Guidecamera bit depth calrification

[nicco...@gmail.com]

Hi all,

I need a bit of help with guidecamera bit depth and Saturation by max ADU value 

I am using a Touptek IMX290 mono camera. I know the sensor is 12bit, but  when connecting it to PHD2 I can choose between 8 and 16 bit, and I choose 16bit.

Saturation by max ADU value is defaulted to 65536

However, looking at the guidestar profile it seems truncated (saturated star) and Peak shows 4116. This makes me think that the data is in fact 12bit.

Nevertheless guiding is good so I am not sure changing the Sat-by max ADU to 4096 would be a good choice

thank you

[bw_msgboard]

How this stuff works all depends on the camera drivers and how they choose to scale the data.  It makes the most sense to have the saturation value reflect the actual max value that can be returned by the camera driver.  Otherwise, you will allow PHD2 to unnecessarily select a saturated star and that can work against you.  

 

Regards,

Bruce

 

---

--
You received this message because you are subscribed to the Google Groups "Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd-guidi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/2ed5a4e0-7c2c-4b9d-b061-ed6b4a4b309bn%40googlegroups.com.

[Niccolò Coli]

Thank you Bruce, how can I be sure of the max value that the driver returns? Looking at the peak value in star profile?

Regards

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/8E81A16489464891BB27EA94B9324BBF%40HomeDesktop.

[bw_msgboard]

Yes, as long as you’re pointed at something that ought to saturate the sensor – for example, a first magnitude star using a 2-second exposure, something like that.  If you’re going to do it this way, *temporarily* stop using the dark library because that can affect what’s displayed.  It’s probably going to be close to 4096 from what you’ve described, there’s no need for great precision.

 

Bruce

 

---

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CA%2BfX3quNC8C4B2dsNT21CGvj_Q2vihuAVkuzEb-80_XPTNAM%3DA%40mail.gmail.com.

[Niccolò Coli]

ok, it looks like despite the "12bit" indication in the driver the image data is rescaled to 16bit. But then a flat top of the istar profile does not necessarily mean saturated star I believe?

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/E604F54008A040648E6F929B9FF351DD%40HomeDesktop.

[Richard Beck]

You need to do the test Bruce described.  If the 12 bits are shifted to 16 by placing zeros in the most significant bits, then 4095 will be your maximum value.  If the 4 zeros are placed in the least significant bits, the maximum value becomes 4095*16 or 65520.  If the bits are assigned differently, the results would be different.

[Niccolò Coli]

It is what I did and concluded the image is scaled to 16bit.

I am asking the question since  I have noticed star profiles with truncated tip but not saturated ADU 

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/d9f649f4-e2b4-4f89-a7f9-ab129210fd57n%40googlegroups.com.

[bw_msgboard]

You can probably ignore the square-top appearance, I would just go with the measured saturation level.  You can always back off a bit on that – for example, I use a saturation value of 65000 on my 16-bit camera.  Particularly if you’re doing multi-star guiding, you are probably in good shape.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CA%2BfX3qsuL_i%2BasqNZ%3Dr9%3D4f9mzAcPQBdYwdTo%3DX46NBQcTHKBw%40mail.gmail.com.