attn DMK camera users...

bird (Anthony Wesley) · #1 10-04-2007, 05:06 PM

Having tried a DMK21AF04 camera last night in comparison to my other cameras I made some observations about the gamma setting that other people might find interesting.

Looks like the minimum displayed gamma value "10" corresponds to a real gamma of about 0.6 or so, and the maximum displayed value "22" corresponds to something like 1.5.

What does this mean, I hear you ask?

Well, the best setting for gamma when capturing is 1.0, ie no change to the image. I'd guess that on the DMK21AF04 this means setting the gamma slider to something like 15 or 16.

If you have the gamma less than 1.0 or greater than 1.0 then you're modifying the image data in a non-reversible manner that loses some information.

I've sent off a query to the manufacturer (TIS in Germany) asking for a formula to calculate actual gamma and gain values from the displayed register values, when I get that then we'll know exactly what the gamma slider should be set to.

regards, Bird

matt · #2 10-04-2007, 05:17 PM

That's very useful, Anthony.

You and I have discussed this gamma issue on numerous occasions.

Will be good to get some "concrete" gamm and gain values for the DMK.

Would be good to have a user interface that makes those values easier to interpret and apply

Well done

davidpretorius · #3 10-04-2007, 07:31 PM

thanks bird, buzz lightyear has been patrolling the gamma quadrant for a while now and appreciates the feedback!!

seriously, we are lucky little vegemites to have you around to actually explain what the hell the rest of us are doing when we move the sliders in apparent knowledgable ways!!!

Dennis · #4 10-04-2007, 08:20 PM

Thanks Anthony - that knowledge will be useful the next time I'm out using the DMK.

Cheers

Dennis

bird (Anthony Wesley) · #5 10-04-2007, 08:28 PM

Quote:

Originally Posted by davidpretorius

thanks bird, buzz lightyear has been patrolling the gamma quadrant for a while now and appreciates the feedback!!

seriously, we are lucky little vegemites to have you around to actually explain what the hell the rest of us are doing when we move the sliders in apparent knowledgable ways!!!

LOL, slide away DP, your guess is as good as mine...

iceman (Mike) · #6 11-04-2007, 06:59 AM

Cheers Bird, I've had mine set at 13 but I don't quite remember it going up to 22. I wonder if our versions of ICCapture are different, or if the .AS version is any different from the standard DMK that I've got.

Oh you're not using ICCapture.. hmm, i'll check my gamma slider next time out and see what it goes to.

Dennis · #7 11-04-2007, 07:16 AM

Mike

Using IC Capture and the DMK21AF04.AS, I can slide all the way from 10 to 22 for Gamma.

Cheers

Dennis

bird (Anthony Wesley) · #8 11-04-2007, 09:03 AM

Mike, I checked with IC capture on one of my Windoze boxes, and the displayed register values are the same as Coriander is showing me under linux. I think these ranges and values are read from the camera.

regards, Bird

bird (Anthony Wesley) · #9 12-04-2007, 11:07 AM

Well, according to the vendor (TIS), the story is:

Gain is a linear value between 0 and 36 db, (register value of 0 = 0db gain, register value of 1023 = 36db gain).

Gamma is N/10 where N is the register value. i.e. register value 10 is gamma 1.0. This has me a bit confused, as it sure didn't look like gamma 1.0 when I was testing it the other night, but now I'll re-test it and make sure.

cheers, Bird

iceman (Mike) · #10 24-04-2007, 07:16 AM

The last few captures for me were at gamma 16 and 18. I much prefer them at this level, and when I set it back to 10 I can see how low it is. I had been using 12, but 18 will be my new "norm".

Anthony, how is the comparison with the DX's going? Anything else to report?

bird (Anthony Wesley) · #11 24-04-2007, 08:40 AM

Mike, I've recorded a fair bit of comparison data, and had a quick look through it. As far as I can see the price of teh camera determines the quality of the image - the DMK is the lowest cost and the DX is the highest, with the DR2 in between, and this is reflected in the results.

The comparison between the DR2 and the DX is interesting - the DR2 probably shows lower noise, but the DX has much more gain available, and double the bandwidth.

Gettting the same image size between the DX/DR2 and DMK was tricky, in the end they are not quite the same but it's close.

I'll try and find time to write this all up in the next few days.

Bird

rumples riot · #12 24-04-2007, 09:52 AM

Hmmm, I wonder if the use of Gamma is more related to the optical system than anything else. I have done some experiments with the gamma and found that even going above the base level damages the information so much that it is un-usable. It looks washed out and cannot be corrected. This was with the c9.25. The final image does not look a patch on images taken on the base level gamma setting.

Whether this is because I have the older version of the DMK or a software related issue I don't know, but I do know that when using the SCT I will not be using gamma. It may be different when I use the 18", that may be another matter.

Just my observations.

Dennis · #13 24-04-2007, 12:29 PM

That is a good observation.

Although I’ve had the DMK for almost 4 months now, I haven’t had the opportunity to use it as extensively as I would have liked, mainly due to uncooperative weather patterns.

On the few occasions I have managed to get out and get it plugged in, I have noticed that I invariably favoured a setting of 10 (slider at far left) for both the Celestron C9.25 (F10) and Mewlon 180mm (F12) when using a TeleVue x2.5 Powermate.

At times, I have set the slider to 12 and the live image looked acceptable, but anything more that 12 definitely degraded the live image to the extent that during the very few excursions I have enjoyed, I didn’t even bother trying to capture an avi stream at anything other than Gain 10, so cannot comment on the quality of a fully processed avi stream at the higher gain settings. Maybe I should have given it a go based on some of the findings here?

Oh well, next time eh!

Cheers

Dennis

bird (Anthony Wesley) · #14 25-04-2007, 02:29 PM

I'd say this has more to do with the display that you're using to view the data - the gamma value should be at 1.0 when you capture, there's no way the optical system can have any effect on it :-)

However it's possible to have laptop displays (assuming we're all using laptops to capture) set up differently so that the image "appears" any way you like...

cheers, Bird

rumples riot · #15 25-04-2007, 04:36 PM

Hmm, I could send you some data with both Anthony. It could be display but I see less data transmission when having the gamma up higher. Just my experience. Better still we could do this test at Wartook Rise in 5 weeks.

Anto · #16 26-04-2007, 07:15 PM

Interesting topic.
I use the gamma low (10-13) for high contrast moon imaging, or (11-13) for Saturn and C ring imaging, or (10-12) for Jupiter , or (12-15) for Venus in UV band.
I think that high values of gamma add so much noise than the gain setting.
I use better the brightness setting in range (0-96).

Regards, Antonello