For 4 years all the rovers managed to accomplish was to stuck a few times in the sand (OK - confirms there is sand there!), take pictures of dust devils and "blueberries" and provide some "tantalising" evidence.

As much as I admire the engineers for making obviously some tough little buggers this just shows inefficient robot space missions are. Bad value for money in other words.

There is no substitute for proper, manned space exploration and hopefully the futility of the rovers aimless roving will help bring some sense into the heads of some people.

VxWorks

as a few seconds on Goole would told you is called vxworks

and of course the real science they do could be found similarly.

"...Linux..."

Wiki: "The rovers run a VxWorks embedded operating system on a radiation-hardened 20 MHz RAD6000 CPU with 128 MB of DRAM with error detection and correction and 3 MB of EEPROM. Each rover also has 256 MB of flash memory."

I wouldn't be too proud of the OS. They had some issue with lack of garbage file collection in the flash memory and the first rover went completely bonkers. They barely had time to figure it out just before the second rover arrived.

I'm not expert enough to know for sure if it was exactly an OS issue, or a rover design issue, but either way the OS could have been slightly more helpful.

I think they are using Wind River's VXWorks as the RTOS

"After only a year NASA reported that Spirit's drill's teeth had worn away after grinding through five times as much rock as expected."

... in contrast with much of today's consumer electrical goods which are cost-squeezed such that they fail immediately after their warranty period or rated number of operations expires - any longer and it would be a failure of the design ...

VxWorks, actually

> There is no substitute for proper, manned space exploration

The Mars rovers aren't a substitute - they're the initial scouting party and not even vaguely futile.

yes, you're right. It would be much better if it were a manned mission so that they could play golf there and hop around like they did on the moon.

I'm not sure you're watching the same channel. In their time on Mars, the MERs have fed back a huge mass of information about its rock composition, surface structure, meteorology, and more - in short, just exactly the sort of data you need in order to plan out a human exploration mission. If you'd like the astronauts to survive the landing, not to mention the takeoff for the return journey, anyway.

Has anyone mentioned yet, they run VxWorks - not Linux?

I doubt we will see manned missions to Mars before we have a fully working space elevator. An elevator will allow putting large masses in high orbit at relatively modest cost, which will make it possible to assemble a large and heavily shielded spacecraft in orbit. Using rockets to lift things into orbit is prohibitively expensive for something of this scale.

An object lesson in how to build an extremely reliable system.

VxWorks for the low level and Java for the Command and Control

( http://www.sun.com/aboutsun/media/features/mars.html )

"their!=there"

And so? Neither of them = they're either. Who were you trying to correct? Were you trying to correct Vladimir in the post before yours? If so you're wrong.

The thing about space elevators that really brings the cost down is when you bring stuff down at the same time as taking something up - like junk or raw materials from asteroids/moon. If you only take things up, your 'haulage' costs still are pretty high.

But you are right, elevators are the way to go, a bit like going from raw meat to cooked meat for early man - saved so much time and energy and enabled man to do so many other things.

..After all, as the Chinese say "The longest journey starts with a single step". Sure, they're maybe not as good as a human (-who'd be creating at his/her union now, not seeing his wife/mistress/boyfriend/Uphill_Gardener and (maybe) kids for 3 3/4 years longer than planned)

But, like Sputnik and Laika, they've paved the way. Wasn't the earlier incarnaiton (running on a couple of AA's- what a bloody shame) called "Pathfinder"?

-Andy

The RAD6000 chips are a modified version of the chips that used to go into the IBM RS/6000s. So a manned mission is now imperative. We need an IBM CE to go there to stick on the e-Server badge, paint it black and (never forget) upgrade the firmware.

It isn't possible. Not today.

We won't be doing manned exploration of Mars for 25 years. There isn't one part of it that we know how to do. Reality is not a sci-fi movie where the director can sweep all the inconvenient stuff under the rug and just beam down for a visit.

The trip is a year out and a year back in zero-gravity. We don't know how to keep astronauts' bones and muscles from atrophying. If we sent a manned mission to Mars today, they wouldn't be able to stand up (even in Mars light gravity) when they arrived.

We don't know how to sustain such a long mission without resupply. We send tons and tons of consumable supplies to the ISS each year to support its crew. We send replacements for machines that break. We sent food and water and fuel and air. Sustaining a 2-3 year mission 60 million miles away in the same way would require boosting a ridiculous amount of mass to escape velocity. We will have to learn to make more reliable gear, and self-sutaining environments.

We have to build a self-launching return vehicle fueled from materials we find on the Martian surface. I assume here that bringing the astronauts home is a mission goal. Otherwise we have to launch from Earth enough fuel for the return mission, keep it cold for a year, and land it on Mars. No way we know how to do that today.

So, lets be glad we have the rovers. Because otherwise we'd know exactly dick about Mars. I like manned exploration way more, when it's possible. But robots work for me until we have a manned capability.

That's all nice if the rovers were meant to be the first step but they were not. They were part of faster-better-cheaper Goldin's way of NASA budget appropriation. And believe me, I admire JPL for persevering and doing a great job on the rovers. The scientists have to do things to survive the political neglect so that when the time comes they can do something really exciting.

But value for money - I still think not. Yes, it's cheap and yes, it's cheaper than it was thought to be as the rovers still manage to shake off enough dust to be able to function. But the return is cheap too.

And no, they are not good at finding a potential landing site for a manned mission. To survey a potential site with a rover you already must know where you want to look. To drop a couple of crawlers somewhere at random and hope they will stumble upon a great camping site it a bit naive - Mars is a big place.

As for human costs, well... how many people die daily in Iraq? People are cheap and plentiful. If you lose a few - there is always more where they came from. OK, you would want to train them to be useful, so there will be an accumulated cost.... But I guarantee, if you annouce a world-wide recruitment campaign for a Mars mission with a 10% probability of returning back safely you will have thousands of applicants in no time.

After all, some people are so bored they are happy to blow themselves up in public places, let alone fly to another planet.

Basically, I am not trying to dimish the achievement of the MER mission (showing how much can be done for a relatively modest amount of money). I am just trying to say that it is not the way forward (at least not a serious one).

It's great that you corrected that guy's spelling.

We do need to maintain standards after all.

BTW loose != lose

it's that the rovers run VXWorks.

Not that it matters - the entire episode will become largely meaningless once we discover our first Stargate.

...then obviously we must send Linus Torvalds.

"The rovers have already expanded our understanding of the red planet beyond expectations."

The usual IT fallacy between data and knowledge here. We have lots and lots of new facts, but most of them stand in need of a convincing integrated explanation. Barring that, it would be truer to say we have now less understanding than we thought we had before. And barring that, it will also have been a waste of time and money.

BTW, I am the only one who thinks these Rovers have such wimpy names? Why not Gog and Magog. Or Scylla and Charybdis. Or Shock and Awe. Or Bill and Steve.

"Not that it matters - the entire episode will become largely meaningless once we discover our first Stargate."

Errr - you will still need an OS to run your rovers, right (the ones you will push through the Stargates)?

@ Vladimir

"And no, they are not good at finding a potential landing site for a manned mission. To survey a potential site with a rover you already must know where you want to look. To drop a couple of crawlers somewhere at random and hope they will stumble upon a great camping site it a bit naive - Mars is a big place."

You're right, that's what the Mars Recon Orbiter did:

http://science.slashdot.org/article.pl?sid=07/10/12/1534201

"Basically, I am not trying to dimish the achievement of the MER mission (showing how much can be done for a relatively modest amount of money). I am just trying to say that it is not the way forward (at least not a serious one)."

So "the way forward" is to send up the crazies who are happy with your 10% return chance at many many many more times the cost of the rovers?

Lets be nice and say sending people to Mars only costs 10x as much that's 8.2 Billion for a 90% chance of failure....good idea!

.... yup and I bet that will be powered by Linux ;)

"Lets be nice and say sending people to Mars only costs 10x as much that's 8.2 Billion for a 90% chance of failure....good idea!"

Aaa, but I did not say 10% mission success probability, it was about 10% safe return probability. There are ways to make sure the mission will succeed even if the crew will fail to return - analyse sample in-situ, transmit the data, send the samples off by an unmanned return vehicle etc.

Those who would agree to go on such terms should not necessarily be "crazies", just ... enthusiasts.

And in perspective - what's 8.2 billion comparing to 150 or so billion a year the US spends on Iraq and Afgan wars?

And I look at that, actually, from this point - we will have to go to Mars (and incur the cost) sooner or later anyway, so playing with robots in the meantime just adds incremental costs to the whole thing.