Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
I can see early release of of New MacPro mini would piss off some early adopters of 6.1 pros
Based on new technology coming out I can see 6.2 to be announced this year and being shipped first quarter of 2015.
for the price
5.2 was cheaper than 5.1 so 6.2 may not be that much cheaper as it will have so much newer technology.
2 drive slots/ports and maybe multiprocessor option to achieve faster processor speed and multicore :)
Time will tell
my dream system will be dual 6 core, 8 memory slots and 2 drive slots.
That will keep me smiling

Just think if Apple is going to update it this year and puts 2 nvidia chips in there instead of amd - buh thats gonna piss some people off which would love the cuda performance
 
Those of you wishing that the next version may have two SSD slots, please be aware that the current nMP already uses every last PCI lane.

There's simply no more room for more independent PCI access. Of course, a future CPU architecture may allow for more/faster lanes, or a different allocation of the existing lanes may be made, etc.

I for one will not be holding my breath for a second SSD slot any time soon. A move towards higher-capacity SSD cards, e.g. via vertical flash memory integration, is much more likely.
I thought the lane allocation was interesting when it came to the networking hardware if Anand got it right:
http://www.anandtech.com/show/7603/mac-pro-review-late-2013/8
There's 1500MB/s worth of bandwidth (3 x1 2.0 lanes) used on (theoretical max) of 422.4MB/s of stuff (2x gigE + 802.11ac), seems like it'd make more sense to at least put the gigE ports together on one lane with a PCIe switch or something. Then there's 2GB/s (4 lanes) for the single SSD that tops out at a little more than 1GB/s.

If they keep stuff mostly the same they can still gain two more lanes cause it looks like Wellsburg will have integrated USB 3.0 and GigE. If Apple uses that functionality they'll have 6 lanes left over for SSDs. Whether they actually do that is another story of course.
I'm also concerned about the thermal capacity of the design limiting enhancements to the graphics cards. More than marginal improvements at the high-end may be a long time coming. The low-end (e.g., D300) has room to grow, though.
Apparently Yosemite has some identifier for the R9 295X (and 290s are working for hackintoshers) so I'm guessing Hawaii GPUs will be coming, but yeah it'll be interesting to see what they can actually pull off with the thermals. The recently announced FirePro W8100 compares favorably TDP wise to the W9000 so that's hopefully a good sign, 220W vs 274W while still providing better performance.
 
iMac, Macbook Pro, iPhone and iPad are their flagship products in their main categories. iPhone overall.

Well, depends what you mean by flagship. Is it the product with the most market share? The most mindshare? Or just their best-in-class? The answer changes with each condition.

I'd say the nMP is their flagship in terms of engineering feats, just like the Bugatti Veyron is Volkswagen's "premier" sports car.

In terms of market share, they sell more iPhones than anything else.

In terms of mindshare... that's tricky. Certainly not the Mac Pro, but I kind of feel like Apple is a bit more monolithic in how people think about it. It's that company with the fruit on everything.

----------

I thought the lane allocation was interesting when it came to the networking hardware if Anand got it right:
http://www.anandtech.com/show/7603/mac-pro-review-late-2013/8
There's 1500MB/s worth of bandwidth (3 x1 2.0 lanes) used on (theoretical max) of 422.4MB/s of stuff (2x gigE + 802.11ac), seems like it'd make more sense to at least put the gigE ports together on one lane with a PCIe switch or something. Then there's 2GB/s (4 lanes) for the single SSD that tops out at a little more than 1GB/s.

If they keep stuff mostly the same they can still gain two more lanes cause it looks like Wellsburg will have integrated USB 3.0 and GigE. If Apple uses that functionality they'll have 6 lanes left over for SSDs. Whether they actually do that is another story of course.

Apparently Yosemite has some identifier for the R9 295X (and 290s are working for hackintoshers) so I'm guessing Hawaii GPUs will be coming, but yeah it'll be interesting to see what they can actually pull off with the thermals. The recently announced FirePro W8100 compares favorably TDP wise to the W9000 so that's hopefully a good sign, 220W vs 274W while still providing better performance.

I'm also concerned about the thermal capacity of the design limiting enhancements to the graphics cards. More than marginal improvements at the high-end may be a long time coming. The low-end (e.g., D300) has room to grow, though.

From what Anandtech said when running specs, it doesn't look like it's possible to hit any bottlenecks or throttling in normal situations with the system at full tilt, so I expect the worst we'll see is a slight downclock to the max GPU like we did with the D700 compared to the W9000/7970. They made a remarkably efficient machine.
 
Well, depends what you mean by flagship. Is it the product with the most market share? The most mindshare? Or just their best-in-class? The answer changes with each condition.

Is the Mac Pro any better engineered than the iPad Air, 15" Retina pro or 5S? It's not really that complex, they just did stuff other companies aren't going to do. It can be the most powerful Mac if you configure it in certain ways, but no one considers workstations or servers the flagship products of Dell and HP (although I guess they don't really have any flagship products). Outside of a small number of Mac Pro customers I really don't think any one thinks of the Mac Pro as being important for Apple, it doesn't scream "Apple" like the sleek, thin, designs of the laptops and iOS devices either. Even the iMac and Mac mini sort of have that going for them. The thin aluminium slab with the white Apple is probably the most undoubtedly product design Apple have put out. You even know when they are trying to disguise it in Film or TV and have Apple's branding working on your subconscious.

I think it's a side project these days, and pretty much always was. Obviously "flagship product" is not an easy thing to assign, but I think you have to go to a stretch to give it to the Mac Pro. They Veyron example is interesting, because they are known for it, but that is because it is this super powerful thing in a huge market place, where the Mac Pro isn't. It's just a powerful thing in Apple's line.
 
Is the Mac Pro any better engineered than the iPad Air, 15" Retina pro or 5S? It's not really that complex, they just did stuff other companies aren't going to do. It can be the most powerful Mac if you configure it in certain ways, but no one considers workstations or servers the flagship products of Dell and HP (although I guess they don't really have any flagship products). Outside of a small number of Mac Pro customers I really don't think any one thinks of the Mac Pro as being important for Apple, it doesn't scream "Apple" like the sleek, thin, designs of the laptops and iOS devices either. Even the iMac and Mac mini sort of have that going for them. The thin aluminium slab with the white Apple is probably the most undoubtedly product design Apple have put out. You even know when they are trying to disguise it in Film or TV and have Apple's branding working on your subconscious.

I think it's a side project these days, and pretty much always was. Obviously "flagship product" is not an easy thing to assign, but I think you have to go to a stretch to give it to the Mac Pro. They Veyron example is interesting, because they are known for it, but that is because it is this super powerful thing in a huge market place, where the Mac Pro isn't. It's just a powerful thing in Apple's line.

Oh, for sure, car analogies are imperfect. I use it primarily because before the nMP was announced that was often the analogy used in the argument for continuing the Mac Pro line--how big car companies took a loss on these powerful halo cars that were primarily for pushing the technology forward to ultimate consumer acceptance. The power certainly isn't the big deal with the nMP compared to its power-per-size, but I'd still say that it's operating in the same realm as these cars, even if it's not the Veyron (perhaps the issue was my choice of car, since the Veyron's main claim to fame is it's the fastest street-legal car. Sadly I'm not a big car guy so I don't know what luxury model would probably be more appropriate? I guess Mercedes-Benz in a way, as they have a tendency to have technology a decade before the common car --computer diagnostics, assisted braking, rear cameras, et al.)

Either way I'm not really arguing vehemently on the position myself. But it's an interesting thought experience to engage with :)
 
Outside of a small number of Mac Pro customers I really don't think any one thinks of the Mac Pro as being important for Apple, it doesn't scream "Apple" like the sleek, thin, designs of the laptops and iOS devices either. Even the iMac and Mac mini sort of have that going for them. The thin aluminium slab with the white Apple is probably the most undoubtedly product design Apple have put out. You even know when they are trying to disguise it in Film or TV and have Apple's branding working on your subconscious.

I think it's a side project these days, and pretty much always was. Obviously "flagship product" is not an easy thing to assign, but I think you have to go to a stretch to give it to the Mac Pro. They Veyron example is interesting, because they are known for it, but that is because it is this super powerful thing in a huge market place, where the Mac Pro isn't. It's just a powerful thing in Apple's line.

For a car analogy, think F1 and the car industry. The Mac Pro R&D could start trickling into the iMac. Once they finally get rid of all the HDDs, imagine a reengineered iMac with dual GPUs. I can already see a shared heat sink with a large, quiet fan turned 90 degrees.

The Mac Pro could be very important.
 
For a car analogy, think F1 and the car industry. The Mac Pro R&D could start trickling into the iMac. Once they finally get rid of all the HDDs, imagine a reengineered iMac with dual GPUs. I can already see a shared heat sink with a large, quiet fan turned 90 degrees.

The Mac Pro could be very important.

Yeah. There is a lot of the MacBook Pro Retina in the Mac Pro, at that.

I also think the Mac Pro is kind of a mind share computer. Not many people will buy one, but it's a nice reassurance that whatever equipment you have had the DNA of a serious computer in it. It's a nice reminder the Mac is not a toy, as so many PC diehards in the early 2000s liked to label it. It's proof that the Mac is on the cutting edge of everything.

It's like a Tesla. Not many people will buy one, but even those who don't buy one recognize Tesla's prowess.
 
I honestly just want the Haswell-EP upgrade. Even if the next refresh doesn't include DDR4 RAM or new GPUs, I'd be happy. Probably going to wait until then to put my order in, since I'll have the funds around the middle of September and it shouldn't be *too* long of wait afterwards for it. Anything else other than Haswell is gravy for me, especially if they somehow make the 6c config cheaper/better performing.
 
Yeah. There is a lot of the MacBook Pro Retina in the Mac Pro, at that.

I also think the Mac Pro is kind of a mind share computer. Not many people will buy one, but it's a nice reassurance that whatever equipment you have had the DNA of a serious computer in it. It's a nice reminder the Mac is not a toy, as so many PC diehards in the early 2000s liked to label it. It's proof that the Mac is on the cutting edge of everything.

It's like a Tesla. Not many people will buy one, but even those who don't buy one recognize Tesla's prowess.

Then why develop it if they thought not many people would be interested in it? I could possibly see what your saying if the nMP wasn't a radically new design. But I don't understand how Apple or any company can build a product they hope doesn't catch on. Apple mass produces products. They're not into specialty products.

As I mentioned in other threads, if the nMP doesn't suit your needs now, then you really need more of a specialty product.
 
Then why develop it if they thought not many people would be interested in it? I could possibly see what your saying if the nMP wasn't a radically new design. But I don't understand how Apple or any company can build a product they hope doesn't catch on. Apple mass produces products. They're not into specialty products.

Because it's still profitable. They don't sell many, but the margins are high. It's the same reason Tesla sells the Model S.

It's also an important lynchpin for software. If you don't have the nMP, you lose FCPX customers, and you lose Logic customers. Once you lose those customers, they'll buy PCs. And then everyone they work with might buy PCs. The Mac Pro is maintaining a halo effect for Apple. You might have a dozen editors in an office of hundreds that use the Mac Pro. But if they leave the platform, the hundreds of people around them might start buying PCs instead of iMacs.

As I mentioned in other threads, if the nMP doesn't suit your needs now, then you really need more of a specialty product.

I think there are quite a few people in agreement in this forum that the GPU performance currently "needs improvement." Especially if you are working at Retina/4k. The iMac is currently beating the Mac Pro for single GPU performance, which is where OS X is anyway.

CPU performance is fine, but I'd rather take the cost savings that comes with Haswell and DDR4 if I'm stuck waiting for new GPUs anyway.
 
Yes and no. The MP6,1 has 4 unused PCIe lanes on the IO board.
Got a source on that? All I know about is Anandtech's breakdown and that's not completely confirmed afaik.
I honestly just want the Haswell-EP upgrade. Even if the next refresh doesn't include DDR4 RAM or new GPUs, I'd be happy. Probably going to wait until then to put my order in, since I'll have the funds around the middle of September and it shouldn't be *too* long of wait afterwards for it. Anything else other than Haswell is gravy for me, especially if they somehow make the 6c config cheaper/better performing.
Things seem to point at Haswell EP starting at 6 cores (from an 8 core die) on the low end, so hopefully the next MP will also bump the base model to 6 cores. I'm hoping the mid range GPU option is more noticeably better than the low end next time. Or just nice GPU upgrades across the board would be fine of course.
 
Because it's still profitable. They don't sell many, but the margins are high. It's the same reason Tesla sells the Model S.

It's also an important lynchpin for software. If you don't have the nMP, you lose FCPX customers, and you lose Logic customers. Once you lose those customers, they'll buy PCs. And then everyone they work with might buy PCs. The Mac Pro is maintaining a halo effect for Apple. You might have a dozen editors in an office of hundreds that use the Mac Pro. But if they leave the platform, the hundreds of people around them might start buying PCs instead of iMacs.


I think there are quite a few people in agreement in this forum that the GPU performance currently "needs improvement." Especially if you are working at Retina/4k. The iMac is currently beating the Mac Pro for single GPU performance, which is where OS X is anyway.

CPU performance is fine, but I'd rather take the cost savings that comes with Haswell and DDR4 if I'm stuck waiting for new GPUs anyway.

Telsa saw a market need (a performance electric car). They just didn't hope they'd be popular amongst Google execs. They wanted to make a new market.

I agree about FCPX. They can forecast sales to some extent just off the software licensing. I would have called it a video rig..similar to a gaming rig. But at this time they consider it too risky labeling it to one specific market. I've worked in multi team environments and it's usually competitive. For FCPX to get verbiage on the nMP website large or small is big...it's subtly there.

I've never really seen an Apple shop.

IMO this thing was released a little early for 4k video. While they chose to market it specifically for 4k, it also implies the power it has with 1080p. Sony only has 50 4k videos for rent. So who's really buying this technology/footage? Where's it needed? 3/4 of my cable channels don't even broadcast in HD yet.
 
IMO this thing was released a little early for 4k video. While they chose to market it specifically for 4k, it also implies the power it has with 1080p. Sony only has 50 4k videos for rent. So who's really buying this technology/footage? Where's it needed? 3/4 of my cable channels don't even broadcast in HD yet.

Didn't say I edit video. Just said I need 4k.

:)
 
, seems like it'd make more sense to at least put the gigE ports together on one lane with a PCIe switch or something.

the latency, added cost/complexity , and one lane doesn't buy much when folks are asking for something that soaks up.....

2GB/s (4 lanes) for the single SSD that tops out at a little more than 1GB/s.

If claw back x1 still x3 short.

If they keep stuff mostly the same they can still gain two more lanes cause it looks like Wellsburg will have integrated USB 3.0 and GigE.

Several problems. First as pointed out above x2 is not x4 . Even if got two back still wouldn't have enough for a SSD on same par as the first. ( dropping in a x2 from the laptop line-up may or may not be useful.)

Second, Intel's integrated GigE consumes a x1 PCIe lane to the Phys connection chip. It is just a cheaper way to GigE (with a cheaper discrete chip that leans on the chipsets for smarts) ... it doesn't reduce the number of lanes (i.e., puts more pins on the PCH chipset). The current C600 (and I think previous) chipset had GigE. I think the hiccup is that is only provides one independent GigE socket. The Mac Pro still requires two. So doesn't really solve the 'two port' problem in a symmetrical way in either the C600 or C610 (Wellsburg) chipset.

Third, the chipset can be configured for four x1 and one x4 , or eight x1 , but I don't think can configuration bundle two x1 lanes into a x2. While there is a somewhat likely a possible kludge around that, Typically x1/x4 PCIe lanes are chopped up by a PCIe switch. Diluting the x4 would impact the primary SSD and the x1 doesn't cut it for a second SSD.

If Apple uses that functionality they'll have 6 lanes left over for SSDs. Whether they actually do that is another story of course.

Part of the problem here is a huge mismatch between what Apple wants to do and what the chipset provides. There are up to to 10 SATA ports that Apple has zero use for. If 4 of those could be flipped into PCIe ports that would be a far more productive 'fix' to solving the problem. Similarly 6 USB 3.0 ports (and 8 USB 2.0) ports. Again if four of those could be FlexIO flipped into x4 more PCIe v2 lanes, that would be handy.

Just like there were versions of the C600 line up that had more SATA ports at the high end there could be a "lower end" version of some C6x0 chipset that tossed some of the SATA/USB interfaces for more generic ones. As PCIe SSD go more mainstream it would only make more sense. (even small blade server designs could leverage a more general I/O than a bucket load of USB/SATA ports they are never going to hook too. )


Apparently Yosemite has some identifier for the R9 295X (and 290s are working for hackintoshers) so I'm guessing Hawaii GPUs will be coming,

If Apple could drop some GPU cards that had the DSP present and active that might go to mute some of the "what does this buy for audio" complaints. ( although I'm sure there will still be folks who complain it isn't the specific DSP they want to use..... )
 
If Apple could drop some GPU cards that had the DSP present and active that might go to mute some of the "what does this buy for audio" complaints. ( although I'm sure there will still be folks who complain it isn't the specific DSP they want to use..... )

Are people actually grumbly about that? The DSP built into the CPU seems pretty good. It's hard for me to see how DSP on the GPU could be any faster.
 
Got a source on that? All I know about is Anandtech's breakdown and that's not completely confirmed afaik.

He is talking about the "unused" x4 hanging off the PCIe switch on the relatively distance IO interface logic board.

It x8 PCIe v3 lanes switched into four x4 PCIe v2 interface. The Thunderbolt Controllers sit on three of those. That doesn't particularly means there is an addtion x4 worth of bandwidth left. Just that there is physical x4 connection, not that there is another x4 of bandwidth.


There is little evidence that would work much better than if Apple slapped a x4 PCIe v2 to two x4 PCIe v2 since on the backplane logic board and just oversubscribed two x4 PCIe SSDs on the same bundle. The TB controllers being so close and the SSD traces traveling over three different logic boards would seem an ripe invitation for latency/switching glitches.
 
Apparently Yosemite has some identifier for the R9 295X (and 290s are working for hackintoshers) so I'm guessing Hawaii GPUs will be coming, but yeah it'll be interesting to see what they can actually pull off with the thermals. The recently announced FirePro W8100 compares favorably TDP wise to the W9000 so that's hopefully a good sign, 220W vs 274W while still providing better performance.

Is the 295X an actual GPU or did you mean the X2?

Regardless, again seems to line up nicely for a nMP release around Yosemite.
 
IMO this thing was released a little early for 4k video. While they chose to market it specifically for 4k, it also implies the power it has with 1080p. Sony only has 50 4k videos for rent. So who's really buying this technology/footage? Where's it needed? 3/4 of my cable channels don't even broadcast in HD yet.

Just because stuff isn't getting to consumers doesn't mean 4K isn't becoming more important as a workflow for the extra flexibility and future-proofing, though. I agree 4K adoption is going to be anemic compared to HD, but the consumer market ≠ professional market in terms of 4K having utility beyond just a delivery format.
 
Things seem to point at Haswell EP starting at 6 cores (from an 8 core die) on the low end, so hopefully the next MP will also bump the base model to 6 cores. I'm hoping the mid range GPU option is more noticeably better than the low end next time. Or just nice GPU upgrades across the board would be fine of course.

I'm hoping that's exactly the case, but I'm doubtful. Part of me thinks Apple might opt to keep the 4-core Ivy Bridge-EP setup, but then offer Haswell-EP/DDR4/enhanced GPUs for the 6c model and up. Maybe differentiate the product lines by labeling the 4c model as its own product. "Mac" vs "Mac Pro", something like that. Perhaps put a price premium on the Pro line for the new technology being introduced and keep the 4c at the same $2,999 or maybe offer a slight price drop on it.

It wouldn't be very "Apple" of them to just drop the 4c and offer the 6c at $2,999. That would be sweet as hell, though. If that happened I'd have an extra $1000 for other things :p
 
I'm hoping that's exactly the case, but I'm doubtful. Part of me thinks Apple might opt to keep the 4-core Ivy Bridge-EP setup, but then offer Haswell-EP/DDR4/enhanced GPUs for the 6c model and up. Maybe differentiate the product lines by labeling the 4c model as its own product. "Mac" vs "Mac Pro", something like that. Perhaps put a price premium on the Pro line for the new technology being introduced and keep the 4c at the same $2,999 or maybe offer a slight price drop on it.

It wouldn't be very "Apple" of them to just drop the 4c and offer the 6c at $2,999. That would be sweet as hell, though. If that happened I'd have an extra $1000 for other things :p

They could possibly still offer the 4c and drop the Mac Pro down to the $2499 starting price it was at before.

Or they could just ignore reality and put the 4c at $3000. :)
 
the latency, added cost/complexity , and one lane doesn't buy much when folks are asking for something that soaks up.....

(bunch of pcie info)
Thanks, I figured complexity and other complications would be reasons but wasn't sure on all the other details.
If Apple could drop some GPU cards that had the DSP present and active that might go to mute some of the "what does this buy for audio" complaints. ( although I'm sure there will still be folks who complain it isn't the specific DSP they want to use..... )
Then there's the Xeon with integrated FPGA line coming but I'm assuming that's a long shot for the MP, if the parts are even coming in the same timeline.
Is the 295X an actual GPU or did you mean the X2?

Regardless, again seems to line up nicely for a nMP release around Yosemite.
I thought there was a standalone 295X but I guess that was just a rumor. Course with a Mac Pro there'd be two GPUs either way, albeit with two x16 lanes rather than one on an X2.
I'm hoping that's exactly the case, but I'm doubtful. Part of me thinks Apple might opt to keep the 4-core Ivy Bridge-EP setup, but then offer Haswell-EP/DDR4/enhanced GPUs for the 6c model and up. Maybe differentiate the product lines by labeling the 4c model as its own product. "Mac" vs "Mac Pro", something like that. Perhaps put a price premium on the Pro line for the new technology being introduced and keep the 4c at the same $2,999 or maybe offer a slight price drop on it.

It wouldn't be very "Apple" of them to just drop the 4c and offer the 6c at $2,999. That would be sweet as hell, though. If that happened I'd have an extra $1000 for other things :p
Yeah I wouldn't be surprised at all if they kept a 4 core around, although if so I also wouldn't be surprised if it was cheaper like goMac's thinking. It's something they've done with other products a bunch of times, like announce a new lower entry price with some annoying caveat(s) in the specs of the new lower end model.
 
Just because stuff isn't getting to consumers doesn't mean 4K isn't becoming more important as a workflow for the extra flexibility and future-proofing, though. I agree 4K adoption is going to be anemic compared to HD, but the consumer market ≠ professional market in terms of 4K having utility beyond just a delivery format.

I'm not flaming but what percentage do you think of the nMPs being sold today will ever run 4k? Don't get me wrong, I get why they marketed as such. But marketing and reality often aren't the same.

I really would like to hear how people are using 4k and selling it to support their purchases. I mean on one hand, every other advice about the nMP given is to go with a iMac because of needs. So why doesn't the same logic apply to all things 4k? Where is the true gain working with 4k over 1080P cameras in the mainstream? Besides Hollywood, who really needs it?

Another interesting fact would be to find out how many people in hollywood or with hollywood needs are shooting with 4k? -100, 1,000, 50,000, 100,000 1,000,000? I suspect the later to be on the extremely high side of production even over the next couple years given how much commercial 4k is out there now.

I've looked at cameras and I could easily afford 10 1080P for the price of one 4k. I can't buy 10 Mac minis for the price of one nMP. So why the uproar over price? Seems it would be right in line or a great deal compared to other 4k technology.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.