Hey,
I'm not highly skilled at soldering, but a friend of mine is, and he also has the proper equipment available at his work.
"I have not looked at any pictures of results of this swap, but would suspect that with non-cooled cameras such as these, increasing sensitivity is going to come with a noise trade-off"
From my experience, the noise levels are the same as before. I often used the highest gain settings before, and have not noticed an increase in noise. I pretty much dare to say that at least 95% of the increase in sensitivity is increase in signal to noise.
"After a quick look at their web site, I would suggest that they would get better noise and durability results by using a secondary PCB and socket for routing signals than by using wires."
I have not done any extensive testing on this (testing sounds easy, but it is actually quite difficult and time consuming to get more or less reliable results... so for now you'll have to do with my 'experience'
). We do have some sockets for this purpose, but I first want to do some testing to see if the wires actually cause extra noise (by my gut feeling says that if it makes a difference, it will only be a very minimal amount, and probably not worth the risk of messing up a resoldering job)
I'm actually thinking of getting the Basler Ace camera within the next few months. Higher FPS, more output bits, same sensitivity, price comparable to the DMK21. But I don't want to be the first to test that camera extensively
I would if I had the funds though
btw; to prove that the DMK21 is still working nicely for planetary images:
http://www.astrokraai.nl/viewimages.php?id=150
http://www.youtube.com/watch?v=TiQHzvxW6Ng
I could probably get some raw data stretched to the same levels to give a better noise comparison. But that will be somewhere next week.