2013 Mac Pro timeframe

[deconstruct60]

Hey there. In simple terms, how do you see the Thunderbolt working on a new mac pro with dedicated graphics cards?

I don't (presuming 'dedicated' means graphics only removable PCI-e cards). Removable cards coupling to Thunderbolt is an expensive, Rube Goldberg solution in search of a problem.

The hugely flawed notion here is that the Mac Pro should only have one GPU. It can very easily have two. The Mac Pro 15" has two. The iMac has two. Why couldn't a Mac Pro have two?

Once stop resisting the notion that there can't be two what happens now when there are two dedicated cards in a Mac Pro. One set of monitors hooked to one card. Another set of monitors hook to the other card.

Now for Thunderbolt the extremely straightforward solution is use embedded ( "solder onto the motherboard") one of those two cards. The 2012 iMacs discrete GPU soldered onto the board. Route just that discrete component's output through the new thunderbolt ports (e.g, drop the two Firewire ports on back perhaps).

Where folks get their underwear in a twist of over how to pump random discrete card output into a Thunderbolt pipeline. The real question there is why bother? Fewer cables? Nope. More standard cables to a larger and more diverse set of monitors? Nope. Cheaper cables? Nope. Multiple video signals on a single cable? Nope ( DisplayPort 1.2 fixes that and Thunderbolt doesn't even do 1.2 ).

If snag some even more expensive fiber Thunderbolt cables can put a bigger distance between monitor and Mac Pro. Really, that's about it and it is a narrow corner case.

Once have a GPU card with a edge for ports readily available the nominally available ports can just be used. In short, they can just use mainstream cards.


Intel likely isn't going to certify whacky solutions where the user can remove the GPU signal completely from the Thunderbolt subsystem. Folks can spead lots of efforts swimming upstream from that...... but it is a huge waste of time.

 

[goMac]

It's difficult to say what Apple will do. Fortunately, it's a problem that is not just Apple's problem. Anyone who is a customer of Intel and builds towers has this problem.

 

[derbothaus]

Hey if you were old school you'd be on a Trackball

True. 1st gen optical is pretty close though.

----------

The razor blade conundrum is a simple one. Whenever sales slow by more than 5%, it's time to add another blade before everyone just decides to grow a beard. With mice much like other consumer marketing, I wonder if they're being realistic with their numbers.

I buy the blade explanation. The mice... not sure really unless people really like to be able to traverse 2650x1600 in 3mm of hand twitching.

 

[Phrygian]

MM is better for me ergonomically because I'm not flicking that scroll wheel as much. You can't hold it like a normal mouse (i.e. it's better if you don't try to rest your hand on it), but once you get over that, it's much better for things like coding than a traditional mouse.

I've got a Naga. I like it, but I actually have complaints about the build quality. Wireless seems flakey, it also feels like I have to charge it far too often, and I'm not sure it's also charging reliably.

I use a wired naga, and i have the 2012 edition. It is an improvement over the first naga i owned.

 

[goMac]

I use a wired naga, and i have the 2012 edition. It is an improvement over the first naga i owned.

Actually, I think it's the 2012 Mamba I have.

The names are confusing on those.

 

[deconstruct60]

It's difficult to say what Apple will do. Fortunately, it's a problem that is not just Apple's problem. Anyone who is a customer of Intel and builds towers has this problem.

Workstation towers? Not really

" ... Data Center Group had revenue of $10.7 billion, up 6 percent from 2011. ..."
http://newsroom.intel.com/community...enue-of-533-billion-net-income-of-110-billion

In the Xeon space, Intel is doing OK and lots of customers are not sitting around on their hands 'waiting for Ivy Bridge'. Some of that upward trend in the Data Center group is the networking associated with servers but the group is on sound footing. The chip/socket revision is slightly longer now but that isn't completely disconnected from the market demand.

Bargain priced ( tablet like average selling price ) towers ? Yes those are in deep trouble. However, they have been in trouble ever since laptops started taking them out 4-5 years ago.

 

[thekev]

I buy the blade explanation. The mice... not sure really unless people really like to be able to traverse 2650x1600 in 3mm of hand twitching.

I was thinking of other things where dpi is used as a measuring tool such as scanners, cameras, and printers. They claim a certain dpi based on output, even if they don't fully resolve things to the level implied by such a value. It's still extremely high. I'm just saying they may be overshooting to achieve a certain level of accuracy in sampling. I'm not really that into gaming. The thing I like about Razer is they make a left-handed mouse, which is otherwise uncommon.

 

[phoenixsan]

Guessing....

1)-New processors capable of run Mac Pros were slated to appear in last quarter of 2012-first quarter of 2013.

2)- I dont see a refresh in the Mac Pro form factor

3)-Evidence in MacOS update builds suggests better graphic supports....

So, what about a wild guessing, and getting a updated Mac Pro in time of the WWDC 2013?

 

[theSeb]

1)-New processors capable of run Mac Pros were slated to appear in last quarter of 2012-first quarter of 2013.

I don't believe so. Not according to any roadmap that I've seen. Announcement during WWDC and then shipping a couple of months later is likely.

 

[chedda]

Best Case Scenario

I have a feeling any re-design will have to be iconic like it's predecessor. The Mac Pro is not just a model it has significance. If it is re-designed apple may want to stick to their system which means it will have a huge life cycle like the current shape.
The pro market & especially my field (3d modelling & rendering) is heating up with the competition between GPU rendering and the Xeon Phi. Apple like one liners such as "the thinnest laptop", "the highest resolution" etc. I am sure they will want to be able to say "the Mac Pro the worlds most powerfull workstation"
Apple will work with intel and launch the Mac Pro with a showcase of 3dsmax with autodesk present. This is one of the last bastions for windows use.

PS i have rose tinted binoculars.

 

[chedda]

News

Just saw this maybe it's helpful:

http://www.tomshardware.com/news/Intel-Haswell-Ivy-Bridge-E-CPU,20590.html#xtor=RSS-998

 

[d-m-a-x]

Just saw this maybe it's helpful:

http://www.tomshardware.com/news/Intel-Haswell-Ivy-Bridge-E-CPU,20590.html#xtor=RSS-998

it's really bad they did not make a Sandy Bridge Mac Pro. Such a huge gap, it is making people think apple does not make towers anymore

 

[JesterJJZ]

it's really bad they did not make a Sandy Bridge Mac Pro. Such a huge gap, it is making people think apple does not make towers anymore

At the very least they should have given us SATA 3 and USB 3 in the "new" Pro.

 

[d-m-a-x]

At the very least they should have given us SATA 3 and USB 3 in the "new" Pro.

not sure about sata 3 or thunderbolt, but usb3 would have been a snap

 

[deconstruct60]

Just saw this maybe it's helpful:

http://www.tomshardware.com/news/Intel-Haswell-Ivy-Bridge-E-CPU,20590.html#xtor=RSS-998

No really.

" ... Later, in the third quarter of this year, Intel will release the Core i3 Haswell and Ivy Bridge-E CPU's. .. "

First, Apple is unlikely to use -E ( Core i7 3xxx labeled ) solutions. Roadmaps dropped months ago tagged Xeon E5 Ivy Bridge coming in early in Q3.

http://www.engadget.com/2012/10/17/intel-roadmap-reveals-10-core-xeon-e5-2600-v2-cpu/

That was back when Haswell roadmaps said they were coming in April. Those have slid to June. It is quite likely that Xeon E5 Ivy is going to slide also. Intel won't announce two products that use the same fab technology process (22nm) to ramp up at the same time. There will be an initial demand bubble and there is no good reason to make two of them overlap if they don't have to. They don't have to ( neither AMD or ARM's parteners are a threat in this space).

The Core i7 39xx v2 models may get preannounced (or extensively leaked ) in June. That would be to keep the high end tweakers from grumbling how their i7 didn't get revised before the laptop i7 did.

 

[deconstruct60]

not sure about sata 3 or thunderbolt, but usb3 would have been a snap

Of the three, SATA 3 (6.0 Gb/s ) is the most likely because it is actually built into the chipset for any Xeon E5 Sandy Bridge option.

http://www.intel.com/content/www/us/en/chipsets/server-chipsets/server-chipset-c600.html

(probably a 602 variant. )

The other two would need discrete controllers. It is unlikely that would change with Ivy Bridge since the same chipset would be leveraged.

There are two 6Gbpw links so two 2.5" drive sleds would be a good match. The other 2-4 would be for HDDs (or perhaps a single ODD).

 

[xgman]

PS i have rose tinted binoculars.

you took the words out of my mouth.

 

[deconstruct60]

Announcement during WWDC and then shipping a couple of months later is likely.

Not likely at all.

1. Recent rumors are reveal grumblings about the iPhone moving back to the late June - early July timeframe. If so that will take over WWDC timeslots just as it did when that was its release time frame. One of primary reasons there was "mac stage time" last year is that the iPhone moved. The iPhone and iPad both in the Fall is a tad crowded. It is likely Apple will open a bit of larger gap between the two. That likely means the iPhone moving back to Summer.


2. Unless there is something significant about OS X 10.9 that requires a Mac Pro it probably wouldn't make the stage. WWDC is primary about iOS and OS X. Where the hardware illustrates new OS feature that demos well ( e.g., Retina MBP 15") it warrants pulling up.

Otherwise new hardware is easier to release on 10.9 once it surfaces rather than iteration that would be superseded in a month or so.

It doesn't look like the Haswell SoC variants will be out by WWDC but a MBA with a SoC may enable some OS X features that demo well.


3. Last year's Mac Pro bump didn't even get a press release let alone WWDC stage time. The highly retweaked iMac got a non-WWDC dog-and-pony show. If there is something supposedly revolutionary about the Mac Pro's physical design that has little to do with OS then Apple can do one of those independent dog-and-pony shows later when it is ready to ship.


4. Intel's slide of Haswell release into June will likely move Ivy Bridge Xeon E5's back slightly. If this is hand waving about announcing IB E5's because won't be available till July ( note that Intel isn't allowing announcements of Haswell systems until close to release. ) that release will also likely slide a bit. Unless WWDC is very late in June

If Intel is artificially sitting on Ivy Bridge Xeon while it is completed testing it could slide forward but that isn't likely but it would remove WWDC timing form the issue.

Given 2012's huge gap between XEon E5 announce and system vendors' release even a July announce might mean a couple month delay for actual systems. That is way too big of a gap between WWDC and shipping. Apple's already partially invoked the Osborne effect on the Mac Pro. It highly unlikley they'd amplify that with a large gap between announce and ship.

Same thing in desktop space last year. Sandy Bridge was suppose to roll out very early Q2 and didn't really start shipping until second half of Q2. That was one reason for the log-jam of systems at WWDC. At this point Apple should have been prepared for that and not scheduled a Mac log-jam.
Intel's roadmaps haven't been extremely accurate for the last 24 months. It is about +/- Quarter sized error bar on the dates they throw up.

 

[CtotheP]

I have a feeling any re-design will have to be iconic like it's predecessor. The Mac Pro is not just a model it has significance. If it is re-designed apple may want to stick to their system which means it will have a huge life cycle like the current shape.

^^^ This

 

[dexum175]

Anyone know when the Apple keynote is on? I'm on the verge of buying a new Mac Pro but I'm afraid the new model is right around the corner.. I'm currently using an iMac but it's really giving me problems now.

 

[theSeb]

Not likely at all.

1. Recent rumors are reveal grumblings about the iPhone moving back to the late June - early July timeframe. If so that will take over WWDC timeslots just as it did when that was its release time frame. One of primary reasons there was "mac stage time" last year is that the iPhone moved. The iPhone and iPad both in the Fall is a tad crowded. It is likely Apple will open a bit of larger gap between the two. That likely means the iPhone moving back to Summer.


2. Unless there is something significant about OS X 10.9 that requires a Mac Pro it probably wouldn't make the stage. WWDC is primary about iOS and OS X. Where the hardware illustrates new OS feature that demos well ( e.g., Retina MBP 15") it warrants pulling up.

Otherwise new hardware is easier to release on 10.9 once it surfaces rather than iteration that would be superseded in a month or so.

It doesn't look like the Haswell SoC variants will be out by WWDC but a MBA with a SoC may enable some OS X features that demo well.


3. Last year's Mac Pro bump didn't even get a press release let alone WWDC stage time. The highly retweaked iMac got a non-WWDC dog-and-pony show. If there is something supposedly revolutionary about the Mac Pro's physical design that has little to do with OS then Apple can do one of those independent dog-and-pony shows later when it is ready to ship.


4. Intel's slide of Haswell release into June will likely move Ivy Bridge Xeon E5's back slightly. If this is hand waving about announcing IB E5's because won't be available till July ( note that Intel isn't allowing announcements of Haswell systems until close to release. ) that release will also likely slide a bit. Unless WWDC is very late in June

If Intel is artificially sitting on Ivy Bridge Xeon while it is completed testing it could slide forward but that isn't likely but it would remove WWDC timing form the issue.

Given 2012's huge gap between XEon E5 announce and system vendors' release even a July announce might mean a couple month delay for actual systems. That is way too big of a gap between WWDC and shipping. Apple's already partially invoked the Osborne effect on the Mac Pro. It highly unlikley they'd amplify that with a large gap between announce and ship.

Same thing in desktop space last year. Sandy Bridge was suppose to roll out very early Q2 and didn't really start shipping until second half of Q2. That was one reason for the log-jam of systems at WWDC. At this point Apple should have been prepared for that and not scheduled a Mac log-jam.
Intel's roadmaps haven't been extremely accurate for the last 24 months. It is about +/- Quarter sized error bar on the dates they throw up.

Yes, that is a well thought out response. I can see now that my predicted timeframes were extremely optimistic.

 

[wallysb01]

It is about +/- Quarter sized error bar on the dates they throw up.

I don't think you need the "-" in that "+/-".

 

[iBug2]

I think you may be spot on.
I don't see Apple releasing any more traditional workstations, their main focus is on iOS, Mac is less than 10% of their turnover.

That 10% is still a multibillion dollar business, and not to forget it drives the ecosystem which sells those iOS devices. Mac has never been more important for Apple than it is now imho.

Mac Pro is another issue. Professional Workstations are needed less and less every day, but since Apple is not abandoning those, yet, they will release a proper workstation, not a mac mini pro.

 

[deconstruct60]

I don't think you need the "-" in that "+/-".

Generally yes but sometimes depends upon the customer and viewpoint.

Intel had their big dog-and-pony release for Xeon E5 series on March 7 2012. Regular server/workstation vendors could start talking about dates for systems publicly after that. ( HP initially targeted April, really May, and Dell July for example).

However, Intel did ship Xeon E5 2600 series products to Supercomputer vendors in Sept-Nov 2011 (to make it into systems for the Nov 2011 Top500 rankings). About a quarter early to the eventual actual general release date. Normal folks couldn't buy them at newegg but they were shipping to those with special classification. So that would be a '-' relative to their revised (as of Q3 '11) schedule for the roll out targeting Q1 '12. [ But yeah according to the '10 and maybe very early '11 roadmaps even that "early release" was a bit late or on time. ]

 

[deconstruct60]

That 10% is still a multibillion dollar business, and not to forget it drives the ecosystem which sells those iOS devices. Mac has never been more important for Apple than it is now imho.

10% isn't as important to them as growth. Growth is a direct, tangible indicator that they are making key products that people like as much as it is helpful in keeping the stock price inflated (or over inflated ).

Apple has less than 10% of the overall historic PC form factor market and they haven't quit selling laptops and desktops yet. The 10% or the "billions" isn't the core issue.

Looking at itunes , iOs , etc is even necessarily. Even inside of that 10% Mac division either something has to dramatically change for the Mac Pro or it likely will get axed.

In the growth respect there is a slight stumble. The sputtered iMac release has Mac sales flat year-over-year. I don't think they are in a panic over that, but if that state continued for a year or so it isn't likely there would status quo of Mac model offerings over time. Non growth models would definitely be axed.

Mac Pro is another issue. Professional Workstations are needed less and less every day, but since Apple is not abandoning those, yet, they will release a proper workstation, not a mac mini pro.

While the iMac is still the primary strategic desktop offering (e.g., main contributor to the poor Mac results this quarter ) I don't Apple wants to put more growth burdens on it. If the Mac Pro could pull its own weight in growth that would be tactically a sound move to make.

If the Mac Pro is in the 100k/yr range then 10k per year growth would be fine. It would not significantly change the total Macs sold per quarter (or per year) but it would demonstrate a positive track.

Professional Workstations aren't needed less and less as much as the number of professions that are getting larger and are covered with value added software that demands a workstation performance is shrinking.

For the users whose workload isn't growing very much there is no need for a "mini pro". A mac mini '14 or '16 will relatively be a "pro" model.