ToUcam RAW Vs. Non-Raw

[gbeal]

OK, here is an interesting concept. Anyone given the RAW treatment a go with their ToUcams?
When the idea first floated a year or so ago, maybe less I tried it but only in a half-hearted way. It didn’t seem to produce those miracle images that all the good guys were knocking out. I soon returned to “Normal”.
Well more recently, I have re-visited this idea, and want to share my meager experiences, as well as see whether others within the group have also tried this.
RAW (if I understand it correctly) removes any of the inbuilt sharpening and other image adjustments that are automatically done within the camera. How these are done escapes me, and really I don’t want to know, but they are. In an effort to extract the very best from your images, it is useful (evidently) to gain all the information, and process/manipulate after the exposure. My Nikon D100 does the same thing when set to the “Raw” setting.
Changing your ToUcam to RAW is simple, using freely available software macros. It takes a mere minute, and is not irreversible. I did mine a few weeks ago.
The first image (of Saturn) showed some promise, so I was off, convinced it would the savior I was seeking.
On the 31st March I imaged, flat out, with both the Mewlon 180, and also my workhorse, the 10” f5 newt, both on the G-11. Target of the night was Jupiter, with Io transiting. Seeing was better than normal, at an estimated 7 or 8/10. In moments where I removed the ToUcam and slipped in an eyepiece the actual view was very good, with the shadow an inky black dot. I was happy, and also tried the 10”, just to guage the difference from an imaging point of view.

The processed results were less than spectacular, I was crushed. I tried it again the next night with similar results. Images in average, to better than average seeing were “soft” with nothing really standing out. Colour also appeared to washed out. In recent posts, (by Exfso I recall) it was suggested that exposure be controlled only by gain, and shutter speed. Gamma will be at zero, and brightness at 50%. I adhered to this, and to be fair the onion syndrome appears to have departed. But the results were crap.

SO……………. I reversed this setting, returning to “Normal”. I wanted to try again, and with similar subject, seeing, and equipment. The 3rd April, a couple of nights ago, I did this, with Jupiter again. Seeing wasn’t quite as good, so so, and about 6/10, but the eyepiece view was not as good as I remembered the 31st. This time similar setting were used, with similar amounts of frames collected. On both evenings, I used 5fps (this is virtually a necessity with Raw, but discretionary with Normal).

The colour is now there, and so is the sharpness, well certainly better than it was. Attached is a composite with both evenings represented. It is not overly scientific, but to me, I must be doing something wrong when I use Raw mode, as others have has extremely successful results. Anyone else tried this, or had any experience with this?

[iceman]

Very interesting Gary, and interesting results.
I've read a bit about the RAW mode over on CN, and lately the thing to do seems to be the OPTIMISED-COLOUR mode. I haven't read into the technicalities of exactly what settings it changes inside the camera, but others have claimed success with that method.

To be honest though, noone has done (that I have seen) a true side by side comparison of raw vs non-raw, optimised-colour vs normal, etc. Yours is a good attempt at a comparison, but even still they are several nights apart, conditions were different, transparency could've been different, settings could've been different, registax might have done things differently, your focusing might've been slightly different. It's really hard to get a true comparison.

In your images, while the colour appears better in the non-raw image, I actually prefer the detail in the raw one. The non-raw one has severe onion ringing, possibly due to lack of exposure/gain (going by Eric Ng's video). Although the raw shot looks over-exposed (slightly) in the middle EQ zone.

I'm not sure if the eyepiece projection method you're using affects it at all either, you're the only one i've seen imaging like that lately. Have you tried at prime focus to see if it makes any difference?

I'm going to try the optimised-colour mode on my ToUcam next time out and then i'll change the settings back in the same night and try and do a comparison also. We need to solve this once and for all and we need as many experiments by as many different people as possible.

Great thread Gary!

[gbeal]

Hhmmmmmmmmmmmm, interesting side affect Mike, and one which I agree, we should kick into touch.
The eyepiece concept is primarily to allow some latitude with seeing. I normally simply grab the 2.5x Powermate (or with the 10" f5 newt, the 5x Powermate) and shoot. This gives about f30, & f 25 respectively, but with slightly different image sizes.
If the seeing isn't up to speed then I can reduce the image scale by using E/P projection instead.
Here's the plan. I will try this again, same night, same scope, same focus, and same Powermate, with the only change being the use of Raw/Non-Raw/Optimised Colour. Then we may get a decent comparison.
I suggest you try this as well, or others. I am actually happy to just play dead, and stick with "normal" (makes a change to have some normality), but then again it bugs me when I can't make it happen.
Hopefully some of the other diehards will have a crack at this as well.
Gary

[[1ponders]]

gary what sort of setup are you using for EP. I've played with it occasionally with moon shots using a tele extender with an eyepiece inserted and the canon. How do you set it up with the ToUcam?

[Robby]

Very interesting Gary,
I haven't done the RAW thing yet, and I'm glad you have agonised over it for our benefit.. Perfect! Full credit as usual.
I may try Jupiter tonight if the conditions stay fine. Will keep you posted. I just checked your website www.star-mate.com/zeissnut and you last years Jupiter shots were stunning. What's gone wrong I wonder? Perhaps you should have kept that Mak! :)
Cheers

[gbeal]

Robber, you stirrer, Aaron and I have just been talking that one through. Don't be surprised if you see another one in the future.
I would have thought you of all people would have tried this Raw thing, as it requires brainpower, something I lack.
Why not do the Raw conversion, and try the experiment, as I have outlined above. The more that try this the better.
Gary

[gbeal]

Paul,
When Rob and I got the ToUcams, the scrooge in us (me mainly) said why not get some ToUcam to 1.25" adaptors made locally, and Rob sorted this, with great effect.
In a flash of brilliance I got mine made with a thread on the front, the side nearest the telescope, and this thread is 24x.75mm. This size is the same size that a Zeiss ortho has threaded on it's eye lens side, once you remove/unscrew the ring on this end.
All I do is screw the eyepiece onto the adaptor, and attach the ToUcam to the other end. Eyepeice fits into the focuser (OK, via a .96"/1.25" adaptor), and away you go. With the likes of the newt (1250mm focal length) an O-8mm gives a "similar: image scale to the 5x Powermate. With the Mewlon (2160mm focal length) the O-16mm gives "similar" to the 2.5x.
What I like is that when the seeing won't allow the O-8mm/5x Powermate I can use the O-10, and accept slightly better image detail, but at a reduced image scale.
Make sense?
Gary

[gbeal]

OK, all you ToUcammers.
For some obscure reason I thought this would cause you all to rip outside and try your own, rather than rely on me to do this for you.
Tonight it was sort of OK, with seeing about 5/10, and a little high cloud which really shifted about, so transperancy about 4/10.
I was busting to try the more scientific approach, since Mike shot the last one down as less than.
Tonight, one scope (Mewlon 180), one barlow (Powermate 2.5x, so f30), one focus, and three avi's pretty close together with about 2 minutes beteen to change the camera from "Normal" to "Optimised Colour", to "Raw Colour". The three avi's comprised about 650 frames each.
All were stacked in Registax, and had a very slight levels tweak in Photoshop, where they were all stuck onto the one piece of black, so you have one, instead of three images. Magic really.
Right that's me, now it is your turn.
Gary

[Robby]

Too easy Gary. Full credit.
I know which one I prefer... and it would seem that Raw is not what is't cracked up to be!
Cheers

[gbeal]

Robber, not sure about that, it could be just the way I went about it. That is the reason I was hoping others would climb aboard and give it a go. It may be a good thing to try over the Taupo weekend.
Gary

[Robby]

I would try Gary, but despite this big Hi sitting over us, we have seen noting but cloud & rain the last few nights.
Heading east tonight to be ready for the eclipse tomorrow morning. Fingers crossed...

[iceman]

lol sorry gary :)

Your new images look good, and a good approach too. Shame about the seeing and transparency.

From your results, the optimised colour looks nicer with the minimal processing you've done. However the raw mode, if indeed it turns off sharpening, might just mean that it requires extra post-processing to sharpen it up post-capture.

Chris Go on the CN forums did a normal vs optimised colour comparison, and his results showed that the optimised colour looked smoother with not as much grain, but not as sharp. So it seems like each method might require specific processing to get the most out of it.

I'll be doing some experiments next time the weather is good and i'm in a toucamming mood.

Nice work gary.

[gbeal]

Hi Mike,
thanks. If nothing else it gets discussion going, and with the level of expertese on this forum we should be able to nail this.
In all honesty, I was wanting the Raw mode to be the answer, and unfortunately the results haven't yet decided this. Perhaps you are right, this mode does indeed need more agressive waveletting, and I will go back and try this.
Following last nights exercise I set the camera to Optimised Colour, and shot 760 frames with the same setup. I will post it in the appropriate forum (I see you shifted this one as well).
Regards Gary
P.S, C'mon guys give it a crack, don't be like Rob, give it a go.

[[1ponders]]

I'd love to give it a go Gary. Would you mind if I sent all this cloud over your way for a while. Just to give us "sunny" queenslanders a break :(

Do you have the links for the RAW and Colour Optimized coversions?

Or email me your bmps or zipped avies and I'll have a play with those.

[[1ponders]]

I guess everyone has their different approach. I really pushed the brightness and contrast down to see what had happened to the colour, detail etc. I think your's have a more natural colour. Couldn't do much with the detail though. I havn't included the raw one. It looked terrible. Onion rings, bad colour balance. I'll stick to either normal or colour optimized I think

[rumples riot]

Interesting stuff Gary, I might get my a** into gear and do something for this thread. However, I tend to find someting that works for me and stick with it. I don't really consider myself expert enough to image in Raw yet. I have heard that it is very hard to get the setting right. But for you I might give it a go, if this cloud ever leaves us.

[iceman]

Here you go Gary.

http://www.iceinspace.com/forum/showthread.php?s=&threadid=1597

[ErwinvdVelden]

Hi everyone,

I tried the RAW mode as well a year ago, but found that it posterizes too much because the 8 bits depth of each frame over the USB causes information loss compared to the 10 bits information that flows between the A/D converter and the in-camera image processor. By contrast enhancing the images before the 8 bits YUV 4:2:0 conversion and USB 1.1 data compression the low contrast information is preserved, resulting in less posterising when performing image processing on the PC.

Keep in mind that this RAW concept was invented to prevent data compression over the USB 1.1 bus, the reality however is that the 10 to 8 bits conversion without contrast stretching is far worse then the USB data compression loss, even at 10 fps.

Even with the optimised mode without the RAW mode you need in-camera processing because of the reasons mentioned above, of course you can fiddle with things like colour balance etc. as long as you apply enough contrast enhancement to prevent posterising when processing the images on the PC.

With the Vesta Pro I found the facory settings still better than the
the 'Bond' settings I tried a year ago:

(see attachment)

or

http://members.dodo.net.au/~erwinvandervelden/jupiter040325br.htm

Maybe the current 'Bond' settings have been improved in the meantime, I might give it another go one day.

RAW makes more sense when the webcam is equipped with a B/W chip, but then it still needs in-camera contrast enhancing.

Cheers,

[iceman]

Interesting comparison Erwin.. Did you process the two images identically, or did you do additional sharpening to the optimised colour image?

From my small experiment I think that the "normal" mode produces a sharper more detailed image, but I also believe that the optimised colour mode image can be processed to make it look like the normal..

So I guess it depends what you're after in your final image.. I guess by not sharpening it in the camera, leaves you more room to do other processing afterwards and the sharpening can be the last step. I don't know.. jury is still out.. just like the IR vs no IR filter comparison..

Have you done a comparison of IR blocking vs no IR blocking filter Erwin?

[ErwinvdVelden]

Mike,

I used addidtional sharpening to the optimised colour mode image. Still I find the normal image slightly better than the optimised colour mode one.

I always use a IR blocking filter because otherwise colour channels are smeared because of atmospheric dispersion.

Cheers,

[iceman]

Thanks for the info Erwin, but



have you ever done a side by side comparison to show the effect? I've yet to see a definitive one..

Also, does it work better on some objects than others? Ie: do some planets/moon give off more IR than others?

Thanks