Luna Landscapes from the 20th April

[Roy Foreman]

Just finished processing these. Seeing wasn't great but some of my captures came out not too bad. On the image containing Ptolemaeus I can see more craterlets on it's floor than I can count, but I doubt these will be visible on these pages. We will see after downloading !

16" Newtonian,  ZWO 183 MM,  Baader FFC at 3x,  15% of 900 frames, Baader Red Filter,  shot in twilight.

Hope you enjoy,

[Roy Foreman]

No, the craterlets are not visible so here is a closer crop - see if that does it !

[CraigT82]

Absolutely fantastic detail there Roy. There's a bit of noise when you look into that expanded image of Ptolemaeus, have you tried stacking more frames to smooth that out a bit? 

[neil phillips]

As Crag said lovely detail that 16 " is pulling out lots on the Floor Ptolemaeus 

[Trevor N]

Excellent detail

[Mr Spock]

Lots of craters there. Very nice.

[WestCoastCannuck]

Wow!  Some great stuff here.  The compressed views are really nice.  And as you say, the craterlets in the expanded view are amazing in number!  Great work!

[Roy Foreman]

8 hours ago, CraigT82 said:

Absolutely fantastic detail there Roy. There's a bit of noise when you look into that expanded image of Ptolemaeus, have you tried stacking more frames to smooth that out a bit? 

Thank you Craig !  The noise is probably due to the fact that I had to turn up the gain in order to maintain a frame rate that wasn't too slow.  Yes I could try stacking more frames.  I did experiment with this a while back and found that when the seeing is not great there was no improvement as I was just adding lower quality frames to the stack. that is why I settled on 15%.  But, I will give it a try on this image and see how it turns out. Thank you once again Craig.

[Roy Foreman]

2 hours ago, WestCoastCannuck said:

Wow!  Some great stuff here.  The compressed views are really nice.  And as you say, the craterlets in the expanded view are amazing in number!  Great work!

 

Thank you very much - glad you like the images !  The seeing was not great and these were the only captures that managed to turn out reasonably sharp.

[Roy Foreman]

7 hours ago, neil phillips said:

As Crag said lovely detail that 16 " is pulling out lots on the Floor Ptolemaeus 

Thank you Neil.  I guess the advantage of using a large aperture is not one of resolution but of gathering more light to enable faster exposure time and freeze the seeing more. Glad to like the images !

[neil phillips]

52 minutes ago, Roy Foreman said:

Thank you Neil.  I guess the advantage of using a large aperture is not one of resolution but of gathering more light to enable faster exposure time and freeze the seeing more. Glad to like the images !

True about the exposure time Roy. But there are numerous examples of larger scopes producing better resolution.  Btw what 16" is it ?

[Luke]

Really enjoyed those views, thanks, Roy, packed with detail and very sharp. I quite like the noise myself in the crater closeup, it has a bit of a feeling of grain to me.

Edited April 23, 2021 by Luke

[Roy Foreman]

10 hours ago, neil phillips said:

True about the exposure time Roy. But there are numerous examples of larger scopes producing better resolution.  Btw what 16" is it ?

Hi Neil - Damien Peach uses a 14" SCT so yes you are probably right about larger apertures and resolution.

As for my 16" Newt - the OTA is from a Skywatcher Flextube Dobsonian, a cheap way of a decent optical set in a light weight tube. It is carried by a massive fork mount that I built myself - apart from the worm wheel sets and electronics.  RA worm wheel is 14" in diameter, and the Dec is 12".  Both are driven by large motors that are connected directly to the worm shaft. No gears, no belts, no backlash. It is fully go to and computer controlled.

The good thing about self building is that if anything goes wrong it is easy to put right. The bad thing is that if anything does go wrong you are on your own !

Interestingly, skywatcher's mirror is conical like an SCT with glass ribs, and held centrally, again like an SCT, so it is very lightweight and cool down time is quite good.

 

[Roy Foreman]

10 hours ago, Luke said:

Really enjoyed those views, thanks, Roy, packed with detail and very sharp. I quite like the noise myself in the crater closeup, it has a bit of a feeling of grain to me.

Hi Luke - so glad you enjoyed the images. It is interesting how views differ on noise and grain. Although normally I like to see clean images, sometimes as you say it can add a certain feel to an image.  In this instance I was pushing the image to its limits to try and show the craterlets, so the noise was inevitable !

[CCD-Freak]

Impressive images !!!

I hope you don't mind but I ran your image through Topaz DeNoise AI and it did a nice job cleaning up the noise.  I try not to push the sharpening too much.  

I find Topaz very useful for lunar and deep sky imaging.

 

[Roy Foreman]

You are right, John, it has done a pretty decent job of reducing noise and, as far as I can see on screen, little or no effect on the sharpness.

thanks for that, and glad you like the images

Roy

[neil phillips]

On 23/04/2021 at 17:32, Roy Foreman said:

Hi Neil - Damien Peach uses a 14" SCT so yes you are probably right about larger apertures and resolution.

As for my 16" Newt - the OTA is from a Skywatcher Flextube Dobsonian, a cheap way of a decent optical set in a light weight tube. It is carried by a massive fork mount that I built myself - apart from the worm wheel sets and electronics.  RA worm wheel is 14" in diameter, and the Dec is 12".  Both are driven by large motors that are connected directly to the worm shaft. No gears, no belts, no backlash. It is fully go to and computer controlled.

The good thing about self building is that if anything goes wrong it is easy to put right. The bad thing is that if anything does go wrong you are on your own !

Interestingly, skywatcher's mirror is conical like an SCT with glass ribs, and held centrally, again like an SCT, so it is very lightweight and cool down time is quite good.

 

Would love to see a pic of that setup Roy. Very cool you built your own mount for it. One way in controlling noise, is to capture twice the amount of frames. So 15% would then amount to 30 % with the same quality. If seeing doesnt vary a lot.

Slowing the exposure down is of course another option. But as you say faster exposures do indeed react quicker to rapid seeing changes in tandem with higher frame rates. 

what was your frame count on these during capture Roy. Because if it was me 10.000 would normally suffice controlling noise. Because  it will allow large stacks like 6 or 700 even 1000 frame stacks. Wavelet control can also minimize noise

 

Edited June 3, 2021 by neil phillips

[PembrokeSteve]

Hi, 

Fantastic images Roy.

I wouldn’t worry about the “tiny”  bit of noise - I prefer the razor sharp look that you have achieved.

Regards,

Steve

[neil phillips]

1 hour ago, PembrokeSteve said:

Hi, 

Fantastic images Roy.

I wouldn’t worry about the “tiny”  bit of noise - I prefer the razor sharp look that you have achieved.

Regards,

Steve

Me too its just he can have hes cake and eat it. and get that razor sharp image but without the noise. And without resorting to software that will soften the image, often with detail loss. If Roy is anything like a lot of us he might like to improve hes technique rather than settling for second best. We can leave  that up to Roy to decide. But for me and a lot of lunar imagers ( the noise is very much unwanted ) Again we can leave that for Roy to decide.  why settle for anything less. When he is putting so much time and effort into it ?

Describing techniques that can help Roy have the best of both worlds cant be a bad thing surely ?

[Roy Foreman]

Thanks guys for the suggestions for reducing noise and improving image quality - I am always striving to do better and am glad to take on board any new or revise technique that might help.

One of the reasons that noise is visible in Ptolemaeus is that it is a pretty substantial selective enlargement from the image above it - and that was taken at a focal length of 5400mm.  But some of it may be down to the camera I am using - a ZWO 183 MM.  This has tiny 2.5 micron pixels and a fairly large sensor resulting in a 20 megapixel output. At full frame it can't go any faster than 19 fps. And the resulting data stream is too much for a conventional spinning disc hard drive. I got myself a half terabyte SSD but it filled up too quickly, so had to replace it with a 1 terabyte SSD !  

It is for this reason that I limit myself to 900 / 1000 frames. Even so it still takes nearly 2 hours to stack each video stream - and that is with an i7 quad core processor. After an imaging run my poor computer is running all day for a few days !

I tried reducing the imaging area, first to a half, then to a quarter, of the sensor area. This allowed 64 fps and 2000 frames. Much quicker to process, but a significant drop in quality.  I also experimented with stacking differing percentages - 10, 15, 20, 25 and 50%.  15% seemed to be optimum, as with greater percentages you are stacking worse quality images.

Once again, I am open to any suggestions, and I am so glad that my images have impressed - I am a deep sky guy really !

Finally, several people have asked to see images of my self built mount. It is extremely difficult to photograph a large telescope in the confines of an observatory. I had to use an ultra wide angle lens with all the distortion a weird perspective that goes with it.  So here are a few of the more successful ones.  Please excuse the clutter and dust in the observatory !  The spring tensioning on the Dec drive is experimental but seems to be working really well, so I am in the process of upgrading the RA as well.  There is an ED 120 Equinox riding on the tube, and it can be swapped out for a 180 Mak or an 8" edge SCT, or in fact any other scope less than 10Kg with a vixen rail.

[neil phillips]

1 hour ago, Roy Foreman said:

Thanks guys for the suggestions for reducing noise and improving image quality - I am always striving to do better and am glad to take on board any new or revise technique that might help.

One of the reasons that noise is visible in Ptolemaeus is that it is a pretty substantial selective enlargement from the image above it - and that was taken at a focal length of 5400mm.  But some of it may be down to the camera I am using - a ZWO 183 MM.  This has tiny 2.5 micron pixels and a fairly large sensor resulting in a 20 megapixel output. At full frame it can't go any faster than 19 fps. And the resulting data stream is too much for a conventional spinning disc hard drive. I got myself a half terabyte SSD but it filled up too quickly, so had to replace it with a 1 terabyte SSD !  

It is for this reason that I limit myself to 900 / 1000 frames. Even so it still takes nearly 2 hours to stack each video stream - and that is with an i7 quad core processor. After an imaging run my poor computer is running all day for a few days !

I tried reducing the imaging area, first to a half, then to a quarter, of the sensor area. This allowed 64 fps and 2000 frames. Much quicker to process, but a significant drop in quality.  I also experimented with stacking differing percentages - 10, 15, 20, 25 and 50%.  15% seemed to be optimum, as with greater percentages you are stacking worse quality images.

Once again, I am open to any suggestions, and I am so glad that my images have impressed - I am a deep sky guy really !

Finally, several people have asked to see images of my self built mount. It is extremely difficult to photograph a large telescope in the confines of an observatory. I had to use an ultra wide angle lens with all the distortion a weird perspective that goes with it.  So here are a few of the more successful ones.  Please excuse the clutter and dust in the observatory !  The spring tensioning on the Dec drive is experimental but seems to be working really well, so I am in the process of upgrading the RA as well.  There is an ED 120 Equinox riding on the tube, and it can be swapped out for a 180 Mak or an 8" edge SCT, or in fact any other scope less than 10Kg with a vixen rail.

I am seriously impressed by both the mount itself, and the skill to build it in the first place.  Awsome Roy. I can see some of your dilemma

Its a difficult assesment . I can only compare to what i am using a Cheap £600 laptop But with fairly fast AMD Ryzen 7 processer

The AMD chips are cheaper than Intel but seriously fast 

Ryzen 7 4700U review: AMD's budget 8-core crushes Intel's 10th-gen chips, again | PCWorld

Your i7 should compare well though. The size of your camera chip will of course impact its performance. Hard to say how to go foward Roy. But my QHY 462C camera can capture at 135 fps at 2.1 Mega pixels Not sure how that compares to what your using ?  I certainly could capture over 8000 frames in 60 seconds at 1920 x1080 thats about one quarter size your chip isnt it.

I find over capturing frames helps. As more can be stacked without reducing quality. I like to get 10.000 frames at full resolution is more than enough to  control quality and noise but far less is also doable.  What wavelets to sharpen do you use ? the top wavelet is the sharpest, But also the worst offender noise wise. I know of a good work around this using software. Send me a PNG of a image you like with a bit of noise over it i can show you something you may or may not like. But i feel its the best technique for noise after the fact. But really i am thinking you need more frames or better wavelet control. Or both. Lets try the cheap and easy option first. And see if you agree. Just pm a image if your interested

[Roy Foreman]

Neil - bit of a delay in responding, I've had some (non astronomical) problems to deal with !

Thank you for the compliments on my self built mount. It helped that I have some metal working experience, and that I am a design engineer by profession, but it was still a huge leap of faith and expense. I was fairly confident that it would work, but I wasn't sure how well.

As far as the lunar imaging goes, I have just remembered that these images were taken in twilight using a red filter which saps 3 stops of light. I had to up the gain a bit to compensate - hence the noise in the selective enlargement.  I take on board all your comments and you are absolutely right in what you say. I am grateful for your offer to work on one of my images, and hope to take you up on this once I have selected a suitable image.

I plan to do and experiment next time the moon comes around.  To image the same area of moon reducing the imaging area each time with corresponding increase in frame rate and frame numbers. This way I will be comparing like with like and can determine the optimum settings. I also intend to experiment with different gain settings to see the effect that has. maybe one of these images will be suitable for you to work on in noise reduction !

You are right - the cheap and easy option is the first one to try !  Incidentally, I never got on with Registax and wavelets.  I use Autostakkert as I find it easier, faster, less artifacts, and better results. It has it's own sharpening routine which I further enhance in Photoshop.

Roy

[neil phillips]

12 hours ago, Roy Foreman said:

Neil - bit of a delay in responding, I've had some (non astronomical) problems to deal with !

Thank you for the compliments on my self built mount. It helped that I have some metal working experience, and that I am a design engineer by profession, but it was still a huge leap of faith and expense. I was fairly confident that it would work, but I wasn't sure how well.

As far as the lunar imaging goes, I have just remembered that these images were taken in twilight using a red filter which saps 3 stops of light. I had to up the gain a bit to compensate - hence the noise in the selective enlargement.  I take on board all your comments and you are absolutely right in what you say. I am grateful for your offer to work on one of my images, and hope to take you up on this once I have selected a suitable image.

I plan to do and experiment next time the moon comes around.  To image the same area of moon reducing the imaging area each time with corresponding increase in frame rate and frame numbers. This way I will be comparing like with like and can determine the optimum settings. I also intend to experiment with different gain settings to see the effect that has. maybe one of these images will be suitable for you to work on in noise reduction !

You are right - the cheap and easy option is the first one to try !  Incidentally, I never got on with Registax and wavelets.  I use Autostakkert as I find it easier, faster, less artifacts, and better results. It has it's own sharpening routine which I further enhance in Photoshop.

Roy

No worries about time taken to reply Roy. I just hope i am helping you focus on on improvements. Your images deserve careful thought As they are top notch. One thing i forgot to talk about was processing time. With my 462c 2.1 mp camera 10.000 frames with my 8 core AMD Renoir chipped laptop. Its not too bad. Certainly not hours, But i am soon going to be using the much larger chipped 178 mono with my 10" F6.3 Newtonian. I suspect when i start running those sizes, large frame counts will no doubt start to become fairly slow.

Even with this 8 core laptop. The problem can become confounded by small 24 box size on Autostakerrt. I may, find like you. I have to find ways to increase speed. So the only option is larger box sizes (fewer alignment points) and fewer Captured frames. High magnification work like Tycho for example is ok to cut back on resolution. (Because its small)  But i will also be doing Full lunar mosaics. I am not buying a large chip just to cut resolution. So like you, i am going to have to find a happy medium. This really does only leave 2 options, less frames. (hence good noise control via software) Larger and less alignment boxes on AS/2.

Its a juggling compromise isnt it. I do believe the software i am using is the best de noise software around. It doesnt smooth the noise away it actually attempts to subtract it. leaving the image in most cases ( depending how severe the noise is ) totally intact. underneath.

But you have to do it correctly. If you look on my p base site it has been used on that.  Any way everything you have said makes sense and i can see your thinking in the right directions to clean up the noise floor on your excellent images.

As i said earlier just hope i am helping. Btw what sharpening exactly are you using on AS/2 I have seen no wavelet control on there. The one sharpening routine i have seen, although good. Doesnt  have any fine control. Which is essential in my opinion. Though i realize you have control on photo shop. I am not going to try and convince you of the power of registax wavelets control. If you have already made up your mind. But i firmly believe you may be writing it off a little hastily. After 15 years of using it myself personally.

Enjoying the conversation Roy. As some of your issues i will face myself very soon when i get the larger chipped 178m

[Roy Foreman]

Neil - I am sure you will get good results from the 178m when you get it, as your imaging skills are certainly up to the task of getting the best out of it.  There will be a learning curve, of course, and yes you will see what I am up against in using the larger chip. Like you I got a large chip camera to do whole disc mosaics, and I also like to pixel peep.  Sometimes, using small pixels, I find I can get better results from doing a selective enlargement than from using optical amplification in the imaging train.  There seems to be so many variables that there is no ' one size fits all' solution.  but it is fun to experiment and see what works and what doesn't.

As for registax, maybe I did write it off a bit too soon.  No matter how I played around with wavelet settings I still could not match what I could achieve in Photoshop. I also found that it produced too many artefacts and triangular blocks.  I am using AS3 which is faster, easier to use, and has an inbuilt sharpening routine - it blends the sharpened image with the unsharpened, and you can adjust the blend ratio, but I found the default of 50% to work best.  I further sharpen in Photoshop, and here I found the best method was to do it in 2 or 3 small steps rather than one large one.  I tend to use a large sharpening amount with a small radius, and use the threshold setting to limit it's effect on tiny features - like noise for example ! 

Thanks for all your input, which I find valuable, and good luck with that 178m - look forward to seeing the results !

Roy