Sandy Bridge-E coming November?

Orcvader · 9 Sep 2011 at 19:54

Source: http://www.tomshardware.co.uk/sandy-bridge-e-core-i-x79-express-processor,news-36393.html

Intel looks to be pushing the release of the Sandy Bridge-E to mid to late November (Nov. 15 - Nov. 27).

According to information coming out of DonanimHaber, Intel is set to release the new Sandy Bridge-E in late November, between the 15th to 27th, which is ahead of the original December date (or the rumored delay of 2012). Motherboards based on Intel's new X79 express chipset should be released shortly before the release of the processors.

Intel will begin the Sandy Bridge-E platform with the release of the Core i7-3960X Extreme Edition 6-core unlocked processor, Core i7-3930K 6-core unlocked processor, and Core i7-3820 4-core processor. The expected prices for the new processors, at release, look to be set at $999 (Core i7-3960X), $583 (Core i7-3930K) and $294 (Core i7-3820).

Now, the question to ask is... Which CPU will hit the market first: the AMD's FX series processors or the Intel's Sandy Bridge-E processors? Feel free to discuss your thoughts in the comments section below or within our forums at Bulldozer rumors or Ivy Bridge / Sandy Bridge-E: News, Rumors & Reviews.

Would be interesting to see how well this performs against the current 1155 Sandy Bridge's... But which would get released first? Bulldozer or this

?

opethdisciple · 9 Sep 2011 at 20:22

Unfourtunatly unless AMD have something up their sleeve, Bulldozer will never reach it potential, becuase its launch when Intel are RE-LAUNHING their next update.

Its going to have to be the cheap and cheerful alternative as usual from AMD.

jonny512379 · 9 Sep 2011 at 20:29

it does not look like the 2 are in the same price bracket, so i doubt that it matters who releases first.
If price is anything to go on i would expect Sandybridge E to easily outperform the AMD BD, even if BD proves to be amazing (server chips aside).
But if the Intel release date has been brought forwards it does make me wonder why and what do they know that we dont?

DragonQ · 9 Sep 2011 at 21:02

As has been said before, Sandy Bridge-E is completely uninteresting to most people here (and this is an enthusiast forum!). I'm sure the chips will be impressive though.

panyan · 9 Sep 2011 at 22:27

DragonQ said:
As has been said before, Sandy Bridge-E is completely uninteresting to most people here (and this is an enthusiast forum!). I'm sure the chips will be impressive though.

i agree, it is really out of our budget range apart from the very largest builds

i still think the 2500k will be the chip of choice aster SB-E

drunkenmaster · 9 Sep 2011 at 22:43

Things to note on the possible bringing forward, Charlies article about pcie 3 not working, and the distinct lack of pci-e 3 in the "features" list.

Drop pci-e 3, launch, bring a fixed version later. pci-e 3 sounds like it was the reason behind the delay, and it looks like they gave up on it, rather than get it working as dropping it for now saves them a lot of effort. Its pretty much entirely pointless now, though if anyone was going to go quad sli/xfire, then a Sandy-e platform is unlikely to be too expensive, and the one situation pci-e might matter at all(for the vast majority of normal users) would be quad sli/xfire with the new cards from both supposedly supporting it.

Its not really a big deal, it really won't effect many, the thing is sandy-e is going to be about as popular with normal users as quad sli/xfire.

Too expensive, £1000 hexcore chips.... slower in encoding than sandy-non-e due to lack of the hardware in the gpu. Bulldozer wasn't ever going to compete with the platform, its going to be a tiny tiny segment, even smaller than the previous hexcore chips as at least those fit in boards people already had. This way, new mobo, quad channel mem, stupid expensive cpu, little real world performance gain in gaming, encoding. As always, if you work for a living using your comp AND you'd make more money having rendering, calculations, something done more quickly it will be great and well worth the cash.

For gaming, encoding and most "home use" stuff Sandy-e is a non starter. Will there be a fixed version with pci-e both mobo /cpu in a few months, or will they wait for Ivy to introduce it, who knows.

pmc25 · 10 Sep 2011 at 02:51

1155 has been the most broken, dysfunctional chipset or platform that I can ever remember.

I'm not surprised Intel dropped PCIe-3 from SB-E and any other frills. 1155 looks like it will continue to cost them and their mobo partners an arm and a leg with the various issues it has. Given that it's not likely to offer much benefit until 8xxx / 7xx graphics cards anyway, this would seem like a good decision.

drunkenmaster · 10 Sep 2011 at 05:58

pmc25 said:
1155 has been the most broken, dysfunctional chipset or platform that I can ever remember.

I'm not surprised Intel dropped PCIe-3 from SB-E and any other frills. 1155 looks like it will continue to cost them and their mobo partners an arm and a leg with the various issues it has. Given that it's not likely to offer much benefit until 8xxx / 7xx graphics cards anyway, this would seem like a good decision.

Retarded, sure with the "p" chipset unable to use the ondie graphics, and the other unable to overclock, but thats more like standard stupid choices by Intel, the sata bug, its not unheard of, but unusual for Intel but thats about it. The strange thing about the generally held belief they knew about the problem before release, and released anyway rather than delay 3 months........ were they worried about Bulldozer, who knows.

But its a good platform in general, very fast chips and Sandy-e isn't replacing it so not sure what difference any of this makes to socket 1155 stuff, especially as we're set to keep that socket for Ivy and probably up to Haswell before a new midrange socket(assuming Haswell won't come on socket 1155) which is looking more likely to be mid 2013.

Freddie1980 · 10 Sep 2011 at 07:20

I think Intel picked up the tab for the Sandy Bridge chipset failures, about a $1 billion if I recall. Also if your in the chipset business you have to expect things like this from time to time so I doubt their partners lost that much sleep over it considering it was fixed within a couple of months. I do agree though Intel messed up with with the chipsets in general with having 3 seperate standards in H67, P67 and Z68 when all they needed was H67 and Z68.

koooowweeee · 10 Sep 2011 at 10:34

Freddie1980 said:
I think Intel picked up the tab for the Sandy Bridge chipset failures, about a $1 billion if I recall. Also if your in the chipset business you have to expect things like this from time to time so I doubt their partners lost that much sleep over it considering it was fixed within a couple of months. I do agree though Intel messed up with with the chipsets in general with having 3 seperate standards in H67, P67 and Z68 when all they needed was H67 and Z68.

This

why didn't they just bring out the z68 straight away. It must have been in the pipe line. would have stopped a lot of customers being unhappy including me

pmc25 · 10 Sep 2011 at 11:57

drunkenmaster said:
Retarded, sure with the "p" chipset unable to use the ondie graphics, and the other unable to overclock, but thats more like standard stupid choices by Intel, the sata bug, its not unheard of, but unusual for Intel but thats about it. The strange thing about the generally held belief they knew about the problem before release, and released anyway rather than delay 3 months........ were they worried about Bulldozer, who knows.

But its a good platform in general, very fast chips and Sandy-e isn't replacing it so not sure what difference any of this makes to socket 1155 stuff, especially as we're set to keep that socket for Ivy and probably up to Haswell before a new midrange socket(assuming Haswell won't come on socket 1155) which is looking more likely to be mid 2013.

Are you calling me retarded or them retarded?

You do realise that the claimed compatibility of the Z series with Ivy Bridge may be totally undoable in some cases. The claim of PCIe-3 may not work at all for some / all current mobos. Then there's the new thing that's come up about needing a return to factory (not return to service centre) BIOS wipe for boards that should theoretically support Ivy Bridge.

The CPUs as distinct from the platform have been excellent and hitch free. 1155 has been and continues to be a total joke.

Re: Intel picking up the tab for mobo vendors, it could only ever be part of the cost. If they'd picked up all the costs associated, it would have been a hell of a lot more than $1bn that they had to put aside.

N00bbuilder79 · 10 Sep 2011 at 13:35

Despite all the negative comments I will be building my new platform around Sandybridge-E, just hope its not a step back from Sandybridge in terms of performance.

drunkenmaster · 10 Sep 2011 at 14:37

N00bbuilder79 said:
Despite all the negative comments I will be building my new platform around Sandybridge-E, just hope its not a step back from Sandybridge in terms of performance.

It most likely will be in anything that can use quicksync, but then a lot of people with discrete cards can't use quicksync anyway. Honestly I forget how it works at the moment, you need a Z or H chipset and at least one screen has to be connected via the mobo's video out?

So you lose that from Sandy to Sandy-e, which is very fast IF you use it.

What reason do you have to go with Sandy -e though, mobo's with a probable 20% markup over z68 chipsets for the quad channel memory, if you have say 2x4gb memory now, then you'd need to buy another set(though considering prices thats hardly a dramatic cost right now), £1000 for a hexcore cpu thats if lucky 50% faster than a chip that costs £240.

Theres no denying it will be the fastest chip around, and I don't know your use, gaming is a pretty good realistic load level for most people on the forums, not everyone. In terms of gaming, Sandy-e won't be faster than a Sandy, a Bulldozer, a old i7, an old i5, and in plenty of situations even a Phenom 2.

I mean speed/value/long term thinking, having a 1155 setup with a 2500k, and then buying an Ivy IF it does have a decent IPC boost, or a Haswell in a couple years , octo core, £250, maybe a 100% boost in performance, is it really worth paying for an ultra high end system that is going to be such poor value for money?

If Haswell was actually on time in Q4 2012 then within a year of Sandy-e you'd have faster chips for 25% of the cost..... I can't see that being on time though with Sandy-e/Ivy both delayed by 6 months.

drunkenmaster · 10 Sep 2011 at 14:38

pmc25 said:
Are you calling me retarded or them retarded?

You do realise that the claimed compatibility of the Z series with Ivy Bridge may be totally undoable in some cases. The claim of PCIe-3 may not work at all for some / all current mobos. Then there's the new thing that's come up about needing a return to factory (not return to service centre) BIOS wipe for boards that should theoretically support Ivy Bridge.

The CPUs as distinct from the platform have been excellent and hitch free. 1155 has been and continues to be a total joke.

Re: Intel picking up the tab for mobo vendors, it could only ever be part of the cost. If they'd picked up all the costs associated, it would have been a hell of a lot more than $1bn that they had to put aside.

Them, calling them retarded, and still most 1155 boards issues are down to stupidity on behalf of the manufacturers and making a board a year + before products are available, before the spec's are final, and trying to support them, it rarely works well.

pmc25 · 10 Sep 2011 at 15:33

The only reason I can think of to go Sandy Bridge-E is if you want an ultra-expensive but consumer-grade dual socket workstation with gaming crossover. Workstation alone an Interlagos chip (or two) would presumably be hugely more cost-effective (even Xeon?).

N00bbuilder79 · 11 Sep 2011 at 11:37

Mine will be replacing my Skt775 Q9650. The computer will be used for lots of video editing and photo editing with a bit of gaming thrown in for good measure.

banja · 11 Sep 2011 at 13:31

pmc25 said:
You do realise that the claimed compatibility of the Z series with Ivy Bridge may be totally undoable in some cases. The claim of PCIe-3 may not work at all for some / all current mobos. Then there's the new thing that's come up about needing a return to factory (not return to service centre) BIOS wipe for boards that should theoretically support Ivy Bridge.

Link please?

Is this likely with the current Z68 boards that are available? ie the Asrock Z68 Extreme4 Intel Z68 or MSI Z68A-GD80-G3??

My plan was to buy one of these boards for a later Ivy Bridge upgrade, but I certainly do not want to have to return the board for a bios update.

Surveyor · 11 Sep 2011 at 13:38

banja said:
Link please?

Upgrading to Ivy Bridge to be blocked by UEFI changes?

The upgrade path Too Ivy Bridge might be blocked by changes to UEFI

FallshirmjAEger · 11 Sep 2011 at 13:39

banja said:
Link please?

Is this likely with the current Z68 boards that are available? ie the Asrock Z68 Extreme4 Intel Z68 or MSI Z68A-GD80-G3??

My plan was to buy one of these boards for a later Ivy Bridge upgrade, but I certainly do not want to have to return the board for a bios update.

In theory, couldn't you just request a new bios chip any way or does the new bios not work like that?

banja · 11 Sep 2011 at 13:43

That's pretty poor news. Surely lots of people will have bought these boards specifically because they boast this upgrade path. The cost reflects it also. Glad I hadn't commited yet.

Are there likely to be newer versions which take this issue into account and therefore are compatible?