Judging by the information we've seen, Intel is having several issues with the new platform, ranging from storage, to PCI Express and apparently even CPU revisions. It's pretty much bad news all around.

On top of all the chipset issues, Intel also seems to have run into a snag with the CPUs, as the C-0 stepping is meant to start shipping to its partners for qualification testing within the next two to four weeks. Intel is apparently waiting for revision C-1 until it feels confident with Sandy Bridge-E.

- Core i7-3960X : 6 cœurs à 3.3 GHz (Turbo à 3.9 GHz), cache de 15 Mo
- Core i7-3930K : 6 cœurs à 3.2 GHz (Turbo à 3.8 GHz), cache de 12 Mo
- Core i7-3820 : 4 cœurs à 3.6 GHz (Turbo à 3.9 GHz), cache de 10 Mo
 
Les processeurs X et K devraient être totalement débloqués au niveau du multiplicateur pour l'overclocking, alors que le seul quatre cœurs, l'i7-3820, ne devrait pas être capable d'aller au-delà du x39 utilisé en mode Turbo. Un bridgage qui serait toutefois moins limitant que sur LGA 1155 puisque la plate-forme LGA 2011 devrait normalement permettre l'overclocking par le bus.

In response to a question regarding Romley - the server platform revolving around Sandy Bridge-E, Mr. Otellini said that the schedule remains unchanged, thus providing a strong response to alleged problems. The product will be in volume production this year. The fact that it isn't already in productions might explain some of the rumors, that Intel needed to roll out another stepping of the chip. So Intel remains committed to deliver this product in time, while not specifying a launch date, as server customers are usually conservative about such things. The CEO also remarks, that its a very exciting product and the performance is quite good.

Bon on sort les grosses pincettes pour cette news, mais le site chinois MyDrivers affirme avoir eu plus de renseignements au sujet des Sandy Bridge E et de la plateforme LGA2011. Selon leur "source anonyme", SBE et donc le LGA2011 dans sa globalité devrait bel et bien être lancé en 2011, à une date non encore arrêtée précisément mais qu'Intel aurait décidé de placer entre les semaines 40 et 49. Cela correspondrait donc à un lancement situé entre le 3 octobre et le 11 décembre.
 
Ils indiquent également avoir pu avoir quelques indiscrétions quant au prix officiel du Core i7-3930K, qui pour rappel sera l'hexacore "grand public" alors que le i7-3960X sera de son côté la version Extreme Edition hors de prix (sans doute 999$ comme ses prédécesseurs). Le i7-3930K serait ainsi à 560$ soit 23$ moins cher que le Core i7-980, qui a récemment prix la place de l'i7-970 à 583$. Pour ceux qui auraient besoin d'un petit rappel des Sandy Bridge E attendus et de leurs caractéristiques, Thibaut vous expliquait tout cela en détail la semaine dernière.

We don't like to be the bringers of bad news, but it's come to our attention that Intel has decided to change its high-end consumer Waimea Bay platform one more time before it launches. The only good news is that we've managed to pin-point which month the platform is expected to launch and that is November and there are several reasons behind this choice.
 
Intel's Waimea Bay platform is starting to become something of an epic saga by now with changed launch plans, chipset changes and what not. We're still not 100 percent sure that Intel won't change anything else in time for the launch, but we can only go on the information we have at hand. As for the November launch, the key thing here for Intel is not to hit the Christmas holiday season, but instead it's to get in with the big system integrators before they decide on next year's system spec. The big boys tend to only change their platforms once or twice a year and Intel has to be in on time when these companies are evaluating what they're going to be selling in 2012, miss this deadline and you have a lot less sales.
 
We're also hearing that Intel is worried about AMD's FX processor refresh which should take place early next year, maybe not in terms of raw performance, but the company needs to have its platform out there to show that the company has yet another high-end platform ready to go. Make what you want of this, but to manage to meet the November deadline, Intel has cut further features from its X79 chipset.
 
If you remember the article from a couple of weeks ago which had the same diagram as the one above, we told you that Intel had gone from the server level Patsburg-D to the Patsburg-B level chipset for the X79. Well, now we're talking Patsburg-A, i.e. no SAS support at all. In other words, we're looking at a chipset very similar to Intel's current P67 chipset, as it will only sport two SATA 6Gbps ports and four SATA 3Gbps. This is starting to make it look less and less like that Intel will simply be able to "enable" these features by a software upgrade in the future and instead we can either look forward to a chipset revision or a new, more expensive chipset for the Waimea Bay platform.
 
And it doesn't stop there, we have more bad news with regards to the X79 chipset, but that's another story…

If you're a regular reader you might've read our article from a few days ago about Intel bringing back adjustable base clock for Ivy Bridge, well, we know have the specifics and they're less complex than you'd think. However, what goes hand in hand is the fact that Intel will have base clock limitations for Sandy Bridge-E, just a few more options on offer than for Ivy Bridge.
 
We were disappointed to learn that Intel will only offer a single base clock option for Ivy Bridge beyond the 100MHz of Sandy Bridge, name 133MHz. For those of you that hoped to be able to overclock Ivy Bridge to your heart's content, well, it ain't gonna happen. Apparently Intel is worried that Ivy Bridge will compete with its LGA-2011 platform if it was to add more base clock options and as such things aren't set to improve much over Sandy Bridge.
 
As for Sandy Bridge-E, well it shares one component with the upcoming Ivy Bridge platform and you can see it quite clearly in the diagram above which is from a YouTube video that we've embedded at the bottom. The little chip marked CK505 at the bottom is the clock generator and this is what controls the base clock. As you can see there's a mention of a PEG/DMI ratio mechanism and this is what allows several different base clocks.
 
Unlike Ivy Bridge, Intel has enabled several options for Sandy Bridge-E and we're hearing that we can expect base clock frequencies of 100, 125, 166, 200 and 250MHz if memory serves. This might still change before the launch of the Waimea Bay platform as Intel tweak things to work as well as possible for overclockers. Still, with four rather large jumps between each of the base clocks we have a feeling that the overclocking community isn't going to be impressed. We might see the X58 platform live on for quite some time next year, despite the fact that many are feeling like it's a platform of the past already.