Intel can't do it right, and the other can do it too!
The other will face the same problems.

They are going to 65nm, like ibm is also going, because of the issues they are having. They believe that, maybe at 65nm these problems will all disapear.

Why do you think AMD doesnt have 90nm ready. How many years they have taken to pass the 2.2Ghz mark. Or intel the 3.2Ghz.
They have passed it but only haded 200Mhz to it recently.

Why is AMD redesigning they core with 512Kb cache and at 130nm yet. To get more yield's of course! 1MB is too much for the processor at 130nm.

AMD processor's perform very good even with little/less cache, they only have those size for marketing proposes. Intel have 1MB so they have it too. If Intel launch a 2MB, we will have to see what they will do.

AMD and IBM already HAVE 90 nano chips that DON'T OVERHEAT, that DO scale properly, and that have way higher IPC than even Pentium M, the only chip InHell has that is worth anything. AMD has chips available in various cache sizes as is best for performance of a given chip. As InHell showed more cache doesn't do much for their chips, so more is NOT always better. The PROPER size cache(s) for a chip architecture is what you want. InHell is in deep sh*t and they are on a media campaign to mislead consumers while they attempt to scrap their defective 90 nano process and pull some new magical 65 nano process out of their azzes...

DON'T hold your breath cause InHell has a LOT of internal problems to overcome, not the least of which is trying to copy the Opteron/A64 designs.

1st - InHell (like you call it) does not copy design's from AMD, but ideas, that's different (Microsoft for example is another), because both processor are way too different. Speedstep for example it's the idea that was copied not the tech from AMD Powernow! X86-64 I don’t know if was licensed, but if isn’t proprietary, who will use it? Intel comes out with their own 64 bit implementation and all those AMD 64bit transistors would go to the garbage, right. Remember 3Dnow vs SSE? Another copy, but who have won?

2st - You are wrong, AMD doesn't have 90 nano chips, and will not have it soon, believe me. You gone see is a new processor (Socket 939) still at 130nm, the difference in fact it's only the capability of working with non registed memory (“normal” DDR ram), and the already announced 1000 MHz HT speed link.

3rd - Have you seen the cooling system (case) of Apple for those processors? How many fans the case have? Notebook’s based on that processor?

4th - Are you sure they scale properly? Remember AMD transaction from .18 to .13, they only had 66Mhz of speed over the .18 versions, and then have to had more metal layers Tbread A and Tbread B? Why do you think Intel have added more metal layers to the new Prescott core?

5th - AMD has showed to us that the cache on their processor doesn't do too much, compare if you can the duron 1.8 vs barton 1.8 with the same ram and fsb and you will see that the difference is very little, and tbread vs barton even less difference. Look AMD64 512kb vs AMD64 1MB there are test's on the net find them.

6th - Remember Intel Nortwood 512KB vs Willamette 256KB? One think I remember amd XP was the best vs Willamette and when Nortwood have come out, every one start questioning AMD PR ratings. And every one was expecting AMD .13 manufacturing process would "kill" intel. Intel takes DDR memory (another amd copied idea), 533 and 800Mhz fsb and became the "leader" again.

7th - Why Intel EE is fast? More cache. Intel processors need more cache, that's a point, more than AMD at least for me.

8th - Intel have many problem new socket, ...., but amd too, ok it's only one less pin on the socket but it's new any way....

We will have to wait.