Scalar CPU clock rates were maybe 25MHz in 1989.
OOO speculative superscalar hyperthreaded multicore CPUs now are 2.5GHz.
Knuth's Challenge: Analyze everything your computer does in one second
11–20 of 52 posts
Re: Knuth's Challenge: Analyze everything your computer does in one second
#12The code running on the CPU isn't the only thing the computer is doing in that one second, the X86{,_64} opcodes we could capture aren't necessarily exactly what the CPU is doing, and code being run by extra controllers and processors isn't likely to be accessible to anyone but the manufacturer.
In 1989, we'd've also been looking at code running on a single core with a single-task or cooperative-multitasking OS (for most home computers, anyhow), with simpler hardware that an individual could completely understand, and it would run at a speed where analyzing a second of output wouldn't be completely beyond the pale.
I've analyzed CPU logs from DOS-era programs and NES games. I certainly haven't analyzed a full second of the code's execution; I'm usually focused on understanding some particular set of operations.
Re: Knuth's Challenge: Analyze everything your computer does in one second
#13Fascinating question, reminds me of this one that made HN almost two years ago: “What happens when you type Google.com into your browser and press enter?”
https://github.com/alex/what-happens-when
Would love to see Knuth's Challenge setup in a repo for collaboration.
HN Discussion on the above link:
Re: Knuth's Challenge: Analyze everything your computer does in one second
#14Fascinating question, reminds me of this one that made HN almost two years ago: “What happens when you type Google.com into your browser and press enter?” https://github.com/alex/what-happens-when Would love to see Knuth's Challenge setup in a repo for collaboration. HN Discussion on the above link: https://news.ycombinator.com/item?id=8902105
You've just made my day. I've had some ideas brewing on this topic for years. It is now at the top of my reading list....along with the other suggestions in this topic.
> Would love to see Knuth's Challenge setup in a repo for collaboration.
Go for it!
Re: Knuth's Challenge: Analyze everything your computer does in one second
#15The code running on the CPU isn't the only thing the computer is doing in that one second, the X86{,_64} opcodes we could capture aren't necessarily exactly what the CPU is doing, and code being run by extra controllers and processors isn't likely to be accessible to anyone but the manufacturer. In 1989, we'd've also been looking at code running on a single core with a single-task or cooperative-multitasking OS (for…
> simpler hardware that an individual could completely understand
Our hardware now is so much more complex, what has it gained us? The quick answer is performance, but is it true? What about correctness? Hard to prove either way, but my guess is we've gained a little bit on performance and lost on correctness.
Re: Knuth's Challenge: Analyze everything your computer does in one second
#16The code running on the CPU isn't the only thing the computer is doing in that one second, the X86{,_64} opcodes we could capture aren't necessarily exactly what the CPU is doing, and code being run by extra controllers and processors isn't likely to be accessible to anyone but the manufacturer. In 1989, we'd've also been looking at code running on a single core with a single-task or cooperative-multitasking OS (for…
> simpler hardware that an individual could completely understand Our hardware now is so much more complex, what has it gained us? The quick answer is performance, but is it true? What about correctness? Hard to prove either way, but my guess is we've gained a little bit on performance and lost on correctness.
GPUs are definitely in the performance at the expense of correctness bucket. It's common for errors to occur in the LSB which isn't a big deal for colors, but might be a big deal for data.
Re: Knuth's Challenge: Analyze everything your computer does in one second
#17The code running on the CPU isn't the only thing the computer is doing in that one second, the X86{,_64} opcodes we could capture aren't necessarily exactly what the CPU is doing, and code being run by extra controllers and processors isn't likely to be accessible to anyone but the manufacturer. In 1989, we'd've also been looking at code running on a single core with a single-task or cooperative-multitasking OS (for…
> simpler hardware that an individual could completely understand Our hardware now is so much more complex, what has it gained us? The quick answer is performance, but is it true? What about correctness? Hard to prove either way, but my guess is we've gained a little bit on performance and lost on correctness.
The reason that modern cores are in the billions of transistors is to get performance with correctness. Tons of speculative work that can be rolled back if some invariants don't hold to be true.
Re: Knuth's Challenge: Analyze everything your computer does in one second
#18Earlier quoted context omitted.
> simpler hardware that an individual could completely understand Our hardware now is so much more complex, what has it gained us? The quick answer is performance, but is it true? What about correctness? Hard to prove either way, but my guess is we've gained a little bit on performance and lost on correctness.
GPUs are definitely in the performance at the expense of correctness bucket. It's common for errors to occur in the LSB which isn't a big deal for colors, but might be a big deal for data.
Aren't the last few generations of GPUs fully IEEE compliant?
Re: Knuth's Challenge: Analyze everything your computer does in one second
#19The code running on the CPU isn't the only thing the computer is doing in that one second, the X86{,_64} opcodes we could capture aren't necessarily exactly what the CPU is doing, and code being run by extra controllers and processors isn't likely to be accessible to anyone but the manufacturer. In 1989, we'd've also been looking at code running on a single core with a single-task or cooperative-multitasking OS (for…
> simpler hardware that an individual could completely understand Our hardware now is so much more complex, what has it gained us? The quick answer is performance, but is it true? What about correctness? Hard to prove either way, but my guess is we've gained a little bit on performance and lost on correctness.
My computer certainly does more than the one 27 years ago did, in the senses of operations it performs in a second, capacity of data storage, increased intercommunication options, increased selection of devices that I can interface with it, and so on. Some of these are just a change in magnitude, and the changes in the software are just as important as changes in the hardware.
There are more parts of the computer (again, both hardware and software) that are undocumented. Taken as a whole, the system is more capable, but closed hardware and software makes me wonder about capabilities in my computer that serve someone that isn't me.
Re: Knuth's Challenge: Analyze everything your computer does in one second
#20Reminds me of how Stanisław Lem went virtual and wrote fictitious reviews of a non-existent books that were far too vast to actually exist in the real world, including one called "One Human Minute" [1]:
"One Minute", for a faux book by J. Johnson and S. Johnson: One human minute, Moon Publishers, London - Mare Imbrium - New York 1985. The book is alleged to be a collection of statistical tables, a compilation that includes everything that happens to human life on the planet within any given 60 second period.
Here are some real reviews of a real book of fictitious reviews of fictitious books [2]:
KristenR rated it really liked it. Shelves: science-fiction, short-stories, male-author.
This volume had 3 essays, each with an interesting concept.
One Human Minute: Lem has styled this piece as a book review...of a book that hasn't been written. One Human Minute is apparently a Guinness Book of World Records that is completely mundane, yet also amped up on steroids. Imagine a book that is full of tables upon tables and graphs and charts about everything that happens on earth per minute. How many babies are born, how many people get struck by lightning, how many people are tortured by electricity, how many orgasms are achieved per minute...
Definitely a philosophical piece, but seemed to be musing about the depravity of the human race. I'm not sure if I missed the point.
The Upside-Down Revolution: The evolution of military and warfare...written under the premise that the author has a history book from the future and publishes it in the present as science fiction. I lost interest partway through this one.
The World as Cataclysm: I have a fascination with astrophysics. I am fully aware that the bulk of it goes over my head and I have near zero retention, but that doesn't stop me from reading/watching anything on the subject that is remotely geared towards the layman. Simply Fascinating. This piece goes into the probabilities of extraterrestrial life. I don't know what Stanislaw Lem's qualifications are, but as I was reading this I was nodding...uh huh, that makes sense...hmmm, I sense a little research project on Lem.
Rich Meyer rated it: really liked it. Shelves: read-in-2014.
An interesting trio of science fiction-tinged essays by one of the great science fiction writers. The title refers to one about a book that covers what happens on the planet every minute, and how the book is updated and computerized until it becomes a power unto itself. The other essays follow the search for intelligent life and the chances for finding it, and a look at the history of warfare from the point-of-view of a book from 2150, and manages to make some pretty accurate predictions.
[1] https://en.wikipedia.org/wiki/Stanis%C5%82aw_Lem%27s_fictiti...
[2] http://www.goodreads.com/book/show/28771.One_Human_Minute