3 ms·

Does anyone know why they decided to go with just a 4 megapixel camera? I guess they could settle for taking more narrow images and recombining later but it se

by lqdc13 11y ago

Does anyone know why they decided to go with just a 4 megapixel camera? I guess they could settle for taking more narrow images and recombining later but it seems better to take one image and crop if they are just interested in small parts. Plus you have the option for digital zoom.

coldtea 11y ago
>Does anyone know why they decided to go with just a 4 megapixel camera? I'd say because they have several other constraints the necessitate that, not because a 50MP wouldn't do. E.g. resistant to cosmic radiation, able to cope with the light extremes it needs to, etc. Here they have some other factors I didn't consider: http://www.extremetech.com/extreme/134239-why-does-the-2-5-billion-curiosity-use-a-2-megapixel-camera http://www.extremetech.com/extreme/134239-why-does-the-2-5-b...
simias 11y ago
Those are not iphone cameras, they need to work under extreme conditions and for a very long time. The camera must also have been chosen years ago.
Cogito 11y ago
It's also worth noting that the camera is quite old. The satellite was built in the late 90s, and was flown more-or-less unchanged when it eventually took to the sky. So not only do you need to account for the space hardening that has gone into the technology (it was designed to keep working through the middle of a solar storm), you also have to take into account the state of the art over 15 years ago.
RogerL 11y ago
They are not even using the 4MB as it is. It takes time to download an image, and apparently they are more interested in what they call 'downlink cadence' - number of images per hour. I don't know enough about the science to say what the desired resolution would be - meaning, would a 50MB sensor actually produce better science for the mission of the DSCOVR (clearly it could be used for other science once you get the resolution good enough). The specifics are here: http://www.nesdis.noaa.gov/DSCOVR/pdf/DSCOVR%20-%20EPIC%20Instrument%20Info%20Sheet.pdf http://www.nesdis.noaa.gov/DSCOVR/pdf/DSCOVR%20-%20EPIC%20In... Also, DSCOVR was initially built for the Triana mission in the late 90s, and it's main purpose was just to serve live-ish images on the internet. (https://en.wikipedia.org/wiki/Deep_Space_Climate_Observatory https://en.wikipedia.org/wiki/Deep_Space_Climate_Observatory)
kjs3 11y ago
It can take a decade to build a satellite. And you have to fix the design before you build it. And you don't usually fly the absolute latest toy; you fly something that's proven reliable. So by the time it's on station, a satellite might be flying 10-20 year old tech.

by lqdc13 11y ago

coldtea 11y ago

>Does anyone know why they decided to go with just a 4 megapixel camera? I'd say because they have several other constraints the necessitate that, not because a 50MP wouldn't do. E.g. resistant to cosmic radiation, able to cope with the light extremes it needs to, etc. Here they have some other factors I didn't consider: http://www.extremetech.com/extreme/134239-why-does-the-2-5-billion-curiosity-use-a-2-megapixel-camera http://www.extremetech.com/extreme/134239-why-does-the-2-5-b...

simias 11y ago

Those are not iphone cameras, they need to work under extreme conditions and for a very long time. The camera must also have been chosen years ago.

Cogito 11y ago

It's also worth noting that the camera is quite old. The satellite was built in the late 90s, and was flown more-or-less unchanged when it eventually took to the sky. So not only do you need to account for the space hardening that has gone into the technology (it was designed to keep working through the middle of a solar storm), you also have to take into account the state of the art over 15 years ago.

RogerL 11y ago

They are not even using the 4MB as it is. It takes time to download an image, and apparently they are more interested in what they call 'downlink cadence' - number of images per hour. I don't know enough about the science to say what the desired resolution would be - meaning, would a 50MB sensor actually produce better science for the mission of the DSCOVR (clearly it could be used for other science once you get the resolution good enough). The specifics are here: http://www.nesdis.noaa.gov/DSCOVR/pdf/DSCOVR%20-%20EPIC%20Instrument%20Info%20Sheet.pdf http://www.nesdis.noaa.gov/DSCOVR/pdf/DSCOVR%20-%20EPIC%20In... Also, DSCOVR was initially built for the Triana mission in the late 90s, and it's main purpose was just to serve live-ish images on the internet. (https://en.wikipedia.org/wiki/Deep_Space_Climate_Observatory https://en.wikipedia.org/wiki/Deep_Space_Climate_Observatory)

kjs3 11y ago

It can take a decade to build a satellite. And you have to fix the design before you build it. And you don't usually fly the absolute latest toy; you fly something that's proven reliable. So by the time it's on station, a satellite might be flying 10-20 year old tech.