How to measure Gain and Read Noise for a DSLR in Siril

Поделиться
HTML-код
  • Опубликовано: 21 ноя 2024

Комментарии • 18

  • @deepskydetail
    @deepskydetail  Год назад +3

    Craig Stark's Article on Measuring Your Camera's Gain and Read Noise: www.cloudynights.com/articles/cat/column/fishing-for-photons/signal-to-noise-part-3-measuring-your-camera-r1929

  • @ReySandu
    @ReySandu Год назад +1

    I'm glad I found your channel mate, keep up the great videos!

  • @kiscsigabee
    @kiscsigabee 7 месяцев назад

    Excellent tutorial, I have been looking for this since I saw your previous video about SNR, but I didn't read the comments, where the link to this video is, up till recently. Maybe you should put a link in the description of the SNR video, or at the end of the video. Thanks!

    • @deepskydetail
      @deepskydetail  7 месяцев назад

      Thank you! Glad it helped. Good idea to put it at the end of the video :)

  • @catpool
    @catpool Год назад +1

    Many thanks for the mathematical background!

  • @kbarth001
    @kbarth001 Год назад

    Great Video. I was looking for a while to get my Sony A7M3 Gain and Noise in e. However I think there is a small mistake in determining the variance / stddev in the flats. Before substracting the flats from each other to one flat a constand number should be added to avoid clipping the negative values to zero and cratinfg a false stdev / var of the resulting image. At least in PI which I used it is like this. Thanks again and hopefully you will do more intresting videos like this on you channel. Greetings from cloudy and rainy Geneva, Switzerland.

    • @deepskydetail
      @deepskydetail  Год назад

      Thanks! In Siril, as far as I understand it, it can have negative values after pixel math (the stdev should be the same either way). So there shouldn't be clipping. But that's good to know in pixinsight.

  • @eacron
    @eacron Год назад +1

    Thank you! This is great resource!! I’ve been trying to do this for quite some time. One question if I have a panel that is pretty bright can I vary the exposure time or does that have to remain constant? Using an ipad don’t you start seeing the bayer matrix?

    • @deepskydetail
      @deepskydetail  Год назад +1

      Thank you! I think you want to keep the exposure length constant. One way you can vary the brightness is by putting paper over the telescope opening. Start with one piece and then keep adding more to vary the brightness. Or if your aperture is bigger than paper, you could try t-shirts or something.

    • @eacron
      @eacron Год назад

      ⁠@@deepskydetailhi! I just wanted to let you know that I’ve tried determining the gain both ways - varying the exposure of the camera and sheet method you suggested - I got essentially identical results. However, I believe my result was half what I expected .2 at iso 800 and .4 at iso 400 for my Nikon d7500. I noticed that the procedure the AAVSO outlines divides the variance by 2 but does not multiply mean by 2. If I divide the variance by 2 I get the result I expect but only if I use the 2x mean column. Any thoughts?

    • @deepskydetail
      @deepskydetail  Год назад +1

      @@eacron Thanks for the follow up! Do you have a link to the procedure you used based on AAVSO so I can compare the two methods?

    • @eacron
      @eacron Год назад

      @@deepskydetail Sure. Hope this helps…www.aavso.org/sites/default/files/publications_files/dslr_manual/AAVSO_DSLR_Observing_Manual_V1-4.pdf
      Two alternate methods are outlined in Appendix A

  • @Ekuy1
    @Ekuy1 Год назад

    i followed your tutorial, but my mean flat brightness and stdev are inverse to each other. is something wrong? so my gain is "negative". my mean values range from 13000-55000 (going up) and my stdev are from 5000-2000 (going down). all my camera settings are manual, and i converted the raw .orf files to 16bit tifs in Ps, then to 32bit fits in siril. then i did exactly the same as you. do you know whats wrong?

    • @deepskydetail
      @deepskydetail  Год назад

      Are you using the same settings for each of the pairs of flat frames?

    • @Ekuy1
      @Ekuy1 Год назад

      @@deepskydetail yeah exact same. what does the stdev calculate? (havent learnt it in skl yet)

    • @deepskydetail
      @deepskydetail  Год назад

      @@Ekuy1 The standard deviation is just how much on average the pixels are different than each other. It's a measure of noise. How are you manipulating the brightness? Adding things (like paper) in between the objective and tablet (or whatever you use as a flat panel)? One thing that it could be is if you're using different exposure times (different exposure lengths would be a problem).
      If none of those things are the problem, could you email me at deepskydetail at gmail the spreadsheet your are using? I can take a look
      Thanks!