Slashdot Mirror
Computers in Space Examined
Wil Harris writes "There's an article about the computers used in space missions over at bit-tech this morning. It covers the processor types and speeds, why space stations are less powerful than the laptops that astronauts take up with them and why tape storage is still de rigeur. An interesting and concise couple o' pages."
The fewer components you have, the less likely you are to encounter a failure.
I remember reading something about most space missions are pre-determined and very straight forward, there's no need for difficult maneuver like one has to execute in a X-Wing.
Having said that, there are still plenty of complicated, unexpected problems in space, but these problems have to be analysed and decision made by people on earth.
I guess it's all circumstantial, I can't even operate my 2001 Toyota electric window if the engine's dead, but my 1989 Toyota has no such problem. So if I crashed into a river, I hope I was driving the '89, but if I'm crashing into another car, I want my '01.
Rock that crushes, Paper & Scissors that don't matter.
That's why i'm staying with my old Pentium 99mhz ^^
Nuff said (but there's something to be said for the butlerian jihad, and Cmdr Adama filling his battlestar with rotary phones and manual typewriters!)
Don't blame Durga. I voted for Centauri.
Doesn't it seem very strange compared to the days where the goverment had super computers and the regular people had no computers? A stark contrast indeed. Now we are...close to the same level? Does this sound realistic, or are aces up their sleeves?
Dave Mills (inventor of NTP) told me that on the last Columbia shuttle mission, they were running some experiments with NTP in space. And, thankfully, they transmitted all their data before landing. But apparently, they were so overworked, they didn't have time to calibrate the machine properly, so sadly, the data is useless.
To make laws that man cannot, and will not obey, serves to bring all law into contempt.
--E.C. Stanton
Cmdr Adama filling his battlestar with rotary phones and manual typewriters!
Something tells me he should focus on adding a few more medics first.
That is still too high tech. Tell me what Cylon attack can defeat a coal-fired Battlestar carrying swarms of pedal-powered Vipers crewed with low-IQ Forrest Gump clones? You can't dare have genius level brains which can be commandeered by the Cylons.
The first manned space flight had a computer on board to control re-entry, but it was basic in the extreme - and locked so Comrade Gagarin couldn't tamper with it. An envelope with the code to unlock the computer was hidden somewhere in the capsule, and should an emergency arise, ground control would tell him where it was. Nice.
:).
Sounds nasty. I would at LEAST want to have some QUICK way of getting to it.
Like a hammer
Foxed Design
The basic gist of the article is "They don't use more than they really need". Unfortunately, this is not a complete answer.
A company I used to work for discussed using some of their technology with Nasa. One of the things they told us was that they preferred processors a two or three years old because they were afraid of random bit-flippings caused by radiation etc.
(Sadly, I wasn't in on this whole conversation, so I doubt I can effectively answer some of the questions that arise. For example, I'm not sure why the processors had to be a couple of years old. I assume it had to do with shielding or something, but I really don't know. If anybody has insight on this topic, I'd really really like to be enlightened.)
"Derp de derp."
Just like Cassini, the Hubble also has on-board solid state recorder (installed during one of the servicing missions), which replaced an old tape recorder. This has been really a nice addition as we can store more data into the solid state device while collecting data bits and dump them when the downlink becomes available. It really helps increase the efficiency of the satellite (and that's a big thing for science mission).
[Note that I've simplified the scheme alot here.]
Though several sections of the device have been damaged by radiation, or something, I hear. So even these things aren't too resilent to the harsh space environment, yet. Something you future engineers should think about as a project.
From the article:
I've yet to discover any spacecraft that employs AMD hardware...
Thats it! Thats why Dell won't switch, obviously their customers cannot use hardware not fit for spacecraft.
If some smart application of physical processes were developed instead - something based on a large and almost limitless power supply - would this not only benefit space endurance but also earth's environment, by reducing the need for harmful chemical production?
I suggest you read Slashdot
Anybody remember RCA's CDP1802, the weird little CMOS RISC-ish 8-bitter used in Voyager, Viking, and Galileo? These things have been running for decades, despite the radiation they've been subjected to.
Now that's engineering!
And why is this relevent? Isn't there atmosphere inside all manned spacecraft?
The real "Libtards" are the Libertarians!
They need more performance than space equipment, and space equipment has power concerns that studio equipment does not, so the equation balances.
I never vote for anyone. I always vote against.
-- W.C. Fields
see http://spaceflight.nasa.gov/gallery/images/station /crew-1/hires/iss01e5127.jpg
You may be onto something. The most limitless energy supply I can think of is found in my spam box right now. The Space Elevator can be made a reality, perhaps, with the propery application of zillions of doses of "lengthening enhancer"? Could the "energy boost" of the illicit HGH herb be applied to rocket thrust? Not only that, I think I can fund my own NASA if I answer every single one of the thousands of Nigerian princes who have been begging me to let them give me millions.
Don't blame Durga. I voted for Centauri.
Because of the density of memory now-day, bit-flipping is becomming a real problem for home PCs and workstations running with an uptime of over a week. Bit-flipping happens all the time and even on your PC. It just may be happening in a region of the wafer that does not currenly have anything important addressed to it...hence not an issue. But someplace, somewhere, a slashdot reader is getting a bitflip causing data-rott once it's commited to the harddrive. By the way, these bit-flips are causes cosmic rays.
If you serious about data integrity and stability, you would be foolish not to use ECC. You may take a 5% performace hit, but it's best to get used to it. If you need that extra 5%, then upgrade your processor to make up for it.
Life is not for the lazy.
Sounds to me more like it is an opportunity for a niche market of companies to fabricate equipment designed to work in space.
With low to no gravity and a high potential for various solar interferences, you need some seriously ruggedized components in order to work. With the amazing success of the Ansari X Prize, I wouldn't be surprised at all to see a few companies take a stab at making spaceworthy computing components.
To me, it seems like quite a few past innovations from the government have been generated as a direct result of wars. WWII, Viet Nam, the first and second Gulf Wars, all have contributed to tech. All we have to show for space related issues are SDI, which doesn't really seem to (a) exist or (b), have contributed anything to the common core of technology.
I am hopeful that the commercialization of space flight will add to our technological advances in (ahem) interstellar computing. The loop is open now to the best and brightest. Let's see what they've got
After all, if an alien civilization captures one of our orbital vehicles and comes to the conclusion that Pong and limited space flights are the height of our technology, we're screwed.
I'm wondering if more powerful computers will ever get up there. As the article states, computing power is minimally needed. Will the space industry ever be able to take higher-powered computers into space? This brings forward the question that if high-powered computers can never get off the planet, how will the future space industry be impacted by that?
There already are a number of companies working the "niche market". The problem is that the market is relatively small, and the costs are high. Hopefully the advent of commercial spaceflight will bring launch costs down enough that more people will launch, and the space market will expand significantly.
Some 9 years ago I worked on some chip design for Hughes and ESA.
Back then, we used 1.2um on 4" (or 6" in the new fab) wafers - and everything was built on a sapphire substrate instead of a silicon substrate to make them radiation hard (when they went through the van allen belt).
It was dull, as every single chip had about 12 inches of paperwork from QA. Every *instance* of every chip had its own paperwork, I mean. It was also dull because they wanted tried and tested tech, not any of this new fangled sub-micron stuff.
That was then. Can anyone let me know how much things have changed?
What the hell are all those little random dots? Is that from cosmic radiation pelting the CCD wafter on the digital camara?
Life is not for the lazy.
There seems to be a fad occuring in journalism that makes it cool to use slang and be insulting to the reader. Maxim is one thing, but when I read this and some of the other recent articles discussing some mostly obscure application of computers I don't feel the need to be insulted because I didn't know. They are like "You thought blah blah blah... but this isn't your standard this or that." Ah well, that seems to be the way amateur hobbiest discourse is heading.
An embedded machine, OTOH, is designed to do one, or a very small range of things, very well, very reliably, and very efficiently. I have had the fortune of working on two space based projects. In the first we used a single board Z80 based space hardened 'computer' to control a simple set of devices. It stored the ASM code in an EEPROM. It was more complex than we needed, as it was a standard issue unit, but much simpler than the Apple ][ we used as the GPC.
On the second project, 10 years later, we were not using incredible different machines on the satellite, though the GPC was now a Wintel machines with 100X the memory and speed. But when your main concern is that things just have to work, processor speed and OS wars have little meaning.
So these stories about how underpowered and behind the times embedded systems are just annoys me. It is just like continuous burns on SciFi shows(kudos to Babylon 5). Perhaps meaningless power is important to the ignorant masses, but we on /. are supposed to know better. I was using a tape drive until at least '87, just because It Worked.
"She's a scientist and a lesbian. She's not going to let it slide." Orphan Black
Unmentioned in the article is why the unlock code was hidden from him.
The Soviets were afraid of a defection, which would be possible if he could run the navigation system himself.
Is it just me, or does using tape storage seem horribly unreliable. I have some bad memories of backing up my 486SX's gargantuan 540 MB hard drive onto 100MB tapes. The process often took 2-3 days, because about 50% of the time, writing to tape would fail, necessitating starting the whole tape over again. How could tape be reliable enough to run space probes and other things that handle huge temperature extremes and tons of radiation?
Orbits are very predictable, and any real-world spacecraft will have a very limited amount of delta-vee with which to maneuver.
Even with realistic sci-fi technology like fusion drive, space battles would still be boring as hell. Read Protector by Larry Niven for a realistic take on space combat.
Why is it that the proponents of "one nation under God" are so eager to get rid of "liberty and justice for all"?
At one point, my data is a bit old, heads 'flew' over the surface of the disk on a thin layer of air. If the air were removed, the head would crash into the disk. Also, I suspect that 'G' forces and vibration aren't good for most hard disks.
Avionics in general, and ground based avionics in particular, tends to have very long life cycles. The basic philosophy is that even if newer stuff is better and even if it might be more reliable, we have much more information about the old stuff. It doesn't matter that our VHF-DF dies if a thunder storm comes within a hundred miles of it. What does matter is that it has NEVER given us false information. NO information is way better than false information. Also, when it dies, we know what it will do. There are no suprises. If all hell is breaking out, you don't need suprises do you.
> Computers in Space
"Awww, come on. Just a quick one."
Jack leaned back in his chair. Normally his patter was much smoother, but he thought he had her nailed. A socially inept appeal was just what the doctor ordered. At this moment. At this place.
"Pop it, yeah, come on baby!"
Crossing his fingers...yes! She lifted her top, exposing her breasts for about two and a third seconds. Nice "heavy C" breasts with large-diameter, dark-skinned areolae. Oval shape. Oval in the vertical direction, although what vertical meant was up in the air, literally, as the woman was aboard the space shuttle, orbiting over the Atlantic somewhere.
Isn't modern technology wonderful? Jack thought to himself.
"They could float thru that door any second. Take another chance! Scary stuff!"
In a few minutes, he'd have her with her top up for a good 30 seconds, half her mind on excitement, the other half paying attention to whether her colleagues would come bobbing in. And 30 seconds was all he'd need to get done. 30 seconds on tape that is, which he could loop over and over. He switched on the recorder.
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
What the hell is an amateur hobbyist anyways? Is that like a tall giant?
I always wonder why you can't overclock the hell out of chips in the cold vaccuum of space...
"Waste not one watt!" - CZ
My experience with space rated equipment isn't all that extensive or current (I was involved in failure analysis of an AP-101 memory card that had an intermittent failure from the STS-2 and had some interactions with the engineers at IBM's old FSD division, which designed the AP-101s and wrote the flight software) but the article misses one very big point that is the really fascinating aspect (to me) of spacecraft computing hardware and I would have to challenge a number of facts in it.
1. The shuttle launch algorithms and orbital maintenance procedures are a lot more complex than the article makes them out to be. There are several hundred parameters that are continually checked, recorded and processed from tens to hundreds of times per second to make sure the flight path is correct and all systems are operating correctly. Along with monitoring the flight path, the computers were/are largely responsible for the data displayed on the astronaut/pilot's CRT displays in the cockpit.
2. It is my understanding, that in the early shuttle missions at least, there were multiple code loads during flight. The original AP-101s had a maximum of 256K words of 32 bit memory, which was enough for a separate launch, orbit and landing image, each which had to be loaded into the AP-101s before the next phase of flight. There have been issues with loading software or receiving and loading new software from the ground.
3. The original AP-101s were designed for the F-15 and could be considered "state of the art" for the early 1970s in terms of processing power and memory size. They are capable of about five MIPs and had a full megabyte of battery backed memory. They were chosen because they had been qualified for the high G-Loads and temperature extremes of the fighters. While the systems used on the shuttle were of the same design as used on the F-15 (and later the B-1B), they were inspected to much higher standards and all failures had to be resolved down to the point of having a test in place to prevent the failure from escaping the manufacturing/test processes as well root cause action plans at the component supplier.
The memory card failure that I was involved with was caused by a solder ball inside a metal RAM chip package. During the shuttle's ascent, vibration caused the solder ball to break free and intermittently touch the surface of the chip inside the package. The problem was extremely difficult to reproduce and was found by placing a microphone on the chip package and tapping the chip with the eraser end of a pencil. Chips with this solder ball defect "rang" differently than ones without this problem. After the ball was discovered and proven (by cutting open the chip package), every chip used in a shuttle AP-101 was tap tested by IBM to ensure no other solder balls were hidden inside the packages.
4. I don't know where that picture of the "Part of the AP-101S" came from as there is no way that is flight qualified hardware for an F-15, let alone a shuttle orbiter. Wire reworks are simply not allowed in high-G, high vibration environments and it looks like the surface mount components are hand soldered into place. I think this is prototype hardware that somebody pawned off on the author.
5. I don't understand where the idea that space systems having to be low power came from. The AP-101s were real power hogs (all their logic is bipolar) and were in fact glycol cooled. A significant fraction of the orbiter power generation is devoted to the compter systems (as well as the spacecraft cooling capabilies).
What is always interesting is looking at how the software for manned spacecraft is developed. A big joke is the Mars Observer and the mix up between English and Metric units, but think about how often you've heard about a software failure on board the shuttle - or any manned spacecraft for that matter. In Apollo, there were none and the software for the CM and LM computers was wire wrapped on a bed of nails instead of being burned into
Mimetics Inc. Twitter
Or just take all the spam from the planet, put it together, and it will create an artificial singularity which can be used as an energy source.
Fight Spammers!
As much as you want to kick around the lowly PC, there are many instances where replacing exotic hardware with commodity hardware (which is by far more tested and debugged and supported due to market penetration than exotic hardware).
make a Beowulf cluster out of all the old space stations in the future. :)
and also Thinkpads!
Also all those people in the control room are getting paid. I'm not trying to be glib, but seriously, if we want to make spaceflight profitable, you are going to have to get some very serious advances in reducing the number of people involved in preparing and supporting these endeavors. Otherwise they will always be pork projects.
Gagarin was simply a PR tool. He was also tuned up by the KGB a couple of times, and there is photographic evidence of his face after the beatings.
The controller for the SSME (Space Shuttle Main Engine) uses a pair of 68000 processors. It is a very critical system. If something starts to go wrong with the engine, it has to detect the problem and shut the engine down before it progresses to a catastrophic failure. It uses two redundant processors for reliability. Each engine has its own controller.
Old microprocessors like the 80386 and the 68000 were the last commercial processors before cache, pipelines and other trickery made timing analysis difficult or impossible. Some people have used DSPs for controllers because they still offer predictable timing.
Mea navis aericumbens anguillis abundat
thats why nasa gets all their parts of ebay for $3 and charges the govt $20000.
Liberty freedom are no1, not dicks in suits.
With today's hardware you could probably replace that entire room with a handheld computer, but what would it require to fully test all possible permutations and guarantee it as acurate as the original?
w00t, it's a ThinkPad! I couldn't tell you which one, but that little red eraserhead is unmistakeable.
Question: The article makes a big deal about standard hard drives not being able to work in Zero G? Why does a laptop HD work in Zero G and a regular hard drive not work? Have I missed something?
Knowledge is power. Knowledge shared is power multiplied.
running on Intel Pentium 4 Extreme Edition 840 Dual corebwith dual SLI Geforce 6800 Ultra 512MB and 4096MB of DDR2-533 memory to run Halflife 2 and Doom3 in space, and of course the latest edition of 3Dmark2005. I am sure those cosmonauts could use some 3D gaming entertainment while waiting for docking. ;)
I've had experience with some of the computers in older government satellites.
Simple processors are preferred because that makes it much easier to figure out the time bounds on a subroutine. You don't want one routine to use up so much time that it keeps something else from being done. Timing information is rigoriously analyzed to make sure that the system won't miss something if lots of things happens at once. Fancy modern archetectures like cache, pipeline stalls, out-of-order operations, etc. make timing analysis very difficult.
Generally interrupts are not used - instead conditions are polled at a regular time slice. One reason for this is that polled data is also down-linked in a telemetry stream for status monitoring and trouble shooting. Also interrupts greatly complicate timing analysis.
They have a very small hole to allow equalization of air pressure between the interior and exterior of the drive.
The hole is covered with a HEPA-style filter (which may be inside the drive), which helps keep out undesirable particles.
The hole is also located in an area where very little airflow occurs (e.g., behind the circuit board), which also helps.
Thus, there is very little air movement between the interior and exterior of the drive, but it does occur.
Unless things have changed in the past couple of years.
Those who sacrifice security to condemn liberty deserve to repeat history or something. - Benjamin Santayana
Those who sacrifice security to condemn liberty deserve to repeat history or something. - Benjamin Santayana
The FPGAs on Spirit and Opportunity seem to be overlooked. NASA's new Reconfigurable Scalable Computer (RSC) Project for space applications is exploring using FPGAs (instead of CPUs) which offer increased performance and radiation tolerance at a fraction of the power consumption.
What's past is NOT ALWAYS prologue for the future!
so from tfa it explains all the hardware is up there because it works. Very reasonable, and I love the simplistic approach in some situations.
I'm no space / gov't expert but here's my 3 million dollar proposal-
Upgrade Torture Sat (tm) (R)(C)(CC)(GPL)
Launch a sattelite with an orbit into the nasty zones of local space packed with modern equipment as a real world torture test.
I bet several chip mfgrs would step up just for the prestige.
You could run the software you wanted for a few years, you could set up special test equipment just to handle the more low level stuff.
Now the space industry is up to 25 years ahead.
Firefox &
There hasnt been a single instance of proven cosmic ray bit flip on ground level.
And for bit-flips of other causes: The bit-failure rate per mbit has dropped a few orders of magnitudes tha last 10 or 15 years.
HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
I thought the same on sdi as you until I ran across a few declassified items in the clinton era.
SDI still exists! Tech for it was still being developed & pitched at a time when it had no noticable public support/funding! I ran across quite a few real badass technologies, demos, and proposals. Everything I saw was done by the big aerospace/military companies. In the same breath I can't assume much has trickled out of these companies yet but i'm sure their funding is back in spades.
Firefox &
The author of the article and some of the comments seem to be surprised that life-vital systems, such as a manned spacecraft, are running on "old" hardware because it's trusted.
I doubt NASA, medical manufacturers and life-saving-equipment (such as airbags, seatbelts etc.) manufacturers are worried about anything other than making sure things work first time, every time, guaranteed and on-time. Severe code analysis is performed on ANY software of this sort, down to mathematical proofs that the program is correct, and you don't want your hardware to do anything you can't predict 100%.
Why is it surprising that custom-built, antique (and therefore every flaw is already public... wanna find out about FDIV bugs, that the interrupt timing is slightly off, or that an instruction that is supposed to take exactly one tick sometimes take more when you're half-way to Mars? I didn't think so.), reliable, low-powered hardware is the norm in any life-critical system?
As anyone who manages code knows, the simpler the better, the more predictable, the easier to debug and test. The Z80 was out and in commercial use for years before all it's bugs were found and documented and that's not the sort of thing you want when the machine has to make a few vital decisions (even if, in some instances, it DOES takes a few microseconds to calculate, verify, and implement them).
Many systems like this even have two or more processors providing the same answers. Any difference between their answers and a warning is signalled indicating a possible hardware failure so that manual control can be assumed.
Secondly, the hardware is considered embedded... you don't need 10 GHz, you don't need masses of cache (if fact it'd probably throw your mathematical correctness right off), and it only needs to be as powerful as is absolutely necessary. My digital watch don't need a GHz, nor does my airbag computer in my car or the engine timing circuits, nor do the life-support machines at the local hospital.
Why would you spend more than you need to, introduce more factors (such as heat, power, airflow, RFI interference etc.) that need to be analysed thoroughly just to put unused hardware into a life-vital system?
The article more than states the obvious and doesn't even bother to point out that most PC's actually waste a vast amount of their time doing unnecessary things. If you can put a shuttle in space with only a few MHz, why do I need at least 1GHz to even be able to LOAD some operating systems within a few minutes?
a 5-point harness :P
Sure, it's DOT illegal, but I keep the original belts installed for inspection time, and the buckle on the harness will undo under load. Bit restrictive, but I really shouldn't be mucking around in the car while I drive anyway.
Facts do not cease to exist because they are ignored. - Aldous Huxley
I worked on BMDO years and years ago. You might remember that project, it ended in the bankruptcy of the Soviet Empire.
Anyway, we used to get gobs and gobs of data streaming in at 3megbit persecond the whole time during the launch of our sub-orbital payload. We would have reels and reels of 2 inch 24 track tape on to which we would record this data.
There were millions for these launches in the budget, but there was no budget to analyze the data!
I once read in an older post on /. (sorry can't find it) that the European Space Agency has implemented a hardened Sparc implementation based on the standard. Sad that the article fails to mention it completely. Here is an interesting link though. I heard those puppies can work under radiation levels that would fry a normal UltraSparc. If anyone has more information that would be very nice.
[rant]And what's more sad is that we are getting to the point where the x86 has become almost a monoculture, which is way very bad. Apparently there are lots of folks who think that the only processors around are named "pentiums". [/rant]
Erfs, this is what i get for not drinking my usual gallon of coffee in the morning before I post.
It's Mauna KEA observatory!
First rule of holes; When in one, stop digging.
The problem's objective is max. performance & min. response time, not less power.
open4free ©
We had an engineer from NASA come speak to the computer science department when I was in school. One interesting factoid not in TFA is that the reason that disk drives aren't used in spacecraft (at least the small ones) is that the spinning platters act as gyroscopes, and will throw off the attitude of the vehicle.
I wish I could remember the other details of his speech, but one thing struck me as unusual, is that the AP101S doesn't use a standard 8-bit word. It's very task-specific, and the designers felt they could justify it based on packaging, weight, thermal, and need of the software.
Chip H.
Well, sort of. The author refers to the fact that the Space Shuttle's AP-101S is based on IBM 360/370 series technologies. Fast forward about 20 years (with two decades of enhancements), and you've got today's IBM zSeries servers -- which can (and do) still run 360/370 software alongside Linux and Java applications written an hour ago.
The zSeries architecture has some unique characteristics for earth bound computing. In both space and in certain industries (notably finance), execution integrity is absolutely critical. (2+2 must always equal 4, never 5, despite cosmic rays or Skippy-the-mail-guy waving his T-Mobile cell phone too close.) In a zSeries server every instruction is executed twice by separate hardware, compared for integrity, and retried if there's a discrepancy. If the retry fails, there's a path to isolate the offending hardware, take it offline, provision a spare, retry, and proceed -- all transparent to software (Linux on zSeries included). There's a straightforward extension called GDPS to allow systems separated by up to 40 miles to participate in this scheme (if you need both integrity and extreme availability), although Linux is not quite GDPS ready yet.
Do you need all this? It depends. If you're playing an MP3 on an iPod and there's a bit flip, who cares? If you're moving millions and billions between bank accounts, of course you need it -- badly.
If you need Linux execution with instruction-level integrity, happy shopping for your zSeries server. And they're a bargain (honestly) "starting under $200,000" each (excluding disk).
IBM 760XD
"Next Generation Laptop" - Windows 2000IBM A31p
The article writes that the Apollo missions lobbed up machines that were marginally more powerful than an i386. By that I assume they mean the decidedly fixed-to-the-ground System/360 Model 91, an ultra-high-end machine at the time, later replaced by a Model 195.
NASAs on-board computer was the ingenious AGC (Apollo Guidance Computer). It's slower than an 8086.
toresbe
The fewer components you have, the less likely you are to encounter a failure. True, however, components add capability to the spacecraft. With fewer components the spacecraft is less capable and the larger the effect of failure to the whole spacecraft if one component of the spacecraft does fail.
Got an old 386 lying around gathering dust? Put it up on E-Bay and maybe NASA will pay you $50,000 for it as an onboard shuttle computer.
Your tax dollars at work...walking the streets for Uncle Sam Pimp-Daddy.
This is done from time to time. For instance, I know half a dozen years ago in one of the European Meteosat satellites, there was a spare card slot inside a computer box that went filled with a special, test-only card.
The purpose was to real-life-test an image-dedicated processor. ESA placed a specific contract to a component manufacturer (Sagem), they devised the card, added a test software that was based on the image-filtering functions that one expected in meteorology at the time (the prime contractor of Meteosat satellites was hired to design the algorithms), and the card did fly.
Now, it has to be said nowadays there is pretty much knowledge about the actual strength of components. Some manufacturers even propose rad-hardened chips, either offspring from military-related developments or even because they themselves develop spacecrafts (eg Motorola with their Iridium constellation have developped, and now market, a special space-hard version of the PowerPC)
Herve S.