Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Cray-1 vs Raspberry Pi

roylongbottom.org.uk

Post reply on HN
Print view View original

181–190 of 193 posts

Re: Cray-1 vs Raspberry Pi

#181

Post by jasonwatkinspdx »

pinewurst wrote: post #170
jasonwatkinspdx wrote: post #156

Earlier quoted context omitted.

It was a later machine than the Cray-1, but I remember seeing a post on the jsoftware forums from someone that ran interactive APL on a Cray. Having a REPL to access a machine like that in that era must have been quite something.

Crays were not good interactive machines for a few reasons, but lacking virtual memory was the most notable.

From what I remember the machine they were using had a Unix workstation that acted as the front end, so the Cray would be just as an accelerator. APL uses a stack based approach to memory management that maps well to bare hardware with minimal OS features.

Re: Cray-1 vs Raspberry Pi

#182

Post by dan-robertson »

camel-cdr wrote: post #178
dan-robertson wrote: post #175

Earlier quoted context omitted.

Thanks for all the detailed information! That answers a bunch of my questions and the implementation of strlen is nice. The instruction I was thinking of is pshufb. An example ‘weird’ use can be found for detecting white space in simdjson: https://github.com/simdjson/simdjson/blob/24b44309fb52c3e2c5... This works as follows: 1. Observe that each ascii whitespace character ends with a different nibble. 2. Make some ve…

I'm actually doing something quite similar in my, in progress, unicode conversion routines. For utf8 validation there is a clever algorithm that uses three 4-bit look-ups to detect utf8 errors: https://github.com/simdutf/simdutf/blob/master/src/icelake/i... Aside on LMUL, if you haven't encountered it yet: rvv allows you to group vector registers when configuring the vector configuration with vsetvl such that vector…

Perhaps another way to do gather with mixed sizes would be with a combination of the vector and packed simd extensions, using a simd register for the lookup table.

Re: Cray-1 vs Raspberry Pi

#183

Post by camel-cdr »

dan-robertson wrote: post #182
camel-cdr wrote: post #178

Earlier quoted context omitted.

I'm actually doing something quite similar in my, in progress, unicode conversion routines. For utf8 validation there is a clever algorithm that uses three 4-bit look-ups to detect utf8 errors: https://github.com/simdutf/simdutf/blob/master/src/icelake/i... Aside on LMUL, if you haven't encountered it yet: rvv allows you to group vector registers when configuring the vector configuration with vsetvl such that vector…

Perhaps another way to do gather with mixed sizes would be with a combination of the vector and packed simd extensions, using a simd register for the lookup table.

I think I'll propose a vrgatherei4, analogous to vrgatherei16, but with 4 bit indices, for LUTs, when I eventually finish my project with a blog post.

Re: Cray-1 vs Raspberry Pi

#185

Post by tzs »

pinewurst wrote: post #170
jasonwatkinspdx wrote: post #156

Earlier quoted context omitted.

It was a later machine than the Cray-1, but I remember seeing a post on the jsoftware forums from someone that ran interactive APL on a Cray. Having a REPL to access a machine like that in that era must have been quite something.

Crays were not good interactive machines for a few reasons, but lacking virtual memory was the most notable.

What do you mean by "virtual memory"?

The Cray-1 had memory management hardware similar to that on DEC PDP-10s that had KA-10 processors and numerous 68K Unix workstations in the '80s, all of which served as interactive time sharing systems just fine.

The memory management hardware in those systems would take the addresses generated by the instructions, compare them to a limit register, and if below the limit would then add them to a base register to get the final memory address. Some systems would have more than one pair of base/limit registers, such as a pair for code and pair for data.

Some people do call such systems virtual memory, because the addresses specified in the code are not the addresses that the memory sees (unless the base register contains 0). But I think it was more common for systems where all the MMU did was simply relocate segments and all of the program's code and data had to be in memory for the program to run to not be considered to be virtual memory. Virtual memory usually meant systems that allowed you to run programs that used address spaces larger than the available memory.

Re: Cray-1 vs Raspberry Pi

#186

Post by mdaniel »

huytersd wrote: post #96
niederman wrote: post #63

Earlier quoted context omitted.

How is this dystopian? Plastic is a really great material -- and way more eco-friendly than metal for building computers. It's only mass-produced single-use plastics like water bottles that are bad for the environment.

I wish we could come up with a plastic that would biodegrade after a fixed amount of time say 200 years.

There's no shortage of those (e.g. https://www.youtube.com/watch?v=h1zDJ1qZTlg and https://youtu.be/F2zm87p8f7M?t=1857 that I just happened to have handy) what there is a shortage of is economical ones (or scalable, the other major obstacle) as compared to the current situation where the negative externalities are not priced into the sales leading to perverse incentives

Re: Cray-1 vs Raspberry Pi

#187

Post by HarHarVeryFunny »

I remember watching a TV program about wave/fluid simulation (real time?) as a wow-wee demonstration of the power of a Cray-1.

In the meantime plenty of colleges and companies were running entire departments on a PDP-11 that had a fraction of the power.

Raspberry Pi faster than a Cray-1 is cool benchmark of how far we have come! The Cray had built-in seating though, which the Pi doesn't! :-)

Re: Cray-1 vs Raspberry Pi

#188

Post by ptek »

buryat wrote: post #92
fastneutron wrote: post #28

When I see comparisons like this, the first thought I have is not the benchmarks, but rather what the most “heroic” real-world calculation of the day would have been on something like the Cray-1, and how to replicate those calculations today on something like a RPi. Weather/climate models? Rad-hydro? The fidelity would almost certainly be super low compared to modern FEA software, but it would be a fun exercise to tr…

nuclear weapons simulations the first machine went to Los Alamos https://www.theatlantic.com/technology/archive/2014/01/these...

I've always been interested in this, I wonder how optimised the code was and if they used LUTs (Look Up Tables) as they did in the 80s for 3D calculations on the home computers.

Oh cool they got CrayOS working.

But still 1MB RAM, I remember getting the slow RAM 512KB update for my Amiga 500 in the early 90s.

Re: Cray-1 vs Raspberry Pi

#190

Post by huytersd »

mdaniel wrote: post #186
huytersd wrote: post #96

Earlier quoted context omitted.

I wish we could come up with a plastic that would biodegrade after a fixed amount of time say 200 years.

There's no shortage of those (e.g. https://www.youtube.com/watch?v=h1zDJ1qZTlg and https://youtu.be/F2zm87p8f7M?t=1857 that I just happened to have handy) what there is a shortage of is economical ones (or scalable, the other major obstacle) as compared to the current situation where the negative externalities are not priced into the sales leading to perverse incentives

That’s just biodegradable plastic. It will last a couple of days out in the open. I actually mean plastic that’s lasts for multiple lifetimes but still starts to break down only after ~200 years.
Post reply on HN