The Mac Pro is Dead - Page 3

That's what the iMac is, and that's why you'll always have to go to someone else to get this.

Unfortunately the iMac doesn't have PCIe slots, extra HDD bays or dual display support.

I can't think of any compelling reason why the single-CPU MacPro has to be a Xeon. Or why it can't sell for $1999

I'll never buy another AIO. I bought the iMac back in 1997 and regretted it a year later. Bought the G4 933 in 2002 and it served me until 2009, when I bought the mini. It's showing its age already, and my next leap will be probably a Mac Pro. Dead? Hardly.

I don't know what Apples mind set is with respect to new Mac models, however they should realize by now that they have turn the Mac Business into a success. They can further drive sales by simply meeting the needs of more users. Frankly bringing more new users into the fold should be a high priority thus the need for a Mac that is attractive to switchers. The Mini is not that Mac.

You see right off the bat you mis the point. Pros don't really care all that much about the Mac Pros giant case, for many it is exactly what they need. However once you step outside of the realm of pros that actually use all of that power the Mac Pro becomes huge and impractical. It is simply to large for many desktops.

In the end a smaller case saves Apple money. Further reducing wasted space means a reduction in other components like fans. In the end you want an easily transported machine that minimizes wasted desk space.

As a side note I've been flirting with an idea that puts a PC into a half rack instrument case. Imagine a HP DVM or similar instrument. With today's components a xMac could very comfortably fit into the case. It would make for a nice standalone PC that could easily be converted into a rack mount. In essence you kill two birds with one stone. Styling is up to Apple but many manufactures make good looking instruments in this form factor.

Well modifications would be required but why not make a motherboard tailored to the platform? I look at it this way if you don't design something that is appealing and capable of being sold in high numbers you might as well stay home. It is sort of the same rational as is used to build the Mini, design to a set of goals.

Those four items should have a major impact price.

I don't really believe you can escape R&D costs because frankly Apple has already tried that with the single processor Mac Pro. You need to work from the ground up to make a cost effective platform. Honestly Apple should be able to hit the $1100 mark with a modestly performing base model. Remember it doesn't cost a lot to build a machine with double the cores of the Mini, better performing cores at that.

As are the rest of us!!! I honestly believe that Apple needs to put effort into this to be successful. Of late the Mac Pro is an example of a platform demonstrating a lack of effort. With the right effort Apple could be rewarded with strong sales.

I just bought the 12 core one recently. If you are a professional and make your living using a Mac, this is the computer to have. The case is old in design but is a marvelous of tech if you look inside. Very well build and designed.
The damn computer is fast as hell.
I just hope at some point Apple will bring back the 30" or plus with a redesigned look and matte screen.

It is dead in the sense that it will inevitably die. Storage is moving towards SSD and can be as small as the tiny cards in the Air. It may even merge with RAM at some point - that's the ideal scenario. So they don't need 3.5" drives.

PCI performance can be matched by Light Peak for add-on cards, in fact USB 3 will suffice for a lot of things.

GPUs are eventually going to be best merged with the CPU and sharing the same memory. The CPUs will keep doubling in core count every couple of years.

I don't expect to see a Mac Pro in 10 years. What I do expect is a $600 16-core Mac Mini with an IGP 4x faster than a GTX 480 that can render higher than the following quality:

http://www.youtube.com/watch?v=Hp9wBqjSCg0

running at 1080p and 120FPS. By then, 256GB SSD should be in the base model, maybe even 512GB and there will certainly be ports fast enough to handle larger external storage.

Now, people may say there's still a need for the Mac Pro even at this stage because rendering 3D graphics animations will still take forever on it. I'd say no. Firstly, this process will be sped up dramatically using the GPU and algorithms will be tweaked to suit. Secondly, there is a point where calculations will rely too much on data from local storage, which will create a bottleneck in a single machine.

The solution of course is to buy 30x Mac Minis, which take up the space of 1 current Mac Pro and then you have a 480-core renderfarm with 7TB SSD and 240GB RAM all for just $18000 and will draw under 1kW of power and you don't hit the IO bottleneck.

Even if the Mac Pro were to reach a similar spec, there's no reason it would remain at the size and price it's at currently.

Long story short: within 10 years, the Mac Pro will get smaller to become the mid-sized tower we are talking about now and the Mini will become fast enough to satisfy most of the requirements for this mid-range tower.

We might also have 16-core iPads though. A 15" 1600x1200 iPad with even 8-core 64-bit ARM, 8GB RAM and 256-512GB SSD would be a pretty awesome little device without some of the iOS restrictions.

However I still see a limited number of people needing a Mac Pro type computer. Frankly they will be the same sorts of people that put the current Mac Pros to work using all of the available cores.

The near future though is likely to be focused on a move to smaller form factor machines. 32nm and smaller feature sizes means that we will see incredibly dense SoC implementations that run very cool. We could potentially see a 4 core 8 thread Mini by next year that uses less power than the current machine. This year is likely to see a Sandy Bridge based Mini, which again should dramatically improve performance and lower power usage. Of course Apple will have to stop treating the Mini like a poor step child but that is another thing all together.

Minis problem is that its size dictates the use of Mobile processors in the lower power ranges. This isn't bad at all as the Mini is a really nice platform for many users. It is a problem however for those that look at computing hardware as a platform for implementing what ever their imagination conjures up. Mini simply doesn't have the flexibility of a machine with easily accessed disk drive slots or a PCI slot or two. However this mythical XMac doesn't have to be a massive tower these days as we have already established that SoC tech means that these machines can be made extremely small. Combine this with far smaller drives, power supplies and basically everything else and you can see where a small but upgradeable machine isn't all that much of a stretch.
We don't need massive tower like machines anymore to make a PC flexible.

Well, their computer business has increased exponentially thanks to their success with consumer electronics. And the majority of their users are just fine with a Macbook or iMac. Their design philosophy works. I doubt theyd change course because a few people still want in-the-box expandability. The iMac suffices for my everyday computing needs, just not for music/audio. But Id wager we are an insignificant portion of the market.



Perhaps I do miss the point, but Im just thinking out loud about how Apple could make a more affordable, expandable machine with minimum investment. Personally, Id much rather have the MacPro in a 2U rackmount case, with detachable mounts for those who dont need it. Towers are sooo 20th century. But designing an Xmac from the ground up means dedicating $$$ and human resources to a project that is not likely to appeal to the masses.

As it is, I dont see why the current entry-level MacPro cant sell profitably for $1999. With the modifications I mentioned in a previous post I figure it would possible to get the price further down, perhaps to $1499


Well, the onus is on us (pun not intended) as Apple computers continue to sell better than ever. But I agree that the latest incarnations of the MacPro seem somewhat half-hearted. Apples product road map so far does not bode well for power users.

Let's hope their iWatch will make up for it

One step at a time.
Imagination Technologies unveils Series 6 PowerVR GPUs (2014)

FWIW, I expect a 15" iPad to have at least 2048*1536 in a couple of years, and probably 3840x2400 WQUXGA (300 ppi) in less than 5 years.

http://www.cringely.com/

If he is right would this do away with the Mac Pro?

I find it laughable that people always bring up the PCI issue with the iMacs. 99% of users are not going to be upgrading their graphics cards.

I find it laughable that people always bring up the graphics card as the only use of PCI expansion.

Seriously, if you're actually working on stuff that requires PCI expansion then paying for a Mac Pro shouldn't be an issue. If it is an issue, then it sounds like you're not billing enough.

Where is this coming from?

There are much more uses for PCI(e) cards than just graphics. Period.

Did I even say I was working on stuff that requires PCI expansion?
Did I mention the Mac Pro? Did I mention $$$?
Seriously, learn to read.

Of course there are more uses for it. All of which are pretty much reserved for professionals or hobbyists.

Did I even say I was working on stuff that requires PCI expansion?

No, you didn't but clearly I was speaking of the proverbial "you."

Did I mention the Mac Pro? Did I mention $$$?

You don't have to mention the Mac Pro. This is a Mac Pro thread. That's what we're talking about.

Again, you're assuming that I'm interested in your justification for the content of your previous post. I'm not. You just didn't need to quote my post to make your statement. That's what made it not clear at all.

.....

For the majority of users the iMac fits the bill better, but I suppose for some pro users who for instance need workstation graphics cards (which the pro is lacking in anyways, but if they NEED OSX) or ECC memory, the Pro is still needed. I can't imagine volume being very high, but if its still around I guess it means people are still buying it.

As for its fate, I'm not certain. Maybe it will be discontinued like the xServe, who knows. That would certainly irk pro users/developers though.

A lot of people think that because Apple doesn't offer Workstation GPUs as BTO that they're not available. That's not true. In fact, there are two Workstation GPUs available on Apple's own site:

http://store.apple.com/us/product/H3...k&s=topSellers

and

http://store.apple.com/us/product/TW...co=MTA4MzU2Nzg

Many professional creators use Macpro's to do their work on and cost is largely irrelevant as it is recovered through business and thus is not aimed at the same people.
With the onset of cloud computing iMacs and Macbook air/pro are perfectly set to become the intelligent domestic user terminals of the future running suitable small apps.

But what professional users need is a re-designed MacPro, maybe half the volume size of todays having just a motherboard with twin zeon multicore processors 32gbt of RAM slots and 8 full PCIexpress slots capacity housing upto six hotswopable 2terabt SSD cards or whatever and a full graphics card for twin monitor workstation use or extra 2 SSD slots for stackable server node use, one design can service all pro users in all studio requirements and the machine can still be bought by funky geeks like me and you wishing to show off OK! and NO despite the wiring nightmare thirty minies just doesn't cut the mustard does it. (that is a rhetorical question btw)

If Apple used such a new sleek streamlined design in their own datacentre it would be quite cost effective to re-develop the Macpro and would also bring professional services back on their cloud nine campus so to speak for the foreseeable future.

I agree:

How can you half the volume and double the number of slots? Also 2TB SSD cards cost about $8,000 each:

http://www.newegg.com/Product/Produc...82E16820227516

4 of them would be $32,000. I reckon it might be a while before that pays for itself unless it's big enough to live in.


You could always have 40 Minis.



The clone wars begun they have.

There would be a few wires hanging around but someone just needs to design a rack for them with the wires concealed.

OK Marvin, guess you didn't understand the word "rhetorical"... good luck with your 40x $1k mini DVD nightmare wired work out of the box rack, suppose someone had just got to do it!.
meanwhile I thought we were discussing the professional future of the MacPro a nice neat modern single box solution for Service studio and business with no mechanical drives as will be all future computers,

22nm zeon and 22nm AMD GPUs and new 20nm SSDs will run much cooler requiring reduced PSUs the whole could easily be housed in a new design case without legacy mechanical drives, for eg: 3 unit hight with attachable 19"rack mount wings built to accommodate 8 PCIe cards affording an unequalled variation of specialist build potential much cheaper than mutiple mini mayhem???, oops that is rhetorical aswell.

Such a PCIe based design can be retro fitted for any specialist requirement as needed such as Photographic studios, recording DAW, or Video editing suite, or server node, two or more units could be totem poled for expansion.
Professional service where the speed of real time productivity counts, the macpro is the bedrock of professional MacOS computing.

What I would like to see is a short 3u tower that could be easily rack-mounted in groups of three so that you could have three machines per 3u. This machine could therefor be used as a desktop, rack-mounted workstation, or as a server. Hot-swappable drive bays with SSD sleds would be as useful in a studio setting as they would be in the server room. Throw in an optional hardware redundancy option for power supplies and Apple would kill two birds with one stone: satisfy high-end pros and people still pissed about the ill-fated XServe. Really, there are so many similarities between a workstation and a workgroup server it only makes sense to consolidate them into one machine.

This.But I want a 2U chassis.

Hi Michael, I can see where your coming from but the Macpro is aimed squarely at studio pro type of business, we live in a 19" rack mounted world as the likes of real world server farms.
short towers are a product of the domestic world that the mini is designed to accommodate, but in a studio we throw about gbts of raw audio and video data compiling and rendering terabts of film footage on the fly in virtual real time which is what the Macpro with its duel zeon processors is designed to do.
the Xserve was technically surplus to requirement, now the Macpro needs to be redesigned to fulfil all things to all pro server and studio users and a multiple PCIe based frame is the way to go, literally everything from SSD to GPUs and break in box cards is PCIe based even legacy harddrives if your so disadvantaged, this is not a platform for the meek, it costs thousands because we make money to afford this level of professional edit suit computing its business.

If you need a domestic computer then stop winging and get an iMac mini or a Macbook because before you could exhaust what you have there will be another better one coming along alot cheaper than a Macpro to satisfy your needs.
But for Macpro professional users we need a whole new generation and designer machine by next year to spend big bucks on for sure and this one design needs to fill the whole hole between workstation and server, are you listening Applemac guys...

When they shrink down the component fabrication though, they just up the performance by doubling the transistor count within the same power usage. The next Mac Pro will have up to 16 physical cores with likely 6 on the low-end. The only reason they'd need a bigger PSU is to support 8 PCI cards because someone would try and fit 8 x 200W GPUs in there and overload the 1kW PSU.


For now it is but it will be upstaged by the underdog. At what point do you think people will stop asking for improvements to the Mac Pro?

- when the Mini has 8 cores in 5 years?
- when the Mini has 16 cores in 10 years?
- when the Mini has 32 cores in 15 years?

The future of the large devices is already in the lower-end devices and the big devices gradually fade out and become more and more specialised.

The iMac succeeds the Mac Pro, the Mini succeeds the iMac, the iPhone succeeds the Mini.

Think back to when the G4 hit 1Ghz about 9 years ago. Now we have this performance in our pockets along with graphics that rival high-end games consoles, 802.11n wifi and 720p cameras. The same will happen in another 9 years. Imagine having the equivalent of an 8-core Mac Pro in your pocket.

Crysis 3 for Cinema, real-time graphics, should be possible to get that quality on IGPs/SoC in under 10 years:

http://www.youtube.com/watch?v=J4nRzeNeka4

The only thing the Mac Pro will be for is scientific computation, not even graphics and even this can be done with distributed means.

In the short term, a smaller Mac Pro would be nice but it will become old and pass away.

guess there needs to be a built in idiot fool-proofing somehow.


I agree but iMacs won't have built in fast 10-20 terabt SSD mem core, it is an intelligent terminal not a prosumer 19" rackmount studio machine, and will be mem limited in the future of cloud computing service.


You may well be right iMacs successors will become powerful computers, but with the advent of cloud computing will they warrant sporting enough ssd and power to become professional workstation class machines to replace a super high res studio editing Macpro and video service server...?

In the meantime can I have a next generation to be going on with until we get there.

Apple won't abandon the Mac pro. It will always be the Mac of choice for the pros in the film, art and design world. Before the iMac became the powerhouse that it is now, the Mac pro straddled that line between high end consumer and the pros. Now I think it's strictly for the pros, and that "niche" as you call it...AINT GOING AWAY. End of story.

Us average consumers and semi pros can do all we need on MacBook pros and iMacs.

The big video editing and graphic design houses will continue to use the mac pros. :-)

I think you are mistaken, if they can axe the Xserve, the MacPro is not sacred either. They high end iMac is already faster than the low end MacPro. People that need more power than the high end iMac provides will use a Mini farm without thinking twice.

You can already fit 2x Mini in a 1U rack. See that Cringley article. Since that was written, OS X Server built into Lion was announced, and we have Thunderbolt. The pieces are already available! And a new Mini is due out real soon!

I've actually tossed this idea around in my head, that is either half or 1/3 rack width PC's of two to three units height. The model or prototype in my imagination is all of those HP instruments that just happen to be half rack wide. The size is very useful for many desktop and not so desktop users. Plus it solves Apples lack of a rack based system.

Actually with the fast advance of high integration devices I would not be surprised to find a viable PC built in a 1/4 rack width and maybe even 1U. That may be a little to compact for what many of us want in the long talked about XMac but I believe focusing on a half rack wide unit would lead to a very flexible design. One that could be used by a wide array of customers.

The "pros" have kept Apple afloat while all the consumers were buying up IBM PC's and Win 95. So sick of hearing about everyones grandma and catering to the "tech illiterate" class. The Mac Pro is staying in one form or another. Apple is not going to relinquish it's 1+ million "Pro" Final Cut users.
The iMac has no I/O for true professional work and sorry a TB connect and external RAID will still have sleep issues, all externals do. I need fibre/SAS attached HBA's, where is that going to go on these non Mac Pro models? The only thing Apple could and possibly will do is move away from content creation entirely and just be a media company. When that happen's I will be warming up to my new friend Billy Gates. Or I'll just pirate the shit out of them.

lol - bothaus

Exactly.
However, screw Billy and Squirter.
Linux and whatever distro takes your fancy would be the way I'd go.
cheers

Yup. Linux for work. Win for games. OS X theme on Debian. May not miss a thing, except all the drivers I need to learn to write

roflmao - that is the Open Source way of course :-)

The iMac spec is plenty for design houses, especially bundling 27" IPS displays.

27" IPS iMac, quad-core, 16GB RAM = $2600
Quad-core Mac Pro, 16GB RAM, 27" IPS = $4275

High-end video editing facilities use multi-TB networked RAID storage, the source media isn't inside the Mac Pros because it's too valuable and needs to be shared/backup up properly so you'd only need enough to have a few active projects like 1-2TB internal per workstation and SSD with a good GPU will cut through HD codecs like butter in a few years.

If SSD drops in price 20% year on year, it will be in the region of today's HDD in 10 years so a Mini can have 4 x 2TB blades in RAID 01 or whatever for < $400.

I very much doubt people will need over 32GB RAM for anything other than a server used for VPS and 32GB would do for a Mini as it's so small.

As much as I am aware that statements are often made about what people will ever need and eventually are untrue, I can't imagine very many cases where anyone would need more than a 16-core Mac Mini with 16-32GB RAM, up to 8TB SSD saturating SATA 6 (by then it will likely be some optical internal connection even) with a GPU that matches today's workstation cards capable of photoreal output in real-time and IO ports running at 100Gbps.

There are just limits to what people need to do with machines because their fundamental purpose is to allow us to create and consume digital content - that statement covers almost everything with the exception of calculations for which you just buy lots of machines or rent server time. If the content that we create or consume doesn't get dramatically more complex then the requirements to deal with it won't either.

Nobody is complaining that 1080p 3D photoreal video isn't good enough so that's the limit and even an iPad will get there eventually.


Not any more though, the iOS devices now make up most of Apple's revenue.


Thunder, thunder, thunder... Thunderbolt:

http://www.youtube.com/watch?v=Hn-qEPNgvJw

Why would it have sleep issues?

You are of course, right Marvin and all that you describe is somewhere down the track in the future, and likely not all that far away.

For now and in the next few years tho' we still work with stuff like AJA and BlackMagic cards and ProTools cards. Those guys working with Red cams have the option of installing Red Rocket <insert your need here> - the point being is you customize it. You simply can't do it with any other machine - except for the MBP's with I/O adapters. Then you have attached storage boxes which just increases the footprint. Apple have always been aware of this so you end up buying your horse for the course at the end of the day given $$$ constraints.

Now we can cut video using smart highly compressed codecs because the horsepowers there on any current Mac on offer. Any of them are powerful enough as they are now and have been for sometime <war story> heck I remember cutting dv on a G3 450 MacBook back when dualie 500's were king (and that's only 10 years ago !) <war story>
You're right the form factor will get smaller.
The big thing about a MacPro is you can have all the innards chocka full of cards, drives, whatever and if you need to take it on the road it's one lift and it's packed.
It's expandable and customizable - can't get past that.

People will always want something they can configure - the MacPro line is it - for a while anyway :-)
cheers

Everyone else here has addressed most of the problems with your post. The part quoted above strikes me as central flaw in your point. You have revealed yourself as a typical American marketing victim. You have accepted the ever present social conditioning that tells you to buy buy buy all the time. And that "new" and "Improved" is the only way to assess the quality of a product. And that a product that is 2 or 3 years old is obsolete. Heaven forbid a product maintain it's external appearance without change for more than a 6 month time period.

As others has stated Apple's competitors have systems that spec out and price out exactly the same. And the exterior design has now pretty much become "iconic"

Word.

by definition creative people create... ever moving forward, to stand still is to cease to be creative and become stagnant and gather cobwebs, Icons are highlights in the timeline of our growing past.
Now we are discussing the state of present /future and change, and the present Icon Macpro has outlived its shape and form but not yet its use,
Marvin speculates future development quite well but it is always dangerous to look too far down the line into too many possibilities, and as creative people we need to grow to the next stage in the here and now frame.

Macbkpros and iMacs are set to become the user thin client interface devices and I would use them in any studio setup but not on a recording desk, not on a pro video edit suite, and not on a stage control video server setup.
ask any engineer if he would prefer lugging around a 40 minis rack instead???

Every accessory comes on PCIe cards and the next development from Intel builds on that specification, its easy to imagine the next Macpro undergoing a new radical design iteration to accommodate these new technological advances,

If I were to design a new studio today I'd want a micro cloud system connected live to the macro cloud dissemination of the web. therefor optimizing a new macpro design to have multiple user configured preferences albeit either a studio desk system or central local server and on even to thunderbolt mutinode internet service provision.

"Time waits for no man" and mechanical drives and the disc have had their time solid state memory cores are here the future has arrived and Intel are adopting PCIe card SSDs as their next standard hence my ideal future macpro being reduced in size but still a studio rack mountable duel Zeon processor M/C having 8 full size PCIe card slots and optical thunderbolt stackable, what more could one want?
In my experience I've inevitably found the simplest neatest solution to be the right one.
and what is to come 10 years hence will have plenty of time for new discussion I'm sure.

This of course does not detract from the 27" top of the range iMacs ability to be a wonderful stand alone client device that equally has many niche uses aswell which can also connect to a businesses micro cloud Macpro based server system the software of which is now included in Lion OS.
or indeed the mini that has many domestic uses ahead of it, but I don't think any studio engineer would want a rack of 40 of them gracing his room in place of a Macpro.

I think the Thunderbolt port is going to be a very important addition and will change all this. One of the problems, especially on the Mac side is that PCI cards are designed for one model of machine. This means they are hard to get hold of, are expensive and can have compatibility issues.

By externalising all 3rd party hardware through a common IO port that fits on any device, it means hardware can be standardised and accessible to every machine.

Over 70% of all shipped computers today are laptops and great for musicians. Being able to plug in Thunderbolt Pro Tools equipment will be a very popular solution:

http://www.jigsawaudio.com/articles/...e-of-pro-tools
http://www.jigsawaudio.com/articles/...video-gallery/

It may be hard to supply enough bus power for some hardware but people build solutions like the mobile Red Rocket:

http://www.maxxdigital.com/mobile-ro...ed-rocket.html

A bit pricey but that's what goes with owning a RED Camera. If that device was standardised to use TB, it would work with both laptops and desktops and could be more cost-effective as they'd sell more of one product.

I think the best way to setup devices is to have all peripherals external even your main data storage with just the OS internal. This way when you want to upgrade, you just unplug the modular processing part and everything else is constant.


There's no real harm in maintaining a design for a while but you can understand it getting stale after 8 years, especially if you weren't a fan of it from the outset. Once a hardware design reaches a certain point though, you can see how it would be hard to improve on. I feel this way about the iPad, the Mini and the Air, which is why they would suggest to me they are the right horses to bet on being the successful designs long term.