IC1848 Soul Nebula

[JonHigh]

Hi All. Took this last night and quickly processed it in Siril and PhotoLab. Trying to work out different settings for the camera. Using HCG mode, higher gain and offset values in APT. 
54 x 3 min subs, gain 300, offset 200. Switched over from LCG to HCG. Although brighter and higher SNR, I think it looks noisier than when I was shooting at gain 100 with an offset of 50 in LCH mode. I assumed it would be less noisier, maybe shorter exposures and more subs? Appreciate any advice. Can I still use LCG at 300 as the switch over is done manually? Also, It didn’t help that my EAF was slightly out of focus, something I need to have a look at. I
will be looking at processing it in the Hubble Pallet this weekend for fun. 
 

 

Edited January 21 by JonHigh

[symmetal]

The HCG mode should normally be used with a lower 'normal' gain setting to get the benefit of lower read noise while preserving dynamic range. This Altair Astro help page should help. It does specifically say that HCG at gain 300 is too high a gain setting, giving more noisy results.

This Sharpcap forum page also goes into detail on the 26C and 26M gain settings which should help clarify things.

I would assume you can use LCG at gain 300, as unlike Zwo cameras, the HCG on/off is independent from the gain setting. I only have Zwo cameras. Altair Astro also don't seem to publish gain and noise graphs etc. for their cameras by the look of it. With Zwo cameras, with the same sensor, there is not much reduction in read noise with increasing gain, once HCG is enabled, so there is no actual benefit to be had using a higher gain setting with HCG mode.

Alan

Edited January 19 by symmetal

[JonHigh]

Hi Alan. 
Thank you very much for your info and help. Really do appreciate it.
Ah. That explains why I have a noisy image. So I was using a Gain value of 900! Blast. 
APT has a manual switch over for the conversion modes. So would I be right in saying that LCG at 300 is equivalent to 100 HCG. If so then the later should give better read noise results but a loss in Dynamic range and well depth?  I found the graphs for the 26c and have attached them. 
thanks again

Edited January 19 by JonHigh

[symmetal]

Hi Jon,

Yes, from your graphs, the gain in e/ADU is 0.25 for gain 100 with HCG, and a similar e/ADU for gain 300 with LCG. Note also that in LCG mode there is little real change in read noise with increasing gain.

I'm not sure what the dynamic range units are in your graph but they aren't so extreme in reality. The graphs for the ASI2600 show it more realisticly in camera stops. On the Zwo HCG is on at gain 100 and higher and off below gain 100. There is an apparent increase in dynamic range when HCG kicks in (not sure how that actually works) so I wouldn't worry too much about the full well figures. For most images you won't really notice any difference for normal gain ranges, and only at really high gain settings which you'll never use is it an issue.

Alan

[JonHigh]

Hi Alan. 
Okay thanks for that. I think I have worked it out in my head now 🙄. Looking at the graphs and of my Sensor Analysis from SharpCap, it’s seems best to have it set at 100 using HCG mode. I will give it a try next time out and hopefully I get a better result?! Let’s face it, either way it will not be as noisy as the shot from last night! 
 

[symmetal]

Your HCG on at 26C gain 100 is the same as the Zwo gain 100 setting (which I use), where HCG is automatically turned on, but the gain meanings used in the the camera drivers are totally different, and I've worked out why the dynamic range goes up on the Zwo at gain 100.

Using the e/ADU values is the only way to give an absolute gain value. When the HCG is turned on the absolute gain increases by 3.08 times. At the 26C camera minimum gain of 100 this is the difference between 0.77 e/ADU at LCG and 0.25 e/ADU at HCG ( 0.77 / 0.25 = 3.08 )

When the Zwo driver gain setting goes from 0 to 100 it's the same as the 26C camera gain going from 100 to 308 in LCG mode.

When the Zwo driver gain reaches 100 ( or 26C gain 308), the internal camera gain setting is reduced by 3.08 times back to the equivalent 26C gain 100 setting so that there is no sudden change in the Zwo camera e/ADU output. This reduction in the internal camera gain in the Zwo, back to its minimum value, increases the effective full well capacity again so the dynamic range increases also. 🙂

The 26C is more versatile but is rather more complicated than the Zwo method.

Alan

[JonHigh]

Thanks again for your time Alan.
Crazy. I don’t understand why the scale values are different from various manufacturers and aren’t just standardized across the board! Oh well.

I use APT for image capture rather than Altair’s own software. I loaded that software up for the first time and indeed it starts at 100 (ZWO 0). Also checked APT and confirms that too, it reverts to 100 if I enter a value <100. So I’ve been using the camera at its base line. Or in ZWO terms Gain 0 in LCH mode all this time.

[JonHigh]

So the auto switch over from LCG TO HCG on the Zwo is at 100 where it’s done manually on the Altair. So I thought I was right in using a gain value of 300 and switching it into HGC mode.  As it’s a rough equivalent value? Better to use 100 in HCG which brings it up to 300 in brightness/sensitivity at a lower read noise level? Sorry about this.

Edited January 20 by JonHigh

[symmetal]

Altair are using a linear scale for the gain, while Zwo are using a logarithmic scale in dB. Their gain scale is actually in 0.1dB steps. dB gain values are a comparison or relative value only and don't specify an absolute value unless reference units and values are given. The response of the ears to sound levels and the eyes to brightness levels are logarithmic in nature so a logarithmic scale is useful to refer to them.

APT will just use the gain values given in the camera driver so using the 26C camera APT gain will start at 100 and you will have a separate HCG on/off switch. If you connect a Zwo camera APT will use gain values starting at 0 with no HCG on/off switch as this is switched automatically.

Altair use linear gain 100 and HCG off as the minimum value while Zwo treat this as the logarithmic 0dB reference gain value, corresponding to an actual gain of 0.77e/ADU and the Zwo gain always refers to the e/ADU scale.  When Zwo turn on HCG mode at their log scale 100, they reduce the actual camera gain to compensate, to prevent a change in the e/ADU camera output when the HCG is turned on.

Altair gain values are not tied to the camera e/ADU output but just the amplifier gain values. The HCG mode gain is treated separately and turning it on will increase the camera output by around 3 times giving a corresponding change in the e/ADU output. To get an actual e/ADU out of their camera, corresponding to a specific effective full well or dynamic range, Altair users have to juggle the amplifier gain and LCG/HCG mode at the same time.

Yes, I would use gain 100 and HCG mode on. This gives the same camera e/ADU output (brightness) as gain 300 and LCG mode on (which is the same as HGC mode off.) LCG on really just means HCG is turned off.

This is equivalent to the Zwo gain setting of 100, though it's a coincidence the gain values happen to have the same number, as explained above they refer to different things.

As you say Altair gain 100 and HCG on gives a lower read noise, and also a higher dynamic range, compared to Altair gain 300 and LCG on, (although they give the same brightness or e/ADU output).

There are only really only two gain settings really worth using on the Zwo.

Zwo gain 0 = Altair gain 100, LCG mode on = 0.77 e/ADU.  Gives highest well capacity and dynamic range, but higher read noise.

Zwo gain 100 = Altair gain 100, HCG mode on = 0.25 e/ADU.  Gives lower well capacity, almost as high a dynamic tange, and much lower read noise.

It's all a bit complicated but hope it's a bit clearer for you Jon. 🙂

For information, Zwo gain values are in 0.1dB steps, so Zwo gain 100 is 10dB higher gain than Zwo gain 0.

10dB is equivalent to a linear gain change of 3.16 times which is very close to the linear gain change when HCG mode is enabled. So Zwo likely chose these values on purpose. 😁

Alan

[JonHigh]

Alan. Thank you so very much for your time and patience. I think it has now gelled. It seems an absolute mine field, but an enjoyable learning curve especially with good people like yourself shedding light on things clearly and thoughtfully. I thought I had an understanding of sensors and how they worked but never realised how different manufacturers use different scales to describe the same thing. Almost like using imperial and metric units I guess! Either way thanks again 👍.  Being an ex-pro photographer with an understanding I assume as the gain levels have risen, I take it that the exposures would need to decrease or should I continue with 3 mins subs for example?

On a side note from looking at your impressive equipment list, I noticed you have WO RedCat-51. I was considering it as a possible next purchase as the Evoguide 50 I repurposed has got me into wide field captures. 

Edited January 21 by JonHigh

[symmetal]

Hi Jon,

The dB (decibel) is pretty much the standard unit for specifying audio and video gain as it's more convenient for expressing large and small gains without getting into large numbers. I worked in BBC Television broadcasting (back when it actually made real programmes 😉) and the dB was the only unit used.

Gain in dB = 20 * log( Vout / Vin ) where Vout and Vin are the output and input voltages. This Wiki dynamic range article links dB and stops which should help. Note it refers to 10 * log as it's referring to power ratios and not voltage ratios as you'd measure in an amplifier. Power is related to the square of the voltage. Power in Watts = Volts ^ 2 / Resistance to make it more confusing. 🙂

With regard to exposure and stacking the only difference as far as noise is concerned between 10 x 1 min exposures stacked, and 1 x 10 min exposure is that the first has 10 amounts of read noise added against 1 for the second. If you expose long enough that the noise from your sky background in each sub is significantly greater than the read noise it makes the read noise added insignificant. Once this point is reached there is no reason to continue exposing and it's better to start another exposure. Sharpcap has a useful utility to work out these optimum exposures if you wish to check it out

That being said 3 mins should be fine to be getting on with, unless you have bad light pollution where shorter exposures would be better, as the sky background noise is greater and the read noise is swamped earlier.

My RedCat is actually a WhiteCat as they had that option at the time and I had to return my RedCat as it had poor star shapes in one corner. The second was worse and the third I ended up with was a white one, which was a lot better but not perfect. I got mine when they first came out and QC was possibly overlooked in the rush. Nowadays they seem to be excellent performers for the price and there are autofocus mods available now for them if you need it.

Alan

PS. Your hubble palette version looks cleaner than your original version. 🙂

Edited January 21 by symmetal

[JonHigh]

Hi Alan.

That's some light reading I will be doing. I love to be educated and gain (excuse the pun) knowledge especially in this hobby that I enjoy maybe too much! So thanks again for your time. I will have a look at the exposure times next time it's clear, whenever that will be. I do have a Pro Licence for SharpCap that I mainly use for PA. I recently did a Sensor Analysis which, to be honest put me on this road in the first place.  But come out the other side with more understanding, so was definitely worth it!  👍 Indeed, I will check out the Smart Histogram function now I have done the SA. I live in Bortle 5/6 which is why I have never ventured further than 3 min subs for the very reason you gave. The camera side I get instinctively. 

Re the RedCat. Sounds like you had an awful time with it but pleased the distributor was good enough to help you!       

Yes agreed, I actually deleted it from this post and started it on a new one as there seemed to be an issue reinserting it as I decided to reprocessed it again.. Many thanks for your comment. 😇

Edited January 22 by JonHigh