Is the new Mac Pro a failure?

[Macsonic]

It does, but my point is that there's an initial sharp drop in price when gear is no longer 'new in box' from a reputable retailer, then a period where depreciation tends to be pretty slow with Apple gear. You seem to be pointing at this initial sharp drop as if it's a reason to be concerned about the resale value of the nMP in particular, when in fact it's entirely normal.

Hi ZnU. My concern also includes the classic Mac Pros, and other Macs as well on the resale value. Somewhat related scenario. A friend of mine was selling his iMac on eBay with "Best Offer" option offered to sellers aside from the buy it now. Most of the offers he received were too low or lowballed offers. The feedbacks he got from prospective buyers was that because several other sellers were selling it at that low a price and they wanted his price to be as low as the others. Let's hope that somehow the resale value of Macs would be at acceptable levels in the next couple of years.

Demand for desktop computers is declining because consumers are increasingly using tablets and smartphones for common computing tasks. This isn't really relevant to the Mac Pro's market segment.

Meanwhile, CPU and GPU performance are not increasing quite as fast as they once where, which could lead to machines holding value better.

Yeah. True. And also, Macbook Pros and iMacs specs and speeds have improved greatly that users are taking this route. I was also surprised on my visit to a design studio that was using a Mac Mini to edit After Effects videos. Have a good weekend.

 

[thekev]

Yeah. True. And also, Macbook Pros and iMacs specs and speeds have improved greatly that users are taking this route. I was also surprised on my visit to a design studio that was using a Mac Mini to edit After Effects videos. Have a good weekend.

That's the thing. They're worth much less used if at the time of sale the buyer has equally viable options involving new hardware at a lower end. Apple doesn't care, because a used sale isn't a sale for them anyway.

 

[drnebulous]

I suppose that you're all right.

I thought that the new Mac Pro was just a Mac Mini in drugs.

 

[simsaladimbamba]

I suppose that you're all right.

I thought that the new Mac Pro was just a Mac Mini in drugs.

For many it is, if they do not know that much about Xeons and ECC RAM and workstation grade GPUs.

 

[ZnU]

Yeah. True. And also, Macbook Pros and iMacs specs and speeds have improved greatly that users are taking this route. I was also surprised on my visit to a design studio that was using a Mac Mini to edit After Effects videos. Have a good weekend.

Competition from then-current lower-end Macs might be a concern for people trying to resell quad core Mac Pros in a couple of years, but it's going to be a pretty long time before a mini is competitive with a current 8+ core Mac Pro. The days of performance doubling every 18 months are well behind us. (For the moment; an eventual move to graphene or other new technologies could return us to the glory days of Moore's Law. But probably not in a time frame relevant to the resale of a current purchase.)

 

[reco2011]

Considering they're in an automated factory, They can make a lot more than one a month.

I hoped you could understand my example was for illustrative purposes and it wasn't intended to be representative of their actual production capacity. Was I wrong in that assumption?

 

[reco2011]

The phrase was "Cannot innovate any more, my ass!", given by Phil Schiller and was well founded IMO.

Personally I don't see anything really innovative about the nMP. Apple has always been one for small size and quiet computing. It started with the Apple /// (when I say this I believe it was the first computer for which Apple, or more appropriately Steve Jobs, made passive cooling a design goal) which was originally built using no active cooling. Apple released several Macs with passive cooling. And several more models which used a single fan. The Mini is just a lowered power version of the nMP (and small in total volume).

I also wouldn't be surprised if the nMP was something Steve Jobs was involved with at one point.

Apple are ahead, it might hurt them slightly now, but they will have it right in the long run. The MacBook air is a similar example. The first MBA was awful, but the concept was right, and Apple defined the ultra book.

Toshiba's Portege 2010 predates the MBA by years (I think about five) and was just as compact as the MBA.

----------

Yep. You're right on that. A friend of mine who works as sales consultant in an Apple reseller shop shared his side. Customers would often ask him if Macs have high resale value. He tries to give careful advise and encourages his customers to buy based on need and usage without depending much on the hope of reselling it for a high value.

Macs tend to have ridiculous resell values. I just sold my iBook G3 600MHz with 640MB RAM, 20GB hard disk, no wireless, USB 1.1, and CD-ROM drive for $60.00. I am currently attempting to sell a Dell D430 Core 2 Duo laptop with 2GB RAM, 120GB hard drive, docking station, wireless, DVD R/RW, multiple USB 2.0 ports for the same price. Only one inquiry on this despite being worlds faster than the iBook.

 

[fuchsdh]

Macs tend to have ridiculous resell values. I just sold my iBook G3 600MHz with 640MB RAM, 20GB hard disk, no wireless, USB 1.1, and CD-ROM drive for $60.00. I am currently attempting to sell a Dell D430 Core 2 Duo laptop with 2GB RAM, 120GB hard drive, docking station, wireless, DVD R/RW, multiple USB 2.0 ports for the same price. Only one inquiry on this despite being worlds faster than the iBook.

I think with old computers it's simple: Macs have aesthetic and collector's value, basic PCs don't.

The observation about Moore's law also points out something else--the costs associated with increasing performance these days are ever-increasing. And the bang for the buck most consumers are getting for those new chips is decreasing (we're a nearing a decade since multi-core computers started becoming mainstream, and yet multithreaded applications are still next to nonexistent outside of professional use cases.) We've sped our computers up with SSDs, PCIe flash, and leveraged graphics cards, but again the hardware is outpacing software (the software than benefits most from these optimizations are also the ones with the ancient codebases and huge legacy costs... hi Adobe!)

Back when my parents were running a graphic design company upgrading every four or five years was an absolute necessity to keep up with demands. Now, I'm still doing billable work on 2,1s and 3,1s with few issues (and my work computers all still have spinning boot drives.) I imagine the next MP I buy will be able to last me six years easy doing pro-caliber work. The number of use cases where you absolutely need the fastest kit available to do ever-more-complex stuff has passed by consumers and is starting to creep into the professional market as well--Jim Dalrymple mentioned how the new Mac Pro isn't for him because all the DA stuff that used to require beefy rigs can be done via a Macbook Air these days. There will always be a market for the latest and greatest, but I'm wondering if that's essentially going to become a tiny market even compared to where it is now--with only the hardcore gamers, scientific number-crunchers, and VFX houses really spending the big bucks on a regular basis.

 

[reco2011]

I think with old computers it's simple: Macs have aesthetic and collector's value, basic PCs don't.

To a degree I can agree. The individual who bought the iBook was a 20 year old girl. Someone who was looking at it for basic tasks. Tasks which my other two PC laptops would be better suited for. But to each their own.

The observation about Moore's law also points out something else--the costs associated with increasing performance these days are ever-increasing. And the bang for the buck most consumers are getting for those new chips is decreasing (we're a nearing a decade since multi-core computers started becoming mainstream, and yet multithreaded applications are still next to nonexistent outside of professional use cases.) We've sped our computers up with SSDs, PCIe flash, and leveraged graphics cards, but again the hardware is outpacing software (the software than benefits most from these optimizations are also the ones with the ancient codebases and huge legacy costs... hi Adobe!)

Back when my parents were running a graphic design company upgrading every four or five years was an absolute necessity to keep up with demands. Now, I'm still doing billable work on 2,1s and 3,1s with few issues (and my work computers all still have spinning boot drives.) I imagine the next MP I buy will be able to last me six years easy doing pro-caliber work. The number of use cases where you absolutely need the fastest kit available to do ever-more-complex stuff has passed by consumers and is starting to creep into the professional market as well--Jim Dalrymple mentioned how the new Mac Pro isn't for him because all the DA stuff that used to require beefy rigs can be done via a Macbook Air these days. There will always be a market for the latest and greatest, but I'm wondering if that's essentially going to become a tiny market even compared to where it is now--with only the hardcore gamers, scientific number-crunchers, and VFX houses really spending the big bucks on a regular basis.

I agree with this but keep in mind we're talking a 600MHz G3 based system with 640MB of RAM and a 20GB hard disk. This system is incapable of doing many of the basic things, such as YouTube, with the standard software. In fact the latest revision of OS X it can run is Tiger...an operating system released in early 2005. A lot of the common software people use today will not run on Tiger.

Regardless she wanted it and she bought it knowing what it was. As long as she's happy.

 

[Radiating]

I want some of what you're smoking.

Please link us to this mythical Mac Pro that can run 17 Titan Blacks via TB.

So 17 Titan Blacks running through 8 PCIE 2.0 lanes? Sounds like moving lake Michigan through a garden hose.

It's my understanding that, there are 17 pcie lanes on the new Mac Pro, excluding the lanes used for the fire pro cards. 5 for 1x cards like hard drives and wifi, and 12 for thunderbolt devices. In theory you could hook up 17 Titan Black cards to the machine. Tests have shown that there is only a 30% hit in performance when running a Titan card over pci express 1x, so this could be a plausible setup for a render farm (it would be wasteful but this just shows that the machine doesn't have as many limits of people believe.

As far as I know, there isn't a single TB box that claims GPU support. Much less the ability to run 17 Titan Blacks.

That's funny because my MacBook had a Titan hooked up to it, and it worked fine through a sonnet echo express pro, and no there wasn't a serious performance hit in any way. Any pcie capable box supports GPUs.

And I can run SLI in old Mac Pro. In fact, I have run 3 @ Titans in Tri-SLI, try that in nMP. In fact I was able to get a 97% in 3D Mark that way, while maxed out nMP with D700s fell by the wayside at 93%. cMP can do Crossfire with no hacks, and SLI with a simple hack.

Care to elaborate? This sounds very interesting...

Does everything work? Do the boot screens function correctly?

That sounds insane.

How "simple" is this hack?

If you are going to post emotional rants, try sprinkling in a few actual facts.

I don't see how this was an emotional rant. The OP basically made up a list of nonsense that shows it was based on absolutely no meaningful real world situation for the specific reason of bashing the Mac Pro.

The fact is that the Mac Pro can do what 95% of people need it to do, and it can do those things with incredible performance and value. The Mac Pro is a great professional machine.

 

[fuchsdh]

To a degree I can agree. The individual who bought the iBook was a 20 year old girl. Someone who was looking at it for basic tasks. Tasks which my other two PC laptops would be better suited for. But to each their own.


I agree with this but keep in mind we're talking a 600MHz G3 based system with 640MB of RAM and a 20GB hard disk. This system is incapable of doing many of the basic things, such as YouTube, with the standard software. In fact the latest revision of OS X it can run is Tiger...an operating system released in early 2005. A lot of the common software people use today will not run on Tiger.

Regardless she wanted it and she bought it knowing what it was. As long as she's happy.

Oh, for sure. I wasn't quite talking about your situation there, more going off comments further up. There are still going to be limits, of course—especially if processing architectures change or they actually start writing software that uses and needs those extra cores or beefy GPUs to compensate. But there's also the consumer computer which is more disposable versus the pro models as well. Those 2,1 and 3,1 Mac Pros certainly fair better than the Core 2 Duo / 1GB iMacs from the same period, and while they can perform capably without any upgrade for those who want the upgrades it's also an option.*

*Which is also another factor—what's the cost/benefit analysis on upgrading versus buying new? In the aforementioned 1990s graphics studio it made far more sense to buy a new computer versus upgrading—the IIci they started with cost $6700 base with a 25MHz processor, 40MB hard drive and up to 128MB RAM. A PowerMac 7100/66 came with storage up to 750MB, a processor at least three times as fast and additional RAM capacity, and cost 43% of the IIci's price. These days I'm not so sure if that's such a slam-dunk, as people have shown you can get very close to the nMP performance with a comparatively cheap set of upgrades. Of course everyone's cost weighing is different, but it does seem like sitting back and upgrading is more attractive now than in the past simply because hardware has matured and software really isn't pushing things as much in many situations.

 

[deconstruct60]

It's my understanding that, there are 17 pcie lanes on the new Mac Pro, excluding the lanes used for the fire pro cards.

Understanding is flawed. It is highly unlikely to be an odd number.

40 PCIe v3.0 from CPU package. 8 PCIe v2.0 lanes from the C602 (IOHub chipset). Total 48. Minus the 32 ( 2 * 16 of the GPU cards ) is 16. But even 16 is flawed.

The other fundamental flaw is that there are more consumers than just the GPU cards. Wifi/Bluetooth , Ethernet , USB 3.0 , and SSD all also consume PCIe bandwidth. The IOHub chipset is really maxed out ( zero available ). Between SSD ( 4 ) , Ethernet ( 2 ) , WiFi-Bluetooth (1) , and USB 3.0 (1) that is 8 right there.

The three Thunderbolt controllers are sharing the lone 'spare' 8 PCIe v3.0 lanes. The only offset is that the TB controllers are x4 PCIe v2 ( where v2 is about half of v3 ). That's what can be farmed out external PCIe cards.


That said, one PCIe lane per GPU card is a joke. Cards that nominally can leverage x16 PCIe v3 on a budget of x1 PCIe v2 is rather limited. Sure could have corner cases where push the data to the card relatively and then do something with the purely cached data on the card. However, from a bang-for-buck perspective, it is a poor imbalanced system.

Chopping down from x16 PCIe v2 to x4 PCIe v2 ( which is what most "GPUs don't really need all that PCIe bandwidth" experiments actually demonstrate ) isn't particularly applicable to a gimmick of thinning out the bandwidth as much as possible.

 

[handsome pete]

The '+' part of that $30K+ was significant. Prices go up rather a lot from there. More to the point... has the discussion really evolved from "FCP X is at best a prosumer app" to "OK, sure, maybe some medium sized facilities are using FCP X, but it hasn't been adopted by Technicolor or Deluxe yet" without FCP X's detractors admitting their characterizations were wrong anywhere along the way? That's some pretty serious goalpost shifting.

Well $100k really isn't anything to flinch at either. I remember the early days of HD where some individual freelancers were buying cameras approaching that range.

And I haven't moved the goalposts anywhere. I've always maintained FCPX to be a fine app that probably won't gain mainstream adoption amongst the medium to large shops (except when it first debuted. That was a mess.). But that doesn't mean there won't be any of them who will use it.

If they've said this lately, they haven't been paying attention. There were some things about the way FCP X handled media storage that were annoying (though not impossible to overcome) in shared storage environments. They mostly went away with 10.1, and 10.1.2 just eliminated the last of them.

Maybe so. I don't know too much about any recent changes to FCPX and shared environments. Last I heard, it had definitely improved greatly, but was still somewhat tricky especially when you start dealing with say 15+ seats in an environment where there are different editors, assistant editors, producers/driectors, etc. all working on projects all at once. This is where Avid still excels in my opinion. For all of their faults (and there are plenty), their media management is still excellent, and there's surely something to be said about having your software and hardware all fall under the same umbrella, where there's quick go to support if needed.

Tests have shown that there is only a 30% hit in performance when running a Titan card over pci express 1x, so this could be a plausible setup for a render farm (it would be wasteful but this just shows that the machine doesn't have as many limits of people believe.

Only 30%? I would translate that as a limitation.

 

[fuchsdh]

Well $100k really isn't anything to flinch at either. I remember the early days of HD where some individual freelancers were buying cameras approaching that range.

And I haven't moved the goalposts anywhere. I've always maintained FCPX to be a fine app that probably won't gain mainstream adoption amongst the medium to large shops (except when it first debuted. That was a mess.). But that doesn't mean there won't be any of them who will use it.



Maybe so. I don't know too much about any recent changes to FCPX and shared environments. Last I heard, it had definitely improved greatly, but was still somewhat tricky especially when you start dealing with say 15+ seats in an environment where there are different editors, assistant editors, producers/driectors, etc. all working on projects all at once. This is where Avid still excels in my opinion. For all of their faults (and there are plenty), their media management is still excellent, and there's surely something to be said about having your software and hardware all fall under the same umbrella, where there's quick go to support if needed.



Only 30%? I would translate that as a limitation.

If you were going to start a post house from scratch these days I'm not sure why you would choose Avid, given that horrible financial straits it's in. Final Cut and Premiere are slowly but surely killing it by eating out its foundations. So it will be interesting to see where the editing environment is in 10 years.

As for GPU limitations, if you can get 70% of the performance from a Titan, that's still a hell of a lot of power if you're running a laptop. The tomshardware guys tested Sonnet's TB1 chassis and their ultimate verdict was that the bandwidth limitations weren't a real issue if you were using a mid-ranged card—so much better than integrated graphics, but you're wasting money putting in a 780 Ti in terms of what you can saturate.

http://www.tomshardware.com/reviews/pci-express-graphics-thunderbolt,3263-7.html

I haven't seen what real-world benefits there are to the TB2 chassis now, but I imagine the same sort of limitations will apply—if only because of power constraints.

Edit: Something just hit me--these chassis are designed for half-length cards, so how would you fit the full-length cards we're talking about here anyhow?

 

[handsome pete]

If you were going to start a post house from scratch these days I'm not sure why you would choose Avid, given that horrible financial straits it's in. Final Cut and Premiere are slowly but surely killing it by eating out its foundations. So it will be interesting to see where the editing environment is in 10 years.

Apple and Adobe have dominated the consumer market, a market that Avid was never really in competition for despite their attempt years back. Of course the blending between consumer, prosumer, professionals, etc. that continues to happen surely doesn't bode well.

Their financial situation is not nearly as dire as it had been made out to be. However it does have a clouded outlook. They will still have a lock on the high-end segment and, with a few good strategic decisions, could come out looking pretty good in all of this.

 

[MacProCard]

It's simple really. Apple tied the nMP to Final Cut Pro X in their marketing campaign. That's their target market. And they list multicam capabilities several times. Because that's where they see the market heading. Too many HD cameras out there nowadays not to push the technology.

Hate to break it to some of you...But what once took a pro, Apple sees teenagers doing it. Tons of clear and concise tutorials out there to do whatever you need to do with your project. And the younger youtuber groups will also be the one's that enjoy the portability design. It won't leave my desk but I can see the kids taking these things anywhere.

If you're a pro and this CAN doesn't work for your needs, you probably should be building your own anyways and it's time to leave Apple's nest so to speak. They have cut your market segment lose and went after a more lucrative one.

https://www.apple.com/mac-pro/performance/

 

[handsome pete]

It's simple really. Apple tied the nMP to Final Cut Pro X in their marketing campaign. That's their target market. And they list multicam capabilities several times. Because that's where they see the market heading. Too many HD cameras out there nowadays not to push the technology.

Of course they tied it to their marketing campaign. Why wouldn't they use one of their pro apps to showcase their most powerful machine? And I'm still not sure why you think multicam is some revolutionary new market direction. FCP7 handled multicam just fine, as do the other NLEs out there., not to mention it wasn't even a feature in the initial release of FCPX. If you're arguing that video editing in general is going to take off even more in the coming years because of the growth of video acquisition, then sure. But multicam is a bit of a niche feature. A feature that some will certainly use, but not the masses. Despite your beliefs, not many people out there have 5+ video cameras sitting around nor the inclination to sync them all up for an edit.

Hate to break it to some of you...But what once took a pro, Apple sees teenagers doing it.

No one disagrees here. We're just saying you don't need a Mac Pro to do it. A quick glance at the FCPX page shows a lot of iMacs and Macbook Pros. And if we're talking about your every day teenager/average Joe, then a $3000+ computer is not really in the cards.

 

[MacProCard]

Of course they tied it to their marketing campaign. Why wouldn't they use one of their pro apps to showcase their most powerful machine? And I'm still not sure why you think multicam is some revolutionary new market direction. FCP7 handled multicam just fine, as do the other NLEs out there., not to mention it wasn't even a feature in the initial release of FCPX. If you're arguing that video editing in general is going to take off even more in the coming years because of the growth of video acquisition, then sure. But multicam is a bit of a niche feature. A feature that some will certainly use, but not the masses. Despite your beliefs, not many people out there have 5+ video cameras sitting around nor the inclination to sync them all up for an edit.



No one disagrees here. We're just saying you don't need a Mac Pro to do it. A quick glance at the FCPX page shows a lot of iMacs and Macbook Pros. And if we're talking about your every day teenager/average Joe, then a $3000+ computer is not really in the cards.

I don't get what you don't understand about marketing. Or why you think I'm wrong? Apple's marketing team, who worked in collaboration with their sales team, wrote it. I just interpret it the way I see it. I guess you can think they were only writing that marketing material for professionals like you. But I disagree.

If you're new to Apple and interested in the nMP for whatever reason, FCPX and multicam technology is right there under performance.

And you don't really need an iPhone 5 or Polo shirts either.

I think you're wrong. Looks like Apple does as well. Maybe the reason you're so disappointed with the nMP is because you don't understand why it was made.

 

[beaker7]

I guess you can think they were only writing that marketing material for professionals like you. But I disagree.

So it's not a machine for professionals?

 

[handsome pete]

I don't get what you don't understand about marketing. Or why you think I'm wrong? Apple's marketing team, who worked in collaboration with their sales team, wrote it. I just interpret it the way I see it. I guess you can think they were only writing that marketing material for professionals like you. But I disagree.

If you're new to Apple and interested in the nMP for whatever reason, FCPX and multicam technology is right there under performance.

Because marketing material surely never tries to paint products in the brightest light possible. You're reading way too much into the whole multicam thing. It's a feature. A feature that was promoted on previous versions of Final Cut as well. It's something that demands a good amount of power the more you throw at it. Other workstations (and the old Mac Pro) do the same damn thing.

And you don't really need an iPhone 5 or Polo shirts either.

Yes, because a $70 shirt or $200 phone are exactly like a $3000 computer. Same thing really.

I think you're wrong. Looks like Apple does as well. Maybe the reason you're so disappointed with the nMP is because you don't understand why it was made.

Where did you ever get the impression I was disappointed with the nMP? Again, you're reading into things that aren't there.

 

[MacProCard]

So it's not a machine for professionals?

Sure it is. Most of them at least. Apple just wants to grow the professional market. Is it wrong for them to re-design it in a way as to appeal to more people?

I believe Apple just told anyone who wants an old, clunky, heavy, noisy, inefficient "pro" box to look elsewhere. And rightfully so in my opinion.

----------

Because marketing material surely never tries to paint products in the brightest light possible. You're reading way too much into the whole multicam thing. It's a feature. A feature that was promoted on previous versions of Final Cut as well. It's something that demands a good amount of power the more you throw at it. Other workstations (and the old Mac Pro) do the same damn thing.

You're in denial. You can't fathom a group of skaters using their iPads and IPhones to display their skillz on the net. That's being very narrow minded imo. And personally, I think you give yourself/profession more credit than you probably should.

 

[reco2011]

Where did you ever get the impression I was disappointed with the nMP? Again, you're reading into things that aren't there.

He seems to do this with anyone who doesn't have anything but 100% positive things to say about the nMP.

----------

Sure it is. Most of them at least. Apple just wants to grow the professional market. Is it wrong for them to re-design it in a way as to appeal to more people?

IMO the nMP doesn't grow the professional market. The nMP is even more of a niche product than the oMP.

 

[fuchsdh]

You're in denial. You can't fathom a group of skaters using their iPads and IPhones to display their skillz on the net.

ance.

And you don't really need an iPhone 5 or Polo shirts either.

I dunno what your problem is, but you come off as having bile dripping from your maw when you talk about "those punk-ass kids".

IMO the nMP doesn't grow the professional market. The nMP is even more of a niche product than the oMP.

Agree. That's not a bad thing really, but it's designed more for certain workflows and pro markets than others whereas the oMP was more of a swiss army knife.

 

[handsome pete]

You're in denial. You can't fathom a group of skaters using their iPads and IPhones to display their skillz on the net. That's being very narrow minded imo.

There you go again projecting your scenarios on the rest of us. Of course I can fathom a group doing just that. In fact, they've been doing it for years now. Where I think you're in denial in that this is some sort of new wave of business that is going to take the video world by storm. You've gone as far as to claim the average person has 10+ video cameras laying about ready to be used for multicam projects.

Video software and hardware has gotten dramatically cheaper and more accessible over the past decade. Why would I ever consider that a negative thing? My bank account has surely benefited from that.

And personally, I think you give yourself/profession more credit than you probably should.

Pot, meet kettle.

 

[MacProCard]

I dunno what your problem is, but you come off as having bile dripping from your maw when you talk about "those punk-ass kids".

Would love to see you tell me that to my face. --Funny how people project their own personal fantasies in the craziest of ways. I hope you don't work with kids.