(pounds head)

[jay]

I just can't find the SPECint95 or SPECfp95 values for a PowerPC 750... from anything other than some Russian website. SPEC itself is no help, even if I know the webmistress well... I know that IBM or Motorola published these in the late 90s.... sigh.

Comments

[brian1789.livejournal.com]

Perfect. That's what I needed... I saw those numbers, actually, but on a Russian site from a Google link. Yours is linkable and useful... thanks. :)

Now I'll tell JPL that I'll need about 2 SPECint95 and 1.5 SPECfp95 -equivalent share, as well as 2MB of memory... that's probably around a 486-class system. Shows you how far the hardened space-rated systems lag behind the commercial market, that the benchmarks for state-of-the-art space processors aren't even published anymore.

[compilerbitch.livejournal.com]

ISTR that the RAD750 isn't equivalent in performance to the PowerPC it was based on. It's a fair bit slower and it uses several times as much power. I vaguely remember reading some Lockheed blurb about it, but it's a couple of years ago now so the details are vague.

(searches the web a bit)

It looks like the NASA Office of Logic Design are running a seminar on the RAD750 in November. I'd strongly suggest that if anyone has the numbers you need, the guy who runs the OLD web site, Richard Katz, will probably be able to point you in the right direction.

I also found this on the http://klabs.org/ web site:

The RAD750™ - A Radiation Hardened PowerPC™ Processor for High Performance Spaceborne Applications

R. Berger, D. Bayles, R. Borwn, S. Doyle, A. Kazemzadeh, K. Knowles, D. Moser, J. Rodgers, B. Saari, and D. Stanley
BAE Systems
IEEE Aerospace Conference, 2001
IEEE Proceedings. ,Volume: 5 , 10-17 March 2001, Pages:2263 - 2272 vol.5
http://klabs.org/rk/papers/processors/powerpc/rad750_2001_berger.pdf

BAE SYSTEMS has developed the RAD750™, a fully licensed radiation hardened implementation of the PowerPC 750™ microprocessor, based on the original design database. The processor is implemented in a 2.5 volt, 0.25 micron, six-layer metal CMOS technology. Employing a superscalar RISC architecture, processor performance of 240 million Dhrystone 2.1 instructions per second (MIPS) at 133 MHz is provided, while dissipating less than six watts of power. The RAD750 achieves radiation hardness of 1E-11 upsets/bit-day and is designed for use in high performance spaceborne applications. A new companion ASIC, the Power PCI, provides the bridge between the RAD750, the 33 MHz PCI backplane bus, and system memory. The Power PCI is implemented in a 3.3 volt, 0.5 micron, five-layer metal CMOS technology, and achieves radiation hardness of <1E-10 upsets/bit-day. This paper describes the implementation of both designs.

[brian1789.livejournal.com]

This is excellent, thanks... and you aren't even here, yet! :)

The software in question was tested on a flight avionics board for the X-34, which was a rad-hard PowerPC 750 (presumably the RAD750), and used about 2-3% of CPU except during control events, when utilization spiked to 20% for a couple of seconds. We could push it down to 10% if we were willing to broaden the high-demand spike somewhat. So I'm telling the instrument avionics guys that we'll need about 15% equivalent of a rad-hard 750. If I went by MIPS ratings, that'd be about 35 MIPS or roughly a i486-class processor. Which is not what they wanted to hear, they were trying to use a old 16-bit board.