base for a years activity a that pond. That would give you a searchable realtime resource. You might be researching deer, someone else might be researching birds, It does not matter what you are looking for, you have a one year searchable database you can use forever.

My data base searches work OK at 500MBs, my 4k downwards are 10 minutes per hour of video at 40mbs. It take about 10 minutes to transcode an hour of 4k videos into the database so it is searchable.
I can quick scan a 4k video with the cursor almost as quick as I can move the cursor.

Now I have a working data base on a thunderbolt network, there is no reason I cannot hang unlimited CPU units off of that database. One to view and edit, on to import and transcode raw video, another one to render edited projects.


The people who built all of this technology look at from the inside, not from as a tool from the outside. If I have 5 years of a raw video database and technology creates a new wizbang, I just plug it into my database. New editor, just plug it into the database.
I can run my database on TB2 TB1 Usb3 usb2. it sucks at the lower speeds but it is usable.

If I stress test my drives, I and run them all at full speed at the same time on my nMP. So I have almost unlimited capability to build a huge video database off of a $3000 CPU. If you take that wildlife video database and let every school use the video database a $3000 cpu and $2000 raid drive, they can search and research a years worth of real time natural activity video from a laptop in real time.

If I look at the value of one years worth of realtime data for a natural event and the ability to search that videos data base, do I really care it take 1 hour or 2 hours to render my edited project? The value is now in the video database, not the rendered end product. That is a paradigm shift in how we deal with video.

While it might make sense to use the cloud to render a final product, there is now you can get the 500+mbs needed for a video data base to work over the cloud today. We were at 10mbs at best on the cloud?

I really do not know if the nMP is the best tool for this job. I really did not expect this functionality, this soon. I sure was surprise to see how well it all works and now I need to restructure my project from a file based video library to a searchable video database. I am lucky because I started my project 5 years ago, knowing we would have the technology to build video databases.

I have a real world application for my nMP, I am in the process of building a realtime video database of a professional 12 year horsemanship apprenticeship. Once I have the raw video in a searchable database, the output possibilities are endless. This is the exact opposite from today's video professionals who are used to working with expensive video and only shoot what they need for their finished project.

This is all great but was all possible before the nMP.

 

[handsome pete]

"Maybe I'm misreading you're whole hypothetical, but why does any of that require a Mac Pro? Why does it need to be portable? If anything, wouldn't such technology move in the direction of direct streaming from the cameras to the server?"

It is the video data base that is the next technology wave. My 500MBs per second raid drive is almost fast enough for seamless searches through 12TB of videos. Videos has now become cheap.
Let say you are monitoring a pond for wild life. you could set up cameras monitor realtime in HD videos and build searchable data base for a years activity a that pond. That would give you a searchable realtime resource. You might be researching deer, someone else might be researching birds, It does not matter what you are looking for, you have a one year searchable database you can use forever.

My data base searches work OK at 500MBs, my 4k downwards are 10 minutes per hour of video at 40mbs. It take about 10 minutes to transcode an hour of 4k videos into the database so it is searchable.
I can quick scan a 4k video with the cursor almost as quick as I can move the cursor.

Now I have a working data base on a thunderbolt network, there is no reason I cannot hang unlimited CPU units off of that database. One to view and edit, on to import and transcode raw video, another one to render edited projects.


The people who built all of this technology look at from the inside, not from as a tool from the outside. If I have 5 years of a raw video database and technology creates a new wizbang, I just plug it into my database. New editor, just plug it into the database.
I can run my database on TB2 TB1 Usb3 usb2. it sucks at the lower speeds but it is usable.

If I stress test my drives, I and run them all at full speed at the same time on my nMP. So I have almost unlimited capability to build a huge video database off of a $3000 CPU. If you take that wildlife video database and let every school use the video database a $3000 cpu and $2000 raid drive, they can search and research a years worth of real time natural activity video from a laptop in real time.

If I look at the value of one years worth of realtime data for a natural event and the ability to search that videos data base, do I really care it take 1 hour or 2 hours to render my edited project? The value is now in the video database, not the rendered end product. That is a paradigm shift in how we deal with video.

While it might make sense to use the cloud to render a final product, there is now you can get the 500+mbs needed for a video data base to work over the cloud today. We were at 10mbs at best on the cloud?

I really do not know if the nMP is the best tool for this job. I really did not expect this functionality, this soon. I sure was surprise to see how well it all works and now I need to restructure my project from a file based video library to a searchable video database. I am lucky because I started my project 5 years ago, knowing we would have the technology to build video databases.

I have a real world application for my nMP, I am in the process of building a realtime video database of a professional 12 year horsemanship apprenticeship. Once I have the raw video in a searchable database, the output possibilities are endless. This is the exact opposite from today's video professionals who are used to working with expensive video and only shoot what they need for their finished project.

You didn't really answer my question. None of what you said is unique to the Mac Pro.

 

[MacProCard]

You didn't really answer my question. None of what you said is unique to the Mac Pro.

Then why did they make the nMP? Why did they make the new iPhone 6? Why does Intel keep making new processors?

What is unique to the nMP? Please tell us your wisdom.

 

[goMac]

Then why did they make the nMP? Why did they make the new iPhone 6? Why does Intel keep making new processors?

What is unique to the nMP? Please tell us your wisdom.

Because some people need faster computers? I'm not really sure how this is relevant though. The above scenario does not require a faster, single computer. Kind of a random question.

 

[MacProCard]

Because some people need faster computers?

Here let me try it. Sorry GoMac, none of that is unique to the nMP.

 

[handsome pete]

Then why did they make the nMP? Why did they make the new iPhone 6? Why does Intel keep making new processors?

What is unique to the nMP? Please tell us your wisdom.

How about you try to refute what I said rather than just pose a bunch of non sequiturs to hide the fact that you don't have a coherent response?

 

[MacProCard]

How about you try to refute what I said rather than just pose a bunch of non sequiturs to hide the fact that you don't have a coherent response?

Don't get me wrong, it's hard to tell what you really mean with you 2 sentence posts of superiority. But according to your logic ...NOTHING is really unique to the nMP. --TB 2(data transmission)? Nope, data transmission has been around since data. Nobody really needs to move data that fast anyways. --Crunching data faster? Nope, crunching data has been around since data. Nobody really needs to crunch data that fast anyways. They could use a mac mini.

 

[Nosferax]

Don't get me wrong, it's hard to tell what you really mean with you 2 sentence posts of superiority. But according to your logic ...NOTHING is really unique to the nMP. --TB 2(data transmission)? Nope, data transmission has been around since data. Nobody really needs to move data that fast anyways. --Crunching data faster? Nope, crunching data has been around since data. Nobody really needs to crunch data that fast anyways. They could use a mac mini.

The case scenario that you and your friend keep arguing about isn't exclusive to the nMP. The nMP isn't even the most useful and powerful computer to use in that scenario. The nMP is a single CPU computer with 2x 3 years+ old GPUs. It is already outclassed by all the other workstation presently in the market where you can get more than 2x bleeding edges GPUs and multiple CPUs.

I have a lab full of HP workstations processing multiple Go Pro HD feeds from helicopters and geolocalizing those feed into ArcGIS in realtime presently. So, no the nMP isn't the be all in this and probably never will. it will always be eclipsed by more open machine not tied to whatever Apple decides you should use hardware wise.

 

[burnsranch]

The issue with new technology is you have to do something new, not just old things better.
I spent 20 years at the top of the tech world, so I understand it,

Have got 5 years of natural horse videos in a video data base. If I want to make a videos of foals under three days of age, I can scan 5 years of videos and cut and past them into a new event. With my 500MBS TB2 diver the process is seamless.

Importing the data is slow and exporting the data is slow, in real world terms.

People claim other platforms work as good or better, but if you test video database functionality, not rendering or editing or exporting.

If you do performance on the three thunderbolt controllers of the nMP, which i have not seen done, I suspect they are leading the industry.

I suspect the video handling of the video database is also right up there.

If I put a rough number to the external drive bandwidth to support a working videos database, not not a video file system, I would put the number at 500MBs.

I have been building a videos database for five years, because I understood the technology would get there. I am isolated from the tech world, but when I research for issues I have to work on for my video database, It is clear to see people don't think in terms of building video databases. They are building video file systems.

The windows platform has huge design issues when it comes to speed.
If you take packet switching, it is a relatively simple task to process packets with a cpu. It is a difficult task to code micro code and silicon to do those tasks quickly.

The basic design of the windows machine only really supports processor speeds, not microcode silicon speeds. It will take a machine like the nMP to support faster speeds.

If the functionality is in the video data base application, there are few people who would understand that functionality, because they have not had the hardware to build functional video data bases.
When I can work faster than real time with 4k videos on multiple external drives, someone has been putting a lot of effort into that functionality.
I can scroll though 60 hours of video in minutes with hardly no hesitation.
Last years library has 100 2 hour events I can scroll through all of them as fast as I can move the cursor whit no hesitation. I can load through multiple years of data, and scroll though all of them with no hesitation.

That is the functionality I am supposed at. I can scroll from slow motion to super fast from a video database, not a program loaded into memory or cached.

 

[MacProCard]

The case scenario that you and your friend keep arguing about isn't exclusive to the nMP. The nMP isn't even the most useful and powerful computer to use in that scenario. The nMP is a single CPU computer with 2x 3 years+ old GPUs. It is already outclassed by all the other workstation presently in the market where you can get more than 2x bleeding edges GPUs and multiple CPUs.

I have a lab full of HP workstations processing multiple Go Pro HD feeds from helicopters and geolocalizing those feed into ArcGIS in realtime presently. So, no the nMP isn't the be all in this and probably never will. it will always be eclipsed by more open machine not tied to whatever Apple decides you should use hardware wise.

I never said it was the fastest anything. Just that it's a quality option for anybody in the market for a powerful desktop computer.

----------

The issue with new technology is you have to do something new, not just old things better.
I spent 20 years at the top of the tech world, so I understand it,

Have got 5 years of natural horse videos in a video data base. If I want to make a videos of foals under three days of age, I can scan 5 years of videos and cut and past them into a new event. With my 500MBS TB2 diver the process is seamless.

Importing the data is slow and exporting the data is slow, in real world terms.

People claim other platforms work as good or better, but if you test video database functionality, not rendering or editing or exporting.

If you do performance on the three thunderbolt controllers of the nMP, which i have not seen done, I suspect they are leading the industry.

I suspect the video handling of the video database is also right up there.

If I put a rough number to the external drive bandwidth to support a working videos database, not not a video file system, I would put the number at 500MBs.

I have been building a videos database for five years, because I understood the technology would get there. I am isolated from the tech world, but when I research for issues I have to work on for my video database, It is clear to see people don't think in terms of building video databases. They are building video file systems.

The windows platform has huge design issues when it comes to speed.
If you take packet switching, it is a relatively simple task to process packets with a cpu. It is a difficult task to code micro code and silicon to do those tasks quickly.

The basic design of the windows machine only really supports processor speeds, not microcode silicon speeds. It will take a machine like the nMP to support faster speeds.

If the functionality is in the video data base application, there are few people who would understand that functionality, because they have not had the hardware to build functional video data bases.
When I can work faster than real time with 4k videos on multiple external drives, someone has been putting a lot of effort into that functionality.
I can scroll though 60 hours of video in minutes with hardly no hesitation.
Last years library has 100 2 hour events I can scroll through all of them as fast as I can move the cursor whit no hesitation. I can load through multiple years of data, and scroll though all of them with no hesitation.

That is the functionality I am supposed at. I can scroll from slow motion to super fast from a video database, not a program loaded into memory or cached.

So what database backend are you using for video? Because I know of one particular real world, big dollar app, that is choking on large image datasets.

 

[handsome pete]

Don't get me wrong, it's hard to tell what you really mean with you 2 sentence posts of superiority. But according to your logic ...NOTHING is really unique to the nMP. --TB 2(data transmission)? Nope, data transmission has been around since data. Nobody really needs to move data that fast anyways. --Crunching data faster? Nope, crunching data has been around since data. Nobody really needs to crunch data that fast anyways. They could use a mac mini.

Look, I don't blame you for however you're choosing to perceive my responses. Tone is often hard to pick up on when reading stuff. But in no way am I trying to take some sort of superior attitude around here. So I wouldn't get too bent out of shape about it.

But I (as well as others) have responded to certain claims that the nMP is a "perfect device" for certain circumstances simply by saying that the nMP is not unique for those instances. There are other machines out there. Possibly cheaper. Possibly more efficient. For whatever reason you seem to take offense to this kind of discussion.

So what is truly unique about the nMP? Well the only things I can really think of are its form factor and the amount of TB ports. Whether or not those are unique enough traits to drive sales remains a mystery. Again, we're all just speculating here. Only Apple really knows their intent behind the Mac Pro revamp. I'm skeptical about its current path, but wouldn't be surprised if it ends up working out. However, I really wouldn't be surprised either if Apple ends up getting out of the workstation segment altogether.

Data transmission? Of course thunderbolt excels here, but that doesn't negate any of the other fast transfer protocols out there. And what did TB really replace? PCIe slots. Other machines have them in droves. Data crunching? Are you really trying to tout this as a nMP strength versus its competitors?

This thread has veered wildly off topic, but that's to be expected since there's really only so much any of us can say about will there or won't there be an update this year. So instead we're all speculating on the future of the device and what Apple's intent is with the device. So lighten up a bit and don't get so bent out of shape when someone offers a dissenting opinion. Just respond with some logic in kind.

 

[Piggie]

I would buy a "Gaming" Mac Pro in a second.

Stick on some overclocked? Intel i7 Skylake? CPU or CPU's.
A Couple of GTX980's or 990's
Couple of half terabyte SSD's
32GB of DDR3 fast Ram.

God that would be so amazingly popular.
I'd love if Apple did that.

Please make that Apple.

They won't of course, but I can dream!

 

[goMac]

Crunching data faster? Nope, crunching data has been around since data. Nobody really needs to crunch data that fast anyways. They could use a mac mini.

Here let me try it. Sorry GoMac, none of that is unique to the nMP.

I never said it was the fastest anything. Just that it's a quality option for anybody in the market for a powerful desktop computer.

Are you just intentionally trolling now?

Saying random double-talk isn't a substitute for intelligent conversation.

 

[beaker7]

If your primary concerns are data crunching and IO throughput for video I have no idea why you'd use a Mac Pro with so many faster options out there.

 

[thedommer]

I kinda hope there isn't a nMP in 2015. Forces me to use my early 2008 one more year. haha. nMP 2016 wooooooo