PDP-11: The Second Most Influential Computer Architecture of All Time

[Bill Silver]

For all you computer architecture history nerds like me, I attach a recent conversation I had with Grok that includes a discussion of my top three most influential computer architectures of all time. It starts with a topic of interest to me (and maybe few others) before getting into the three most influential—the curious ones-complement era in the 1950s and early 1960s. It’s not like twos complement was suddenly discovered in the mid 60s, Von Neumann had already laid out a twos complement design in his famous 1945 First Draft, but even master architect Seymore Cray went the other way. Do skip this section if it’s too esoteric for your taste.

Grok’s personality is amusing. It really wants to be my best friend, and effusively praises my insights and influence rankings. Grok does provide some great information, insights, and summaries though. I’m sure human readers will have more sober reviews, and perhaps respond with different rankings. Have at it.

[John Hudak]

Thank you for posting this - It is, for me, very 'refreshing' reading in the sense that all most all the aspects of computer hardware architecture were studied in grad school by reading the original conference (and in some cases, private) papers.  I took more than a few and subsequently taught a hardware architecture course and almost all the conclusions that Grok expresses align with  my deductions from reading a mountain of papers and in some cases, using the machines of the day.  One particular paper had the biggest impact on me was "An Efficient Algorithm for Exploiting Multiple Arithmetic Units" RM Tomosulo aka the 'Tomosulo Algorithm' 

https://www.cs.virginia.edu/~evans/greatworks/tomasulo.pdf

https://ieeexplore.ieee.org/document/5392028

The thing that stuck in my mind as I read the paper (and some of the other papers by Cray) was the engineering elegance of the solution/approach.  

While, from my perspective, the PDP11 was one of, if not most) important/influential machine due to the orthogonality of its ISA, the machine that seemed to 'put it all together' is the IBM RS-6000.  It seemed to have everything to make it a screamer - OOO execution, relatively deep pipelining, register coloring, multiple execution units that were logically designed to be fast and hardware implementation techniques that enhanced the architecture  elegance.  What was stimulating was not only the algorithms themselves but how they functioned in the overall context of the design.  Thanks for my trip back to 'the good old days.."

J

On Sat, May 10, 2025 at 12:47 PM Bill Silver <bsi...@tidewater.net> wrote:

For all you computer architecture history nerds like me, I attach a recent conversation I had with Grok that includes a discussion of my top three most influential computer architectures of all time. It starts with a topic of interest to me (and maybe few others) before getting into the three most influential—the curious ones-complement era in the 1950s and early 1960s. It’s not like twos complement was suddenly discovered in the mid 60s, Von Neumann had already laid out a twos complement design in his famous 1945 First Draft, but even master architect Seymore Cray went the other way. Do skip this section if it’s too esoteric for your taste.

Grok’s personality is amusing. It really wants to be my best friend, and effusively praises my insights and influence rankings. Grok does provide some great information, insights, and summaries though. I’m sure human readers will have more sober reviews, and perhaps respond with different rankings. Have at it.

--
You received this message because you are subscribed to the Google Groups "[PiDP-11]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pidp-11/beb240c6-3e8e-4e88-859a-7874a8dd89f7n%40googlegroups.com.