This is not a "Pro" machine.

[subsonix]

However people generally NEED storage beyond an SSD boot drive. You just do. So you are spending more to hook up an expensive box via Thunderbolt, and now your supposedly great looking desktop becomes a spaghetti mess of cabling.

If all you are doing is replacing the internal drives of the current Mac Pro, you could do with just an USB3 enclosure. It would mean exactly 1 more cable going out of the back of the Mac Pro.

From some pictures I have seen of the new Mac Pro it also looks as if there may be an option for a second PCIe SSD along the line (possibly even on the release).

 

[deconstruct60]

This is a red herring.

No. The post and quote I responded to was about Apple going after everyone. They aren't. This...

Of course pros "care" about what it looks like,

is a red herring. My reference " For the pros who... " is a subset of pros. This unqualified pros is a snare all pros while

many people lauded the previous Mac Pro design as having a beautiful case.

is again a subset.

However people generally NEED storage beyond an SSD boot drive.

There is nothing about this new Mac Pro that prohibits past a single boot drive. Even in the pictures availabe can see an outline for a SSD drive connector on the second card.

If Apple only offers this in all configurations as one and only one drive that would be weird even for Apple. The 2012 Mac mini and iMac both have two drives. It would not be hard to put it then into equivalence on drive count with the others in a BTO config ( just like the others go to 2 drives primarily in a BTO configs).

If just one very high priced , higher capacity SSD drive is the only internal capacity expansion option then yes this is likely a troubled product offering.

Having a boot SSD and a workings space SSD isn't all that weird though.

So you are spending more to hook up an expensive box via Thunderbolt

There is nothing in the 2012 design that necessitates that the bulk storage has to be Thunderbolt. Or that a Thunderbolt device is the storage device.

, and now your supposedly great looking desktop becomes a spaghetti mess of cabling.

Implicit presumption that the storage is actually on the desktop. It doesn't have to be.

What IS Apple's objective? It's almost like they don't have one aside from making the most different looking thing they can, function be damned.

If function is being a NAS box then yes not 100% aligned with function. Don't need two GPUs for a NAS box? Not hardly.

but the two workstation class GPUs is a huge clue as to what the objective is. It is leveraging those is a highly effective way.

Seriously, what was the goal? It's a completely weird machine. They could have made it much smaller still and included the ability to put in drives.

Weird ( relative to machines from 10-15 years ago ) doesn't mean it doesn't have a goal. It is far more likely the goal is optimized general target market realization 2-5 years from now rather than the general target market 10 years in the past.

Actually not much smaller if going to keep computational performance constant. The height of the three daughter cards and the RAM DIMMs pretty much drive that height. The fan drives the diameter. Hence cylinder. It is functional.

That is quite different from making standard SATA drive containers part of the requirements. If they were in then that would contribute. So obviously they are out so not a driver of the shape. If the intent is only is only the fastest storage internal then there any strong requirement driver for standard HDD containers.



Any kind of SATA 2.5" (let alone 3.5" ) storage would make it bigger not smaller. You could chuck computation chops for massively increased storage capacity, but what you would have is upper end NAS box. If anything this design presumes you already have something from the broad class that kind of device belongs to.


The design here approaches internal storage very much like a L3 cache on in a CPU design. There is no general purpose RAM in a CPU package either.

 

[deconstruct60]

If all you are doing is replacing the internal drives of the current Mac Pro, you could do with just an USB3 enclosure.

Not really. Just two of VelociRaptors sequential file streaming at 200MB each

http://www.storagereview.com/western_digital_velociraptor_1tb_review

would likely swamp USB 3.0. Random access sure, just fine but streaming would have problems. Three of those and definitely have problems.

For 1-2 drive sled deployments of primarily random access HDD workload, USB 3.0 works. 4 sleds and Raptor or high end streamer and it isn't going to fly.

 

[subsonix]

Not really. Just two of VelociRaptors sequential file streaming at 200MB each

Eh, using striped VelociRaptors as an example is an edge case, not a typical setup for what most people have in their internal bays, at all. And fwiw, it should be right at the edge of what USB 3 can handle.

 

[blackhand1001]

In the short term, the Mac Pro 2013 with bootcamp , Windows , and certified graphics might actually make that penetration bigger.

Not take over that whole segment, but perhaps help sell as many (or more) new Mac Pros they formerly sold into the upper 30th percentile in the "soon to be old" dual processor space.




Yet another potential growth market with the expanding interest of 3D printing/manufacturing. Things like that where more folks are pouring into a market rather than to a market with a relatively fix number of players and extremely high barriers to entry (e.g. mass distribution media) that are largely stagnant.

Your better off with an hp z420/620/820 workstation. The new machine is a joke and the form factor will make it much more expensive than competitors.

 

[OS6-OSX]

They aren't trying to design something for everyone. The objective to put devices that make a difference to the people they do target.

Then who are the people "they do target"? Will it be a pool of folks larger than the "niche" market they claimed were the old MP crowd? The same crowd that did not purchase enough MP's comparatively speaking to their ithings?
"What can we do to sell enough MP's to continue the line?"
"Let's give them some icandy but zero real world comparisons!"
"Let's compare it to the HP z820". "Well that's not our targeted market"
"Let's compare it to a modified/updated 5.1". "Well that's not our targeted market" "Plus we don't want to embarrass ourselves!"
"Well, what do we compare it to?" "Nothing!"
"Let's just give them some internal numbers without comparison."
"But these are "Pro's" that will see through our smoke and mirrors."
"Who cares, they are not our targeted market!"
"Let's throw it out there and hope that it sticks" "But hopefully we don't get stuck with them in the warehouses"

I'm not against the MP 6.1 for those who want one. I just not in that line.
Only time and more importantly the price will tell!

 

[CaptHenryMorgan]

I can't understand why Apple is calling a non-expandable, no disc drive machine a "Pro" machine.

"Pro's", aka Professionals, do NOT want external expansion boxes cluttering their desk just because Apple deems it necessary to not allow people to expand it's internal components. ("expansion is external")

Professional users also want to burn media (videos, photos, etc) to DVDs and Blu Rays. Now, on top of a ~$3000+ machine, they have to purchase an additional external drive just to do that. Apple is not gonna undercut the price of the high end ridiculously priced 15" rMBP. ($2799)

Don't get me started on the internal storage. WHY can Pros NOT expand the internal storage? Yes, there are external drives, but when it comes down to it, it's yet ANOTHER external expansion box, cluttering an already cluttered desk.

Speak for yourself. As someone who works with all Thunderbolt drives and external cards and components, this is the PERFECT machine for the post house I work in! Not to mention I can toss it in my backpack and bring it to another site if need be.

 

[subsonix]

Your better off with an hp z420/620/820 workstation. The new machine is a joke and the form factor will make it much more expensive than competitors.

A 12 core Ivy Bridge-EP and 7000 Tflops worth of workstation class GPUs and a PCIe SSD will not exactly be "cheap" what ever brand you buy.

It seems like the new case has reduced the amount of parts (and material) used compared to the current Mac Pro.

 

[blackhand1001]

A 12 core Ivy Bridge-EP and 7000 Tflops worth of workstation class GPUs and a PCIe SSD will not exactly be "cheap" what ever brand you buy.

It seems like the new case has reduced the amount of parts (and material) used compared to the current Mac Pro.

The base model on the new mac pro is not going to have any of that. Also the hp z series workstations still have dual sockets so you can load in 2 8 cores or 12 cores and get more cores. You can also load in 2 6 cores which is a lot cheaper than having 1 12 core and you get twice the bandwidth as you have two QPI links. It also has real expandability and comes with a much better warranty than the apple machine including on site service included for 3 years standard. The simple fact of the matter is that apple is not competitive in the professional market. The compromises Apple made to get a form factor that the target market simply doesn't even want or need is too great. A slimmed down mid tower that still offers expandability would have been a much better option. Preliminary tests on these new mac pros show that it has thermal throttling issues which is a big problem for a machine that would be used for rendering. Overall its not what the market wanted and it seems more like the purposely made this and then will blame the bad sales to give them a way out of this market.

 

[subsonix]

The base model on the new mac pro is not going to have any of that. Also the hp z series workstations still have dual sockets so you can load in 2 8 cores or 12 cores and get more cores.

When the Sandy Bridge-EP was introduced, dual 16 core configurations was close to $8k without any graphics card for the HP workstations. So even though it's possible it's probably not what most will go for.

You can also load in 2 6 cores which is a lot cheaper than having 1 12 core and you get twice the bandwidth as you have two QPI links.

But it's divided on two CPUs.

It also has real expandability and comes with a much better warranty than the apple machine including on site service included for 3 years standard. The simple fact of the matter is that apple is not competitive in the professional market. The compromises Apple made to get a form factor that the target market simply doesn't even want or need is too great. A slimmed down mid tower that still offers expandability would have been a much better option.

That remains to be seen I think... The compromises are likely done on the basis of Apple researching their market and a realization of what is possible with native CPU/GPU today and around the corner.

 

[blackhand1001]

Explain to me how the new mac pro is a better fit for a professional than something like this. Apple let down a lot of people with this new mac pro. They let people down when they refused to update the machine for over 3 years. That simply doesn't cut it in the pro market. Apple simply decided it was more interested in dicking around with the form factor and making huge compromises than actually serving the pro market. They will use the poor sales of this new mac pro as an excuse to exit the market.

This machine is much smaller than previous mac pro yet still offers expandability and good thermals and noise levels.

HP also offers an even larger z820.

 

[subsonix]

Explain to me how the new mac pro is a better fit for a professional than something like this. Apple let down a lot of people with this new mac pro. They let people down when they refused to update the machine for over 3 years. That simply doesn't cut it in the pro market.

If you say so, I guess it must be true.


Edit: you should use moar and larger images (4608 × 3979) is not enough!

 

[linuxcooldude]

They will use the poor sales of this new mac pro as an excuse to exit the market.

I believe you got this part wrong. If they are looking to end the Mac Pro they would simply do it. Just like the X-Serve. They would also not spend all this for research & develop to build a new one. They would simply end it.

 

[martinX]

Preliminary tests on these new mac pros show that it has thermal throttling issues

I wasn't aware of that. That sounds interesting. Can you share a link to these tests?

 

[goMac]

I believe you got this part wrong. If they are looking to end the Mac Pro they would simply do it. Just like the X-Serve. They would also not spend all this for research & develop to build a new one. They would simply end it.

Yep.

If they wanted to end it, why would they spend all this time and R&D just to kill it?

 

[deconstruct60]

Preliminary tests on these new mac pros show that it has thermal throttling issues which is a big problem for a machine that would be used for rendering.

While that wouldn't be surprising for the ultra Max BTO configuration ( two W9000 , 12 core E5 v2 , Max RAM ) and several bus powered devices dangling off of USB/Thunderbolt, it would be interesting to see the numbers.

Likewise numbers for the the more mainstream configurations.

Overall its not what the market wanted and it seems more like the purposely made this and then will blame the bad sales to give them a way out of this market.

It probably is what a subset of the market wanted. Most likely Apple has to resigned to initially targeting a fraction of the market they once covered so this won't initially rack up all-time record Mac Pro numbers in the first year. The more important factor is if it get back on a growth path. It it takes 3-4 years to catch and pass the old "record" Mac Pro units says Apple has the time and patience to do it.

The subset of the market that is stagnant and/or shrinking they cut loose because all those will do is get the Mac Pro canceled in future years anyway. So not that much of a bigger gamble to make a beat that changing direction to cover future growth patterns now.

Since it is initially a smaller market, it is probably not going to be hard to cover the Osborne Effect suppressed and EU Market withdrawal sales slump with about as large numbers.

But for the "We have committed to group shared storage" market this design is a good fit. There is a high and low price market segments ( those with SAN/NAS and those with sneaker net DAS boxes that get moved from machine to machine ). In that market this is a good fit. It is not a one-man-show so don't need a 'stuff everything into a single box" solution.

Whether that subset is going to grow faster than the one-man-show is up in the air. Especially in the media space where storage requirements are on a hyper growth path. Ever bigger RAWs , ever bigger archives... all that leads to increasingly not fitting in a single box over the next 2-4 years.

 

[Tesselator]

Judging from the responses so far I guess everyone is reading your post the same as I. Pretty much all non-sense.

Explain to me how the new mac pro is a better fit for a professional than something like this.

It ships with and natively runs OS X. It has 6 TB2 ports - each of which is about four times faster than SATA III and can connect to external PCIe cards of very many types.

Apple let down a lot of people with this new mac pro.

I like the response above: "If you say so... it must be true". Hhehhee...

They let people down when they refused to update the machine for over 3 years.That simply doesn't cut it in the pro market.

I guess all the "pro" users who bought those machines disagree and disagree with their wallets.

Apple simply decided it was more interested in dicking around with the form factor and making huge compromises than actually serving the pro market.

I guess that's one way to look at it. I think it's a really close-minded and hateful way of looking at things tho. From the little info we actually have it looks to me like they are trying to cut costs by eliminating past standards which either are now or soon will be obsolete - and by streamlining several important aspects of fabrication, assembly, and logistics. It looks to me like the MP6,1 will be an extremely well balanced "workstation" - balanced for the vast majority of current MacPro users. The base price is the only question left in determining how popular it will be.

They will use the poor sales of this new mac pro as an excuse to exit the market.

Yeah, that's too silly to even address.

Image

This machine is much smaller than previous mac pro yet still offers expandability and good thermals and noise levels.

HP also offers an even larger z820.

Does it have six TB2 ports? No.
Does it have any TB2 ports at all? No.
Is it stuck with the antiquated or soon to be antiquated SATAIII for I/O? Yes.
To expand it's PCIe capabilities does it require extraordinarily expensive expansion systems? Yes.
Is it about six times bigger than the MP they showed at WWDC? Yes.
Does it draw more power and therefor is more expensive to run? Probably yes.

About the only advantage the z820 has over the MP6,1 is the number of RAM slots available. The smaller box above has no advantages at all and in fact is about half the spec of the MP6,1.

Preliminary tests on these new mac pros show that it has thermal throttling issues...

I call BS!

 

[goMac]

Is it stuck with the antiquated or soon to be antiquated SATAIII for I/O? Yes.

I think for pros this is going to be the killer right here, especially if the new Mac Pro offers dual SSDs. With drives apparently moving past (!!!) SATAIII speeds those SATAIII bays are going to be become not useful.

True, on a PC you could add a PCIe SSD card, but then why do you need the SATAIII bays?

 

[deconstruct60]

I think for pros this is going to be the killer right here, especially if the new Mac Pro offers dual SSDs. With drives apparently moving past (!!!) SATAIII speeds those SATAIII bays are going to be become not useful.

Not. They are useful for HDDs , SSHDs (hybrids ) , and affordable, high volume SSDs ( at least for now high volume.).

I'm kind of surprised none of the HDDs vendors have gone to a 2.5 drive in 3.5 container and pack lots more Flash in the gap SSHD drive.

SSHDs won't move past SATA 6Gb/s over a broad range of workloads any time soon but will increasingly over time get better at filling that up more often. SATA isn't going to disappear everywhere. ( SAS hasn't given up the fight http://www.lsi.com/solutions/Pages/12GBS.aspx . That will trickle down to SATA over time. )


It is questionable whether all chipsets will come with 6-10 SATA lanes as opposed to adding more v2 ( or possibly small number of v3 ) lanes in exchange for dumping SATA controller bandwidth and implementation space. Either as embedded controllers or just leaving it free.

For example if Intel could have one model in theE5 v3 (Haswell) chipset that drops the 10 lane + RAID SATA complex all together and puts in a 10GbE link, x3 worth of bandwidth USB 3.0 controller, or another plain x2-3 v2 PCIe lanes that would fit with the new Mac Pro's direction.

Honestly, it is pretty big waste when Mac laptops have just one SSD and the chipset has 4-6 SATA lanes ... for what? If dumped those for x4 PCI-e v2 lanes could make those single storage drives all PCI-e ( OK SATA Express over PCIe ) and the chipsets would be more aligned with what the are actually being tasked to do.

Not all the chipset variants of a given generation... maybe just one or two. There is a fork coming in the mainstream where the bulk of stuff is going to be socket less BGA and the chipset that goes with that dominating paradigm is probably going to be different. Similar thing happening in server space. Tons of servers are being stuff into large data centers and connectivity to the internet is something this is higher priority than doing more than one directly coupled storage device. ( OS boot drive attached to server which then goes to find iSCSI/SAN/etc. for data. )

Part of the problem getting the Xeon E5 launched in 2011 was the large and complex SATA/SAS subcomponents of the chipset.

True, on a PC you could add a PCIe SSD card, but then why do you need the SATAIII bays?

2.5" and 3.5" storage containers are still getting faster even if drop pure SSDs from the picture.

 

[deconstruct60]

Does it have six TB2 ports? No.
Does it have any TB2 ports at all? No.

This somewhat boils down to whether looking backwards or forwards.

If it is how big is my PCI-e card sunk costs, then Thunderbolt isn't going to play a big role.

If it is we have a largely all Mac shop and what will play well with the other Macs we have then TB has alot more traction.

Different groups are going to be at different stages of just how paid off their sunk costs are and will lead to very different viewpoints. It has nothing to do with pro or not ( unless haven't been professionally paying off the sunk costs. )

Is it stuck with the antiquated or soon to be antiquated SATAIII for I/O? Yes.

SATA 6Gb/s will be supersede in next year or so by 12Gb/s SAS/SATA but that actually presents a problem for Thunderbolt also. Thunderbolt v1 can't handle a 12Gb/s SAS controller. Thunderbolt v2 isn't going to do much better.

To expand it's PCIe capabilities does it require extraordinarily expensive expansion systems? Yes.

PCI-e external standard boxes aren't all that more expensive and maybe be cheaper in some cases.

Is it about six times bigger than the MP they showed at WWDC? Yes.

About the only advantage the z820 has over the MP6,1 is the number of RAM slots available.

Not really. It has more x16 sockets also. You could actually hook up a real "world leading speed" I/O card to a z820. Can't do that with a MP 2013.

Or can put 2 GPGPU cards in along with a card to primarily run displays.

There are some advantages.

The MP 2013 is more about how much is "good enough" for 80+ % of targeted users.

The MP 2013 doesn't try to chase that last set of topmost 1-2% percenters as much as the older box did. Even then current Mac Pro isn't chasing the topmost 1-5% of folks.

 

[blackhand1001]

I think for pros this is going to be the killer right here, especially if the new Mac Pro offers dual SSDs. With drives apparently moving past (!!!) SATAIII speeds those SATAIII bays are going to be become not useful.

True, on a PC you could add a PCIe SSD card, but then why do you need the SATAIII bays?

Every external hard drive enclosure right now still uses a sata interface internally and then is converted to thunderbolt or usb or whatever interface it uses. That means that it will never perform as well as a native running sata 3.0 device. The only exception to this is esata drives as they never have the signal converted. Also nothing is preventing someone from installing pci express ssd into the z820. Hard drives don't even come close to saturating a sata 3 drive and with external ssd's the drive controller to convert it to thunderbolt will cause extra latency which dramatically effects random read and write speeds. Thunderbolt is not a replacement for pci express or sata 3.

People also seem to think that thunderbolt is the solution to everything. Last time I checked the number of accessories for thunderbolt was pretty low and the ones that do exist are overpriced and very limited. Thunderbolt doesn't come anywhere close to the performance of a 16x pci express slot.

 

[goMac]

Every external hard drive enclosure right now still uses a sata interface internally and then is converted to thunderbolt or usb or whatever interface it uses. That means that it will never perform as well as a native running sata 3.0 device. The only exception to this is esata drives as they never have the signal converted. Also nothing is preventing someone from installing pci express ssd into the z820. Hard drives don't even come close to saturating a sata 3 drive and with external ssd's the drive controller to convert it to thunderbolt will cause extra latency which dramatically effects random read and write speeds. Thunderbolt is not a replacement for pci express or sata 3.

To clarify, I was talking about the SSDs internal to the Mac Pro which are NOT on a SATA3 bus, because the SATA3 bus is too slow.

I also mentioned the PCI Express SSD on the PC bit.

Not. They are useful for HDDs , SSHDs (hybrids ) , and affordable, high volume SSDs ( at least for now high volume.).

Sure, but all that could also be moved over to the Thunderbolt bus or 10 gig ethernet. With how cheap hard drives are these days, it's a better deal to buy a large number of them, stick them on the network, and have a large amount of capacity and bandwidth shared between users. Not to mention the implied redundancy keeping the data safe.

 

[slughead]

SATA 6Gb/s will be supersede in next year or so by 12Gb/s SAS/SATA but that actually presents a problem for Thunderbolt also. Thunderbolt v1 can't handle a 12Gb/s SAS controller. Thunderbolt v2 isn't going to do much better.

Exactly--but PCIe can. This can't be said enough about the new Mac Pro: Thunderbolt 2 is no replacement for PCIe drive controllers over 2GBps

 

[goMac]

Exactly--but PCIe can. This can't be said enough about the new Mac Pro: Thunderbolt 2 is no replacement for PCIe drive controllers over 2GBps

Wait, why wouldn't Thunderbolt 2 support the full throughput of a 12 gigabit/sec drive?

 

[slughead]

Wait, why wouldn't Thunderbolt 2 support the full throughput of a 12 gigabit/sec drive?

Sorry, my mistake, I misread his post (Gbps vs GBps). Obviously TB2 can support 2GBps.

The LGA2011 chipset has 40 lanes for 40GBps (with PCIe 3.0). At most, those TB ports will be able to handle 12 GBps, and only divvied up into 2 GBps cables.

The argument often used here is that TB is way better than SATA because it's not "locked in" to something that may get phased out. Fair enough, but meanwhile, TB2 isn't even enough to handle a four port SAS controller--and this is supposed to replace PCIe ? (not saying you made that statement, but others have)

TB2 is not nearly as future-proofed as PCIe (ironic, considering it doesn't even exist in the market yet), though people are making the argument that SATA should be dropped in favor of TB for that reason.