PIER-14 Shall we change back to native Focal length?

To be fair, I am definitly the bossy one lol

Steve I didn't know you were away. I wish that when I was out to work that I'd had such a generous boss as Phil. I'll reply after you return, there is no urgency on this. Switch your phone off and go enjoy yourself. 

Cheers, 

Ray 

Quote "If you are going to make new pier with large FOV , it is better to go back to native on pier 14 ,  during the galaxy seasons it will be more usefull

i still have difficulty to understand how work your scheduler as some jobs like 1160 hasnt started since two month on pier1"

Steve, 

This thread is about pier 14 and you have given us valuable guidance on submitting jobs in general. Following up on this now as there does appear to be weaknesses in the way the scheduler is dealing with some jobs, particularly with the pier 1 queue. Florent in an earlier post was finding difficulty in trying to understand how the scheduler algorithm prioritises jobs, ditto. Using a scheduler is the only way to run a shared setup, but anomalies exist which may be why questions are being asked such as "where is this data?". Florent did draw attention to job 1160, see above, and to which I had drawn attention to when another submission for the same target was made. When you replied, obviously you were fully concentrated on pier 14 and understandably believed this was being caused by the mosaics. Job 1160 is on pier 1.

So, Ngc660 galaxy submitted August 12th still not started, similarly, the other submission for this target, #1233. In addition to this the Sculptor galaxy, Ngc253, was submitted August 13th, and I think most of us know this does not rise very high. Both of these are also broadband images and ought to have a higher priority over narrowband particularly around new moon. Another important thing that appears to be absent in the scheduler is the job submission date. 

So to take last night for instance, I think time was concentrated on 2 jobs. Ngc772, a galaxy, and sh2-132. The galaxy transits at 01:32 at 71°, and the diffuse nebula at 21:49 at 72°.
The 2 aforementioned jobs not yet started :
Galaxy Ngc660 transits at 01:16 at 65°, but of more concern with the moon now waxing is Ngc253. This transits at 00:17 at just 26°.

Two situations, one is that no priority is being given to a job submitted 2 months ago compared to one more recently, the other obviously is that an object with a short imaging window has also not been started despite also being summitted 2 months ago, and effectively being trumped by targets that ought not to be. 

I realise that you have limited control over the scheduler, but what is being witnessed does somewhat fly in the face of your guidance to us on submitting jobs. 

Just playing devils advocate Steve on this, and none of the jobs mentioned are mine. 

Cheers, 

Ray 

hi Ray

As you know I am away at the moment with the Mrs so I will respond more fully when I get back to work, in the meantime I have a suggestion I was already looking at behind the scenes anyway, it will unfortunately have the side effect of removing some of the reliability in response to a more rigid system but you cannot have your cake and eat it so to speak

Scheduling set to:

1, our standard moon parameters 
1, standard pier horizons 25/30degrees etc 
3, rising plan settings rather than meridian or west 

Prioritisation (higher the number the higher the priority) set to:

priority 0 = Narrowband
priority 1 = RGB
priority 2 = Luminance

its a simple system that will work but it will remove any flexibility I had to cheat the system for special or low objects ortherwise it is pointless for me to implement the suggestions above if I break the rules every time lol

if user put objects in that are 15, 20 degrees etc the system will just ignore them and likewise putting objects in that are well over into the west.

it means we have a system in place, that everybody knows and understands that we all stick to 

please come back to me with your thoughts

Steve 

Quote "If you are going to make new pier with large FOV , it is better to go back to native on pier 14 ,  during the galaxy seasons it will be more usefull

i still have difficulty to understand how work your scheduler as some jobs like 1160 hasnt started since two month on pier1"

Steve, 

This thread is about pier 14 and you have given us valuable guidance on submitting jobs in general. Following up on this now as there does appear to be weaknesses in the way the scheduler is dealing with some jobs, particularly with the pier 1 queue. Florent in an earlier post was finding difficulty in trying to understand how the scheduler algorithm prioritises jobs, ditto. Using a scheduler is the only way to run a shared setup, but anomalies exist which may be why questions are being asked such as "where is this data?". Florent did draw attention to job 1160, see above, and to which I had drawn attention to when another submission for the same target was made. When you replied, obviously you were fully concentrated on pier 14 and understandably believed this was being caused by the mosaics. Job 1160 is on pier 1.

So, Ngc660 galaxy submitted August 12th still not started, similarly, the other submission for this target, #1233. In addition to this the Sculptor galaxy, Ngc253, was submitted August 13th, and I think most of us know this does not rise very high. Both of these are also broadband images and ought to have a higher priority over narrowband particularly around new moon. Another important thing that appears to be absent in the scheduler is the job submission date. 

So to take last night for instance, I think time was concentrated on 2 jobs. Ngc772, a galaxy, and sh2-132. The galaxy transits at 01:32 at 71°, and the diffuse nebula at 21:49 at 72°.
The 2 aforementioned jobs not yet started :
Galaxy Ngc660 transits at 01:16 at 65°, but of more concern with the moon now waxing is Ngc253. This transits at 00:17 at just 26°.

Two situations, one is that no priority is being given to a job submitted 2 months ago compared to one more recently, the other obviously is that an object with a short imaging window has also not been started despite also being summitted 2 months ago, and effectively being trumped by targets that ought not to be. 

I realise that you have limited control over the scheduler, but what is being witnessed does somewhat fly in the face of your guidance to us on submitting jobs. 

Just playing devils advocate Steve on this, and none of the jobs mentioned are mine. 

Cheers, 

Ray 

The BSI sensor by SONY in pier 14 (IMX455) is far more sensitive than the Panasonic sensor in Pier1 (90% QE compared to 60%), and it has much better dynamic range (around 2 stops better) than 1600MM. We cannot assume the same integration time can be applied to these two piers (as many of us subscribed to both). IMO, we collect excessive data for Pier 14 and not enough data for Pier 1.

I agree - lets get it sorted!

Regards

Mark

here[/url] from Astrodon. 

Not sure how you would implement this in Pixinsight.

However, Surely the PhotometricColorCalibration tool derives the relative weights of the R and B channels relative to G, directly from a linear image (see here). Could we use that on a near zenith sub?

Straying outside my competence here, so just guessing.

Awesome, anything that gets you 90%+ is good enough in my opinion for now, so if you have a simple method you can apply and understand, then
go for it

Re G2V

Mine was just a suggestion to bring the issue to the front of the queue so to speak, the actual tool used to
create better weightings is neither here or there. I was just hinting
in my usual not very subtle way LOL

Let me know if my guide to Scheduling helps by the way, I tried to keep it super simple and none
technical, With this I hope once and for all everybody gets it finally

Steve

"Broadband imaging

Can one of the members put forward a G2V star selection job so you can calculate filter weights for the pier, this will make your LRGB data acquisition more efficient and you will get many many more images"

Steve - happy to. Not something I have done before, so let me know what is wanted. I see a tutorial video here from Astrodon. 

Not sure how you would implement this in Pixinsight.

However, Surely the PhotometricColorCalibration tool derives the relative weights of the R and B channels relative to G, directly from a linear image (see here). Could we use that on a near zenith sub?

Straying outside my competence here, so just guessing.

"If we remove the reducer , I l afraid that we will tend to do a lot of mosaic and had even fewer complete dataset provided by this pier.very thought decision....

The problem with very large field of  is that there isn't so much target to shoot.   You capture all in one time "

The problem we have been having with this scope with the reducer is that we are pushing the size of the field further than it can comfortably go. As a result, we get coma at the edges and the star shapes go off. Not a huge problem for a single image, as with the huge number of pixels, unless you are pixel peeping you don't notice. But for mosaics, we get some pretty nasty overlap areas to process, unless we increase the overlap and trim the panels - effectively moving towards native FL.

For that reason I would vote to remove the reducer. Not such a big field of view, but better star shapes at the edges, and easier mosaics if you want to do them. 

Do's[/size]

1. [size=150]Always put in objects that are rising, about to rise or rising in the future for up to 3/4/5/6 months time[/size][/*]
[size=150]Use the whole sky as it rises  N, NNE, NE,SE,S SSE
[/size][/*]
3. [size=150]Take account of the moon rise times alter over the course of a month in respect to the moon phase and spread jobs about so it continues to gather dat through the moon phases and rising times[/size]
   
[/*]

[size=200]Dont's[/size]

1. [size=150]unless you have no choice then don't choose objects already on the west (as it gets dark) unless they are at the meridian, as objects you have previously choosen that are rising will eventualy reach the west all on their own [/size][/*]
[size=150]dont pick so many objects in one quadrent of the sky[/size][/*]
[size=150]dont just use the SSE SE and South parts that are rising as it limits what will get imaged over the course of the month
[/size][/*]
[size=150]Last but not least - stop planning data collection like you would at home ie whats over your head now![/size][/*]

[size=150]Above all else plan much further ahead as our softwhere will accomodate you
better and be more productive. it is quite happy to accept jobs that are
5-6 months time and as they rise it will gather the data


May i also make other very important suggestions that will help with productivety
[/size]
[size=150][size=200]Broadband imaging[/size]

Can one of the members put forward a G2V star selection job so you can calculate filter weights for the pier, this will make your LRGB data aquisition more efficient and you will get many many more images
[size=200]
Narrowband imaging[/size]

In my opinion you all gather far far too much hydrogen data and not enough of the other filters, this chip is very sensitive across the board so again its a worthwhile exercise to get it working effienctly and gathering better data

I really hope this helps clear things up and make it a more enjoyable experience for all of you

Steve[/size]

hello All

if you are going to make new pier with large FOV , it is better to go back to native on pier 14 ,  during the galaxy seasons it will be more usefull

i still have difficulty to understand how work your scheduler as some jobs like 1160 hasnt started since two month on pier1 

For CMOS and acquisition time , i think that it will be a good thing to schedule a test  so everybody can judge how much time it need on a objet 
may be to have a target imaged    for 1h (with 30 x120S)  , 3h (with 30x240s)  for each filter (SHO) and so we could compare

i will check what target we have already done ,and schedule a quicker version of the same target

Hi Steve,

Beginning to feel your pain.

Guilty to a lot of what was mentioned.   Now I'm assuming that we are going native next month so a lot of what you mentioned hopefully will take place as a natural consequence of the change. 

In my/our defence the enormous fov led to the temptation to go full out on capturing extremely large areas of the sky with more detail than the more commonly used approach of using extremely short focal lengths. You have at the same time now  offered a practical alternative to these mosaics by way of a change to pier 11. This adds the narrowband capability which is absent on piers 4 and 5 and therefore a good alternative to pier 14 but at a lower resolution. 

A reluctance to submit more jobs right now may be down to the imminent change, which would impact on any incomplete capture. This is the not the first time we've been in this situation and I believe partly justifies what is happening. We can submit future jobs, particularly those more suitable for the native setup, however if these start to be captured this could be detrimental to pending jobs resulting in those not being finished in November. 

There appears to be no straightforward single solution other than to monitor almost daily the current queue and take a degree of manual control. Would it not be better towards the time of the visit to submit a mix of short jobs and top ups to any completed job where a weakness is found in part of the data. 

Apologies for the verbal diarrhoea, perhaps this will encourage some more reasoned feedback. 

Cheers, 

Ray 

If we remove the reducer , I l afraid that we will tend to do a lot of mosaic and had even fewer complete dataset provided by this pier.very thought decision....

The problem with very large field of  is that there isn't so much target to shoot.   You capture all in one time

Good morning

As many of you may have already realized, I am not one for undue procrastination, so I thought I may make a few points to bring the shall we shan't debate a bit closer to an end.

Pier-14
Some may not feel that this pier supplies a lot a data sets but when most of them are huge multi pane mosaics you may feel like a 4 pane mosaic is "1 image" where in fact it is 40-60 hours of data in some cases that's been collected, this on most singles objects with a CMOS sensor would be 5-7 singles subjects. You as the members have decided to do mosaics
As I have also stated on Pier-1 previously, at times member are forgetting that this is a fast CMOS based system, and it does not require the many many hours of subs we are all used to for CCD cameras. Except for really special objects like the Dolphin, Squid, Spaghetti and other hard dim objects. The sooner "you" as the users of this pier start using it fully, it will be far more productive.
Whilst I understand we may be going native in 5/6 weeks time, there are still not that many jobs in from what I can tell. January-March jobs also need to be put in now as the system will start them far earlier that you realize, I promise you. I don't want to sound like a broken record either


So now to the blunt part

Pier 14 at native although slower will in my opinion be more productive as it will

1, frame most objects better with the smaller FOV as well as be better across the whole 12-month period, eg galaxy season
2, mosaics will in fact have no loss at all due to the increase in star shape and less other aberrations which means you will not have to do such large overlaps

From a business perspective it is my job to promote this pier and in the end one day make a little money, the best way for me to do this is by showing the pier at its best with it producing a lot of data for the members. I believe my needs have a certain amount of symmetry with yours I stated above LOL

I am honestly here to help and not rain on your parade but I would like to see the pier used to it's potential. If you need any info or help or anything just ask

Ps I mentioned the Dolphin so you may ask for it to be done with this Pier, you will need lower horizon limits to do it but HA OIII it would be epic!

Steve

If we remove the reducer , I l afraid that we will tend to do a lot of mosaic and had even fewer complete dataset provided by this pier.very thought decision....

The problem with very large field of  is that there isn't so much target to shoot.   You capture all in one time

I agree we should remove the reducer. 
I would also vote for a super widefield setup like Sigma 105 art /Canon 200mm F2 + 6200mm as an additional synicate. Pier 4/5 is great but sometimes we might need to add some Ha and/or Oiii to the image.

Beautiful image Vikas and I agree to remove the reducer.
Takahashi is well known for the outstanding performance but the reducer is degrading the image quality.
If you want a bigger field of view than you have to get a bigger sensor or another high end astrograph with a shorter native focal length.
A reducer is only second best.
Imaging in prime focus is always preferable above using a reducer.

Mike van den Berg

Beautiful image Vikas and I agree to remove the reducer.
Takahashi is well known for the outstanding performance but the reducer is degrading the image quality.
If you want a bigger field of view than you have to get a bigger sensor or another high end astrograph with a shorter native focal length.
A reducer is only second best.
Imaging in prime focus is always preferable above using a reducer.

Mike van den Berg

hi mark. I use photoshop actually to add the signature though i think there is a way to do it in Pixinsight too. a quick google search revealed this though i have not tried it...PI sign


u can get a custom made signature from here....sign

i find it easier to manipulate in PS

vikas

Vikas

I am sure I should know this, but how do you add the signature?

Regards

Mark

Hi Vikas, 

That's very interesting to know. I know that Steve says he's bored seeing the same objects repeatedly being shown on astrophotography web sites and magazines, but not about his attitude towards Top Picks, APOD and the like.

I won't say what I was going to on that and regarding his reply, but instead relate a story of someone where I worked. 

Around mid morning would be the time for some of us to browse the daily newspapers while partaking of a beverage. Occasionally when a picture  taken at an event appeared this person might then comment on any member of the public who was close to the action and got  included in the picture. They'd be accused of "trying to get their picture in the papers".  They had a word for them but I cannot remember what it was right now. Anyway, one day that person came into the office and immediately proceeded to open up a newspaper they had with them, open it up to the page where they'd inadvertently been snapped while at a concert. Everyone there had to see it of course, much to the amusement of everyone.  

I'd be very proud (chuffed) if I ever produced anything that I felt could be submitted somewhere, let alone then be picked out. I too like to look at and enjoy them, it certainly helps and inspires. 

Cheers,

Ray 

Ray, My issue with APOD is it seems to have diluted in its nature in so much of how broad a spectrum they use for "astronomy" picture of the day. It also seems become very political with the winners hat being used in a very Masonic fashion across an ever decreasing gene pool.

I love seeing well processed images ,different takes on things not to mention some of the stunning wide-field and Aurora shot, a lot of other sites still seem to be carrying the mantle forward I am happy to say. A lot of skill goes into this hobby and its great to see recognition for that skill.

I hope that expands a little on my thinking

Steve