Hacker News new | past | comments | ask | show | jobs | submit login
Homebrew Cray-1A (chrisfenton.com)
107 points by nilicule on Oct 13, 2014 | hide | past | web | favorite | 22 comments



Chris Fenton here - someone did actually send me an 80MB 'disk pack' with software, and I was able to modify a super old disk drive to actually read it:

http://www.chrisfenton.com/cray-1-digital-archeology/

And thanks to some awesome help from others, I was able to actually recover a copy of COS (Cray OS):

http://www.chrisfenton.com/cos-recovery/

Andras Tantos and I have been collaborating on this together for a while now, and he actually has a downloadable Cray X-MP datacenter simulator he wrote, where you can run the COS image I found:

http://www.modularcircuits.com/blog/articles/the-cray-files/


Maybe you should link these pages together better


Website design is not a strength of mine =) Most of my random projects just get their own page on my site whenever I get around to doing a write-up.


This isn't about web design. You should just add links to these articles on the article you submitted. Good for google ranking, increases the chances of someone reading them, makes user happy :)


Now if only someone can get this guy some Cray-I software. Even the Computer Museum, which has an actual Cray-I they use as a piece of furniture, apparently doesn't have any.

The Cray-I is a rather simple machine at the logic level. There are 64 of some registers, but they're all the same. The instruction set is small.


To be fair, booting the Cray would require a good sized power substation. The power bill would be... impressive.


The 1s used between 100kW and 150kW. So a small substation. The XMP used 200-300kW. The 2 used 150-200kW.

There was a YMP model (the EL) that could be power off normal mains (220v).

Energy efficient was different back then.


I wonder if anyone's asked the National Cryptologic Museum/NSA


How did they code it, in assembly?


We used Fortran, although asm could be used and apparently there was a Pascal, but I never saw it. We actually did the dev cycle on a front end system (VAX/VMS or Unix) and submitted the program as a batch job (COS was a batch OS, the multiuser UniCOS didn't show up till much later).


Wow, thanks for replying.

I must have been an incredible opportunity working with a supercomputer. Did you ever compute a mathematical constant or an algorithm to a large precision out of curiosity?


Yeah...I got lucky. My third year or so at Georgia Tech a professor offered a "special topics" (e.g. not in the course catalog; his personal research interest) undergrad class in supercomputing. GaTech was good for little bonuses like this (I took Micheal Barnsley's first IFS class, too).

In addition to Crays, we got time on a CDC Cyber 180/990 and a 205, an ETA machine, and had cursory intro to NEC SX, BBN Butterfly, Multiflow Trace and a few other weird things noone has ever heard of. We wrote a handful of numerical methods programs for each (so, yes, computing mathematical constants and such, but more vector/array manipulations), and looked at how the architecture of the machines (short v. long vectors, register based v. memory based, vector v. MIMD v. VLIW) effected the speed of the programs and what optimizations/refactorings resulted in large speedups.

Good times.


This is so awesome.

But what I'd really like is that to-scale case and put a small PC in it. Even if one does not care at all about its historic significance, the Cray-1 was one of the most visually appealing computers I have ever seen. Having a miniature version of it sitting on my desk would be extremely cool.


Well, just saying:

http://www.thingiverse.com/thing:26671

On a slightly larger scale, I will, eventually, in my infinite space time, have a full size wooden cray I in my rec room as a sitting couch. I have the carpentry skill and tools and structural experience to pull this off. For many years I have periodically worked on dimensioned drawings. My strategy revolves around making eleven identical 22.5 degree segments with a framework and panels mounted in slots and two "about half" segments and then bolting the column detail from one to the internal framework of the other. This makes it very easy to build and very easy to move (well, sure, 13 total segments and maybe 10 bolts each means a lot of wrench time when you move it, but no segment is individually too big for a person to move.)

To warn other people, this is a typical "large machine in a giant data center looks small, but it looks huge in a normal size house" kind of problem. Its about six and a half feet tall and about eight and a half feet across so its a rather substantial investment in space. Then again you can use the enclosed volume for storage. Essentially you have eleven small full height coat closets and under seat storage also.

I have a basement rec room / craft room / lab that runs the length of my house so it will be more or less in proportion other than height.


Nice! Thanks for the link!


Please upload any working cray image to bitsavers.org.

I wrote an emulator for a Motorola Exorciser (6800 development system: http://exorsim.sourceforge.net/ ), and am totally in debt to whoever uploaded the MDOS disk images.


I'm wondering about the relative performance between this and the real Cray-1A. Surely, this thing beats it on a FLOPS/Watt basis, at least. :)




Anxious to see how this turns out. Very interesting!


I really like the idea and implementation.


But can it run Crysis?

Or even Doom?




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: