Feeling the strain
But before it did, there was work to be done on process technology, and the introduction of the first of the three major post-scaling technologies that Mark Bohr talked about: strained silicon.
In a highly simplified nutshell, strained silicon involves the material being stretched – or strained – in such a way as to pull the individual silicon atoms apart from one another. Doing so frees up the electrons and holes in the material, increasing their mobility substantially, thus allowing for lower-power transistor designs.
Although strained silicon had been under investigation at MIT and elsewhere, the early techniques were was biaxial – that is, the entire silicon lattice was stretched. Intel's breakthrough was the development of uniaxial stretching. Biaxial straining was good for nMOS but bad for pMOS, both of which need to be balanced for good transistor performance.
Biaxial straining also had problems with source drain and defects, Bohr told us – "not a very manufacturable technology". The uniaxial approach, however, could be applied "just to the pMOS device," Bohr said, "and it didn't have any significant yield issues, so it turned out to be both a high-performance solution and a good manufacturing solution."
But back to the departure and then the return of P6.
The follow-on architecture to P6 was NetBurst, and it was not exactly Intel's finest hour. By the time P6 had evolved into the Pentium III, its pipeline was just 10 stages long; NetBurst doubled that to 20 stages in the Willamette Pentium 4 in 2000, and increased that "Hyper Pipelined Technology" to 31 stages in the Prescott Pentium 4 in 2004 – which, by the way, was the first processor to use Bohr's 90nm strained silicon process technology.
According to Pawlowski, the reason for the deeper pipeline was "frequency, frequency, frequency". In a bit – well, more than a bit – of an oversimplification, deep pipelines require higher frequencies to achieve the same performance as architectures with shorter pipelines.
Next page: Seduced by the 'Megahertz Myth'
What's that in brontosauri?
Succeeded despit etechnology not because of it
The story of teh scusess of intel microprocesors is that commercial and not technical factors dominate.
The 8086 was very much inferior to the 68K and the 16032 it was probably on a par with the Z8000. I rember Intel trying to sell to me at that time and they always emphasised price, the agreement with AMD that gave guarantee of supply and assurance on pricing, and support. They never tried to sell on performance or technical aspects because it was well behind Motorola.
The PC then came out and things changed very rapidly. Intel broke the AMD arrangement and the price of the first non-agreement part the 80287 sky rocketed. Technically intel parts were still very much second best but they sold fantastic numbers o fparts. The 80286 retained the awkward segmented architecture extended withprotected mode performance was still very poor. The 386 finally had a sensible memory architecture but still had the nasty special purpose registers and complicate dinstruction set and performacnce was still very poor compard to other micros. It was probably not until the pentium that Intel gained parity with other microprocessors.
None of these technical things mattered, one design decision by IBM made Intel the dominant microprocessor company with massive reources despite not because of their technical design.
Ahh, those were the days...
...when bytes were were real bytes, Motherboards could be fixed with a soldering iron, "intellectual property" meant you'd paid off your Encyclopedia Britannia, and 'programming' meant hand coding raw MC. Maybe assembler if hung over.
And yes, counting every damn clock cycle.
God, I feel old.... <sniff.>
"After the 8086/8 came the 80286..."
No, it didn't. After the 8086/8 came the 80186/8, which was then followed by the 80286.
I remember coding in 80186 assembly on my dad's Tandy 2000...
A few corrections more...
1. IT was not built on the Intel 4004 or its successors. The information technology industry started in the 1950s with pioneering data processing applications leveraging emerging computing technology. Remember LEO, and the IBM 1401? They were certainly information technology systems. You'd have to use a pretty discrete and tortured definition of IT to claim the 4004 was its first building brick.
2. You use the phrase 'first processor' to describe the 4004. Here comes more pedantry... This is not true either. It was the first commodity, commercially available microprocessor -- which is to say an IC with all the traditional components of a CPU. Computer processors in modern sense date back to at least 1949 and EDSAC. The Digital PDP-11, a direct contemporary of the 4004, certainly has a processor, as did all it's ancestors. What it didn't have was a single chip 'microprocessor.'