Email or username:

Password:

Forgot your password?
Top-level
Ken Shirriff

The flip-flop is a very important circuit, a more controllable latch. Here's one type of standard-cell flip-flop in the Pentium. With standard cells, the designer doesn't need to worry about this complex wiring; the circuit is pre-designed in the library. 13/21

13 comments
Ken Shirriff replied to Ken

The software for standard cells is very complex. "Automated place and route" software first places the cells in rows to minimize the distances between circuits. Then the wiring is routed between the cells as densely as possible. Both are NP-complete, so heuristics are used. 14/21

Ken Shirriff replied to Ken

The Pentium was a BiCMOS chip, using CMOS like current processors but also fast Bipolar transistors which made some signals 35% faster. BiCMOS doesn't help nowadays for digital circuits, but is commonly used in analog integrated circuits, especially high-speed ones. 15/21

Ken Shirriff replied to Ken

This cell is a BiCMOS buffer, sending a signal at high speed to other parts of the chip. The large boxy transistor in the upper left is a bipolar NPN transistor. Its construction is very different from the long, thin CMOS transistors. Bipolar is fast, but power-hungry. 16/21

Ken Shirriff replied to Ken

Here is a BiCMOS inverter, using two NPN transistors, visible at the top. 17/21

Ken Shirriff replied to Ken

With its high performance, the Pentium processor had a large impact on the computer market. The Pentium even made it into pop culture; you can rock out to the song "It's all about the Pentiums".
youtube.com/watch?v=qpMvS1Q1so 18/21

Ken Shirriff replied to Ken

Intel started using standard cells with automated place and route software for the 386; designing the chip manually would have been too slow. One of the engineers working on this was Pat Gelsinger, who is now Intel's CEO. Maybe standard cells can make you a CEO too! 19/21

Ken Shirriff replied to Ken

It's hard to explain circuits in detail on Twitter, so if you want more details on the standard cells in the Pentium, see my blog post:
righto.com/2024/07/pentium-sta 20/21

Ken Shirriff replied to Ken

My earlier Twitter thread on standard cells in the 386: x.com/kenshirriff/status/17538 21/21

Kroc Camen replied to Ken

@kenshirriff Thank you for posting on Mastodon; I will catch this on the blog. Twitter links are effectively useless unless you are a logged-in Twitter user (try it!) so consider preservation there-of nil

Tom Forsyth replied to Ken

@kenshirriff By the time we dusted off the P54C to use as the heart of Larrabee/Knights/XeonPhi, the whole thing had been converted through about three different design languages. Also, Larrabee was one of the first/largest test projects for automated synthesis at Intel. Some of the SIMD stuff was still done by hand, but the P54C core was all automatic synthesis and doesn't look nearly as pretty. There were lots of teething problems! But it worked in the end.

Ken Shirriff replied to Tom

@TomF Interesting! Can you say more about the different design languages? Also, do you know why the chip was called the P54C? P5 makes sense for the Pentium but P54C seems random.

Tom Forsyth replied to Ken

@kenshirriff There's also P54CS and P54CQS, but I don't know what the letters mean, sorry - way before my time.

The original P54C was done in Intel's in-house design language, which was then retired in the 90s I think. Fortunately the design had been ported to more modern design languages like VHDL or Verilog in the meantime, so then we took that version, ported it yet again to the absolutely newest language that Intel was using.

http :verified: replied to Ken

@kenshirriff Interesting song. And funny how Bill Gates has a cup with "think similar" on his desk.

Go Up