Shortly after I had a TiBook (PowerBook G4) that was _only_ 1-inch thick! Compared to 1.75” Dells my coworkers had, it seemed like the future. DVD drive, modem, Ethernet, full sized DVI port, FireWire, WiFi, Bluetooth, optical audio in and out, gigantic display with a bezel that was unrivaled for years, even among Macs. What a beast!
(I know you meant 2020, but it’s fun to think about the air in 2000).
https://top500.org/lists/top500/list/2000/06/
https://www.notebookcheck.net/Apple-M1-GPU-Benchmarks-and-Sp...
Only true if all you're running is matmul (supercomputer has general purpose CPUs so more flexible than M1 GPU) - also those flops are probably FP64 in supercomputer ratings and FP32 in M1.
As a smart man I knew used to say, supercomputers are about I/O not raw compute. Those have terabytes of RAM not 8GB.
Throughput-wise, the supercomputer is competitive because it has a lot of local RAM connected to lots of independent nodes, which, in aggregate, is comparable to modern laptop's RAM throughput (still much more than disk) with a caveat, that you can only leverage the supercomputer bandwidth if your workload is embarrassingly parallel running on all nodes[1]. Latency-wise, old RAM still beats NVMe by two or three orders of magnitude.
[1]: there's another advantage that supercomputer has which is lots more of local SRAM caches. If the workload is parallel and can benefit from cache locality, it blows away the modern microprocessor.
It was also used by the PS2 for local multiplayer between multiple consoles. Although Sony eventually removed that port.
All our intel MacBooks now run Linux just fine. The oldest is 2012, with 4Gb but most are 8 or 16Gb.
I would always recommend more RAM first over a faster processor; back when I would build desktop machines for Windows, I would use the second best CPU and put the savings into RAM.
It isn't the most powerful computer in the world but I never ran into any problems... so it's probably an OK compromise for most people, especially in the world where RAM is scarce because of AI datacenter buildouts.