Gain on 533mc pro with fast lens?

[ONIKKINEN]

14 hours ago, 900SL said:

It's complicated by 14 bit to 16 bit conversion, and offset.

@BrendanCand others, i think this part was skipped in the above calculations.

The 533MC is a 14-bit camera but as far as i know every software will report measured values in 16-bit. The output value must be converted back to the original to have meaningful e/ADU conversion rates and other measurements regarding noise amd whatnot (divide by 4).

Offset of 50 equals 500 in 14-bit values (the only thing the camera cares about) and becomes 2000 when converted to a 16-bit number. The above calculations come out as 237e if converted to 14-bit which sounds a lot more reasonable than 950e.

[symmetal]

@BrendanC I've produced Excel sheets for some cameras giving exposure times for specified read noise swamping factors like this. If you wish I can make one for the ASI533. You just need to input your mean bias ADU value ( and camera bit depth if not 16 bit) and optimun sky background ADU values are calculated. Swamping factor 5 is what I use, as vlaiv has recommended this in the past. With the calculator top right you can get an exposure time for a specified background ADU by inputing the values from a test exposure, or a previous sub you've taken if the sky darkness is similar.

I also use zero gain when using luminance filters with the RASA 11 as at gain 100, it will swamp the read noise in 15 seconds, (in bortle 3). At gain 0, around 60 secs is required which is better to avoid 100s of subs.

Alan

Edited April 11, 2023 by symmetal

[ONIKKINEN]

Here is a sharpcap sensor analysis result sheet from the 533MC:

Should contain all the details one needs to figure out what number to put where in the read noise swamping formula.

So at Gain 200 we have 1.42 read noise and 0.32 e/ADU rate (ADU=camera measured value, in this case 14-bit). Offset was already defined as 50 which is 500 ADU or 2000 in 16-bit.

So for a read noise swamp factor of 5 we get this: 1.42*5 = 7.1 which when squared comes out to 50.41e. 50.41e divided by the e/ADU conversion rate of 0.32 comes out to 157.5 in 14-bit values. Multiply by 4 to get 630 in 16-bit values which is what your software using the camera will report. This is the number you are looking for to have on top of your offset.

Your offset being 2000 in 16-bit means @BrendanC you are looking for a number of 2630 given by APT/NINA/whatever else when out in the field in order to swamp read noise x5 at gain 200.

I think in your shoes i would rather shoot at gain 100 as the drop in read noise is really not that much. Unless you are scraping the barrel so to speak in terms of mount performance. With gain 100 we get: 1.63*5=8.15 squared to 66.42e, so only 16 electrons more compared to gain 200 but you get triple the full well. In the case of gain 100 you are looking for a number of 2260 reported by the capture software to swamp read noise x5.

Edited April 11, 2023 by ONIKKINEN

[900SL]

34 minutes ago, symmetal said:

@BrendanC I've produced Excel sheets for some cameras giving exposure times for specified read noise swamping factors like this. If you wish I can make one for the ASI533. You just need to input your mean bias ADU value ( and camera bit depth if not 16 bit) and optimun sky background ADU values are calculated. Swamping factor 5 is what I use, as vlaiv has recommended this in the past. With the calculator top right you can get an exposure time for a specified background ADU by inputing the values from a test exposure, or a previous sub you've taken if the sky darkness is similar.

I also use zero gain when using luminance filters with the RASA 11 as at gain 100, it will swamp the read noise in 15 seconds, (in bortle 3). At gain 0, around 60 secs is required which is better to avoid 100s of subs.

Alan

That would be great for the 533MC Pro

It's 14 bit, unity gain is 100 (1 e/ADU), noise is 1.5e.

Default offset is 70 or 2800 ADU (bias median)

Can you tell me the formula for calculating the sky background ADU value based on a swamp factor of 5?

 

[vlaiv]

6 minutes ago, 900SL said:

an you tell me the formula for calculating the sky background ADU value based on a swamp factor of 5?

(read_noise * 5)^2 / gain + offset_ADU

(adjust to taste for bit count)

It is better to compare achieved and target background in electrons then converted to ADU units that are reported by imaging software - as you can easily calculate needed exposure without guessing.

Just divide the two and that will give you number to multiply your current exposure by. For example if you measured 50e and you calculated that you need 75e background level - then increase exposure by 75/50 = 1.5 times. You can't do this once you add offset and convert to ADU units.

 

 

[symmetal]

@900SL. No problem. I'll make the spreadsheet. I get all the info needed from the curves published by Zwo. I'll use the offset/bias figures you've quoted for an image you can download. I'll give a link to download the excel file too so you can make any changes you want to print your own sheet.

The formula used is on the third line of the image I posted. Vlaiv provided the formula as an improvement on the one I originally used and he approved the latest spreadsheet contents too when I last posted it. 😊

The gain value used is the e-/ADU figire of course and not the camera gain in 0.1dB steps.

I can use the data provided by @ONIKKINEN to provide another sheet using actual values to see how it compares to the published specs.

Alan

Edited April 11, 2023 by symmetal

[ONIKKINEN]

11 minutes ago, symmetal said:

@900SL. No problem. I'll make the spreadsheet. I get all the info needed from the curves published by Zwo. I'll use the offset/bias figures you've quoted for an image you can download. I'll give a link to download the excel file too so you can make any changes you want to print your own sheet.

The formula used is on the third line of the image I posted. Vlaiv provided the formula as an improvement on the one I originally used and he approved the latest spreadsheet contents too when I last posted it. 😊

The gain value used is the e-/ADU figire of course and not the camera gain in 0.1dB steps.

I can use the data provided by @ONIKKINEN to provide another sheet using actual values to see how it compares to the published specs.

Alan

I should mention that i found this off google image search and its not something i have made myself.

Should you google 533MC sensor analysis you find a number of results, some of which have slightly different values so for the purposes of creating a spread sheet you should probably take an average of the found measurements instead of one of them at face value (although the differences are very small). I have done the sharpcap analysis on all of my cameras a few times and the measurements vary a little bit. Both vary between different runs of the tool and between the measured and camera manufacturer stated amount i mean.

[symmetal]

Here's the chart for the ASI533 using ZWO's figures, and a link to download the Excel spreadsheet to make your own charts. The graphs are plotted from the data entered so you can quickly see if you've made an error. If you want a different swamping factor just change the factor number at the top of the column. To quickly enter all the e-/ADU gain values just enter the top value and the others will be calculated from that.

ASI533, Sky Background ADU (SGL).xlsx

Note that the camera offset contributes to the bulk of the ADU values calculated for most gain settings, as almost all of the bias ADU is from the offset.

@900SL With regard to your initial question, I would say that for the ASI533 there's no point in using a camera gain setting below 100 (unity gain). Below unity gain you're effectively throwing away incoming photons. At gain 0 you need to receive 3 photons to register an increase of 1 ADU. If you receive 5 photons you still only register 1 ADU so you're throwing away 2 out of every 3 photons received. Also the read noise is higher and there's no real improvement in dynamic range.

With the ASI 6200 or ASI 2600, gain 0 can be useful as I mentioned before, as it hasn't achieved unity gain at gain 0, so every photon received gives a corresponding increase in ADU.

Here's the calculation for the ADUs shown in Excel format. 🙂

Alan

Edited April 12, 2023 by symmetal

[900SL]

@symmetal Many thanks for that Alan, I'll be trying out those ADU targets (at 100 gain!) in the near future. And thanks for the spreadsheet, downloaded    🙂

 

Edit: Can I assume this is pretty much filter invariant and suitable for OSC sensor?

I assume the filter will just extend the exposure required to get the target ADU, but wasn't sure if there is any difference between Mono and OSC in terms of target sky background ?

Edited April 12, 2023 by 900SL

[BrendanC]

Re the calculator - why would I shoot to a required ADU? How would I establish what that should be? And what is it actually calculating - as in, why specify a sample exposure time, and then it calculates the exposure?

@vlaiv - I don't know where to begin with what I find confusing about all this! 'you take a calibrated sub (you should really calibrate your subs for stacking, so there is no reason not to have calibrated value) - you measure median value on patch of background in the image - multiply with a number and compare to reference value you calculated from read noise.' - sounds easy, but what I'm missing is steps, and how I would do this. As in 'step 1, 2, 3', with guidance actually on how to get these values. Then we're talking ADUs, electrons, gain, offset, 14-bit, 16-bit, whatever else can be thrown into the mix. I still think DSLR was easy - in APT, I could just use the histogram and shoot to 1/3-1/2 way across it. Job done.

I'm just not that great with maths, and I find it very frustrating that I cannot find a simple, quick, easy-to-follow 'recipe' that will allow me to figure out how long I should shoot for, actually while at the scope. For example, using my figures, you say I'm shooting x19 longer than I should. If I'm shooting 120s BB, and 300s NB, then that implies I should be shooting 6s  BB and 15s NB! I'm shooting at F4.5 in a Bortle 4. Which brings me to another point: all the calculations and spreadsheets seem to be about getting the minimum exposure to swamp read noise, with no consideration as to what an optimum, real-life situation would be for exposures. Again, this is not as easy as 'expose until 1/3 to 1/2 way across the histogram', which I used with no problem whatsoever with my DSLR.

I did once find this, which I thought would work: 'Simply take a Bias and look at the AVG ADU.  Generally you want the AVG ADU of your LIGHT to be 400-600 above your Bias (for monochrome).  If you are shooting OSC, then you will want to be ~1200 above the Bias  (2-3 times higher because of bayer matrix).  For example, if your Bias Avg ADU is 500, then your target Light Avg ADU will be around 1000 for monochrome, or 1700 for osc.  Take a 30 second Light image, and adjust your exposure until you reach your Target Avg ADU.' (https://www.cloudynights.com/topic/781537-understanding-adu-reading/?p=11256105).' This implied that I would shoot to an ADU of 3,200, which worked for 300s NB exposures at gain 200, but for gain 101 BB, it came out at 540s! So, nice 'recipe', easy to follow, works in the field - but gives results that are far too long rather than far too short. Another dead end.

So, I'll bow out of this thread now. I'm clearly lacking the brainpower to understand this. I'll just stick with 60s or 120s BB at gain 101, and 300s NB at gain 200, without really knowing why, because it seems to work.

Thanks, Brendan

PS FWIW, I've attached my 'Exposures calculator for dummies' spreadsheet which shows the kind of thing I'd be looking for. Perhaps it's right? Who knows?

Exposure calculator for dummies.xlsx

Edited April 12, 2023 by BrendanC

[vlaiv]

27 minutes ago, BrendanC said:

I still think DSLR was easy - in APT, I could just use the histogram and shoot to 1/3-1/2 way across it. Job done.

That is advice for properly exposing flats - not optimum exposure for shooting lights. It is therefore completely wrong in this context.

I'll try to explain in simplest terms possible why we do this calculation and how it works, and then provide you with easy to follow steps. Only issue is - you can't perform these steps while you are shooting unless you have some sort of software that will do it for you (or can be mix of different software + spreadsheet for calculation).

We determine minimum exposure length so that it will not significantly impact our final result in terms of noise. Shorter we go - noisier final result becomes. This is because there is one noise source that does not behave like all the rest noise sources - and that is read noise. Read noise comes per exposure. All other stuff that is important - signal and various other noises (thermal noise, shot noise, LP noise) behaves differently than read noise and varies with time. They all only depend on total imaging time and not number of exposures.

What makes the difference between many short subs and few long subs is the only thing that does not depend on total imaging time - and that is read noise.

We expose for long enough so that read noise per exposure becomes insignificant - or increases total noise by indistinguishable amount. This happens when read noise is 5 times less than the largest other noise source and that is most often sky background noise. That is why we compare the two. Here is step by step, easy to follow, check list how to calculate optimum exposure length. It will be tailored for ASI533 and Gain 200 - but I will put differences to all other models in brackets.

1. Take single exposure of patch of the sky where your target is. Expose for some amount of time - let's say 60 seconds

2. Measure median ADU value of the empty patch of that exposure (select background with as few stars as possible - ideally without any objects and stars)

3. From that ADU value - subtract mean bias or dark sub value of same exposure length (you either need to take these in advance, or cover scope - take one sub and measure it)

4. Divide this value with 4 (use different number here for other models, 4 is for 14 bit ADC, 16 is for 12 bit ADC and 1 is for 16 bit ADC) and multiply it with 0.3162 (use actual e/ADU value for the camera and gain you are using)

5. Divide number you got in previous step with 50 (read noise for your camera and gain setting multiplied with 5 and then squared)

6. Divide 60 seconds (or whichever exposure length you used in step 1) with number you get in step 5 and that is exposure length that you should be using for your sky conditions and your setup.

45 minutes ago, BrendanC said:

I'm just not that great with maths, and I find it very frustrating that I cannot find a simple, quick, easy-to-follow 'recipe' that will allow me to figure out how long I should shoot for, actually while at the scope. For example, using my figures, you say I'm shooting x19 longer than I should. If I'm shooting 120s BB, and 300s NB, then that implies I should be shooting 6s  BB and 15s NB! I'm shooting at F4.5 in a Bortle 4. Which brings me to another point: all the calculations and spreadsheets seem to be about getting the minimum exposure to swamp read noise, with no consideration as to what an optimum, real-life situation would be for exposures. Again, this is NOT as easy as 'expose until 1/3 to 1/2 way across the histogram', which I used with no problem whatsoever with my DSLR.

That x19 is wrong - because I assumed you are familiar with bit counts and ADU values and that you adjusted for 14 bit of camera. I was wrong, you don't know these concepts - and it was pointed out by @ONIKKINEN in one of his replies above:

14 hours ago, ONIKKINEN said:

@BrendanCand others, i think this part was skipped in the above calculations.

The 533MC is a 14-bit camera but as far as i know every software will report measured values in 16-bit. The output value must be converted back to the original to have meaningful e/ADU conversion rates and other measurements regarding noise amd whatnot (divide by 4).

That and unknown offset value makes wrong number in the end - but procedure presented was correct.

44 minutes ago, BrendanC said:

PS FWIW, I've attached my 'Exposures calculator for dummies' spreadsheet which shows the kind of thing I'd be looking for. Perhaps it's right? Who knows?

I took a brief look at the calculator - and I think it is in essence correct - but that calculator is made so you can estimate exposure length without measuring it.

You don't need to take test exposure with that calculator - but you do need to know SQM value of your sky. It is only estimation to get in the ballpark (which is probably just fine and can be used). Above method is relying on actually measuring things for your location - taking test exposure and measuring sky brightness that way and not relying on SQM.

It is also different in that you don't need to know details of your equipment (that is also measured) - which you normally know anyway (scope focal length, pixel size) - and some that you don't necessarily know (optical efficiency of your scope, QE of your sensor and so on ...).

 

[BrendanC]

This is incredibly useful, now I think I'm starting to get somewhere, thank you.

I've built this into a rough spreadsheet, attached, with notes.

The ADU for the bias is estimated - I don't have a value for a 60s bias to hand, but I'll try and get one later today.

I'm also taking the e/ADU value from the Sharpcap graphic above. I've also got my own results for that somewhere, again I'll look it up when I get the chance.

So, with those figures plugged in, I'm getting about 100s. This almost starts to look about right, and gives me a basis for understand this. If you could quickly look at the spreadsheet I'd appreciate it.

Thanks again,
Brendan

vlaiv calc.xlsx

[vlaiv]

3 minutes ago, BrendanC said:

If you could quickly look at the spreadsheet I'd appreciate it.

That all looks good.

Yes, you are right - I rounded up to 50 because I was not sure read noise was exactly 1.4 - someone measured it to be just a bit higher than this value and ended up with 50.something - so I've chosen 50 to be the the figure - nice and round

(it does not change the result very much if you use 50 instead of 49 or 51 - but if you can - go with exact figure).

Just a note - for bias removal - ideally just get one dark for the same exposure length as your test exposure (so 60 seconds) - but do pay attention that you have all the settings the same (gain, offset, temperature - the lot) as you would when taking regular darks for your regular light exposures. Then measure average of that and use in calculation.

 

[BrendanC]

Brilliant, thanks! I just realised I do have 60s darks (should have been obvious) which are at the same gain, offset, temp etc, and they're also at 1997 so that part's still correct.

Now I'm going to dig around for the e/ADU figure from Sharpcap and plug that in.

Again, thanks for laying this all out for me. I tend to learn by seeing examples and then reverse engineering them, so this is ideal.

[symmetal]

6 hours ago, 900SL said:

Edit: Can I assume this is pretty much filter invariant and suitable for OSC sensor?

I assume the filter will just extend the exposure required to get the target ADU, but wasn't sure if there is any difference between Mono and OSC in terms of target sky background ?

Yes, it's the noise from that sky background that does the read noise swamping. Different filters would just need different exposure lengths to achieve those same ADU values. If using separate RGB filters it's easiest to use one exposure for all three colours and choose an exposure that perhaps works for the 'dimmest' filter, probably Blue. Red and green will then be a little overexposed compared to the 'ideal' exposure but not enough to make any difference. It saves having to have separate darks for each colour.

For OSC the same applies as you're 'stuck' with one exposure for all three colours. For a test image with no significant nebulosity and not satutated with stars the mean of the image will be close to the average sky background ADU for all three colours which should be fine to use. You can check sample pixels on your capture program to see the actual difference between RG and B pixels for more accuracy but it's not necessary to be spot on with ADU values as no two nights will have the same sky background darkness so a compromise needs to be reached that's fine for the 'average' night sky.

These exposure calculations are really only of use for broadband imaging. For narrowband imaging the exposures required to reach these ADU values will be much longer than 10 or 20 minutes commonly used, so for NB it's best to expose for as long as you can get away with and accept that read noise will have some impact in your final image.

Alan

[vlaiv]

2 minutes ago, symmetal said:

For narrowband imaging the exposures required to reach these ADU values will be much longer than 10 or 20 minutes commonly used, so for NB it's best to expose for as long as you can get away with and accept that read noise will have some impact in your final image.

This really depends on level of read noise. It is certainly true for old CCDs that had 7e, 8e and even 10e or more of read noise.

Modern cmos sensors have very low read noise, and if you have something like 1e or less of read noise, and you image in city center (as is often the case - people in heavy LP usually go for NB to have any chance of imaging) - then you might be surprised by needed exposure length - it can be as low as few minutes, especially for wider band filters (some really narrow band - like 3.5nm filters will still need longer exposures even with small read noise).

BTW - with narrow band, it's worth examining dark current noise - it can become most significant noise source over LP noise for some cameras, especially in summer months if one decides to cool down to -5C or -10C versus for example -20C

[900SL]

4 hours ago, BrendanC said:

Snip.....

I did once find this, which I thought would work: 'Simply take a Bias and look at the AVG ADU.  Generally you want the AVG ADU of your LIGHT to be 400-600 above your Bias (for monochrome).  If you are shooting OSC, then you will want to be ~1200 above the Bias  (2-3 times higher because of bayer matrix).  For example, if your Bias Avg ADU is 500, then your target Light Avg ADU will be around 1000 for monochrome, or 1700 for osc.  Take a 30 second Light image, and adjust your exposure until you reach your Target Avg ADU.' (https://www.cloudynights.com/topic/781537-understanding-adu-reading/?p=11256105).' This implied that I would shoot to an ADU of 3,200, which worked for 300s NB exposures at gain 200, but for gain 101 BB, it came out at 540s! So, nice 'recipe', easy to follow, works in the field - but gives results that are far too long rather than far too short. Another dead end.

Oddly enough shooting a test frame to check that ADU is a set amount above the bias ADU is a simple way to do it in the field, although 1200 value quoted above is incorrect for 533MC Pro. .

 

For my 533MC Pro using AsiAir pro, my bias is 2800 ADU (70 offset) and at gain 100 for swamp factor of 5, I need (5 * 1.5e)^2 = 56 which is then converted to 16 bit so 4 * 56 = 224

add 224 to 2800 and exposure median should be above 3024, which matches Symmetals spreadsheet value above. I shoot to get around 3100 to 3200 as a rough guide, I usually set an exposure then check ADU and histogram 

For 50 offset this should be around 2224, which is much shorter exposure than you quote above

 

 

Edited April 12, 2023 by 900SL

[BrendanC]

OK, I actually followed that and actually understood it, so I might be getting somewhere. Thanks. I've got @vlaiv's method captured in a spreadsheet but will also look at shooting to 2224 ADU to achieve swamp factor of 5, and see how I get on.

[900SL]

I usually go a little over on ADU to allow for the fact that the ADU median is not the sky background, it is the median of the entire image including highlights and nebulosity etc. Try 2400 see how it looks?

[symmetal]

1 hour ago, vlaiv said:

This really depends on level of read noise. It is certainly true for old CCDs that had 7e, 8e and even 10e or more of read noise.

Modern cmos sensors have very low read noise, and if you have something like 1e or less of read noise, and you image in city center (as is often the case - people in heavy LP usually go for NB to have any chance of imaging) - then you might be surprised by needed exposure length - it can be as low as few minutes, especially for wider band filters (some really narrow band - like 3.5nm filters will still need longer exposures even with small read noise).

BTW - with narrow band, it's worth examining dark current noise - it can become most significant noise source over LP noise for some cameras, especially in summer months if one decides to cool down to -5C or -10C versus for example -20C

Thanks vlaiv. In my previous mentions of this subject I did say that in heavy light pollution the exposures in NB may be achievable, but didn't mention it here.  In Bortle 3, I found I would still need around 2 hours NB exposure even with the RASA 11.

6 hours ago, BrendanC said:

Re the calculator - why would I shoot to a required ADU? How would I establish what that should be? And what is it actually calculating - as in, why specify a sample exposure time, and then it calculates the exposure?

It looks like you've sorted it out with vlaiv's help, but to recap.

You shoot to a required ADU as given in the table, as at that ADU the sky background noise sufficiently exceeds the read noise, making the read noise contribution to your image  insignificant. You can then start another exposure, rather than continuing with the first exposure, and so avoid blowing out the highlights more than necessary, or risk tracking errors or aircraft etc. spoiling your image.

All my calculator does is avoid you having to do multiple test images to achieve that ADU value. Say you need to achieve 3000 sky background ADU as read from the table using your gain setting. Checking a previous image of say 2 mins exposure, (at that same gain setting), you read off the sky background as being 2600 ADU. Putting these values in as the sample image, and 3000 as the required ADU the calculated exposure is the exposure needed to achieve 3000 ADU with your setup. It just subtracts the bias value from the ADU values entered and proportionally adjusts the ADU/exposure values. Nothing fancy.  If you put the sample exposure in seconds the result will be in seconds. If the sample is in minutes the result is in minutes. 

Alan 

[symmetal]

1 hour ago, 900SL said:

I usually go a little over on ADU to allow for the fact that the ADU median is not the sky background, it is the median of the entire image including highlights and nebulosity etc. Try 2400 see how it looks?

I use SGP and it includes a useful 21x21 sample under the cursor. The whole image mean and median is quite close to the more accurate sky background sample under the cursor as the vignetting tends to counteract the highlights, if there's plenty of sky background in the image. 🙂 This is a OSC image. If the mean and median are close the whole image values are pretty good to use.

Alan