AmigaOne X1000: best potential mini-space shuttle controller?

[dreamcast270mhz]

@HaywirePC

The 1.8GHZ dual core PA6T should be equal to a C2Duo at 3Ghz. WHy? PPC executes more data per cycle than x86 due to its smaller pipelines. And unless you have a smart phone, most Amiga Browsers way outclass your typical mobile web browser. More on this:http://en.wikipedia.org/wiki/Megahertz_myth

And as Karlos said clock rate and CPU power is irrelevant to Aviation and Space applications. For instance, the F-14 Tomcat used i386 and i486 CPUs in the F-14D variant, and that was in use until 2006. And the F-22 Raptor I believe uses a 604e derived CPU. Game consoles are also a good example. The Gamecube used a 400mhz CPU when CPU power in desktop PCs was 1-2.5 Ghz in range, yet it runs many games a PC never could? It has to do with application entirely.

[Karlos]

It says 20Mhz and not 33Mhz depending on what link you look at.

The RAD6000 was rated at up to 33MHz, that's not to say it's used at that speed in all applications.

http://en.wikipedia.org/wiki/Comparison_of_embedded_computer_systems_on_board_the_Mars_rovers

Anyway I'm guessing robustness is only a part of it.

Power usage is also a factor, but radiation hardening is a big deal for circuitry that is going to be used outside the atmosphere / magnetosphere of Earth. It only takes one high energy cosmic particle to pass through your processor substrate to flip the state of one or more bits in a register / memory cell etc.

[dreamcast270mhz]

What do they use for radiation shielding for circuitry? Lead?

But anyways heat may be another concern. In a vacuum there is no air to cool a CPU, and the higher a CPU's clock rate versus a lower clock rate, the more cooling is needed for the higher clock.

[Karlos]

The 1.8GHZ dual core PA6T should be equal to a C2Duo at 3Ghz. WHy? PPC executes more data per cycle than x86.

I wouldn't be so sure about that. Modern x86 hardware is pretty damned good at the old instructions-per-cycle count. The Core2 is no slouch.

[Karlos]

What do they use for radiation shielding for circuitry? Lead?

Nope, that would be too heavy. The key to radiation hardening is to use larger process sizes. Although it makes your transistors bigger and therefore increasing the chance of a particle hit, it decreases the effective amount of charge that particle can dump into the silicon comprising your transistor, thus reducing the chance of a bit flip. Basically, the smaller your components, the more vulnerable they are to the damaging effects of ionization. The second thing you need to do is harden the chip circuitry itself, eg putting ECC on everything.
 

But anyways heat may be another concern. In a vacuum there is no air to cool a CPU, and the higher a CPU's clock rate versus a lower clock rate, the more cooling is needed for the higher clock.

Well, without air, the CPU still cools. There's thermal contact with the circuit board and furthermore, radiative heat loss. However, low power 33MHz parts don't get that warm. For devices like the Mars Rover, they actually have to put electrical heaters in the control unit to ensure it doesn't freeze, since your average Martian night is bitterly cold.

[dreamcast270mhz]

I wouldn't be so sure about that. Modern x86 hardware is pretty damned good at the old instructions-per-cycle count. The Core2 is no slouch.

Well on a dual G5 1.8 versus a C2D Mini 2.4 I can definitely beat it playing UT2004 in terms of Framerate at a given setting. Also, Doom 3 and photoshop do better against it , but then again the Mini doesn't have an impressive GFX card.

PPC seemed to always be ahead of any Netburst CPUs and the 604e did very well versus a Pentium Pro

[smerf]

Hi,

@Karlos,

"these devices will withstand radiation levels that would literally kill you within minutes:"

Hey Karlos, could you check that out and do a timing survey, so we would really know how much time would go by before you expire.

Uhhhh! on second thought, don't do that, I wouldn't have any one to bust, send Kesa instead.

smerf

[dreamcast270mhz]

Naw, I have a few people I know on here but wont say their names that I would volunteer, Kesa is pretty much harmless but I have some traitors and trolls in mind.

[Kesa]

Hi,

@Karlos,

"these devices will withstand radiation levels that would literally kill you within minutes:"

Hey Karlos, could you check that out and do a timing survey, so we would really know how much time would go by before you expire.

Uhhhh! on second thought, don't do that, I wouldn't have any one to bust, send Kesa instead.

smerf

Naw, I have a few people I know on here but wont say their names that I would volunteer, Kesa is pretty much harmless but I have some traitors and trolls in mind.

 

[commodorejohn]

Naw, I have a few people I know on here but wont say their names that I would volunteer, Kesa is pretty much harmless but I have some traitors and trolls in mind.

The Purge is coming...

[dreamcast270mhz]

Yeah it is if I ever can bait the trolls into getting banned (dont worry nobody here is on my list :p)

@Kesa

You've been harmless, because the worst you do is post when drunk.

[Tripitaka]

What do they use for radiation shielding for circuitry? Lead?

But anyways heat may be another concern. In a vacuum there is no air to cool a CPU, and the higher a CPU's clock rate versus a lower clock rate, the more cooling is needed for the higher clock.

Plasticized gold foil is used in radiation shielding, you would be hard pressed to find anything better. As for cooling, electromagnetic radiation will move quite freely through a vacuum (otherwise we would get no heat from the sun).

[Duce]

Enjoying the education on hardened circuits, Karlos.  Never really thought much about it, did some reading and it's all very interesting.

[psxphill]

PPC seemed to always be ahead of any Netburst CPUs and the 604e did very well versus a Pentium Pro

Intel up to p4 and Intel after p4 may as well be two different companies.
 
PPC may have been competitive in performance back in the pentium pro days, but it's not competitive now. Which isn't necessarily a problem of course, performance is only one reason you choose a processor.
 
p4 was designed to run at high clock speeds with no regard to actual performance, the same as the new ppc chips that are in the PS3/360.
 
If your benchmarks are true then it's probably not the cpu execution speed that is the limiting factor in the intel mac mini.

[itix]

The 1.8GHZ dual core PA6T should be equal to a C2Duo at 3Ghz. WHy? PPC executes more data per cycle than x86 due to its smaller pipelines.

PPC requires more instructions to do its job. Everyone can create winning benchmarks.

[takemehomegrandma]

What matters for these applications is robustness rather than speed. For example, the Mars rovers use PPC based processors that run at only ~33MHz that were made when desktop machines were in the high hundreds. However, unlike your current 2-4GHz processor, these devices will withstand radiation levels that would literally kill you within minutes

Military Spec, I *knew* it! Well I'm sure the PA6 is that, since it was used by the military, so that shouldn't be a problem. And since the Xorro and Xena are XuperDuper, it should be clear that the X1000 is up for the job!