DSS-what am I doing wrong?

[David JM]

I'm struggling with DSS.

The following are the 30s ISO 12800 images from last night. Left is a single pic from the camera, middle is the stack [17 subs] in RegiStax 5 and right is the same stack with darks, flats, biases in DSS.

Any ideas what I am doing wrong? I'm using the jpeg format from the camera, not raw.

[Zakalwe]

You wont need to use bias files...the data is included in the darks. In effect you are subtracting the noise twice, which means that you end up adding noise to the image.

Also 30 seconds is quite short, especially at ISO 12800. When using short exposures you might find that the read-out noise is swamping the signal. When I used to use a DSLR for imaging I found the best results were with ISO800 and lower. Any higher and I was getting loads of noise, especially on warmer nights.

Ohh, and use RAW, not JPEG. JPEG is a lossy compression file which means that some data is discarded.The camera also applies some internal processing to boost saturation, sharpening, contrast etc. RAW will give you the "true" data from the sensor, which is what you want.

[Shibby]

Also be aware that the right-hand image has been stretched, which is why the noise is more obvious. DSS does this automatically in the preview, but it's best to always save "without adjustments applied" then post-process yourself.

[David JM]

Also be aware that the right-hand image has been stretched, which is why the noise is more obvious. DSS does this automatically in the preview, but it's best to always save "without adjustments applied" then post-process yourself.

That wasn't a preview, it was the final save in DSS.

[Shibby]

...with adjustments applied?

[David JM]

IU can't seem to find any settings for the adjustments, where is it?

[Jannis]

First of all, that't not a bad result, so i would absolutely say you're doing good so far.

But like mentiond, always use raw, Always. Once you get used to it, and see the quality difference, you'll never use JPG again for almost anything...

Also, i would try not to shoot at that high ISO, as it's only a simulated ISO anyway. The 500D's sensors true ISO is maximum 3200. What you do at 6400 and 12800 is simply boosting the exposure (you can do the same in PS lightroom or canon DPP and get same result) and making it look as if it was a higher ISO. This Can be useful from time to time, but not for astro-photo. You're loosing a lot of dynamic and colours, and are not really gaining anything anyway if you're stacking the raw files. I would recommend to lower the ISO to no higher then 3200, and increase the exposure (if possible) to at least around 90-120 sec, as 30 sec is rather short exposure to capture such a faint target.

After this, i'd stack in DSS (it takes the camera raw files), then go to adjustments and increase the saturation to about 15 to get the colours back. After this, don't do much more, only save file as an uncompressed file (fits/tif) and contineue in PS, GIMP, or what you prefer. Here, start with aligning the colour channels if needed (if not already done by DSS), then start with the basics (curves and levels)

I'm no pro at this, but that's what I would have done to begin with at least.

[RikM]

DSS doesn't subtract bias twice. It uses a different stacking routine to other software packages. If you have flats, you also need the bias but you need to take enough of them to get a nice smooth result. Aim for at least 30 of each, more if you have time. The final screen in DSS is a preview with an auto stretch applied. I always choose the option to save the tiff with settings embedded but not applied. Then do all the post processing in proper editing software. RAW is much better the jpg.

Sent from my iPod touch using Tapatalk

[David JM]

Also, i would try not to shoot at that high ISO, as it's only a simulated ISO anyway. The 500D's sensors true ISO is maximum 3200. What you do at 6400 and 12800 is simply boosting the exposure (you can do the same in PS lightroom or canon DPP and get same result) and making it look as if it was a higher ISO. This Can be useful from time to time, but not for astro-photo. You're loosing a lot of dynamic and colours, and are not really gaining anything anyway if you're stacking the raw files. I would recommend to lower the ISO to no higher then 3200, and increase the exposure (if possible) to at least around 90-120 sec, as 30 sec is rather short exposure to capture such a faint target.

I thought that I had read that the camera 'grouped' pixels in the higher ISO settings so that there were only 1/4 of the pixels but they were 'bigger' in the same way that higher ISO films have larger grains to have more chance of capturing light?

[Jannis]

Some cameras have that option, it's called pixel binding. But the 500D doesn't have it as far as i know. In fact, none of the canon EOS xxD / xxxD cameraes have it as far as i know (unless the new 650D have it).

The 500D only increases the amplification of the signal that the sensor have captures to make it look brighter as if you have had a more sensitive sensor.

Correct me if i'm wrong.