Intel's Sandy Bridge Getting Accelerated Launch

They'd best not even think of doing anything even more retarded with their naming scheme!

The name keeps reminding me of Sandy Balls

I thinking i9 for the naming scheme as with the rumours and if this chip is as disgusting as I think it will be. My phenom 1055T 4.0Ghz setup will be sold to the highest bidder and I will join the blue team, I love how the industry is at the moment earlier this year I hated Nvidia with their overpriced chip rebrands, hot products and late turn to market however I ended up getting a Palit GTX 460 Sonic Platinum as soon as the benchies were released. Also Intel we all know at the moment they make the better chips and I read somewhere the AMD bulldozer will be a serious let down in comparison to Intel's offering.
You know how it is when building a PC, you buy a product and then a couple of months down the line something cheaper and more powerful comes out. Which roadmap are you guys taking Intel or AMD and why???

@ Herr_Koos your right the comment is speculative at best, but lets not forget Intel are easily, one generation ahead of AMD at the moment and in order for the bulldozer to be anything worthy of consideration for Joe the PC builder, AMD must make a leap of FX proportions which is unlikely. Seeing as AMD have outlined that they will be investing more time into chips which combine GPU & CPU Llano etc, of which Bulldozer is not.

I really hope for the sake of the customer that their offering is competitive but considering the fact that it will most likely be released on the AM3+ platform which will be backward compatible with AM3 may mean no triple channel memory controller. Either way this is all speculation and I would love to be proved wrong.

Three things I wish to point out...

1) The performance difference between the i5-750 and i7-920 is very slight outside of heavy multitasking; with the i5 you're giving up a triple channel memory controller plus Hyperthreading, and only gaining a more aggressive Turbo implementation, however it doesn't make too much difference in the end. So, HT isn't as massive as Intel would like people to believe, nor is a triple-channel memory controller. Both are very nice to have, but as we've seen with HT, it can actually negatively impact performance in some areas. This could be down to the poorly threaded nature of a lot of today's software, plus maybe triple-channel was overkill, at least for now.
2) Bulldozer's implementation of SMT is more aggressive than Nehalem's because the only thing they haven't duplicated is the FP resources - each module is literally two cores. Also, software is getting more and more threaded, so this should validate their decision to implement SMT. At the same time though, you'd expect this to benefit Intel's HT-based CPUs as well - we'll see.
3) AMD is developing the AM3 R2 based around the AMD 990 chipset. There's very little information about it even 7 months after it was first talked about, except that the assumption that you'll be able to use Bulldozer on AM3, but it'll run better on R2. I expect AMD to let us know about compatability concerns in due course, but it would help to know now and not later.

In the end though, benchmarks will tell all.

The countdown has officialy started then.
Can't wait now ...

@ Silverblue

Thanks for the heads up on the SMT implementation on Bulldozer, I came across this fascinating article backing up your claims of aggressive SMT implementation.

http://tech.icrontic.com/articles/ [...] cat-llano/

Good article, though it's got me thinking about how Intel plans to improve HyperThreading.