With a color camera each pixel has a red, green, or blue filter over it. With monochrome each pixel also has a red, green, or blue filter. It seems that the "less sensitivity" argument at 5:00 needs more explanation.
With a colour camera there is a bayer matrix over the pixels, usually with 2 x Green, 1 X Red and 1 X Blue configuration. Therefore a green photon has a 50% chance of falling on Red or Blue filter and not being recorded. For Red and Blue it's a 75% chance of not being recorded for obvious reasons. With monochrome all the photons of the colour being allowed through the filter have the opportunity to be recorded, and will be, outside of other efficiency losses.
I think in part the explanation can be found starting at 8:40 regarding the filter light throughput and the sharpness of the dropoff around each color's nanometer range.
I know its easy to get muddled with presentations, I certainly did! You need cryptic notes. It is offputting when the presenter can't remember which camera was used or which shadow fell on what. Not fair, I agree, as you clearly know your subject, but that's how it was in the Business I was in. This is a long video and you need to hold the viewing audience attention from the start as 'The Expert' who does not stumble - a very rare creature! Bob.
How to de-rotate/stack the 3 interpolated sets of R G B? 1) Make one video concatenation of the 3 videos of each channel (say R1+R2+R3) then derrotate the R_total_video in WinJupos and stack the R_total_video_derrotated? 2) Or de-rotate each of the 3 videos individually, stack one image from each set (imageR1, ImageR2, imageR3) and then de-rotate the 3 images combining in one imageR_derrotated? Thank you.
Awesome Chris! question. are you focusing on Jupiter each r g b? so how long does this take? Are you auto focusing or manual? Also what laptop would you recommend? and disk type space ?
From speaking to Chris Go at this and other talks, we can relay that he uses parfocal filters and carefully focuses for the green filter. At long focal ratios approaching f/30-40, the red and blue filters will be at the same focal plane as the green, but the green data is critical for alignment of the frames. You have to manually focus on the planet, as current autofocus software uses point sources (stars) to calculate focus, which does not work consistently on extended objects like planets. For a laptop, Chris Go recommends a fast processer with a lot of RAM and a standard (non-SSD) hard drive. In the past, SSDs could only take around 1000 re-writes before the sectors began to be corrupted, though in the last year or so there are some capable of 100,000+ re-writes that may change this recommendation. The fastest CMOS cameras can record 2-3 GB of data per second at maximum frame rate and bit depth, so a large hard drive (1 TB+) and a bigger portable drive are key to handling that much data.
Outstanding presentation Chris.
Thanks so much Christopher!
This is great! - lots of great info and advise
Is there any software that uses the GPU to process the data instead of the CPU? CUDA or the ATI programming?
Thanks a lot ! Detailed video from one of the world's leading amateur planetary imagers !
This was great.
With a color camera each pixel has a red, green, or blue filter over it. With monochrome each pixel also has a red, green, or blue filter. It seems that the "less sensitivity" argument at 5:00 needs more explanation.
With a colour camera there is a bayer matrix over the pixels, usually with 2 x Green, 1 X Red and 1 X Blue configuration. Therefore a green photon has a 50% chance of falling on Red or Blue filter and not being recorded. For Red and Blue it's a 75% chance of not being recorded for obvious reasons. With monochrome all the photons of the colour being allowed through the filter have the opportunity to be recorded, and will be, outside of other efficiency losses.
I think in part the explanation can be found starting at 8:40 regarding the filter light throughput and the sharpness of the dropoff around each color's nanometer range.
I know its easy to get muddled with presentations, I certainly did! You need cryptic notes. It is offputting when the presenter can't remember which camera was used or which shadow fell on what. Not fair, I agree, as you clearly know your subject, but that's how it was in the Business I was in. This is a long video and you need to hold the viewing audience attention from the start as 'The Expert' who does not stumble - a very rare creature! Bob.
12:52 Three cheers for sacrificing for science!
This presentation is brilliant just like Christopher's images. Thank you for uploading was very informative.
How to de-rotate/stack the 3 interpolated sets of R G B?
1) Make one video concatenation of the 3 videos of each channel (say R1+R2+R3) then derrotate the R_total_video in WinJupos and stack the R_total_video_derrotated?
2) Or de-rotate each of the 3 videos individually, stack one image from each set (imageR1, ImageR2, imageR3) and then de-rotate the 3 images combining in one imageR_derrotated?
Thank you.
Awesome Chris! question. are you focusing on Jupiter each r g b? so how long does this take? Are you auto focusing or manual? Also what laptop would you recommend? and disk type space ?
From speaking to Chris Go at this and other talks, we can relay that he uses parfocal filters and carefully focuses for the green filter. At long focal ratios approaching f/30-40, the red and blue filters will be at the same focal plane as the green, but the green data is critical for alignment of the frames. You have to manually focus on the planet, as current autofocus software uses point sources (stars) to calculate focus, which does not work consistently on extended objects like planets.
For a laptop, Chris Go recommends a fast processer with a lot of RAM and a standard (non-SSD) hard drive. In the past, SSDs could only take around 1000 re-writes before the sectors began to be corrupted, though in the last year or so there are some capable of 100,000+ re-writes that may change this recommendation. The fastest CMOS cameras can record 2-3 GB of data per second at maximum frame rate and bit depth, so a large hard drive (1 TB+) and a bigger portable drive are key to handling that much data.
Excellent presentation and lots of info to absorb. I'll be watching this again.
nice talk , if would be nice if you gave the links to all the programs products you mentioned
Is there a newer color CCD cam that has the features to overcome the advantages to use a monochrome for planetary?
Thank you very much for this informative video on Planetary Imaging.
Great video Chris !!!
Thanks for the presentation Chris!
Thanks !