Webcam Settings

[Clayton]

No worrys Clayton i need to stop assuming that others know what i mean by being broad in my terms. But obviously one wouldnt drop frame rate for no reason that would just lose data, which is why i assumed others would know what i mean. But looking back it is confusing i agree. again apologies.

I think i will try captures at lower frame rates and different exposures other than 30 fps. to see if this wasnt a one off. as i already said it may have been a effect of better seeing at that moment, or my focusing may have been tighter, just to understand if im assuming right about these quality estimation errors. but if i am right, then the combined effects of these errors are so damaging that unless a way can be found to remove them quickly then smaller stacks with smoother frames seems a advantage.

Which i admit is at odds with what ive always belived and probably only under good seeing would it work any way. I need to experiment at 1/23rd 1/19th 1/15th to see the effects. as per my usual 30 fps 1/30 secs settings at the magnifications ive been trying to achieve, gain is pretty much maxed out with those settings. Apologies Gavin think ineed to be discussing this elsewhere

I hope this is OK with you Neil

I have several questions.

1. What is the final f ratio that you are using? It looks like it might be quite high if the image of Jupiter that you posted earlier was full size.

2. What camera?

3. What capture software? (I'm guessing K3CCD from earlier comments)

4. How large is a typical stack & why do you have smaller stacks at lower frame rates ? (the first part of this question may provide the answer to the second if the stacks are quite large)

Just to let you know. I also believe that Registax arranges quality in ways that at times don't agree with our/human ideas of what is best, and can lead to less than best results. For example imagine if our object was a perfect circle with 2 dots in the middle very close to each other. Depending on the seeing the dots may merge or seperate, the planet may appear squashed in one direction or the other or circular. I feel that Registax is very good at sorting the sharp dots from the blurred dot/s, but not so good at dealing with a squished planet with sharp dots. I don't know this for sure but it's an impression I have gained.

[neil phillips]

Hi Clayton F ratio difficult to judge i have a 2" Orion adapter on the focuser, i also used a 3 inch tube extension i used on that shot.

Made from a old rubbish barlow that came with a celestron firstscope, So about 5" extension with a TV 3x barlow scope f6 im guessing about about a 5x barlow, so f30, scope is 1800mm focal length so 5 x that is 9000 mm focal length pretty healthy for 20 degrees,

DFK21a f04

IC capture

OK cross wires again, when i say small stacks i mean half the frames captured when going from 30 fps to 15 fps hence a smaller frame stack , not a file size comment i was reading Mike salways tutorial where he mentioned this

quote

A faster framerate is great if the object is bright enough, but there’s no point capturing at 30fps (1/30s exposure) if your histogram falls off at under 120 (see image on the left below). The dim image will need to be stretched more during post-processing, which will reduce the depth and quality of the final image. Ideally you should change back to 15fps (1/15s exposure) and get more light in with each frame if the seeing is very good, as there won’t be much movement within each frame so the longer exposure won’t hurt your resolution

end of quote

Because of the high magnification on this shot, gain at 30 fps at 1/30 secs was maxed out,. but the image was still fairly dim.

Seeing for 20 degrees was quite stable this night, so combine these facts with the reduction in quality registration errors. By stacking fewer frames at 15 fps rather than 30 fps, it seems no surprise this was the best capture.

How much was due to the better quality avi at 15 fps, or the reduction in quality estimation errors, I cant honestly say. but i suspect both improvements helped Clayton

[Clayton]

Have you thought about dropping the image scale back a little to reduce the f# and so allow a faster shutter speed, without overdoing the gain?

Personally I would prefer to use a larger image scale than I do, but $$$ at the moment are scarce

I did some calc's tonight. Jupiter, depending on the source, is currently around 40 arcseconds across, it covers about 100 pixels on my 640 x 480 sensor that is = to 0.4"/pixel when I operate the 200mm f6 with 2x Barlow. If you were to run your 300mm with a 3x barlow that would be = 0.1"/pixel. If you add the extension to run at f30 then it is about 0.05"/pixel.

I feel (don't know for sure) that as seeing is rarely going to be sub arcsecond and resolution of any detail won't be lost by for eg dropping the scale to half (0.2"/pixel @f9) but the gain in available light will help you to freeze seeing and keep those stacks smaller with smoother frames?

This is all just theoretical. I dont really know but I believe it is worth a try:icon_salut:

[neil phillips]

Maybe worth a go Clayton but its a little late in the day for doing much more with jupiter i feel, not sure about the seeing at my new location yet, it might be tad worse though, but was pretty bad last night. I find my setup works well with the tv 3x barlow but this larger secondary does not. so going to have get the smaller one on as soon as i can, under calmer conditions i think it is worth staying at 3x as this image shows with the smaller flat fitted. under calmer air

This has about a 1 mm clearance with the primary inside the secondary. if i can get very precise collimation with this. then the central obstruction is about as small as it can get with a 12" f6 but contrast doesnt seem to be the only benefit as the images appear sharper with this secondary. Why i dont know.?

I must admit my brain doesnt like going down in scale, im constantly pushing the other way which can often back fire on me. But just pretty pictures isnt the way to go i feel.

resolution i think does suffer a little with smaller scales. even when seeing is terrible often a image will show more than a smaller scale version. but doesnt always look particulaly nice, pro and cons i guess.

But some of the better shots ive pulled off over the years have been because i think like this. and dont forget high scale perfecting both in terms of collimation, camera settings focusing ect. is really very good training for those nights that are waiting to be pushed.

If one doesnt train. then i find it hard to suddenly just pull all that together on a good night.

Through repatition one can be prepared. Mars for example in the coming months is going to be high. but small at 14 arcseconds so would be nice if i could get the system and myself working well at large image scale for that.

But i must admit often it is worth dropping back a bit, your shot recently is a prime example. I thought it was very very nice indeed. both in terms of detail and colour. processing ect. But would still like to have seen what you would have got if you had gone up a little. I noticed you said it was about 6/10 seeing which surprised me. as the image quality looked far better than that.

If your location is that good Clayton i will play devils advocate with you, and suggest under those conditions. or better, as you stated seeing was 6/10 then i suggest you could go higher and still get the nice sharp detaill you had on that shot. showing much finer detail.

But i will think about what you have said and maybe experiment. It never hurts to experiment does it. Especially when someone suggests something to me, when ive seen them produce something that impresses me like your shot did. very much so.

Now i need to think about ridding software of registration errors. Oh no hours of manually going through frames argh

[Clayton]

That image is certainly a nice one. I have just gotten over big secondary mirror drama's myself (mine was undersize for where it was placed along the tube. I ended up moving the primary back as it seemed easier than relocating the secondary and the focuser forward).

One thing I found in my investigations is that figure on secondary mirrors is not always that good so it is possible to get a dud every now and then. But I don't think its all that common to get a really bad one. A star test should show the difference between the two.

Mars here will be a fizzer this time I think as it is going to be very low in the North. Being in the middle of the city on a summer evening doesn't help the seeing. But I will have to give it a go

I would have liked to up the scale a bit on that Jupiter shot as well, but as I don't have the gear yet :( The seeing was what we locals call "fast" which is where the whole planet goes from fuzzy to sharp and back, but doesn't wallow around like looking at the bottom of a swimming pool (I have a theory that Registax has trouble with the latter.) Most of the time it was fuzzy which is why I called it 6/10 and why the images ended up with 52, and 83 frames respectively from 1200.

My 200mm Newt is limited to about .5" resolution if it were perfect. I think there might be some gains to be had by me pushing scale to around .2"-.3" per pixel, but it is only a feeling at this stage and I won't have the time or gear to try until Jupiter is long gone Maybe nex year :D

I know what you mean by the horror of manually sorting frames.

I uset to get 15sec avi's of the Moon before I had a driven mount. Manually sorting through even just 150 frames tested me:icon_eek:

[Clayton]

Whoops I think I just shot myself in the foot re. the seeing.

When I checked it was the other way around With a sharp but wobbly view That'll teach me to rely on an old man's memory Plus it sort of shoots my Registax theory down as well

[Clayton]

Some interesting reading here on your camera Neil

http://www.theimagingsource.com/downloads/icx098bq.en_US.pdf

Check out page 9 which may explain the image degradation at 60fps.

I'm not sure if I am interpreting this correctly ?

[neil phillips]

Sorry for the late reply Clayton ive recently just got back on the net. Ill check this link out thanks for the info