I don't think too many people are on the edge of their seats waiting for the 45nm phenoms. They are waiting for AMD's response to Nehalem. AMD needs to respond, and respond quickly. The 45nm Phenoms are not the answer.

No one's aware of this 45nm Phenom preview that shows a huge drop in power draw? Sure, 2.3GHz isn't exactly pushing the envelope, but that goes for the 65nm model, too. It's at least promising for the more affordable models, and I guess it gives them (a lot) more breathing room to push the clocks at the high end.

http://en.hardspell.com/doc/showcont.asp?news_id=3858&pageid=3150

AMD cost cutting and mistakes has lost it almost 12months of competitive advantage. Less competitive products means you can't sell as much or sell for less. Less revenue means more losses. More losses means more cuts. More cuts means slower progresses and fall behind even further and leads to a downward spiral.

> Don't expect any miracles from a power consumption
> perspective, though: X-bit labs projects thermal
> envelopes of 125W, the same as that of the existing
> Phenom X4 9850.

Is it too much to hope that 45nm Phenom is keeping the 125W power envelope because the clock speeds will be super high?

This is Xbit Labs saying they think the TDP will be 125w, not AMD. They're guessing. AMD has, however, said in the past that the K10.5 will have higher IPC efficiency due to a few tweaks. Basically they said they'll keep more of the CPU busy per clock and it has something to do with the extra L3 cache. I read this on the shortbread earlier in the year like in Jan or Feb.

Also, those 'new Core i7" chips will need to come with a brand-spanking-new $200+ dollar motherboard.

Your new Phenom will be socket-compatible with the old one.

Price is a consideration for some of us.

I think 45nm Phenom must have changes like ATi did to 3800 series to give HD 4800 series which is a clear winner (at the beginning at least) to number-superior GTX2x0 if they have performance and price concern in them.

The roll out of the new 45nm processor is really slow, so slow that if 45nm parts have no dramatic efficiency change to get pass 45nm Q9x50 series from Intel, i don't think they will stand a chance of surviving and cannot let hopes of gaining CPU share like they did to that of nVidia's GPU.

But, still, Tomshardware has (link: http://www.tomshardware.com/news/phenom-amd-4ghz,6249.html) a good news that the current 9950 breaks 4GHz with LN2. No applicability to normal users, but what a relief to us the AMDers. They stick the HT to around 2GHz, but the (so called) FSB is up to aorund 250. HT has really been a problem. Hopefully 45nm parts will have HT v3.1 and see if the situation changes.

Its good news if AMD doesn't have any 45nm CPU in 2008, it means that they are selling them all to the server market for a better profit margin.

At most I'd expect a couple of FX CPUs in 2008.

Any news about Phenom FX ? There are rumors out, which says Phenom FX 80 will clock 4 GHZ and FX 82 will clock 4.4 GHZ... O-o

Link: http://www.reviewage.co.uk/images/stories/phenom-cpuz.jpg

This is just guessing... Anyway, will be faster than current phenoms, and at least less power hungry than today processors, plus having a mature platform... will be a good all terrain platform with the current chipsets and video chips from amd. There is nothing to to worry about i think. (Just remember the times of a power hungry K6 that didnt perform any better than P2, amd survived...)

That said, recent rumors suggest the first desktop Core i7s will cost upward of $284, so AMD could undercut Intel and see some success at lower, higher-volume price points.

Er, Intel has the luxury to price the i7 (Nehalem) processor family as they choose. You can guarantee that they will price at competitive levels that ensure little or no per-unit production profit for AMD - considering that AMD's initial 45nm per-unit production costs are likely to be far higher than Intel's i7 on a now-mature 45nm process. Thus no money for AMD to pay off the engineering development, production start-up and marketing costs incurred bringing their 45nm processors to market, let alone money to invest in follow-on products. The Intel production machine can slowly throttle AMD to death, since the AMD 45nm CPUs bring zero tangible performance benefit over the Nehalem family, which will be in full production by the time AMD's 45nm offerings hit the market. The delay to January 2009 may be fatal to AMD. Seems as if IBM's historic inability to speedily transition a silicon process from prototype/low-volume to high-volume, high-yield mass production may have hit AMD squarely between the eyes. A great silicon partner -- NOT !!

Not good enough, AMD. I guess it's yet another year of not being competitive, not even at the mid-high end. Shame. Is 45nm going to be another failure like their 65nm?

To be fair, if they're just delaying the mainstream 45nm chips to allow for large shipments of server CPUs, then that's not a problem - the mainstream for the most part can survive on the current offerings.

And if there is a 45nm Phenom X4 FX chip at higher clock speeds, that would be good.

I just don't feel comfortable thinking about possibilities like that when it comes to AMD these days.

They will start producing them in the second half of this year, it will be December 31st is all ;)

Bad news for AMD. They lost the CPU race a long time back. Their only ace was the "platform" of their excllent 780G/790GX chipset with the low power CPUs which were more than adequate for nearly everyone that didn't game.

Disappointing, hopefully those 45nm phenoms can atleast hit 3.6ghz on air. I guess AMD needs to focus on their next architecture "bulldozer" since phenom didn't triumph core2s as many expected

"In an official announcement, AMD said Bulldozer will be its first substantially new CPU core since the original Athlon 64 processor of 2003. And it'll be the first ground up architectural redesign since the K7 Athlon of 1999."

Goddamnit, AMD.