I suppose it's more surreal than comparisons I made back then because there's very little difference, really, in the operating system and popular languages.
I did some back of the napkin math and realized that was roughly equivalent to my Zen 2 box.
So of course my failure to replicate the author's success was not the fault of my hardware.
In the end it seems GA/GP was an "overoptimizing" for very non-convex/non-linear search spaces (much more than real problems) and that we're better off with NN/SGD representations of problems
I have had a lot of fun this year building a neural network from scratch as well as genetic programming. For real tasks, it does seem that the former has won out.
I have a layman's suspicion that GP will find itself useful in some way and have a resurgence in the future, but it might just be naïve wishful thinking.
I once worked for a server manufacturer that did a setup with hundreds of 1U size, dual socket Pentium 3, 1.0 GHz CPUs. In the era when the only way to have more than one core was a dual socket motherboard.
[1] https://www.gene-expression-programming.com [2] https://en.wikipedia.org/wiki/Gene_expression_programming [3] https://ccc.inaoep.mx/archivos/CCC-17-009.pdf
(famously used in the Cray 2)
Similarly expect similar prices since nothing there really changed either. Yields are a very known quantity at this point.
But the problem isn't so much with the Core Count but the TDP per Core. Imagine it is 3W per Core, 128 Core gets to 384W. And that is excluding the IOD.
Hopefully this changes with Zen 3 looking like an awesome consumer chip.
I'm pretty sure double the core count at 90% of the speed will get you better performance in _a lot_ of scenarios.
And we'd still have be able to make processors with lots of slow threads of course for applications where that's cost or power effective, that's comparatively very easy.
These days however, the difference between fastest core performance on low-core processors and fastest core performance on high-core processors is rather a lot less than 50% (Comparing like for like, eg AMD to AMD and Intel to Intel).
This kind of machine is nice to look at from afar but for most apps trying to get anywhere near a 64 fold speedup on your application on this would mean scrapping the code base and doing 1 failed rewrite, followed by 1 marginally successful one, taking up ruinous amounts of calendar time and engineering resources.
But of course it's diferent now than in 2005, because today we can't do any better.
I was skeptical so I downloaded some SPECint benchmark results ( https://www.spec.org/benchmarks.html ) of T1s, Power and Xeon, compared them and thought "mh, probably a bad idea" and then I ended up having to invest quite some time to convince my management to keep using "normal" servers.
On the other hand, later, I had a bit of fun hearing stories from other colleagues telling me how slow those T1-machines were once they started running on them their normal DB-workloads => after months of everybody complaining about bad performance, everybody went back to normal servers => a lot of time & money (& nerves) spent for nothing.
Normal IT politics, I guess.
Lunch driven procurement.
In all seriousness, someone is trying to make an app to benchmark software[0]. You might be interested in that thread. Thanks for the link.