The Mars Rover Has a Crappier Camera Than Your iPhone. Why? Government Sucks. (Also, Engineering Is Hard.)

For the growing number of Mars porn addicts, the quality of the images released by NASA from the newly landed Curiosity rover has been frustratingly low. To paraphrase an old saw: If we can put a rover on Mars, why can't it take pictures that are at least as good as those phone photos we all took of ourselves in the bathroom mirror last week?

Luckily, Mars rover project manager Mike Ravine is here to answer all your questions (well, not here, but at Digital Photography Review):

A number of factors led to the use of 2MP sensors in the main imaging cameras used on NASA's Mars rover, Curiosity, says the project manager responsible for their development. The slow data rates available for broadcasting images back to Earth and the team's familiarity with that family of sensors played a part, says Malin Space Science Systems' Mike Ravine, but the biggest factor was the specifications being fixed as far back as 2004. 

Got that? The whole thing has to be done according to laboriously approved government specs, and that means what happened in 2004 sticks with the team for the rest of the project, even when technology changes. 

'There's a popular belief that projects like this are going to be very advanced but there are things that mitigate against that. These designs were proposed in 2004, and you don't get to propose one specification and then go off and develop something else. 2MP with 8GB of flash [memory] didn't sound too bad in 2004. But it doesn't compare well to what you get in an iPhone today.'

I think I speak for space geeks around the world when I say: Head. Desk.

To be fair, Ravine goes on to explain that there are also good scientific reasons why the payoff from a newer, fancier camera probably wouldn't be as impressive as you might think. Two of the scarcest resources in a project like this are experience and data about how technology will respond to extreme conditions, and bandwidth to transmit information back to Earth. So using old camera tech that the engineers were already familiar with was useful. And with photo transmission capacity of just 250 megabits per day, there's only so much we were going to be able to see anyway.

And the rover does have the capacity to produce the most iPhone-like picture of all. A semi-misleading photo of itself, shot from a flatteringly elevated angle:

Still, James Cameron is pissed.

Comments

[kinnath]

I think I speak for space geeks around the world when I say: Head. Desk.

As a practicing engineer, I can say you need to get a real life.

[Tman]

Seriously, NASA has some serious flaws but the cameras on the Curiosity Rover are not one of them. They haven't even started transmitting the shots from the Mastcams yet, so give it a rest already.

Sorry KMW, this is a terrible post.

[CE]

The first pictures coming back were from the really low resolution hazard avoidance cameras, so it's not surprising that the quality was low. These are the pictures people were laughing at for being so primitive.

The hi-res pictures have started to arrive now, but even those are well behind current commercial technology, since the project was designed years ago and has to survive in extreme conditions.

[yonemoto]

Not just that, but the images are transmitted via the deep space network, which has a data bandwidth limited by relays on probes that were launched even before 2004.

[nova3930]

As another practicing engineer, I hate it when people write about these topics without knowing a thing in the world about them.

When you're building something that can't be repaired once it's out of the lab, rule #1 in designing that something is that it has to work first time everytime which generally necessitates generating a specification and sticking with it so you can sufficiently integrate the components.

Changing equipment repeatedly as technology changes is a great way to get something that just won't work when the time comes because you didn't have sufficient time to integrate and test the total system.

[Tman]

Also, NASA should have told Cameron to FUCK RIGHT OFF. 3D cameras are not as cool as he thinks they are.

And even a good 3D camera can't make up for terrible writing and gigantic plot holes.

I'm extremely pleased with NASA knowing that Cameron was not able to add anything to Curiosity.

[kinnath]

This is one of those cases where I know someone who knows someone who knows someone who destroyed a bunch of cameras during environmental testing to find one that would work on the surface of another fucking planet.

Instead of bitch about 2 megapixels, we should be applauding a camera that actually works.

[James Anderson Merritt]

This^^^ (Kinnath's post of 8.9.12@3:26PM)

and this vvv

"Two of the scarcest resources in a project like this are experience and data about how technology will respond to extreme conditions, and bandwidth to transmit information back to Earth. So using old camera tech that the engineers were already familiar with was useful. And with photo transmission capacity of just 250 megabits per day, there's only so much we were going to be able to see anyway."

[Auric Demonocles]

Being a software guy, I'm aware of changing specs being a pain in the ass, but they probably should have been a bit more forward thinking with their initial specs.

[kinnath]

As KMW points out, the limiting factor is bandwidth on the return pipe. You can take a couple of super high resolution photos or you can take lots of medium resolution photos.

So a 2 megapixel camera with 8 GB local storage and at 250 megabit per DAY data channel seems like an appropriately designed system.

By the way 3 kb/s seems like a pretty fast pipe for an RF link to Mars.

[Auric Demonocles]

I think that medium res photos would be the default, but it's stupid not to have the option of higher res photos for images of interest. Just because you can take a 10 megapixel image doesn't mean you have to (or should) for every one.

[kinnath]

Circuit boards, processors, memory, sensors, optics, etc -- you have to decide the most likely, most desireable operation of the system and then design the lowest weight, lowest power, etc package that fulfils that most likely, most desireable operation.

You don't put in a 10 megapixel camera to take 90% of you photos at 2 megapixel resolution because that changes all the other components in the system too.

[Stilgar]

Other than being nicer to look at, is there anything that the white coats are going to see in a 10MP photo that they wouldn't in a 2? I doubt it.

Nasa probably has also purchased a radiation hardened sensor which is not run of the mill consumer off the shelf. They have probably also done extensive lab testing to see how the part holds up in the various extreme environments it has and will yet go through.

In addition to a very limited uplink, power requirements are probably also a small consideration. Just like a leaky faucet, even drawing milliamps adds up.

[neoteny]

3 kb/s

3 kbit/s you meant.

[kinnath]

The com guys have drilled into my head that 3 kbps is 3 kilobits per second. So I screwed up and put in / instead of p.

[neoteny]

Right.

[CE]

They mean the same thing.

[SKR]

That's what I was thinking too. Small b is bit and big B is byte, right?

[kinnath]

I know

[Jerry on the road]

My guess is that they opted for a camera chip for which the process technology was well known.

[Auric Demonocles]

That would be the most reasonable explanation. Which makes me assume that it's not the truth since this was government employees.

[HazelMeade]

The standard procedure is it has to be flight-tested and radiation hardened.

My guess is not a lot of people make radiation-hardened camera chips, which limits the selection.

However, NASA's requirement that things be flight-tested first (someone else has to have used it in space before), is inherently limiting. That's why NASA never flies cutting edge technology. It always flies stuff that's at least 10 years old.

Not that there aren't valid reason for that, but there are somethings that are reliable enough you shouldn't have to coerce someone into flight-testing the unit.

[kinnath]

It is my understanding that they selected a camera that passed environmental testing.

[The Immaculate Trouser]

NASA has a highly risk-averse culture, both because it is a government agency and because its high-profile missions have fallen short in the public eye. A conservative design is slightly disappointing, but better than a failure and not entirely unexpected.

Hopefully as time progresses, either NASA or a private corp will be more adventurous in design, but I can't fault them too much for having a functional, if conservative, design.

[CE]

I'm still waiting to see live hi-res video as a rover lands on the surface, without waiting for NASA to edit out the Martian plants and wildlife.

[Loki]

Changing equipment repeatedly as technology changes is a great way to get something that just won't work when the time comes because you didn't have sufficient time to integrate and test the total system.

This. People don't seem to realize how much integration and testing goes on behind the scenes during the development process of spacecraft. Changing the cameras would have necessitated completely re-testing the new cameras to ensure they would survive the launch environment, the space environment on the way to Mars, re-entry, and still work once you got there. Not to mention some of the effects that might have on other parts of the overall design.

[Mr Whipple]

Well, after the Hubble fuck up, what do you expect?

[The Immaculate Trouser]

Came here to say just that.

Different device, different purpose; therefore, different design. Changing the rover at last second would have been costly and unnecessary; no good engineer would advocate for a complete redesign of a one-shot rover for a minor update.

[Episiarch]

It pains me to defend these people even a little, but you need to remember that they are designing something that has to operate a gagillion miles away by itself. Any bugs, anything goes wrong, and the mission is fucked and all that money is down the drain. Things have to work, there can be no bugs, no unanticipated problems. They can't just push out a patch (or maybe they can?) or replace a piece of hardware that suddenly starts malfunctioning. It's a completely different situation from most terrestrial engineering and software coding.

They probably exacerbate the ramifications of this situation, but it's still going to have some serious effects on specs and planning.

[Hugh Akston]

That money was down the drain at T- 0. Whether it works or not doesn't matter.

[Episiarch]

That is a different discussion.

[Hugh Akston]

Well now we're discussing it, so who wins? I'll give you a hint: it's me.

[Episiarch]

"You win again, gravity!"

[Auric Demonocles]

That only came up because you watched the episodes on before the new one.

Which is also why I went straight to snusnu in this morning's thread.

[tarran]

Also, the semiconductors have to survive the nasty radiation flux outside the protective cocoon of the Earth's magnetic field.

The tinier the semiconductors, the more likely they are to be ko'ed by some high energy proton.

[Tim]

Them high energy croutons is bad.

[Pro Libertate]

No argument here, though I will note that if, say, a few astronauts were walking around on Mars, we could've changed at least a couple of cameras close to the last minute before they left.

Musk is throwing out fifteen years as a possibility for getting people to Mars. Yeah, maybe he's wrong, but don't bet the house on it. NASA has been unable to leave LEO with humans on board in 40 years. Forty years.

[Tim]

Forty Ears?

[SIV]

20 Eyes

[Auric Demonocles]

NASA has been unable to leave LEO with humans on board in 40 years.

My friends from grad school all think I'm insane because this is my logic for not giving a shit about the shuttle program being shut down.

[Pro Libertate]

Yeah, LEO is for pussies. You didn't see Armstrong hanging out there for long.

[Auric Demonocles]

Also, where have all the women who have been in space hung out?

[Pro Libertate]

Male hegemony keeps women in LEO, not choice.

[Auric Demonocles]

This seems to contradict your "LEO is for pussies" theory from above.

[Pro Libertate]

No, I'm male.

[Auric Demonocles]

Ah, I see. You weren't describing a fact, you were outlining a policy goal. An admirable goal, at that.

[Pro Libertate]

It's all instinct. I have no choice but to oppress women.

[Loki]

They can't just push out a patch (or maybe they can?)

In some cases they can. Assuming the flight SW resides on a reprogrammable EEPROM that can be commanded to be re-written they can and have patched flight SW on spacecraft before. They don't like doing it, because one thing goes wrong, like lose contact with the spacecraft in mid-upload, and you're fucked, but it can be done.

[Pro Libertate]

Also, aliens.

[CE]

Yeah, they might re-program your probe and send it back to Earth to destroy you.

[Pro Libertate]

ERROR! ERROR! STERILIZE!

[The Derider]

Yeah, I see zero evidence for the "government sucks" proposition here, and a lot of evidence for "engineering is hard".

[Episiarch]

Fuck off, joe, you little pissant coward.

[Warty]

But he's said he's not joe. Why would he lie?

[Episiarch]

Is that lying, or moving the goalposts?

[Warty]

I don't need to dignify your wild allegations with a response. I'm too busy presenting evidence and winning all the debates.

[Hugh Akston]

Come on you guys, joe said he would never come back here, and you know he would never lie about that. Plus he's way too busy being a lawyer (or was it a city planner?) to bother trying to argue with us libertards.

[T]

Urban planner. I dimly recall he had some interesting points about planning as it actually works once, but felt compelled to justify the worst excesses of the system because democracy or something.

[Trespassers W]

FACT PWNED

[Lurker Kurt]

I also do not believe this is Joe. I believe I once saw The Derider admit he was wrong. Joe never did this.

[Mike M.]

That's the rover? It looks more like an overshot shot of Myst Island.

[np]

terrestrial OT:

http://www.tomshardware.com/ne.....16620.html

Over the past couple of years 3D printing has become more and more impressive, capable of quickly and efficiently creating a large range of objects. But one professor from the University of Southern California has dared to dream even bigger, developing a 3D printing system that could effectively print an entire home in less than a full day.

TEDxOjai - Behrokh Khoshnevis - Contour Crafting: Automated Construction

Called Contour Crafting, the process involves utilizing a gigantic 3D printer that is placed overhead an empty lot where the home will be built. The machine builds walls with multiple layers of concrete, adding plumbing and electrical wiring as it goes and eventually leaves a complete home that only needs doors and windows to complete.

[The Hammer]

Could it print out a Mars Rover with a better camera? Maybe a 3d camera?

[Auric Demonocles]

I like the idea. Send a 3D printer to Mars with enough supplies to print the stuff to gather/make new supplies. Then print some sweet exploratory equipment. And more 3D printers.

[Res Publica Americana]

And more American flags, for fuck's sake. The one we left there has FADED.

[Auric Demonocles]

We'll eventually cover the surface with 3D printers and American flags.

[The Hammer]

So does a 3D printer qualify as a Von Neumann machine?

[Pro Libertate]

If they're independent and self-replicating, sure.

[Auric Demonocles]

Depending on its resolution and the specs that it has, yes.

[CE]

Why not just print out a whole new planet, closer to Earth and more hospitable?

[Pro Libertate]

We can only print planets in black and white with the current state of the art.

[Tim]

I've seen photos: the houses look like a giant first grader made them with soggy clay.

[CE]

The singularity is near.

[Tonio]

I think I speak for space geeks around the world when I say: Head. Desk.

Yes, but not in the way you think, Kitty. You're really, really out of your element here. Please give up the space program beat to prevent further embarrassment to Reason (and yourself, of course).

Kinnath and Epi have valid points which an honest writer would take to heart. Also, COTS hardware such as an iPhone is not radiation hardened and has a nasty tendency to fry in the deep space environment; surprised that a space expert such as you missed that.

[JW]

And the rover does have the capacity to produce the most iPhone-like picture of all. A semi-misleading photo of itself, shot from a flatteringly elevated angle

I've never seen a robot do duck lips before.

[Almanian's Evil Twin]

Man - you gotta get OUT more! Robots be doin' duck lips EVERYWHERE these days!

"Self shot and HOT", said the robot.

[HazelMeade]

I think it's wearing a push-up bra.

[Agammamon]

Yo, fuck James Cameron.

[Brian Sorgatz]

Slightly off topic: Thank you, Katherine, for writing "laboriously approved," not "laboriously-approved." I'm annoyed by the trend of attaching adverbs to adjectives with an unnecessary hyphen. With the possible exception of well, no adverb requires this.

[The Hammer]

Well-said?

[LTC(ret) John]

Well met, indeed.

[Tim]

Wall-Mart?

[Trespassers W]

Tractor pulls?

[Res Publica Americana]

I was under the impression that it was actually grammatically erroneous to do that, not that it was just a matter of style.

Are you saying all that I've known in life is a LIE?!

[Spoonman.]

This post is bad and you should feel bad.

[Tim]

Girls should be writing about fashion and ponies, not mars landers.

[Tonio]

Spoonman is right.

No, Tim, it's not about "girls," it's about competence. And the credibility of Reason and the libertarian movement.

There are lots of valid critiques of a goverment funded space exploration, but this is not one of them.

[Tim]

Don't make me release the Kraaken Tonio.

[RBS]

And the credibility of Reason and the libertarian movement.

Really? One post about the Mars Rover and the entire movement is destroyed?

[Citizen Nothing]

This is why nobody takes libertarians seriously.

[Citizen Nothing]

Also, I denounce the Mars Rover.

[Apatheist ?_??]

Nu-uh the usage of "cunt" is why nobody takes libertarians seriously you stupid cunt.

[Spoonman.]

Has nothing to do with her sex, just that if you refute your point in your post, you should Ctrl+A+Del.

[LTC(ret) John]

Maybe bext time they could duct-tape the latest iPhone, wrapped in Tyvek to the lander?

[Res Publica Americana]

"I'm Katherine Mangu-Ward, Captain of the Federation Starship _________."

Any candidates for the name?

[Tim]

Free Enterprise

[Generic Stranger]

Win

[Almanian's Evil Twin]

Free Willy

[neoteny]

No Free Lunch.

[Jesus H. Christ]

"Pussy Riot"

[rac]

All you "practicing" engineers, kiss my ass. They didn't not put a better camera because they were "generating a specification and sticking with it so you can sufficiently integrate the components". they didn't put it in because to change the specification would have fucked with their compartmentalized "practicing engineer" minds. Thus, launch when O-rings are frozen...etc.

[Tim]

If I am going to drive my car over a bridge, I want a bridge designed by an engineer.

[rac]

Two words, Tacoma Narrows.

[Tim]

That's four words.

[rac]

You must be a practicing engineer.

[CE]

Like this one?

http://en.wikipedia.org/wiki/I.....ver_bridge

[sarcasmic]

they didn't put it in because to change the specification would have fucked with their compartmentalized "practicing engineer" minds

I doubt that.

Likely the spec didn't change because attempting to do so requires formal submitting of the request to bureaucrats who know jack shit about engineering, who then refuse to grant any changes because if something went wrong they would be held responsible.

[Res Publica Americana]

But government aerospace is badass. MSNBC told me so.

[kinnath]

See above -- the pipe is the limiting factor.

[sarcasmic]

I'm just musing from personal experience at dealing with government specifications.

[Tim]

Usually what happens is that the engineer specifies, say, stainless steel pins and then somewhere an MBA decides that aluminum is lighter and cheaper, and then Vinny from from the pipe fitters local 78 installs wooden dowels and charges for aluminum.

[The Hammer]

They were too cheap to spring for Hughesnet?

[HazelMeade]

Or because changing the spec means you have to go through the entire massive testing process again, because that's what the "process" is.

It's a very valid argument. The whole government procurement has these burdensome and rather stringently enforced procedures so that if you change something, you have to reset it at every level, component, subsystem, systemtests, fuck all. Thus it is essentially impossible to make changes to the design late in the process, even if a rational engineer could say "this component works exactly like the previous component and fits all of our requirements, but has twice the megapixels."

It ought to be possible to just retest the interface between the component and the subsystem it interacts with, but unfortunately the process often prohibits that sort of thing.

[Loki]

^^THIS^^

[kinnath]

Thousands of people at dozens of companies and agencies working to deliver a functioning system to another planet -- the entire point of the process is to strangle innovation.

[kinnath]

fuck off

[rac]

And astronauts die.

[kinnath]

Again fuck off

[Auric Demonocles]

Using systems engineering principles is a big part of the reason Spirit and Opportunity lasted so much longer than the initial plans called for.

[Auric Demonocles]

Which is not to say you should never change specs. Just that you should be careful about doing it, and do a thorough eval of the impact.

[Pro Libertate]

Fuck that--just wing it with some duct tape. The only reason all of these hardcore specs are needed is that it costs so danged much to get to orbit, let alone to other planets.

In the future, planetary science will be done with homeless crack addicts/astronauts.

[Auric Demonocles]

They could really solve the spec problem by just launching a shitload of rovers at once.

[Pro Libertate]

Exactly. Or just stranding a colony of convicts there. Like a hundred thousand or so. With regular supply ships, with their manifests based on how many pictures and other data the convicts provide.

[Auric Demonocles]

Ooooh, only 1 tb of data this time? I guess I'll have to keep these new power supply units.

[Pro Libertate]

That's right. A prison run by scientists. It's never been tried before, which means it has to be done.

[RBS]

I thought that was the plot of Alien Resurrection.

[Pro Libertate]

That's fiction, believe it or not.

[RBS]

Look, you can't expect people to actually read anything before commenting.

[rac]

Stop making fun of Australians.

[HazelMeade]

Yeah, people have had this idea of "swarm" robots that self-organize to explore terrain. I'd love to see someone do something like that.
That way you don't have to design them to function perfectly with a 100% guarentee. If a few break, you still have a bunch more.

[Pro Libertate]

Look, I've proposed a mechanism for exploring Mars and preparing it for habitation, using sentient, self-organizing devices that we have already in large supply. Building robots is a waste of money.

To reduce the moral horror of stranding thousands of people on Mars, I suggest we just send those convicted of capital crimes. If it turns out any of them were unfairly convicted, we'll bring them back.

[rac]

Why do you hate Australians so much?

[SugarFree]

Fuck sending involuntaries. Tell the hipsters that Mars is the next Redhook and they get 100 vintage Tees for every TB of info they produce that is not in a form suitable for a one-man underground theater show or an acoustic set at a coffee shop.

[rac]

It ain't no place to raise your kids.

[Pro Libertate]

They'd be easier to manipulate than hardened criminals, true.

[CE]

Just send a 3D printer and plenty of toner, and then transmit the specs for a state-of-the-art rover after the printer lands.

[Pro Libertate]

If we could cheaply transmute elements, most of our problems would be solved.

[James Anderson Merritt]

"And all the science, I don't understand. It's just my job, five days a week..."

Do you remember the public school music textbooks that contained such labor-oriented folk-songs songs as "Erie Canal"? "Rocket Man" will be in the 22nd century versions of those books (whatever form they take). Mark my words.

[Tonio]

Sorry, rac, but you're outside your area of competence.

[rac]

I am not an engineer. But I've seen some played on TV. And, people died and bridges collapsed, etc. I'm not saying anyone expects perfection. I'm just saying that everyone on this site knows why these pictures from Mars suck. And all the gobbledygook can't cover up the fact that when you work for the government, you have to accept incompetence. That, I do know something about. The fact is, THE PICTURES COULD BE BETTER. For $2.5 billion, why not?

[Almanian's Evil Twin]

Whatever. Enjoy the dorkfest, geeks.

[Tim]

Nelson is that you?

[Res Publica Americana]

Don't fucking call me a geek. Do you wanna get knifed? DO YOU WANNA GET KNIFED?

[Pro Libertate]

I play the Ratchet Clank PS3 games with my daughter. In one of those, there's a big-brained alien that warns Ratchet prior to attack, "I'll have you know that I'm a level 60 wizard with melee abilities!" Then Ratchet kicks his ass.

[Pro Libertate]

AND! Fuck!

[Loki]

AND! Fuck!

You fuck your daughter too? You sick bastard.

[Pro Libertate]

Ye gods. I was talking about the ampersand.

[darius404]

DO YOU WANNA GET KNIFED?

Is that code for something?.... Nevermind, I don't want to know.

[Jesus H. Christ]

Oohh, bitchy engineers. I think the article makes a good case of why the camera quality is what it is, without going into a lot of detail. I imagine there are many people out there who wonder why the pics are so lo-fi, and were probably hoping for better images. Perhaps "space geek" should have been "space tech newbies" or some such, so as not to imply a level of expertise that would get the engineer's panties in a bunch.

[kinnath]

No KMW again displays here utter ignorance of how any technical project really works. This is not the first stupid post in her career at Reason.

[The Hammer]

And Jesus was saying that it's really not a stupid post, you're just a dick.

[kinnath]

Actually it is a stupid post. The fact that you don't understand the technical reasons why is not important.

[John]

I still don't get why they are not in color. And who cares about slow data transfer. The thing is up there twenty four seven. Does it really matter if it takes a while to download a totally kick ass high resolution color photo? It is not like it is going anywhere.

[Tim]

It is going somewhere John, it's a rover.

[HazelMeade]

There will be color photos coming.

They just wanted to get the FIRST photo quickly.

[Almanian's Evil Twin]

They just wanted to get the FIRST photo quickly.

So Mars photos are like H 'n' R commenters: "FRISTED!!"

[HazelMeade]

Absolutely. NASA is all aout FIRST!

[Tman]

John, to answer your question, they will have full res color pics eventually.

Here's the breakdown of the cameras-

http://www.nasa.gov/images/con.....-4-673.jpg

There are several cameras on the rover, but the only ones they've used before today were the NavCams and the Hazcams. Today they released a low res shot from the Mast cams, and eventually we will get full res shots from the MastCams.

Don't worry, we'll get some cool shots soon enough.

[CE]

It would have been nice if the article had explained this.

[Loki]

It is not like it is going anywhere.

The Earth does, as does Mars. The ground stations on Earth are only in contact with the rover for finite amounts of time. Plus IIRC the rover uses the MRO and Mars Odyssey satellites to relay the data back to Earth. Also, the bandwidth available isn't just used to send back pretty pictures. There's other payload data from the other instruments as well as rover telemetry. So it's not really plausible to have a 10 MP camera on board.

[Pound. Head. On. Desk.]

Head. Desk.

Can't. Quite. Place. Those. Two. Words.

[HazelMeade]

Do you people have any idea how long it would take to download a high-resolution color photo from Mars at available data rates?

We're talking about a planet that is 40 light-seconds away.

[Pro Libertate]

We should run a T-3 between here and Mars.

[Joe R.]

You didn't build that.

[Pro Libertate]

Well, whoever builds things should build that. I suppose there would have to be some sort of relay sitting way outside the ecliptic to keep the cable from running into the sun. Unless we can sunproof the cable or something.

[Generic Stranger]

It wouldn't change the transmission rate, even if we could do it. The limiting factor is the speed of light, which cables don't overcome.

[Generic Stranger]

Err, strike that. For some reason I got latency and bandwidth confused.

[Citizen Nothing]

Is that less than it takes to make the Kessel Run?

[DaveSs]

Quite a bit more than 40 light seconds.

Its about 14 light minutes now, and increasing slightly each day.

[Pro Libertate]

Yeah, I meant to mention that. I recall a 30-minute turnaround in communications was faked in Capricorn One. Yes, that's where I got all of my space program facts.

[HazelMeade]

Maybe I was thinking about the moon.

[DaveSs]

That would be closer to 3 light seconds.

[Pro Libertate]

That's no moon.

[Trespassers W]

I like Star Trek.

[Pro Libertate]

I thought it was Star Track.

[Fluffy]

I guess my problem is I fundamentally don't understand how the distance matters.

Once the signal is sent, it doesn't take any additional power to send it further. It just keeps going at the speed of light. So shouldn't it take the same amount of power to send a signal to Mars as it takes to send a signal across my front yard? Is signal degradation over such a short distance that big a problem?

[HazelMeade]

The signal is not a laser. Power delines with the cubed root of distance.

[HazelMeade]

Er inverse cube.

[kinnath]

Yes, inverse cube law.

Directional antennas help to focus the beams, but that's a really long fucking way to broadcast a signal.

[HazelMeade]

Basically the energy is spreading out in a volume around the origin.

Well, I;m guessing they probably use directional antennas so it may not be that bad, but you still lose power over distance, because the signal is never going to be focused precisely on the destination, unless it's a laser. That's why you need a very big antenna at the other end. (Good luck trying to get a laser to point at a specific spot on Mars).

[kinnath]

Drop a pebble in a lake. A wave radiates out in all directions (making an ever growing circle). The length of the wave (the circumference of the circle) grows as a function of r-squared (the distance from where the pebble hit the water). The same amount of energy is spread over the ever increasing circumference of the circle.

The farther from the origin of the wave that you sample the wave, the less power you will receive.

Broadcast a signal in space, you get an ever growing sphere. Power at any given point in the sphere drops by r-cubed.

[kinnath]

I shouldn't have listened to Hazel. The wave in the lake grow by a factor of r (2 pi r). The surface of a sphere grows by r-squared. So the power of a radio wave drops by r-squared.

volume would be r-cubed.

[CE]

So increase the spec on the data rate.

[Fluffy]

Can one of you engineer guys help me out and explain the 250mb data pipe?

Is it a power thing?

Because that seems pretty low to me.

They've got plutonium batteries on the thing. I realize Mars is far away, but there's also no magnetic field there, no competing radio sources - and we can come up with receiving stations as sensitive as we need.

Me no understand lectricitie good.

[Pro Libertate]

Resistance is futile.

[Citizen Nothing]

Voltage/Current is futile? Wow. It all becomes clear.

[Pro Libertate]

Precisely. I was worried I'd have to explain, but I forgot where I was commenting.

[Citizen Nothing]

A magical place where even journalists know about Ohm's Law.

[kinnath]

Data rate is a function of radio frequency

Power is a function of radio frequency

Higher data rate requires higher radio frequency consumes higher power.

And as Hazel points out, there is a budget. There is other data beside pictures to transmit.

[CE]

But the taxpayers just want to see the pictures. In hi res. And color. And video.

[DaveSs]

Power and distance factor into a low speed data rate.

You also have
Limited availability of the large dishes necessary to receive the transmission (the Deep Space Network supports a lot of different missions and IIRC has only three sites)

Mars rotates meaning the rover can't be seen from Earth for ~12hrs 15m at a time. Can't see the rover, it can't send you data.

The rover can only see satellites orbiting Mars for a few hours each day, so it can't always use them as relay stations either.

[DaveSs]

Correction: the rover can only see the satellites for 7 to 8 minutes a day.

[Citizen Nothing]

Seems to me that the engineers on this board are a lot of excuse-making whiners. "We couldn't put a better camera on the rover. It would be too hard. Waaah."
Remind me not to hire any of you to build my death ray.

[RBS]

You can hire me. I have zero engineering background so I won't worry about all this nonsense. As an added bonus I have interns/clerks during the summer to use as test subjects.

[SugarFree]

I'll hire them to build my death ray, knowing that I'll be happy when all I get out of it is a death handgun.

This balance of terror just. got. personal.

[Stormy Dragon]

You can't just put a normal consumer camera on a rover. Between the vacuum and the radiation it was exposed to on the way to Mars, your iphone's camera would be dead by the time it got there.

[CE]

I'm betting an iPhone would still work, based on how I've treated mine.

[Stormy Dragon]

An IBM PowerPC 601 processor costs $500
An IBM RAD6000 (the radiation hardened version of the 601) costs $300,000

Commercial satellite manufacturers wouldn't be paying a 60,000% markup if the consumer version would still work in space.

[DaveSs]

One thing to keep in mind is that a lot of the huge images you may have seen from the Spirit and Opportunity rovers are actually dozens (or hundreds) of images taken with a long focal length lens that are stitched together back here on Earth into one giant image.

So, even though its only a 2MP camera, the fact that they have a 100mm lens on one of the cameras means gigantic high resolution images are very achievable.

[CE]

And they also doctor the data to avoid showing the blue sky on Mars, because people expect it to be red.

[Joe R.]

We should have just sent Ansel Adams, except that he's dead.

[Jesus H. Christ]

This administration would probably send Jackson Pollock, and he is also dead.

[Proprietist]

Ok, while I agree with those here saying that they can't just swap components mid-testing because technology advanced, KMW's point still stands, as this camera quality is pretty awful based on digital imaging technology that existed in 2004. We already were planning to spend exorbitant money on this mission, they surely could have invested in more state-of-the-art photo quality to make the expense of the trip more worthwhile?

[kinnath]

The classic engineer's nightmare is to design a complex system to meet all the stated requirements for the system, to complete the project on schedule, and to have the system work properly when delivered . . . and then the first user says "Why didn't you do such and such?"

The problem is you don't get to beat the bastard to death while shouting "Because it wasn't part of the fucking requirements!"

[kinnath]

We already were planning to spend exorbitant money on this mission, they surely could have invested in more state-of-the-art photo quality to make the expense of the trip more worthwhile? so it was critical to make sure the fucking system worked when it got there!

[Metazoan]

Maybe the iPhone isn't rad-hard? Same reason they use ridiculously expensive and "slow" by our standard rad-hard processors on commercial satellites.

[fearsomepirate]

You know who's an even bigger space geek than you? Every single engineer who works at NASA. There is a lot of bureaucratic BS that stops the space program from running efficiently. You could talk about it all day, from the Senate Launch System to the white elephant known as the International Space Station. Putting a 2MP camera on a rover bound for Mars isn't one of them.

[Armed Texan]

This is typical government at work.

1) Flash Memory: Replacing 8GB of flash with 32-128GB (or more) is easy. The newer, larger capacity flash adds nothing in weight or complication to the design.

2) Photo Sensor: If the design relied on one single sensor working without flaw and not being damaged through 4,152 kN of thrust, months in vacuum, months of bombardment by solar winds and radiation, and a landing sequence that would make Rube Goldberg wince, then the design was flawed. There should be two or more sensors with a higher capacity sensor as the primary and a less finicky back up.

3) Bandwidth: This is typical government thinking along the lines of freeways. When confronted with congested freeways, government planners often choose to lower speed limits, raise emission standards, and dedicate lanes for HOV when what is needed is more lanes or new routes. Being able to move between two points faster is the best way to increase safety and reduce emissions, but that is not how government planners think. And here we have the same issue. When confronted with limited bandwidth, the answer is...take away performance! Why are we limited to 250Mb per day? Would the FCC not give permits to use more spectrum because it might interfere with a radio station playing muzak on the moon? Its the receiving satellites launched before 2004 you say? That's like designing a new car with modern components and carbon fiber, but using wooden spoke wheels with hard rubber.