Question about the Smart Histogram (The Brain)

[umasscrew39]

I have had great success using my ZWO ASI cameras with SharpCap and essentially using gain/exposure/brightness settings either by trial and error or those suggested by others. The past two nights I tried the recommendations from the Brain for my 294MC Pro and was disappointed. For example, I usually use a gain of 300 and a brightness of 4 for exposures any where from 60s-180s which are live view stacked for 30 mins or 1 hr. I then post-process those using PixInsight. This way, I get the advantage of both live views and a better quality AP image afterwards. A disadvantage to this approach is that I sometimes get over-sampled/bloated stars depending on the object. On the other hand, the Brain recommends 121 gain with exposure time of 134s for my area. When I do this, the live views are barely noticeable unless the object is unusually bright, making the live viewing aspect worthless. Should this be the case or am I doing something wrong?

[admin]

Hi,

The thing to remember is that regardless of the camera settings in terms of gain and exposure, the camera will detect the same number of photons in half an hours imaging from a particular target – if you're using gain 300 and 120 s exposures then the images will look brighter than those taken with the recommended settings from the smart histogram brain, but the same image data is basically in both sets of images.

Moving from Gain 134 to Gain 300 on a ZWO camera is an extra 16.6db of gain - basically making each image 6.7 times brighter with 300 gain than at 1.34 gain for the same exposure. Remember that there is no real difference in the amount of image data by using extra camera gain - you are just multiplying the image strength in electronics in the camera.

When using the suggestions from the brain, you should be able to compensate for the weaker individual frames by stretching the final image more (either in live stacking or in tools like PixInsight)

Hope this helps, Robin

[umasscrew39]

Thanks, Robin. ok- I think I understand the science behind the cameras (i.e., photons, pixels, etc) but I am still confused over 2 things. Based on what you said, it sounds like there is no difference if I do what I have been doing vs. following the Brain's recommendations. I would have thought that shorter exposures with lower gain would potentially produce less bloated stars or cores of nebulae like Orion. Am I still not getting the advantage of using the Brain's recommendations?

Thanks,
Bruce

[admin]

Hi Bruce,

If you go shorter exposures than the brain's recommendations for the same gain then you will be missing out on final image quality. If you go longer exposures then you won't be gaining on image quality, but you won't lose out either. What you will be doing is making your mount's tracking/guiding work harder to keep those long exposures with pin sharp stars. If your mount is up to this then you have no worries.

Cheers, Robin

[umasscrew39]

hmmm....

That makes sense but the data are telling me something else.

When I measured my sky brightness using my Optolong L-enhance dual narrowband filter (H-alpha, OIII), I get a value of .21e/pixel/s. The Brain said the optimal gain is 121 with exposures of 134 sec with my ZWO ASI294MC. I have been using a gain of 300 with exposures of 120 sec, so actually less exposure time which makes sense given the higher gain. Again, maybe splitting hairs as in the end when I post-process the fits files they are good. One question is: should actually use no filter to measure sky brightness vs. the NB filter which, of course, darkens the sky dramatically?

[admin]

Hi,

You should definitely use the narrowband filter to measure the sky brightness – that's how the brain knows to give you longer exposure suggestions when using narrowband, because the filter reduces the sky brightness measurement considerably.

On the ZWO cameras, the difference between gain 121 and gain 300 is 18 dB, which is nearly a factor of 10. The two different exposures that you mention are close enough to make little difference. This explains why your individual frames are much brighter with the high gain approach – you are amplifying them more.

Cheers, Robin

[umasscrew39]

Hi Robin

OK- got it. Thank you for the updated information/clarification.

Bruce

[Blade71]

Hi Robin,

Just read over this thread and I also have an ASI294MC and was wondering, is the smart histogram readings dependent on the filters and telescopes I use? I read the the sensor analysis isn’t dependent on scope or filter but wonder if that’s different for the smart histogram?

Also is it normal that smart histogram gives a different unity and dynamic range gain than what is set by the manufacturer? The unity/max dynamic Range gain for the ASI294MC is 120 and zero respectively but the results stated unity was 220 for my filters and varied for max dynamic range.

Thanks

[admin]

Hi,

The brain calculations and the smart histogram automatically take account of any filters by virtue of the fact that the amount of light reaching the sensor is reduced by the filter that is in place and whichever of these you are using will base its calculations on the reduced amount of light being detected.

Unity gain for that particular sensor is approximately 120, but the calculations that the brain performs only aim to get as close as possible to the unity gain – if you have particular dark skies are using a narrowband filter then what may happen is that the calculated exposure value at unity gain would be longer than the maximum exposure selected in the drop-down – in that case the calculation is adjusted to use the maximum exposure chosen and an increased gain value (the smallest increase that will still give optimum results).

The calculations for maximum dynamic range are different to those performed by the manufacturer – the manufacturers figures are only considering a single frame whereas the brain calculations consider the total dynamic range of the final stacked image and also takes account of the fact that shorter exposures allow more frames to be captured in the same amount of time.

Cheers, Robin

[LRSibb]

I ran Smart Histogram in my light polluted back yard with my entire filter suite - very impressed.

ZWO 1600 MM Pro, Celestron Edge HD 8

results are below:

SkyGlowLRS.JPG (27.86 KiB) Viewed 2449 times

Everything makes sense - I have been doing Narrowband with 10 to 20 min subs - going for finer details on faint emisson Nebulae. I am pushing the tracking and guiding with a hypertuned CGEM, but the extra detail seems to come with longer subs. Sky is too bright for LRGB generally but it was great to quantify sub length requirements.

I have two questions:

1) Re Sky brightness Calculation units: is a specific gain assumed for the electrons/pixel/second values in the output? or is a unity gain assumed where one photon arrival is equivalent to one electron in this value?

2) The detection threshold limit plot is useful for seeing the flattening for normal targets. is there an ability to get raw output that has more information - or at least an expanded vertical scale that can show the sensitivity at the far end of the curve that appears flat, but may not be when zoomed in at the longer exposure times? This relates to a more general question: What are the limits of the optimal exposure calculation for very faint targets in Narrowband?

Thanks again for the tool - it has revealed a lot.