Jump to content

Banner.jpg.b83b14cd4142fe10848741bb2a14c66b.jpg

Nikon D3200 movie crop mode?


Recommended Posts

I would like to attempt some planetary imaging using my refractor and DSLR, but have some questions about video modes. According to this article...

http://www.skyandtelescope.com/astronomy-resources/planetary-imaging-with-your-dslr-camera/

"The trick to capturing the highest-resolution, planetary imaging details with a DSLR is to use a mode that allows you to record the image off the camera’s sensor at its native pixel resolution",

i.e - 1:1 pixel ratio where the video is not down-scaled from the full sensor resolution, 6016x4000 pixels in my case.

To do this, it states that some cameras have a crop mode at 640x480, for example, that use the central 640x480 pixels instead of down-scaling the full sensor image.

My camera has a movie mode of 640x424 which appears to fit the bill, but how do I know if this is down-scaled or recorded natively using the central 640x424 pixels?

Link to comment
Share on other sites

This is a good question. I'd love to hear the answer as well.

Using 1920x1080 mode, I'd have to use a 4x barlow just to get the same image scale as I'd get from a regular 6000x4000 image.

Link to comment
Share on other sites

Take a short video of anything using normal mode and repeat using crop mode. Playing them back if the crop mode video has a much smaller FOV ( in your case about 1/10 FOV) compared to the normal video then it's genuine crop mode. If the FOV is similar then it's a scaled down full sensor image. The crop mode video if genuine  should be a small centre cut-out view of the normal video, with the same image scale.

If you have a live preview, this should show you without having to record and playback a test video.

Alan

Link to comment
Share on other sites

Thanks Alan,

I knew there'd be a simple way, just couldn't quite get it straight in my head.

I just tried it and unfortunately the 640x424 shows the same, in fact larger, fov due to 4:3 ratio, whereas the 1920x1080 just letterboxes the same fov down to 16:9.

So, down-sampling it is. Blast.

crop.jpg

Link to comment
Share on other sites

Being as the camera is down-sampling the full sensor resolution, if I were to attempt planetary imaging, would I be safe to assume the following then:

1. I will need 2" T adapter and barlow instead of the 1.25" ones I was eyeing up, to prevent vignetting?

2. I may as well use the 1280x720 @ 60fps mode to get more frames?

Also, would I use a more powerful barlow to increase image scale, or should I stick to the F20-F30 rule resulting in a smaller image scale? Scope is F7.5, so whilst a x5 barlow would give a nice image scale, it would be imaging at F42.5!

EDIT - Actually, x2.5 is the most powerful barlow in 2" format I can find anyway! However, I read that most barlows operate at higher mags with cameras attached due to spacing, so a x2.5 may give me x3.3ish.

Link to comment
Share on other sites

8 minutes ago, parallaxerr said:

Being as the camera is down-sampling the full sensor resolution, if I were to attempt planetary imaging, would I be safe to assume the following then:

1. I will need 2" T adapter and barlow instead of the 1.25" ones I was eyeing up, to prevent vignetting?

2. I may as well use the 1280x720 @ 60fps mode to get more frames?

Sounds like a plan Jon.

Don't know if Back Yard Nikon would be any help, folk use BYEOS to video planets.

Dave

Link to comment
Share on other sites

Just now, Davey-T said:

Sounds like a plan Jon.

Don't know if Back Yard Nikon would be any help, folk use BYEOS to video planets.

Dave

Good shout, I believe the Canon method is to record zoomed live view to laptop which achieves the 1:1 pixel ratio requirement. I am, however, operating sans laptop!

Link to comment
Share on other sites

Jon, Pity your crop mode video is scaled down. My EOS700D didn't have crop-video either, but luckily the Magic Lantern firmware 'upgrade' provides it. I managed fine with a 4X Powermate on a f6.3 scope with the Canon. You might find the image is too dim at f42 though.

Alan

 

Link to comment
Share on other sites

Just now, symmetal said:

You might find the image is too dim at f42 though.

Yes exactly, a little extreme I think. Unfortunately, Nikon didn't release a SDK for the D3200, so the software can't be hacked.

Realistically a 2" x2 or x2.5 @ 1280x720 & 60fps is the best I think, with actual multiplier values being a bit higher due to spacing. I should see something like this (FoV wise I mean)...

astronomy_tools_fov.thumb.png.fca6e1e62317f7f4b4b9f068f2061e2b.png

Link to comment
Share on other sites

The 12dstring site gives a bit smaller Jupiter image than your last post Jon.

As the Dawes limit of your 120mm scope is 0.97 arc-seconds and your imaging here at 0.47"/pixel which is  at the optimum nyquist sampling rate (although whether the nyquist theory applies to astro images is debatable) you shouldn't get any more detail by going to a 'bigger' barlow

JupiterFOV.png.ef9025d17674b3d97d162ef0e4e80ef6.png

Alan

Link to comment
Share on other sites

6 hours ago, symmetal said:

The 12dstring site gives a bit smaller Jupiter image than your last post Jon.

As the Dawes limit of your 120mm scope is 0.97 arc-seconds and your imaging here at 0.47"/pixel which is  at the optimum nyquist sampling rate (although whether the nyquist theory applies to astro images is debatable) you shouldn't get any more detail by going to a 'bigger' barlow

JupiterFOV.png.ef9025d17674b3d97d162ef0e4e80ef6.png

Alan

Thanks for the info Alan, so many factors to consider.

Just one question, the D3200 has 3.9um pixels not 6.1um as in your calcs above. Does this affect anything?

Cheers, Jon

Link to comment
Share on other sites

You're right Jon. I picked the wrong camera. ? The smaller pixels means that the D3200 640x480 is in effect a smaller chip size than for the D3000 so Jupiter will occupy more pixels and appear larger though you probably won't resolve more surface detail. You're now sampling at around 3 times the Dawes limit. so the finest detail resolvable will be spread over 9 pixels or so.

This is not a bad thing as you can always down-sample (effectively binning) your image to get a 'sharper' less noisy result. It could help too with having to record  in compressed video as some of the compression artifacts would be smaller than your image resolution so could be reduced by low pass filtering (gaussian blurring) your result slightly. The compression artifacts make the stacking in Autostackert or whatever you use less effective at cleaning up your image. The stacking to reduce noise assumes the noise is random but the unwanted artifacts from compression are image dependent. If you have a choice on the camera opt for the highest quality recording, less compression and a larger file size.

This is all theory and results may be different, but it's one of the things that makes this hobby interesting. ?

Alan

Link to comment
Share on other sites

1 hour ago, symmetal said:

You're right Jon. I picked the wrong camera. ? The smaller pixels means that the D3200 640x480 is in effect a smaller chip size than for the D3000 so Jupiter will occupy more pixels and appear larger though you probably won't resolve more surface detail. You're now sampling at around 3 times the Dawes limit. so the finest detail resolvable will be spread over 9 pixels or so.

This is not a bad thing as you can always down-sample (effectively binning) your image to get a 'sharper' less noisy result. It could help too with having to record  in compressed video as some of the compression artifacts would be smaller than your image resolution so could be reduced by low pass filtering (gaussian blurring) your result slightly. The compression artifacts make the stacking in Autostackert or whatever you use less effective at cleaning up your image. The stacking to reduce noise assumes the noise is random but the unwanted artifacts from compression are image dependent. If you have a choice on the camera opt for the highest quality recording, less compression and a larger file size.

This is all theory and results may be different, but it's one of the things that makes this hobby interesting. ?

Alan

Thanks for the comprehensive post Alan, really appreciate it :)

So, it looks like I don't have the "perfect" gear for imaging planets, but I'll give it a go. I managed to source a used Altair Astro 2" X2 ED Barlow and a 2" T2 nosepiece for cheap, so no big expense. I already have a stack of Altair 2" extensions of varying lengths, so I may be able to tune the barlow to get best results and prevent unnecessary oversampling.

I will bear the compression thing in mind, but may start in 720p mode because that allows 60fps as opposed to 30fps at other resolutions.

 

Link to comment
Share on other sites

According to Imaging Resource Review D3200 Video you have a choice of High/Normal video quality on all modes so you need to select 'High' which gives  24Mb/s video bit rate (less compression) on the three higher resolution modes. You get the 30 or 60 fps when you select NTSC video mode. In PAL mode you get 25 or 50fps. NTSC or PAL shouldn't affect the MOV video file quality, apart from a different number of FPS, it changes the video encoding format on the 'Video Out' socket on the camera for plugging into a PAL or NTSC TV.

At 60 fps the longest exposure possible is 1/60 sec, while 30fps can allow as long as 1/30 sec. It's best to try and use as fast a shutter speed as you can without the image being too dim to try and 'freeze' the 'seeing'. Increasing ISO to allow a faster shutter speed may be beneficial in some viewing conditions. Good luck. ?

Alan

Link to comment
Share on other sites

13 hours ago, symmetal said:

It's best to try and use as fast a shutter speed as you can without the image being too dim to try and 'freeze' the 'seeing'. Increasing ISO to allow a faster shutter speed may be beneficial in some viewing conditions

So basically do the complete opposite to everything I've learnt about DSO imaging!!! 

Just had a play with the camera, shutter speed vs iso. At 12400iso I can get to very fast shutter speeds but that's in daylight of course. Looking forward to trying it out under dark skies.

Also tried live view zoom to see if I could achieve 1:1 pixel ratio but unfortunately it defaults to no zoom when record is pressed.

Link to comment
Share on other sites

  • 2 weeks later...
On 27/05/2018 at 09:58, parallaxerr said:

So basically do the complete opposite to everything I've learnt about DSO imaging!!! 

Just had a play with the camera, shutter speed vs iso. At 12400iso I can get to very fast shutter speeds but that's in daylight of course. Looking forward to trying it out under dark skies.

Also tried live view zoom to see if I could achieve 1:1 pixel ratio but unfortunately it defaults to no zoom when record is pressed.

Hi Jon

Doing DSLR Planetary is something i dabbled in a few years back. I had one of the very few Canon models that actually had the special 'video crop' mode you mention, it was a 60D. It let me record just the central 640x480 pixels at 60fps and i got semi-decent results when paired with my C8 (see below for the best i managed on Jupiter). It is true that Magic Lantern opened this up to most models, plus BYEOS also allows a near exact 1:1 recording of the pixels, but alas these 2 options are for Canons only. And unfortunately Nikon's idea of what Live View means is very different to Canon's, so there's no such luck for using BYN in the same way as BYEOS. 

Alan has given you some great advice on minimising the compression and maximising the fps. I would also probably try the 720p mode over 1080p, for no other reason than because 1280 divides into 6016 better than 1080 does. Who knows, it might mean slightly better downsampling?!

 

Jupiter (small) v2.3.jpg

Link to comment
Share on other sites

You may be interested in Nikon Hacker. They do video hacks for the D3200 but not a crop. See here - http://simeonpilgrim.com/nikon-patch/nikon-patch.html

If you have firmware 1.03 then I would advise not to use the Beta 64 mbps. Beta releases are a gamble. Stick with the lower rates. I don't know if this is useful but you can have a look anyway.

I have used Nikon Hacker but not for the D3200. Seems ok to me,

Before you start it may be best to read this - https://nikonhacker.com/wiki/Patching_Help

Dave

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.