Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Status
Not open for further replies.
Heh. So my NDA now proves worthless. I have found Google search results referencing a "Tylersburg WS" for workstations, that includes other things, like "Tylersburg-D36".

I can honestly admit that I had not heard about those. Nor would I have had a reason to hear about them. (So my lack of knowledge of them should not be taken as confirmation of lack of existence.) So while I can't confirm it, I can't deny it either. And it sounds perfectly logical. I can say that the only search results I see for "Tylersburg-DP" are this thread, though. It looks like it's just "Tylersburg-D##"

Of course, this means I will have to go look this all up when I get in to work tomorrow; which means I won't be able to post in this thread again. :rolleyes:

Intel naming schemes, gotta love um.

From what I understand there are:
Tylersburg-EP: High end servers (dual socket?)
Tylersburg-EN: Low end servers (single socket?)
Tylersburg-WS: Workstations (dual socket, dual IOH?)

Then 24S, 24D, 36S and 36D refering to PCI-E 2.0 lanes and quickpath connections (S being one, D being two).

I don't know how much the EP, EN and WS stuff is marketing compared to particular features needing to be present.
 
Intel naming schemes, gotta love um.

From what I understand there are:
Tylersburg-EP: High end servers (dual socket?)
Tylersburg-EN: Low end servers (single socket?)
Tylersburg-WS: Workstations (dual socket, dual IOH?)

Then 24S, 24D, 36S and 36D refering to PCI-E 2.0 lanes and quickpath connections (S being one, D being two).

I don't know how much the EP, EN and WS stuff is marketing compared to particular features needing to be present.

I think people are confusing nomenclatures here...

Nehalem-EP is the Gainstown dual-socket variety of CPU. AFAIK, the EP suffix does not apply to chipsets... only CPU's.

Tylersburg comes in four configurations with either 24 or 36 lanes and one or two quickpath links as noted above. The 36S is currently marketed as the X58.

While Apple could implement a dual 36D chipset configuration, it would be costly and completely unnecessary. What on earth does one need 48 pci-e 2.0 lanes for... especially with OSX where peripheral support is hardly noteworthy. Quad crossfire or Tri-SLI are just not considerations here.

The most likely dual socket implementation would include a single 36D chip with an added PCI-e switch to add additional lanes if Apple felt the need.
 
The most likely dual socket implementation would include a single 36D chip with an added PCI-e switch to add additional lanes if Apple felt the need.

So what would that give us? Two 16x and one 4x.

Oh, sure, "add" another one... Apple being Apple, I can see them not and just giving us three PCIe slots.

I don't know; so 36D is the most likely board? I'll make the change.
 
Also, due to circumstances beyond my control, the funds I had dedicated to my Mac Pro will not be going toward a Mac Pro. Therefore, I will be stuck with this Penryn MacBook Pro until such time as I can get the money to afford a Mac Pro again.

O, come now, Tallest, new shoes for the baby *can* wait until next year. This is *important* !! :apple:
 
a little mess about last night; a simple play on the existing styling. I personally don't think they will be changing the form of the upcoming model and i do not expect there to be any black finishes on the case.

3273604411_87de450ea6_o.jpg
 
Hey Tallest, thanks for keeping up the thread anyway. Here's hoping that one Haswell will be provided by an employer for development... and that you'll have your own mac pro at home.

what a bummer, though.

edit:

very cool mock up... but i doubt we'll see 3 fw400's up front, if any.
 
This thread is great. Thanks for all the info an updates in the OP, Tallest Skil.

The mock-up above is pretty interesting, but most people overlook the handles/portability Apple built into their floor models since 1999. I don't think they are going to get rid of that. Personally, I love the current Pro case, but the handles are not ergonomic and hurt an ungloved hand after a few seconds. Having a fold out or mechanical handle would be foolish and costly.
 
It's a little more important than that, plus I have no children.

Well, also I don't have any children. That I know of.

Seriously, just trying to be a little light. (Does that sentence actually work?) Hope you turn up an unexpected windfall and can go for whatever you like -- Mac Pro + twin 30" ACDs? Best regards. :apple:
 
Well, also I don't have any children. That I know of.

Seriously, just trying to be a little light. (Does that sentence actually work?) Hope you turn up an unexpected windfall and can go for whatever you like -- Mac Pro + twin 30" ACDs? Best regards. :apple:

Twin 30" LED Cinema Displays. The updated kind. :D

We'll see what happens.

Oh, I think that "light-hearted" works better. :cool:
 
I think people are confusing nomenclatures here...

Nehalem-EP is the Gainstown dual-socket variety of CPU. AFAIK, the EP suffix does not apply to chipsets... only CPU's.

Tylersburg-EP and WS are dual socket platform names announced by Intel some time ago.
 
So what would that give us? Two 16x and one 4x.

Oh, sure, "add" another one... Apple being Apple, I can see them not and just giving us three PCIe slots.

I don't know; so 36D is the most likely board? I'll make the change.

36D seems the most likely yes. How Apple will provide (at least) the other two x4 we don't know. Either by what VirtualRain is suggesting or with dual IOHs.

Dual 24Ds would provide two PCI-E 2.0 x16 and two PCI-E 2.0 x4 slots. Or maybe one x4 and one x8 I don't know. WeE don't really know what combinations are going to be used or how pricing makes them viable, just that Intel plan to have many combinations on all platforms. We also don't know Apple's ideas for PCI-E 2.0 bandwidth, other than PCI-E is the only expansion card method they use.
 
36D seems the most likely yes. How Apple will provide (at least) the other two x4 we don't know. Either by what VirtualRain is suggesting or with dual IOHs.

Dual 24Ds would provide two PCI-E 2.0 x16 and two PCI-E 2.0 x4 slots. Or maybe one x4 and one x8 I don't know. WeE don't really know what combinations are going to be used or how pricing makes them viable, just that Intel plan to have many combinations on all platforms. We also don't know Apple's ideas for PCI-E 2.0 bandwidth, other than PCI-E is the only expansion card method they use.

Using two Tylersburg chipsets is easy, but like you, Apple's thinking might be difficult to guess. Then there's the cost.

We could see anything from a single 36D to combinations of 2x 24D, 24D+36D, or even 2x 36D. Given Apple has never supported CrossFire which requires no licensing fees, I'm Leary of actually seeing anything greater than two 16x slots for graphics.

Costs may dictate the use of a switch to provide anything greater than the lanes available on the 36D in a dedicated layout. The down side, is anything attached to it, is competing for bandwidth of the lanes it's connected to.
 
If they do make the case longer, I could see them shortening the stand and handles to minimize the over-all height. I did a search for "36D" and got all kinds of interesting hits. :D I'm not sure If I like dual 24D's or just one 36D. :p But one 24D and one 36D would really freak me out.
 
Just like Intel PC motherboard manufacturers for the Windows market, Apple could implement any combination of PCIe expansion they want.

This is a common chip used by such manufacturers to extend the PCIe lane configuration beyond that provided by the chipset...
http://www.idt.com/?catID=6264187
http://www.idt.com/?id=162


My personal thinking is that Apple will provide 4 16x slots (physical) with two of those being x16 (electrical) from the Tylersburg and the other 2 slots being 8x (electrical) or if they go overboard (for marketing purposes and be somewhat wasteful) all 4 slots might possibly be 16x the latter two of which would be implemented through a switch like the IDT I linked to above through the ICH10 (which has a 4x link to the CPU through the Tylersburg.

Given the Mac's lack of support for multiple graphics cards, all of this is completely excess. Not even top of the line graphics cards running in SLI can saturate a x16 2.0 bus. So rest assured, your new Mac Pro will not have any expansion bus bottlenecks.
 
Would making the handles more rounded (like recent Mac designs) help?

Yes...at least from how I see it...

Actually, I have sliced my finger open pretty bad on one of the handle edges. In fact apple ended up giving me a small compensation for this. Some of the edges of the case can be very sharp especially with lateral movement. In any case I wrapped all of my edges in black car door edge guard after that little fiasco as you can see in the pic. And as long as we are on the subject of ergonomics, what would be so wrong with some nice little wheels? Isn't rolling better than dragging?
 

Attachments

  • macpro1avatar.jpg
    macpro1avatar.jpg
    10.5 KB · Views: 614
New topic... CPU Cooling...

What does Apple use for CPU coolers on the Penryn chips now? Do they use stock Intel coolers? If not, is it custom? Are they noisy?

Has Apple ever offered water cooling?
 
Not even top of the line graphics cards running in SLI can saturate a x16 2.0 bus. So rest assured, your new Mac Pro will not have any expansion bus bottlenecks.
I was thinking in terms of lane availability (simultaneous request for access), not saturation.
Actually, I have sliced my finger open pretty bad on one of the handle edges. In fact apple ended up giving me a small compensation for this. Some of the edges of the case can be very sharp especially with lateral movement. In any case I wrapped all of my edges in black car door edge guard after that little fiasco as you can see in the pic. And as long as we are on the subject of ergonomics, what would be so wrong with some nice little wheels? Isn't rolling better than dragging?
I've cut myself too many times to count, even on user friendly cases. There always seems to be an unrounded edge of stamped metal somewhere. ;) And I always seem to be the person who finds it (cable routing :rolleyes:). :eek: :p
 
Status
Not open for further replies.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.