Back to forums

ToupTek CMOS OSC Camera (IMX533 sensor) - Unity Gain and High Conversion Mode (RisingCam, Altair, Omegon, etc.)

ToupTek ATR3 CMOS 09000 KPA CMOS Help Needed
Tommy Mastrodkamenandrea tasselli
•26 replies•1.2k views
Tommy Mastro avatar
Tommy Mastro avatar
Hi Astro Community,

I recently bought the ToupTek CMOS ATR3CMOS09000KPA (IMX533) camera.  These are the same camera rebranded as RisingCam, Altair, Omegon, and others.  I am looking for assistance interpreting the data to find unity gain. 

The manufacturer  claims unity gain is at 300 but never responded when I asked if that was in High-Gain-Mode (HGM) or not.  The user manual states the gain ratio is 3.05.  I am assuming this is a multiplier?

Looking at the data below, it looks to me the Unity Gain is probably close to what  ToupTek says—300, but the data doesn't clarify in what mode.  So, one might assume in "normal" mode (I.E. without checking the boxes for HGM or HFW).   

So if that is the case, and if the gain ratio is indeed a multiplier of 3.05, then perhaps unity gain is 100 when in High-Conversion-Gain (HGM).   

In other words . . . gain set to 100 in HCG Mode  X  3.05 (gain ratio) = New Gain of 305— close to what the manufacturer recommends.   My conclusion is that unity gain is 300 without HGM engaged . . . and 100 with HGM engaged.

Thoughts? 

Data follows . . . 




Well Written Helpful Insightful Engaging
andrea tasselli avatar
andrea tasselli avatar
I'd use SharpCap sensor analysis tool to derive the actual camera parameters whether in High Gain mode or otherwise.
Tommy Mastro avatar
Tommy Mastro avatar
andrea tasselli:
I'd use SharpCap sensor analysis tool to derive the actual camera parameters whether in High Gain mode or otherwise.

Thannks Andrea,

I thought it a good idea, so I gave it a shot.  Unfortunately, SharpCap wouldn't recognize my camera.  I Googled if SharpCap works with ToupTek and this is what I found . . .

andrea tasselli avatar
andrea tasselli avatar
Tommy Mastro:
andrea tasselli:
I'd use SharpCap sensor analysis tool to derive the actual camera parameters whether in High Gain mode or otherwise.

Thannks Andrea,

I thought it a good idea, so I gave it a shot.  Unfortunately, SharpCap wouldn't recognize my camera.  I Googled if SharpCap works with ToupTek and this is what I found . . .


That's sad news. You could try the ASCOM driver but not sure the results would be dependable.
Well Written
Tommy Mastro avatar
Tommy Mastro avatar
andrea tasselli:
Tommy Mastro:
andrea tasselli:
I'd use SharpCap sensor analysis tool to derive the actual camera parameters whether in High Gain mode or otherwise.

Thannks Andrea,

I thought it a good idea, so I gave it a shot.  Unfortunately, SharpCap wouldn't recognize my camera.  I Googled if SharpCap works with ToupTek and this is what I found . . .


That's sad news. You could try the ASCOM driver but not sure the results would be dependable.

No luck with the ASCOM driver.  I am attempting the DirectShow driver, which shows the camera in the camera list but will not connect to it.  I keep getting an error message.
andrea tasselli avatar
andrea tasselli avatar
If you have PI you can use the script BasicCCDParameters to evaluate the performance hence the gain at any point and it only requires darks, bias and flats. Isn't all automated as in Sharpcap but at least gives you a potential validation of your basic assumptions.
dkamen avatar
dkamen avatar
There are discussions around the internet, bottom line:

No HCG: unity gain ~ 300 (actually somewhere between 301 and 309)
HCG: unity gain close to 100

I use 101 with HCG and works great.
Tommy Mastro avatar
Tommy Mastro avatar
There are discussions around the internet, bottom line:

No HCG: unity gain ~ 300 (actually somewhere between 301 and 309)
HCG: unity gain close to 100

I use 101 with HCG and works great

Thanks Dkamen.  

I'm starting to agree with the web voices as well, but I read a lot of conflicting info. I emailed tech support of the manufacturer (ToupTek) and hope to hear back soon.

I will keep everyone posted.

In the meanwhile, I'll be shooting at gain 100 in HGM and low noise mode (LNM or Ultra Mode in NINA).  At least I know the affect of Low noise mode because it was included in the manufacturer's sensor analysis.
Marc V avatar
Marc V avatar
Considering the Gain ratio of 3.05… do you think that LCG Gain 305 = HCG Gain 100 ?

I use HCG Gain 100 with my duoband filter and LCG Gain 100 for broadband.
Tommy Mastro avatar
Tommy Mastro avatar
Marc V:
Considering the Gain ratio of 3.05... do you think that LCG Gain 305 = HCG Gain 100 ?

I use HCG Gain 100 with my duoband filter and LCG Gain 100 for broadband.

Hi Marc - from what I'm reading, I would still use HGM with broadband - or choose gain 300 in LGM for broadband. 

Instead, I would use different exposure times to compensate for the different filters/targets.  But I'm no expert so take my advice with a grain of salt!
Well Written Respectful Concise Supportive
Milen Markov avatar
Milen Markov avatar
There are discussions around the internet, bottom line:

No HCG: unity gain ~ 300 (actually somewhere between 301 and 309)
HCG: unity gain close to 100

I use 101 with HCG and works great.

That's a good guideline, what offset are you using?
Well Written Respectful
dkamen avatar
dkamen avatar
Milen Markov:
There are discussions around the internet, bottom line:

No HCG: unity gain ~ 300 (actually somewhere between 301 and 309)
HCG: unity gain close to 100

I use 101 with HCG and works great.

That's a good guideline, what offset are you using?

50. It is a little pitched, could have used 10 on most occasions but it does no harm and with offset it is better to be safe than sorry. 50 means the same darks work with all my scopes, from the reduced Newt which is f/4.9  to the Mak which is f/12.

Cheers,
D.
Milen Markov avatar
Milen Markov avatar
Milen Markov:
There are discussions around the internet, bottom line:

No HCG: unity gain ~ 300 (actually somewhere between 301 and 309)
HCG: unity gain close to 100

I use 101 with HCG and works great.

That's a good guideline, what offset are you using?

50. It is a little pitched, could have used 10 on most occasions but it does no harm and with offset it is better to be safe than sorry. 50 means the same darks work with all my scopes, from the reduced Newt which is f/4.9  to the Mak which is f/12.

Cheers,
D.

Hi again! I asked the manufacturer and he recommended an offset of 400... It seemed strange to me, so I'm looking for more information...
dkamen avatar
dkamen avatar
Offset only becomes a problem if you set it too low, so you can't go wrong with 400.

400 would work for higher than unity gain (it would also work for unity) so it gives you more flexibility.
Tommy Mastro avatar
Tommy Mastro avatar
I calculated my offset to be around 325, so not too far off.  Offset is easy to figure out.  Just Google how to set offset.  There is a procedure for it that takes 3 minutes.

An accurate unity gain is a little more difficult to figure out IMHO.
andrea tasselli avatar
andrea tasselli avatar
Sergej Kopysov avatar
Sergej Kopysov avatar
I have analysed the sensor of my camera. Might be useful to give an example to others. I always use HCG 100 gain mode, оffset 300. Low noise mod
Tommy Mastro avatar
Tommy Mastro avatar
Sergej Kopysov:
I have analysed the sensor of my camera. Might be useful to give an example to others. I always use HCG 100 gain mode, оffset 300. Low noise mod

I use the same setting but with offset of 315.
Gernot Schreider avatar
Gernot Schreider avatar
andrea tasselli:
Tommy Mastro:
andrea tasselli:
I'd use SharpCap sensor analysis tool to derive the actual camera parameters whether in High Gain mode or otherwise.

Thannks Andrea,

I thought it a good idea, so I gave it a shot.  Unfortunately, SharpCap wouldn't recognize my camera.  I Googled if SharpCap works with ToupTek and this is what I found . . .


That's sad news. You could try the ASCOM driver but not sure the results would be dependable.

SharpCap does support ToupTek cameras since Nov 2023, vers 4.1.11388,
https://forums.sharpcap.co.uk/viewtopic.php?t=7222
adrian-HG avatar
adrian-HG avatar
There are discussions around the internet, bottom line:

No HCG: unity gain ~ 300 (actually somewhere between 301 and 309)
HCG: unity gain close to 100

I use 101 with HCG and works great.

 i use the same parameters  , i have the omegon branded

with a offset off 30
Brandon Flores avatar
Brandon Flores avatar

After doing some experimentation, I’ve found that unity gain for my ATR533M in high conversion gain, high full well, and low noise (ultra) mode is actually 200 or so.

(I didn’t use Sharpcap, I just took a bunch of light frames along with a bias to compare the ratio of the pixel values to their variances to find unity gain.)

Helpful Concise
Read noise Astrophotography avatar
Read noise Astrophotography avatar

Hi, TS has a fantastic video about the various settings and how to set these cameras up properly or… depending on use correct settings…

https://www.youtube.com/watch?v=xg_AoeYBvSo

I hope it helps !

BTW Awesome choice of camera I love mine!

Willem Jan Drijfhout avatar
Willem Jan Drijfhout avatar
Why the focus on unity gain? I get that there is a theoretical quantisation error benefit, but is that in today's sensors with 14bit digitisation still relevant?
Any gain setting higher than the minimum is reducing full well depth and dynamic range. Result is more blown out highlights, and less HDR malleability in post.
Most of my cameras switch automatic to HCG at a particular gain setting. So if you want to use HCG (e.g. narrowband), that's the setting you choose. Otherwise gain 0 is the best choice. 

It looks like for the Touptek you choose LCG/HCG separately. So choose whichever you prefer for that sequence, and set the gain to the lowest setting (100 for this camera).
Well Written Helpful Insightful Engaging Supportive
dkamen avatar
dkamen avatar

Willem Jan Drijfhout · Nov 26, 2025, 05:08 PM

Why the focus on unity gain? I get that there is a theoretical quantisation error benefit, but is that in today's sensors with 14bit digitisation still relevant?
Any gain setting higher than the minimum is reducing full well depth and dynamic range. Result is more blown out highlights, and less HDR malleability in post.
Most of my cameras switch automatic to HCG at a particular gain setting. So if you want to use HCG (e.g. narrowband), that's the setting you choose. Otherwise gain 0 is the best choice. 

It looks like for the Touptek you choose LCG/HCG separately. So choose whichever you prefer for that sequence, and set the gain to the lowest setting (100 for this camera).

Hi,

the way I understand it, 14-bit does not help with quantization error when gain is too low. If you need 5 electrons per data unit, that means incoming signals of 0, 1, 2, 3, 4 electrons are all recorded as 0 ADU. Same for 5-9, 10-14etc (ignoring for the moment other issues such as read noise that are also affected by gain). It doesn’t matter that you have quantization “resolution” of 16,000 values, you are actually being very lossy, especially where it hurts the most, i.e. fainter signals: recording 9 electrons as 1.0 instead of 1.8 is 44% error, recording 60,994 electrons as 1398 instead of 1398.8 is no big deal.

The only thing that benefits from gain < unity is stuff that would satutate at gain >= unity. Typically stars and maybe the Trapezium. Things you are (usually) better off exposing separately.

For the faint stuff, you need unity or above, to ensure a captured quantum is reflected in measured value. And unity is favoured as a reasonable default for traditional exposure lengths (1-5 minutes) because it minimises saturated pixels.

Helpful Insightful Engaging
Willem Jan Drijfhout avatar
Willem Jan Drijfhout avatar
Willem Jan Drijfhout · Nov 26, 2025, 05:08 PM

Why the focus on unity gain? I get that there is a theoretical quantisation error benefit, but is that in today's sensors with 14bit digitisation still relevant?
Any gain setting higher than the minimum is reducing full well depth and dynamic range. Result is more blown out highlights, and less HDR malleability in post.
Most of my cameras switch automatic to HCG at a particular gain setting. So if you want to use HCG (e.g. narrowband), that's the setting you choose. Otherwise gain 0 is the best choice. 

It looks like for the Touptek you choose LCG/HCG separately. So choose whichever you prefer for that sequence, and set the gain to the lowest setting (100 for this camera).

Hi,

the way I understand it, 14-bit does not help with quantization error when gain is too low. If you need 5 electrons per data unit, that means incoming signals of 0, 1, 2, 3, 4 electrons are all recorded as 0 ADU. Same for 5-9, 10-14etc (ignoring for the moment other issues such as read noise that are also affected by gain). It doesn’t matter that you have quantization “resolution” of 16,000 values, you are actually being very lossy, especially where it hurts the most, i.e. fainter signals: recording 9 electrons as 1.0 instead of 1.8 is 44% error, recording 60,994 electrons as 1398 instead of 1398.8 is no big deal.

The only thing that benefits from gain < unity is stuff that would satutate at gain >= unity. Typically stars and maybe the Trapezium. Things you are (usually) better off exposing separately.

For the faint stuff, you need unity or above, to ensure a captured quantum is reflected in measured value. And unity is favoured as a reasonable default for traditional exposure lengths (1-5 minutes) because it minimises saturated pixels.

This is all very true. I'm just wondering what the difference is in real life, after you take into account the shot noise (which should be several times higher than the read noise), and then any signal measurable above that. Wouldn't any error not become negligible?
On the other hand, saturated pixels cannot be undone. For any fast-ish system with several minutes exposure, saturation is not limited to just the trapezium. And even if structures are not saturated, almost full wells will quickly become issues during processing where it will saturate after-all.
Helpful Insightful Respectful Engaging