×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

Touptek Camera SDK Major Update

  • Posts: 21
  • Thank you received: 9

Replied by max on topic Touptek Camera SDK Major Update

I think I can continue with the driver implementation. I will issue the pull request on sunday.

I have a question concerning the rebranded Touptek cameras like the OMEGON. Will the OMEGON driver and the other brands also get an update soon or should
the fullwell mode only be implemented for the original touptek cameras?

CS Max
1 year 2 months ago #89618

Please Log in or Create an account to join the conversation.

  • Posts: 29
  • Thank you received: 3
OEM camera hardware and drivers are exactly the same as touptek, theoretically you only need to duplicate a copy of the SDK, Kstars-Tools-Customer Drivers could also duplicate a one.


INDI - Remote is also my favorite feature, I use ssh -R to publish 7624 on a public domain, but unfortunately it is not possible to encrypt it, it would be perfect if I could ask for a password when the client connects
Last edit: 1 year 2 months ago by lanco.
1 year 2 months ago #89625

Please Log in or Create an account to join the conversation.

  • Posts: 12
  • Thank you received: 1
I could provide the OEM branded SDK later.
1 year 2 months ago #89626

Please Log in or Create an account to join the conversation.

  • Posts: 239
  • Thank you received: 38
I installed 22.04 LTS and installed indi 1.9.9 and it worked. However the driver did crash first then upon reconnecting it worked both on my windows client and Ubuntu Kstars
1 year 2 months ago #89627

Please Log in or Create an account to join the conversation.

  • Posts: 12
  • Thank you received: 1
Build on Ubuntu 22.04.1 LTS with INDI Library: 2.0.0 and Kstars 3.6.3 and it is OK.

I think if there is an issue with SDK it is probably not the high full well capacity mode function. Our guess is the compatibility of the SDK, unfortunately, we could not reproduce the crash.

If anyone could provide detailed instructions on building the newest INDI library it would be really helpful. :D
1 year 2 months ago #89628
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 21
  • Thank you received: 9

Replied by max on topic Touptek Camera SDK Major Update

hi,
the driver is almost ready. What I don't like is that the label is cut off. See attachments: "High ... Mode" shall be "High Full Well Mode".

What do you think about abbreviating it: "HFW Mode" for the High Full Well Mode and "LN Mode" for the Low Noise Mode ?

CS Max
The following user(s) said Thank You: lanco
1 year 2 months ago #89631
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 12
  • Thank you received: 1
I think HFW makes sense and it makes the button looks clean :D
1 year 2 months ago #89643

Please Log in or Create an account to join the conversation.

  • Posts: 29
  • Thank you received: 3
The names are rather all right.

Finally, I hope that by the observation of the histogram after the real shot, we can select the combination of modes whose noise distribution is closest to the Gaussian or Poisson distribution.

For example, low noise mode and HCG mode may both destroy the regularity of the distribution, and whether the switch HFW mode improves the regularity of the distribution. This all determines the best combination of shooting modes, and the result of this combination can be called a Dark Nebula Photography Mode or QualityMode defaults.

Conversely, the optimal efficiency tuning, with HCG on + LowNoise on + HFW mode off, is the most efficient SpeedMode defaults for outdoor exposure.

A full well capacity is only one aspect. Because when the full well is particularly large, the average of an electron is not converted into one adu. too the pursuit of expanding the full well, the nebula dark part directly can not shoot things. What is more important is the regularity of noise distribution and the selection of balanced egain when conforming to that regularity
Last edit: 1 year 2 months ago by lanco.
1 year 2 months ago #89647
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 21
  • Thank you received: 9

Replied by max on topic Touptek Camera SDK Major Update

I have issued a pull request for the new feature.
Please find attached two histograms with LCG and HFW on/off.

CS Max
The following user(s) said Thank You: Sonny Cavazos, Bernd Limburg, lanco
1 year 2 months ago #89650
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 145
  • Thank you received: 15
Hello,

I built the latest Toupcam driver and tested the HFW switch with a DeepSkyPro 2600c IMX571 color camera.

Old SDK, LCG (no HFW obviously):


New SDK, LCG, HFW on:


Old SDK, HCG, no HFW:


New SDK, HCG, HFW on:


(The values for the R channel are bad because my flats used here had too low red ADUs, so best look only at the G channel)

Fullwell capacity is approx. doubled, but e/ADU has doubled as well. Driver gain was 100 (lowest) in both cases. So unity gain in LCG+HFW on mode must be at a higher driver gain than 100. Has anyone figured this out? Or maybe a gain curve?

CS, Bernd
Last edit: 1 year 2 months ago by Bernd Limburg.
1 year 2 months ago #89685
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 29
  • Thank you received: 3
Actually toup has several sets of gain calibrations that I have measured in the past because the corresponding egain changes after the HCG switch. The first floor is Ascom's test, which could be a reference.

I will build and try HFW tonight
Last edit: 1 year 2 months ago by lanco.
1 year 2 months ago #89691

Please Log in or Create an account to join the conversation.

  • Posts: 114
  • Thank you received: 17
I have questions about current stable indi drivers for Touptek cameras. What are these options ("Dual gain" and "Dual gain mode") for? And why "Dual gain mode" is "High" by default?
1 year 2 months ago #89698
Attachments:

Please Log in or Create an account to join the conversation.

Time to create page: 1.394 seconds