Review of the Touptek IMX585c Cooled camera Generic equipment discussions · AstroShed · ... · 209 · 8608 · 120

NonkelErik 0.00
...
· 
·  1 like
·  Share link
@Tobiasz

What is the newest firmware version with the fix for 16 bit hdr?
With whom did you communicatie from Touptek to obtain this non public version please?
Regards
Erik
Like
StewartWilliam 5.21
Topic starter
...
· 
·  2 likes
·  Share link
Erik Derboven:
@Tobiasz

What is the newest firmware version with the fix for 16 bit hdr?
With whom did you communicatie from Touptek to obtain this non public version please?
Regards
Erik

The version I have is 4.36, and I am in contact with there sales department as I have one of there 585 cameras for review on loan from them…
Like
NonkelErik 0.00
...
· 
·  2 likes
·  Share link
Today I received the new firmware 4.36 together with the Ascom Touptek dll.
After FW upgrade i did a Sharpcap sensor analysis:

At gain 100: 0,52 read noise! 
15.75 STOPS

I'll post a screenshot later!
Like
NonkelErik 0.00
...
· 
·  2 likes
·  Share link
sharpcap TPK 585~2.png
Like
TiffsAndAstro 1.81
...
· 
·  1 like
·  Share link
Erik Derboven:
sharpcap TPK 585~2.png

something looks odd as there is no unity gain? Also 585 is 12bit not 16bit.
Edited ...
Like
Emission 2.11
...
· 
·  1 like
·  Share link
Erik Derboven:
sharpcap TPK 585~2.png

Wow! Absolutely mindblowing. I am so excited to use this sensor under the stars
Like
TiffsAndAstro 1.81
...
· 
·  Share link
Tobiasz:
Erik Derboven:
sharpcap TPK 585~2.png

Wow! Absolutely mindblowing. I am so excited to use this sensor under the stars


I'm not certain this image/test  is correct as it doesn't show unity gain? I'm deffo not an expert. Struggling to link a screen shot of my result from my 533 to compare though
Like
StewartWilliam 5.21
Topic starter
...
· 
·  1 like
·  Share link
TiffsAndAstro:
Tobiasz:
Erik Derboven:
sharpcap TPK 585~2.png

Wow! Absolutely mindblowing. I am so excited to use this sensor under the stars


I'm not certain this image/test  is correct as it doesn't show unity gain? I'm deffo not an expert. Struggling to link a screen shot of my result from my 533 to compare though

The 585 has an HDR mode that up until now did not work, that is what this whole thread is about, the HDR mode means high dynamic range mode, it works by adding 2 modes together to produce a 16 bit mode, and now with this new firmware it seems to work….
so the test is correct 👍🏻
Edited ...
Like
TiffsAndAstro 1.81
...
· 
·  Share link
AstroShed:
TiffsAndAstro:
Tobiasz:
Erik Derboven:
sharpcap TPK 585~2.png

Wow! Absolutely mindblowing. I am so excited to use this sensor under the stars


I'm not certain this image/test  is correct as it doesn't show unity gain? I'm deffo not an expert. Struggling to link a screen shot of my result from my 533 to compare though

The 585 has an HDR mode that up until now did not work, that is what this whole thread is about, the HDR mode means high dynamic range mode, it works by adding 2 modes together to produce a 16 bit mode, and now with this new firmware it seems to work….
so the test is correct 👍🏻


I Read that screenshot of the test such that at 100 gain it needs 2 1/4  electrons to gain one adu.
Like
Emission 2.11
...
· 
·  Share link
Armin Lukas:
Would love to see some test results in the night sky and a comparison HDR on/off. But looks promising!

You wont see much of a difference, because HDR is a separate mode like LCG or HCG. It is called HDR because the sensor captures two pictures at once, one in low gain and one in high gain and merges them together on the sensor directly. In a very simpflied way you could say the sensor combines the best of both worlds: the low gain full well and low read noise from high gain. 

Orion nebula core for example. You should be able to expose for longer without clipping (because of the full well) but you dont have to because the noise is so low that faint details are already above the noise floor. 

The IMX585 sensor but especially the Touptek HDR algorithm gives you so much more room "to play". As far as I know the only camera of this type and vendor that utilises the HDR mode properly. Every other brand delivers only the LCG/HCG mode. 

regards
Like
Emission 2.11
...
· 
·  1 like
·  Share link
TiffsAndAstro:
Erik Derboven:
sharpcap TPK 585~2.png

something looks odd as there is no unity gain? Also 585 is 12bit not 16bit.

585 is 12Bit in LCG/HCG Mode. HDR output is 16Bit.
Like
StewartWilliam 5.21
Topic starter
...
· 
·  1 like
·  Share link
So what is unity gain then, as that confuses me too, it’s not on that chart, or is it a lower setting like a gain of 50 or similar….?
Like
TiffsAndAstro 1.81
...
· 
·  Share link
Tobiasz:
TiffsAndAstro:
Erik Derboven:
sharpcap TPK 585~2.png

something looks odd as there is no unity gain? Also 585 is 12bit not 16bit.

585 is 12Bit in LCG/HCG Mode. HDR output is 16Bit.


I'm really wondering how this works. I need to read more into this, apologies. Is the temp set to plus 10c ? Can't see a minus sign.
Edited ...
Like
NonkelErik 0.00
...
· 
·  1 like
·  Share link
Yes +10 deg C
Like
Emission 2.11
...
· 
·  1 like
·  Share link
TiffsAndAstro:
Tobiasz:
TiffsAndAstro:
Erik Derboven:
sharpcap TPK 585~2.png

something looks odd as there is no unity gain? Also 585 is 12bit not 16bit.

585 is 12Bit in LCG/HCG Mode. HDR output is 16Bit.


I'm really wondering how this works. I need to read more into this, apologies. Is the temp set to plus 10c ? Can't see a minus sign.

Temperature should not matter, but Touptek uses minus 10 degrees in their manuals.

If you want to read more about it, then Sony's Starvis 2 and "ClearHDR" are the terms you are looking for. 

No need to apologise.
Like
StewartWilliam 5.21
Topic starter
...
· 
·  1 like
·  Share link
So I ask the question again, what is the unity gain of this camera when using the HDR mode, as there is not one on that chart….??/
Like
Emission 2.11
...
· 
·  Share link
There is no unity gain in HDR mode, because the sensor has already 0.44e/ADU in base gain, which is 100. Regarding the touptek manual of the 585, unity gain (1e/ADU) can be achieved with High Conversion Gain mode and gain 100. 

You are free to use HCG/gain100 but the HDR mode is and will be superior to HCG regarding full well, read noise und dynamic range.
Edited ...
Like
TiffsAndAstro 1.81
...
· 
·  Share link
here's my test of my 533 couple weeks ago for comparison in case it helps any. 533 has a 14bit adc and sharpcap reports it as 14bit, so im a tad wary if sharpcap says 585 is 16adc. is it possible the revised driver converts 12 bits to 16 bits and could that account for the lack of unity gain showing? i think both sensors output 16bits to nina when imaging, but the 533 doesn't seem to in my sharpcap, so i have to wonder why the 585 says 16bit in that test?

533 at -5 compressedforastrobin.jpg
Like
Emission 2.11
...
· 
·  2 likes
·  Share link
TiffsAndAstro:
here's my test of my 533 couple weeks ago for comparison in case it helps any. 533 has a 14bit adc and sharpcap reports it as 14bit, so im a tad wary if sharpcap says 585 is 16adc. is it possible the revised driver converts 12 bits to 16 bits and could that account for the lack of unity gain showing? i think both sensors output 16bits to nina when imaging, but the 533 doesn't seem to in my sharpcap, so i have to wonder why the 585 says 16bit in that test?

533 at -5 compressedforastrobin.jpg

Yes, the 533 has a 14bit ADC which combined with the other sensor specs results in measured 14Bit Output. Regarding NINA and 16Bit, I think NINA does some bit scaling with images.

As far as I understand Sharpcap does not report the ADC resolution from the sensor directly but calculates it based on the images and measurements it takes. 

I can only repeat. The 585 does ClearHDR imaging, which takes two images at the same time and merges them on the sensor directly. The sensor is doing that, what we are doing with astrophotography: Stacking/Merging images to gain a higher dynamic range.

Edit: Now that I see your screenshot and Gain0/50. You have a SVBONY camera correct? Most brands do not have modes like "Low Conversion Gain" or "High Conversion Gain", the switch is done "on the fly" while raising the gain. You can see it in your sharpcap result between Gain 50 and 100. There is a great fall off of read noise and full well, because the camera switches over to high conversion gain. 

Touptek cameras don't do that, you have to know which mode you want to use first and choose it via dropdown in NINA directly. And regarding the 585 you get a third option, which is "HDR". With this camera I would forget about using LCG/HCG and simply use HDR forever.
Edited ...
Like
StewartWilliam 5.21
Topic starter
...
· 
·  2 likes
·  Share link
TiffsAndAstro:
here's my test of my 533 couple weeks ago for comparison in case it helps any. 533 has a 14bit adc and sharpcap reports it as 14bit, so im a tad wary if sharpcap says 585 is 16adc. is it possible the revised driver converts 12 bits to 16 bits and could that account for the lack of unity gain showing? i think both sensors output 16bits to nina when imaging, but the 533 doesn't seem to in my sharpcap, so i have to wonder why the 585 says 16bit in that test?

533 at -5 compressedforastrobin.jpg

I think it’s been explained multiple times why the 585 shows 16 bit in the chart, this is a superb feature of the camera and apart from the smaller sensor is superior to the 533 in many respects, so your chart has no real bearing on this camera as it’s a different beast altogether…
The HDR mode is very interesting and clever how it works and maybe something that could be built into other 12 and 14 bit Touptek cameras in the future…
Like
Emission 2.11
...
· 
·  1 like
·  Share link
AstroShed:
TiffsAndAstro:
here's my test of my 533 couple weeks ago for comparison in case it helps any. 533 has a 14bit adc and sharpcap reports it as 14bit, so im a tad wary if sharpcap says 585 is 16adc. is it possible the revised driver converts 12 bits to 16 bits and could that account for the lack of unity gain showing? i think both sensors output 16bits to nina when imaging, but the 533 doesn't seem to in my sharpcap, so i have to wonder why the 585 says 16bit in that test?

533 at -5 compressedforastrobin.jpg

I think it’s been explained multiple times why the 585 shows 16 bit in the chart, this is a superb feature of the camera and apart from the smaller sensor is superior to the 533 in many respects, so your chart has no real bearing on this camera as it’s a different beast altogether…
The HDR mode is very interesting and clever how it works and maybe something that could be built into other 12 and 14 bit Touptek cameras in the future…

As far as I know the ClearHDR mode is only available on Starvis 2 Sensors, which is the security camera branch of Sony. The image merge is done on sensor, which means it is a hard-and software feature. Enabling it via firmware only won't be possible, I think the sensor hardware must support it, too. 

I guess the next big milestone would be Sony adapting the ClearHDR feature into the next generation of bigger sensors like 4/3, APSC and FF. But this will depend on the demand of the brands that produce sony-based handheld cameras like Sony themselves or Nikon, ... and so on. If the sensors become widespread available, Touptek can/will get their share. 

But the 585 is still fine though, smaller sensor and FOV does not need a high focal length or big corrected field to work, so you have less constraints regarding the telescope. With a smaller telescope you dont need a bigger mount. It has its advantages

Regards
Like
TiffsAndAstro 1.81
...
· 
·  1 like
·  Share link
Tobiasz:
TiffsAndAstro:
here's my test of my 533 couple weeks ago for comparison in case it helps any. 533 has a 14bit adc and sharpcap reports it as 14bit, so im a tad wary if sharpcap says 585 is 16adc. is it possible the revised driver converts 12 bits to 16 bits and could that account for the lack of unity gain showing? i think both sensors output 16bits to nina when imaging, but the 533 doesn't seem to in my sharpcap, so i have to wonder why the 585 says 16bit in that test?

533 at -5 compressedforastrobin.jpg

Yes, the 533 has a 14bit ADC which combined with the other sensor specs results in measured 14Bit Output. Regarding NINA and 16Bit, I think NINA does some bit scaling with images.

As far as I understand Sharpcap does not report the ADC resolution from the sensor directly but calculates it based on the images and measurements it takes. 

I can only repeat. The 585 does ClearHDR imaging, which takes two images at the same time and merges them on the sensor directly. The sensor is doing that, what we are doing with astrophotography: Stacking/Merging images to gain a higher dynamic range.

Edit: Now that I see your screenshot and Gain0/50. You have a SVBONY camera correct? Most brands do not have modes like "Low Conversion Gain" or "High Conversion Gain", the switch is done "on the fly" while raising the gain. You can see it in your sharpcap result between Gain 50 and 100. There is a great fall off of read noise and full well, because the camera switches over to high conversion gain. 

Touptek cameras don't do that, you have to know which mode you want to use first and choose it via dropdown in NINA directly. And regarding the 585 you get a third option, which is "HDR". With this camera I would forget about using LCG/HCG and simply use HDR forever.


yeah i watched cuiv's video some time back. i'd imagine other 'manufacturers' using imx533 will be implement it on their versions if they wanted to. ill go find out how the imx533 is able to turn 12bit into true 16bit on chip as someone said happens, above. perhaps touptek make use of the ddr buffer to store both images and then some processing capacting to merge them or something.
Like
TiffsAndAstro 1.81
...
· 
·  1 like
·  Share link
Tobiasz:
AstroShed:
TiffsAndAstro:
here's my test of my 533 couple weeks ago for comparison in case it helps any. 533 has a 14bit adc and sharpcap reports it as 14bit, so im a tad wary if sharpcap says 585 is 16adc. is it possible the revised driver converts 12 bits to 16 bits and could that account for the lack of unity gain showing? i think both sensors output 16bits to nina when imaging, but the 533 doesn't seem to in my sharpcap, so i have to wonder why the 585 says 16bit in that test?

533 at -5 compressedforastrobin.jpg

I think it’s been explained multiple times why the 585 shows 16 bit in the chart, this is a superb feature of the camera and apart from the smaller sensor is superior to the 533 in many respects, so your chart has no real bearing on this camera as it’s a different beast altogether…
The HDR mode is very interesting and clever how it works and maybe something that could be built into other 12 and 14 bit Touptek cameras in the future…

As far as I know the ClearHDR mode is only available on Starvis 2 Sensors, which is the security camera branch of Sony. The image merge is done on sensor, which means it is a hard-and software feature. Enabling it via firmware only won't be possible, I think the sensor hardware must support it, too. 

I guess the next big milestone would be Sony adapting the ClearHDR feature into the next generation of bigger sensors like 4/3, APSC and FF. But this will depend on the demand of the brands that produce sony-based handheld cameras like Sony themselves or Nikon, ... and so on. If the sensors become widespread available, Touptek can/will get their share. 

But the 585 is still fine though, smaller sensor and FOV does not need a high focal length or big corrected field to work, so you have less constraints regarding the telescope. With a smaller telescope you dont need a bigger mount. It has its advantages

Regards


i don't think imx533 is starvis2, which is a shame. i think its great touptek can improve on a 12bit sensor by doing stuff like this. pretty amazing.
Like
Emission 2.11
...
· 
·  1 like
·  Share link
TiffsAndAstro:
Tobiasz:
TiffsAndAstro:
here's my test of my 533 couple weeks ago for comparison in case it helps any. 533 has a 14bit adc and sharpcap reports it as 14bit, so im a tad wary if sharpcap says 585 is 16adc. is it possible the revised driver converts 12 bits to 16 bits and could that account for the lack of unity gain showing? i think both sensors output 16bits to nina when imaging, but the 533 doesn't seem to in my sharpcap, so i have to wonder why the 585 says 16bit in that test?

533 at -5 compressedforastrobin.jpg

Yes, the 533 has a 14bit ADC which combined with the other sensor specs results in measured 14Bit Output. Regarding NINA and 16Bit, I think NINA does some bit scaling with images.

As far as I understand Sharpcap does not report the ADC resolution from the sensor directly but calculates it based on the images and measurements it takes. 

I can only repeat. The 585 does ClearHDR imaging, which takes two images at the same time and merges them on the sensor directly. The sensor is doing that, what we are doing with astrophotography: Stacking/Merging images to gain a higher dynamic range.

Edit: Now that I see your screenshot and Gain0/50. You have a SVBONY camera correct? Most brands do not have modes like "Low Conversion Gain" or "High Conversion Gain", the switch is done "on the fly" while raising the gain. You can see it in your sharpcap result between Gain 50 and 100. There is a great fall off of read noise and full well, because the camera switches over to high conversion gain. 

Touptek cameras don't do that, you have to know which mode you want to use first and choose it via dropdown in NINA directly. And regarding the 585 you get a third option, which is "HDR". With this camera I would forget about using LCG/HCG and simply use HDR forever.


yeah i watched cuiv's video some time back. i'd imagine other 'manufacturers' using imx533 will be implement it on their versions if they wanted to. ill go find out how the imx533 is able to turn 12bit into true 16bit on chip as someone said happens, above. perhaps touptek make use of the ddr buffer to store both images and then some processing capacting to merge them or something.

I think you meant the 585 sensor, because the 533 is no ClearHDR sensor. Other vendors like ZWO or QHY can also implement the HDR mode with their 585 sensors, but it will take R&D resources to develop the right HDR algorithm that works. Depending on this whole HDR escalation situation in this thread Touptek needed several weeks the develop their own algorithm. I'm afraid this is an investment not every vendor wants to do.

There are other HDR variants that could be utilised by non-clearHDR-sensors, but there you have the same problem: the cost of development. 

Regards
Like
Emission 2.11
...
· 
·  1 like
·  Share link
TiffsAndAstro:
Tobiasz:
AstroShed:
TiffsAndAstro:
here's my test of my 533 couple weeks ago for comparison in case it helps any. 533 has a 14bit adc and sharpcap reports it as 14bit, so im a tad wary if sharpcap says 585 is 16adc. is it possible the revised driver converts 12 bits to 16 bits and could that account for the lack of unity gain showing? i think both sensors output 16bits to nina when imaging, but the 533 doesn't seem to in my sharpcap, so i have to wonder why the 585 says 16bit in that test?

533 at -5 compressedforastrobin.jpg

I think it’s been explained multiple times why the 585 shows 16 bit in the chart, this is a superb feature of the camera and apart from the smaller sensor is superior to the 533 in many respects, so your chart has no real bearing on this camera as it’s a different beast altogether…
The HDR mode is very interesting and clever how it works and maybe something that could be built into other 12 and 14 bit Touptek cameras in the future…

As far as I know the ClearHDR mode is only available on Starvis 2 Sensors, which is the security camera branch of Sony. The image merge is done on sensor, which means it is a hard-and software feature. Enabling it via firmware only won't be possible, I think the sensor hardware must support it, too. 

I guess the next big milestone would be Sony adapting the ClearHDR feature into the next generation of bigger sensors like 4/3, APSC and FF. But this will depend on the demand of the brands that produce sony-based handheld cameras like Sony themselves or Nikon, ... and so on. If the sensors become widespread available, Touptek can/will get their share. 

But the 585 is still fine though, smaller sensor and FOV does not need a high focal length or big corrected field to work, so you have less constraints regarding the telescope. With a smaller telescope you dont need a bigger mount. It has its advantages

Regards


i don't think imx533 is starvis2, which is a shame. i think its great touptek can improve on a 12bit sensor by doing stuff like this. pretty amazing.

Unfortunately, the IMX533 is not a Starvis 2 sensor and therefore not capable of doing ClearHDR imaging. But the IMX533 has other advantadges like the bigger FOV, so it depends on the customer what he or she prefers. 

Because I like zoomed in FOVs, thats why I use smaller sensors like the 585. My 10 inch RC utilises an IMX432. 1,1" Sony sensor with 9µm pixels. At 2000mm it is a nice photon sucking sensor. 

Regards
Edited ...
Like
 
Register or login to create to post a reply.