not sure how they could integrate sata3 and TB on the current boards but Im not an engineer so...
Sent from my iPhone using Tapatalk
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Mac Pro 2011 "Sandy Bridge E" everything we know
- Thread starter Amethyst
- Start date
- Sort by reaction score
not sure how they could integrate sata3 and TB on the current boards but Im not an engineer so...
Sent from my iPhone using Tapatalk
SATA 6Gb/s via a discrete chip connected through PCIe. Thunderbolt has to be done the same way in both, either a separate TB PCIe card or a TB chip connected though PCIe.
You have 36 PCIe 2.0 lanes to play with on the platform. X79/Patsburg is supposedly compatible with LGA 1366 as well.
Thunderbolt on the iMac and the notebooks uses a PCIe switch when you have to deal with a discrete GPU. That leaves x8 for the GPU, x4 for the Thunderbolt controller, and x4 that appear to be unused. (I do not have the block digram with me right now.)
Sounds interesting! I guess a few days(?) wait and we will see what apple has to offer this time. Altho I do see a slight chance of an entry level MP equipped with SP SB xeon according to all these rumors across the boards, but unfortunately I'm not a prophet
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
From my recollection no Mac Pro has ever had something like this, it's always been Intel chipsets and whatever they support natively.
In my opinion it's X79 or nothing.
Even if the X79 is attached to an LGA1366 socket (recycle/select faster version of currently available parts), a TB chip would still have to be connected via PCIe lanes, as that's not included in the X79 (nor is USB 3.0, if decided upon as a feature that will be included).
It would however, add SATA III natively.
But if it doesn't include lanes for slots (i.e. 8x for discrete peripheral chips, such as TB only), then there wouldn't be any lanes for PCIe slots, as the LGA1366 parts do not contain a PCIe controller on the die as the LGA2011 parts do.
i7 extreme edition 3.47 wouldnt be a bad "entry level" MP. I know it's not Xeon and that people could just go and buy Dell/HP/etc systems instead but after seeing those Bare Feat tests I really dont know what to think
Sent from my iPhone using Tapatalk
If i7-990X is Entry level Mac Pro, Where are the High-end?i7 extreme edition 3.47 wouldnt be a bad "entry level" MP. I know it's not Xeon and that people could just go and buy Dell/HP/etc systems instead but after seeing those Bare Feat tests I really dont know what to think
Sent from my iPhone using Tapatalk
The Xeons? (Possibly the DP's once they're out? just a theory...)
Sent from my iPhone using Tapatalk
That's correct. I was just answering the question that SATA 6Gb/s and Thunderbolt could be achieved with the current hardware. However, TB is a discrete controller and will most likely be in Mac Pro as well.
i7 extreme edition 3.47 wouldnt be a bad "entry level" MP. I know it's not Xeon and that people could just go and buy Dell/HP/etc systems instead but after seeing those Bare Feat tests I really dont know what to think
That CPU is $999. Entry-level Mac Pro would suddenly cost $3499 if Apple used that CPU.
There is evidence of an 8-core E5-2400 series engineering sample running at 3GHz with 150W TDP. Unlikely to see it in the Mac Pro with that TDP (after not even dual 130W the past 2 years), but shows that there will be high-clocked 8-cores and that Sandy-Bridge EP is the family branding for the LGA 2011 dual processor platform.
Last edited:
Do you have any sources? I would love to give them a read (no, I'm not doubting you, but I always like to see the source).
Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_3 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8J2 Safari/6533.18.5)
I think we've all been looking in the wrong direction to figure out when the new MP's will drop.......
Once the AMD Radeon HD 6980's become obsolete, we'll see the new MP's with their freshly-obsolete video cards!!
(only half-joking)
I think we've all been looking in the wrong direction to figure out when the new MP's will drop.......
Once the AMD Radeon HD 6980's become obsolete, we'll see the new MP's with their freshly-obsolete video cards!!
(only half-joking)
You are absolutely correct!
Wouldn't it be nice to be wrong for a change and the Sandy Bridge E Mac Pro will launch with a HD 7970?
Oops, sorry, I was dreaming there for a second. Now, where was I?
.
Perhaps some more details from ASRock.
http://www.techpowerup.com/148972/ASRock-Giving-PCI-Express-3.0-A-Big-Push.html
http://www.techpowerup.com/148972/ASRock-Giving-PCI-Express-3.0-A-Big-Push.html
Some news on the Sandy Bridge EP/EN side:
http://vr-zone.com/articles/intel-might-yet-get-sandy-bridge-e-out-this-year/12991.html
Looks like not only the CPU has some problems, but the X79 chipset has run into issues, too. If the story is true, we won't see the next Mac Pro anytime soon.
http://vr-zone.com/articles/intel-might-yet-get-sandy-bridge-e-out-this-year/12991.html
Looks like not only the CPU has some problems, but the X79 chipset has run into issues, too. If the story is true, we won't see the next Mac Pro anytime soon.
Thanks for the link.
There's already been mention of a possible delay past Q4 2011, but at least this one gives an indication as to why (seems sound, given the additional complexity of these parts).
I hope they wait rather than pull what they did with the Z68 (late release) to get something shipped. Unfortunately however, such a situation seems feasible (additional features released in a later part).
As per the CPU, there must be some notable stability problems (bugs) if they're waiting for yet another stepping (one area where shipping early/rushing it could be a disaster). I'm also uneasy about the issue with PCIe 3.0 shipping or not as premature hardware may not be fixable via a firmware update (no idea of the current situation of those circuits = passing specs set by PCI-SIG or not).
In theory, if Intel's circuits fully pass these specs, as does any PCIe 3.0 card installed in such a slot, any issues that would arise would be fixable (firmware or drivers). It could still make some users nervous though, particularly in the enterprise market where stability is a very high priority (LGA2011 is primarily aimed at servers and workstations afterall <where Intel will get the majority of their sales figures from on these parts>).
http://www.techpowerup.com/149241/Sandy-Bridge-E-Model-Numbers-Clock-Speeds-Surface.html
Core i7-3960X - 3.30 GHz, 3.90 GHz turbo boost, 6-cores, 15 MB L3 cache
Core i7-3930K - 3.20 GHz, 3.80 GHz turbo boost, 6-cores, 12 MB L3 cache
Core i7-3820 - 3.60 GHz, 3.90 GHz turbo boost, 4-cores, 10 MB L3 cache
The euro pricing that is floating around indicates that the ~$300, ~$600 and ~$1,000 price points will be maintained. Nothing amazing we didn't really know.
Core i7-3960X - 3.30 GHz, 3.90 GHz turbo boost, 6-cores, 15 MB L3 cache
Core i7-3930K - 3.20 GHz, 3.80 GHz turbo boost, 6-cores, 12 MB L3 cache
Core i7-3820 - 3.60 GHz, 3.90 GHz turbo boost, 4-cores, 10 MB L3 cache
The euro pricing that is floating around indicates that the ~$300, ~$600 and ~$1,000 price points will be maintained. Nothing amazing we didn't really know.
Last edited:
4 grand for the top one is insane. I'll keep my hex westmere processors for awhile. Be interesting to see how the benchmarks do with 15mb L3 cache.
Hopefully the pricing will hold due to the die shrink, but we don't know for sure yet (SP versions). Where I'm really anxious however, is with the DP variants. The indications of various issues/delays doesn't particularly fill me with optimism either, as they tend to translate into additional development costs.
So I'll continue to wait for Intel's Quantity Pricing to be announced before breathing a sigh of relief.
http://vr-zone.com/articles/intel-mulling-pci-express-x2-interface/13030.html
I wonder why they did not do this earlier. x1 lanes started getting over saturated ages ago.
I wonder why they did not do this earlier. x1 lanes started getting over saturated ages ago.
Not sure of the credibility, but pretty much what I would have expected. We already know Sandy Bridge clock for clock performance differences over Nehalem/Westmere and the memory bandwidth was obvious when you are looking at triple channel 1333MHz vs. quad 1600MHz.
I don't think it'll be that big of a deal for the majority of the PC enthusiast/gamer market as they have already adopted well to LGA 1155. For the Mac Pro it should mean a decent step up though over what has been available, especially on the lower end. For those who already have 6-core or 12-core systems or looking to get one I don't think there will be too much regret though.