I5 preview, good gamer, and cheaper too

I'll add this quote :
"Lynnfield, today, looks very good. Enable all of its turbo modes and I believe Intel has another winner on its hands. When Nehalem first launched I complained that the move to a smaller L2 cache kept it from significantly outperforming Penryn in some applications and games. With Lynnfield's turbo modes I believe my complaints will be addressed; need better performance in games? Turbo mode solves that. In many ways, Lynnfield could end up being even more significant than Core i7 ever was.

Penryn was always good, Bloomfield was nice to talk about but Lynnfield may end up being the one you marry."

http://www.anandtech.com/cpuchipse [...] =3570&p=11

Damn, you posted it before me. I just deleted my thread. Try something more "bombastic" with the title, jay, such as "Anand Previews Lynnfield: Crazy Turbo, i7 Performance". I completely missed this thread before checking for the third time.

Anyway, it looks far more interesting than i7 to me. And it looks like the integrated PCIe controller has paid off.

let's please wait till this thing comes out in September before everyone starts beating it like a dead horse.

My take is, unless you want all those app extras, especially ones where tri channel helps, i5 wins hands down, as well as P2.
Other reasons would be, you simply want to brag about owning the "best", or, you game multi card setups, where the i7 does have an advantage over P2, and i5 just cant go theres decently, not with the newer cards coming out, anyways, cause 8x per pci lane wont be enough

^ well in far cry 2 they were certainly different, around half an hour before I read that article I was reading up on the 955 BE and it was getting 14fps above the i7 in the same benchmark with both using the same kit, on that article the 955 got absolutely owned by the i7 and i5.

edit: also it would kinda suck if when you turned on SMT the CPU downclocked 3 multipliers like the article made it out to be.

Negated the Down rating for you due to proper usage of the word "Bombastic".

I'm waiting until my own PC dies before I upgrade. I hardly see the point in upgrading anymore.

Intel isn't shooting itself in the foot. i7 is still marginally superior.

Sweet!

^ I thought i5 had locked base clocks aswell? if so then they wouldn't be able to OC at all.

and isn't 22nm going to be released in 2011?

Riiiiight... AMD with its solid business acumen and gorilla financial strength is definitely in the driver's seat in determining poor Intel's revenues.

I guess in your view, a $400M+ loss last quarter was great, and slashing prices to drive even higher losses is fantastic. Yep, Intel shouldn't even try to compete with crafty ol' AMD - they'll just start giving away them CPUs and drive Intel right out of business!

That has more to do with yield per wafer. 32nm allows Intel to get more i5 CPUs per wafer and thus they can supply for the demand of "main stream" CPUs and make a profit. It also enables Intel to fix any bugs in the 32nm process as i5 is less complicated than i7 to make, once the bugs are fixed with the 32nm process they can move on to the i7 and have a very good yield of i7s per wafer.

It's amazing that we are almost to the point where we are stressing the limit of Si (Silicon). New material will probably be needed by the next gen CPUs.

Well, AT did the cpu, heres a mobo preview

http://www.bit-tech.net/hardware/2 [...] -preview/1

I wonder what they'll do when they reach the atom barrier.

No, I meant if it was possible to create transistors from something THAT small.

Whether the GF spinoff was genius or idiocy, only time will tell, esp. since it was Hector's idea and we all know how those have turned out previously . But what's undeniable is that it was a necessity, due to AMD management's remarkably inept stewardship and inability to correctly predict the future with their ill-timed ATI purchase for which they overpaid by about 9X. If AMD continues to make blunders of the same size, it'll take a supertanker full of petrodollars to keep them afloat. So let's hope AMD is as you say, better managed nowadays, because they are not 'much more sound than they were a while ago' at least financially - they just found a new banker who's willing to gamble on them for a while longer.

AFAIK, GF has said nothing about pushing 32nm on SOI forward, so they'll still be a year behind Intel on CPU process nodes.

And yes I agree AMD has a better lineup across the board, than they did a year ago. But so does Intel. And I still think bottom-line profitability still applies, even with ATIC holding the purse strings.

I don't think they'd be 'semiconductor transistors' since that is a 'bulk' phenomena - orderly lattices with doped impurities, far away from surface effects, so that some valence electrons become unbound and free to wander about, leaving holes wandering about in the other direction - sorta like my neighbor's dog . Instead, these would be more like atomic electron switches, no holes need apply.

As for making them, it'll probably take some sort of nano or molecular self-assembly machinery (nano-bots), since moving atoms around with things like scanning electron microscopes would probably take maybe a minute per switch depending on how fast and how accurate your SEM can place them and how many atoms make a switch - 30 maybe? So for a 1 gigaswitch CPU, that'd be one billion minutes which is about 16.6 million hours or 694,444 days or 1,902 years. So even though your're quite young, you might wanna get started pretty quick .

Hes got AIDS.

Word, Playa.

On the above test, keep in mind, this was being done with a ES, so HT was enabled, which wont be the case on a 2.66 Lynnfield