M45

[Stardust]

I'm only really getting going with the dark art of astrophotography, I only have about 4 pictures that I'm happy with so far.

Here is my M45 and would be keen to get feedback on how to improve or things I should have done.

It was 11x10 mins with darks bias & flats, plus my trial and error in photoshop

Unmodded Canon 400D on a TV 85 with reducer all mounted on a 10" LX200 with phd guiding

[ollypenrice]

Hi,

This is a good capture. You are in focus and have round stars. You have also achieved a flat field. I think the data could give more at the post processing stage. Two things jump out at me first of all.

1) The colour balance is off, with too much green and not enough blue. As you stretch the linear data I'd suggest you cut back the black point (Levels, move the left hand slider to the right) so as to get the top left of the histogram peak aligned in each channel comme ca;

Be careful never to clip too much off. A final cutting back is the very last thing I do in processing. See below;

If the blue channel still looks a bit thin you can give it a slight lift in Levels (mid point slider to the left) or Curves. 

2) I can see the noise reduction instantly. It gives a 'vaseline on the lens look,' to quote Steve Loughran. I do sometimes use noise reduction, but very selectively and only on the darker parts of the image. The bright parts don't need it and do carry the detail which NR destroys. Two good reasons for not applying it to the bright stuff. There are various ways to select the regions for NR. You can use the Colour Select tool experimentally (the fuzziness slider decides how wide a range you select with one click) and apply NR to the selection. You can do it this way but to a layer underneath and then use the eraser on the original top layer to let the NR through. You can make a mask between layers based on brighness. I just use colour select. I use masks far less than most people but I'm not making a virtue of it, it's just what I do.

Note that this reply is in response to a PM from the OP. I'm not trying to rip his image to pieces!!

Olly

[Xplode]

Without noise reduction this image would look much better.

Do you use dithering when guiding? I found that dithering seems to remove a lot of the pixed pattern noise when images are stacked.

[Stardust]

Thanks guys this is just what I wanted, any advice from people that know is much appreciated.

I will have another play but to be honest my photoshop skills aren't great, best make an effort to understand what going on.

I have seen the link to Anna Morris so will take notes from that.

I'll have to check the settings, dithering does ring a bell

As Olly said, I asked for some constructive criticism, don't think he's been mean to me. 

thanks again

David

[lensman57]

I'm only really getting going with the dark art of astrophotography, I only have about 4 pictures that I'm happy with so far.

Here is my M45 and would be keen to get feedback on how to improve or things I should have done.

It was 11x10 mins with darks bias & flats, plus my trial and error in photoshop

Unmodded Canon 400D on a TV 85 with reducer all mounted on a 10" LX200 with phd guiding

That is a great first try, well done. If you wish drop the raw fits or tiff file in to drop box without any adjustments whatsoever and we have a go at the data to see what you have there.

Very nice capture indeed.

Regards,

A.G

[Stardust]

That sounds like a plan, I have just created a drop box so IF I've got it right the tiff file should be here

https://www.dropbox.com/s/24iveeippglmqos/Copy%20of%20m45stack.TIF

Yes I'd be very interested to see what experienced folk can do with my data and to know if what I'm capturing is good or otherwise.

That will be one step forward, play away.

[lensman57]

That sounds like a plan, I have just created a drop box so IF I've got it right the tiff file should be here

https://www.dropbox.com/s/24iveeippglmqos/Copy%20of%20m45stack.TIF

Yes I'd be very interested to see what experienced folk can do with my data and to know if what I'm capturing is good or otherwise.

That will be one step forward, play away.

Hi,

I have done  a quick process in StarTools, just to see the level of data. The data is good but very blue, which is expected I guess. You could probably improve on the image during developemant by using a robust star mask, which I didn't do and try and mask off the nebulasity and give it a series of small stretches in PS as the background is noisy due to the use of a DSLR and needs to be well isolated. Great capture anyhow.

Regards,

A.G

[Stardust]

That's great, much cleaner,  I can now see where I've gone wrong, in some part at least. In an attempt to smooth out the background noise I have indeed made it all tooo smooth. I have tried again and whilst improved that aspect the image has gone very blue.

I use a CLS filter, would that account for the blue cast and what setting could I use to counter that. It has been suggested to use tungsten as the colour balance setting, would this help.  

[lensman57]

That's great, much cleaner,  I can now see where I've gone wrong, in some part at least. In an attempt to smooth out the background noise I have indeed made it all tooo smooth. I have tried again and whilst improved that aspect the image has gone very blue.

I use a CLS filter, would that account for the blue cast and what setting could I use to counter that. It has been suggested to use tungsten as the colour balance setting, would this help.  

Yes the CLS filter is notorious for painting everything blue, I try not to use mine anymore, I find the IDAS a much more palatable proposition. Unless you are shooting Jpegs the colour balance has no bearing on the RAW files so long as DSS does not apply adjustments to the final fits or tiff files. Shooting jpeg for AP is a big no-no. The less molestation is done in or out of camera the better before the files reach the stretch software.

Regards,

A.G

[lensman57]

For the sake of being even handed here is a version processed in Pixinsight. I am a real novice with PI so perhaps others could do a better job particularly with noise reduction.

Regards,

A.G

[Stardust]

Well here is my revised effort, apart from the rather blue tinge I have improved on other areas I think

[Xplode]

Better, but still a lot of details removed by noise reduction.

Which iso was used? The 400D was released 7 years ago so it has a lot of noise

Any idea what's causing those weird spikes on a few of the stars?

[Stardust]

Time flies, hadn't realised the age of the camera, what would be a significant improvement that wouldn't be a huge cost? I suppose the 1100D should be better as it's quite new or would I be better off putting the money towards a CCD. Now that's a bit outside my comfort zone at the moment.

ISO used was 800 and no idea why there are spikes on just some stars, any thoughts on that?

[Xplode]

1100D is a good option, maybe you could buy an astro modified one from Cheap Astrophotogtraphy?

For something better i think there's 2 options, either the Canon 6D for color or go all the way with mono CCD and filters.

The spikes are weird since they are in the middle of the image, but not on stars on the edges. I have no idea what causes them

[lensman57]

Time flies, hadn't realised the age of the camera, what would be a significant improvement that wouldn't be a huge cost? I suppose the 1100D should be better as it's quite new or would I be better off putting the money towards a CCD. Now that's a bit outside my comfort zone at the moment.

ISO used was 800 and no idea why there are spikes on just some stars, any thoughts on that?

Please do not buy in to this new sensor is better than the old sensor fallacy. One of the most popular OSC CCDs at an affordable price is the QHY8L which I believe uses the same sensor  that Nikon used in their revival camera against Canon, the super popular D70. Can anyone remember that long ago, less than 10 years? With proper cooling and considering that it is an OSC, many imagers are producing beautiful work with this sensor, I think that it maybe the same sensor in the Starlight Xpress M25C and no one ever complained  about this camera using old sensor ( please correct me if  I am wrong ). If I am right then the question is not about  how old the sensor is but how good are the designed  electronics and cooling  for AP. DSLRs are always at a disadvantage not becuase of their sensor technology, they do use the most up to date  ones, but how good the electronics are for AP imaging. DSLRs are designed to operate at very short exposures, up to a few seconds at the longest, CCDs are designed to work at exposures lasting into hours and be stable. You can buy a Canon 5D , original, for less than £500.00 now and it does have a full frame sensor, low noise and a decent pixel dimension. But If I had the money and I could choose I would still go for an Atik 11000, even if it costs a few quid less than £4000.00, atleast it is made for the purpose. The other point to consider and it is a very important one, is that the larger the sensor the more demand is put upon the optics of the telescpoe. Even the new SW Esprit range can only cope with a sensor dimension of less than 40mm diagonal. A full frame sensor is over 43 mm in diagonal.

A.G

[Stardust]

Now that I know I can track well enough I would consider an upgrade, but maybe stick with what I've got to learn more in the way of processing, or am I just making it hard from the beginning? Would I get better results from something else.

I had discounted CCD on a  cost basis but I will do some research, the QHY8L isn't as expensive as I'd imagined. I'll have to see what would match up with the TV85 as a primary scope but could use the LX200 later as well.

I'd probably go for a colour camera given time restraints and the lack of clear nights to be able to use anything at all at the moment. 

[Xplode]

Usually a new sensor is better than an old sensor when comparing same technology and there is no denying that the 1100D is much better than the 400D , it has lower noise and is more sensitive. Less noise means less imaging time is needed to reduce it to an acceptable level after stacking.

Of course an one shot CCD can produce great results, but in my opinion their cost is just too high for what you get when one can get pretty much the same results with a cheap astromodified 1100D as a cooled OSC CCD that costs 1000£

Fun fact: The 6D has double the sensitivity of the 5D, 25% QE for the 5D vs 50% QE for the 6D. Senstivity has doubled after 6-7 years for full frame, for crop cameras it has gone from around 30% QE to 45% for the 70D in the same amount of time. Reason for the smaller jump in crop sensor sentivity is the size of the pixels have become much smaller for crop sensors.

[Stardust]

OK that's making things clearer. I spent some time searching images till late last night taken with the modded 1100D and a QHY8. I must admit that I was struggling to see what the extra £600 was giving.

There were subtle differences but I was left wondering if it was really worth the extra cost for where I am at the moment.

If the 1100D will give similar results that is a very attractive buy at under £300 for an astro modded camera.

If the QHY8L is not such a great improvement at £900 and I need to spend more, say £1200-£1500, TBH that's something I can't afford, not sure £900 is at the moment either.

[lensman57]

OK that's making things clearer. I spent some time searching images till late last night taken with the modded 1100D and a QHY8. I must admit that I was struggling to see what the extra £600 was giving.

There were subtle differences but I was left wondering if it was really worth the extra cost for where I am at the moment.

If the 1100D will give similar results that is a very attractive buy at under £300 for an astro modded camera.

If the QHY8L is not such a great improvement at £900 and I need to spend more, say £1200-£1500, TBH that's something I can't afford, not sure £900 is at the moment either.

You are right about holding off spending money. The difference between a modded DSLR and a CCD is not in what is captured on the chip but the amount of noise that is generated  during the capture by the chip itself and the electronics ( mainly due to heat build up ) and yes a modern ICX 674 sensor has a QE of near 80% ( above 65% in Ha) and this helps a lot . The set point cooling feature alone will make a huge difference, add to that the extra sensitivity and ability to use dark frames from a library ( set point cooling allows the build up of darks library) makes for a very efficient work flow, I do not use dark frames with either of my 314L or 428Ex cameras and this saves a lot of time. I have 2 modded DSLRs, 1000 d and 1100 d, and 3 CCD cameras, all Atiks and I do use the DSLR for its wider FOV but if I were financially in a position to afford a real fast short FL astrograph such as BORG or a TAK I would sell all the DSLRs. They are really slow to get going compared to a CCD. There is no denying though that a  modded DSLR represents unrivalled value for money to start AP and if finance is an issue as it is for a lot of us then DSLR is the way to go. Good luck with your choice and have a good Christmas.

Clear skies to all,

A.G

[Stardust]

I've made a decision and ordered an astro modded 11ooD and will be very interested in the images from that. In the mean time I will be reading up on PS .

Thanks for all your input