PIER-14 Starting from scratch

Dear fellow pier 14 subscribers,

It's likely that in 2 to 3 weeks time our pier will be up and running using the new camera and filters. In the meantime as much of the outstanding jobs will continue to be collected, subject to the weather and the moon's position. We've recently received a lot of jobs back from our pier and these ought to be checked asap and any replacement data requested and added to the job queue.

What will happen to the existing list of our submissions I don't think has yet been mentioned. I'd like to see it cleared down just as soon as the existing setup is disabled. Assuming that this does happen then we will then be able to start to submit the new jobs based upon the revised fov.

I believe that now is a good time for us to start suggesting targets. We are effectively starting from scratch with a smaller fov but being able to use almost all of it without mis-shapen stars. For several individual targets a mosaic will now be necessary, so perhaps we can discuss what percentage overlap of the panels we should use.

The new joblist needs to keep the pier busy imaging from the start. Initially there may well be a short period during the night around the full moon when there will not be any narrowband jobs. I think we've all been made aware that future jobs will need to be continually added to the list, but maybe we should first concentrate on what is current and you would like to have. 

One point that has been discussed on occasions but never given any backing is the number of hours submitted for a single job. With the change to how soon we have our data sent to us, that is the team no longer does a quick scan of the data beforehand, there will be many more instances of us having to submit a new job to replace bad subs.

For example, if you put in a job for 15 hours where there is a relatively short imaging window, then unless you get it back sooner, they'll not only be a shorter imaging window time remaining to collect the replacements, but there could be many more subs required. 

With that recent change I personally think we should stick with a maximum recommended time for a single submission. It might be 8 to 10 hours as a range or 10 to 12 hours, plus a small tolerance.

This does not mean you can't go beyond that for your target, only that you'd need to split it into 2 or more jobs. If it involves broadband and narrowband filters then this might be a good way to split the job up. Personally I think that going beyond 20 to 24 hours needs to be discussed on the forum to let people know beforehand. Chances are you'll always get backing from other members of the Syndicate. It's not that you cannot submit whatever you like, it's simply being considerate to others. 

I've rambled on for far too long as usual. Any thoughts, and all constructive criticism welcomed. 

Cheers, 

Ray 

 

 

Hi Ray this is a really good idea if I am honest :)

The part I have highlighted in red, I presume you are talking about forward planning ?

if so I agree that as well as putting new jobs in to get the pier up and running, now is the time to instigate forward planning with job selection as the software works best when objects are Rising or rising in the future.

Not to mention that:

• This gives the pier longer to collect the data
• A greater chance to image throughout the month as the Moon moves East through its different phases, as you would drip feeding objects that rise at different times, they will always vary in offset position to the Moon so moon avoidance will also work better
• The best part of our sky is North East, East, South east then South all the way up to the meridian, so catching objects early will have image quality benefits for all users.

Just to reiterate (hopefully without sounding like an old record) our software works on a rising algorithm so is most productive if you forward plan using rising objects across varying sky positions

HTH

Steve

Hi Ray,

thanks for staring the conversation with the key topics you bring forward. Here are a couple of thoughts for your consideration:

a. About mosaic overlaps: I actually like 15%. I think that it is better than 10%, at least in my experience. Indeed, I found 10% to be borderline in some instances. What is your experience?

b. About job lengths: My natural tendency would be to favour data quality over number of targets. Indeed, as a scientist, I always aim for the best possible input data, as this makes post-processing (I think in terms of data analysis) much more robust and fruitful. Indeed, data that was not collected, will not be revealed by post-processing, whatever the skill-level of the astrophotographer. I fully acknowledge that this may mean a few less targets, but would lead to better quality images ... and targets come back year after year in a predictable manner (fortunately).

c. About job splitting: I like the idea to split jobs in a way that enables us to review the data quickly and then agree to collect more data if needed in a more nimble manner. We may also decide among ourselves about the relative priority of such "supplemental request" over other jobs in the pier queue and provide the RoboScope team with this information.

d. About sub duration: What I have yet to figure out is how to determine "necessary and sufficient" sub durations (which of course enables overall job duration optimization) on the basis of e.g. 5-10 subs across a range of durations of the weakest signal in the target. We may wish to do so, i.e. reach a point where we collect such data as "PreJobforAnalysis" and agree among ourselves of the ideal job. With my own setup, in my backyard, it take such subs and then define what "seems OK" based ore on gut feeling than a reasoned data-driven approach. I would much appreciate your thoughts and experience on this topic.

As I am new to RoboScopes, I am of course note completely familiar with how the pier-community works, but may I suggest that an occasional via-voice conversation could be beneficial? Just a thought. 

With my best wishes

Manuel

Hi Manuel, 

Many thanks for the very comprehensive and fast reply. My immediate response follows so I may have missed something out. 

a. Mosaic overlap.

I agree that 15% would be a good recommendation and does allow for a small bit of rotation between frames. I often wondered about targets near the pole and if that has any affect on mosaics. I cannot offer a scientific explanation as to why I think it might be relevant, just a gut feeling.

b. Quality over quantity.

Totally agree with you on this Manuel. :) My main reason is to avoid as much as possible having to deal with a lot of noise. I'm pretty useless when it comes to dealing with it in post processing. There will always be some, but the less the better.

Therefore, maybe we should not be specific about a  number of hours without it being necessary to discuss it in the forum.

Submitting a 30 hour job on a faint emission nebula would not be considered unreasonable, while a long job on what appears to be a far less challenging object, ought first to be put to members with a specific reason for the integration time. This often is to allow the capturing of faint details that do not appear in the majority of images that we get to see. Once explained everyone is satisfied. 

c. Prioritising resubmission after discussion.

That's a great suggestion, however, I don't think Steve will be able to incorporate that into the scheduling given what it may involve them in. I'm sure he'll soon let us know his thoughts. :) Having said that, there is a case for such a thing and that relates to the next point. 

d. Submitting a "PreJobforAnalysis" .

That would be useful and something I have occasionally done, but not on this pier. The issue is getting such a job prioritised, the previous point. 

I'll try and discuss this with the top man, failing that I'll have a word with Steve. :) 

Interactive discussion would be so beneficial, however, this is always dependent upon people's free time and their time zones. Emails might be a part solution if a response is urgently needed. 

Cheers and thanks again for your very helpful reply, 

Best wishes, 

Ray 

A quick follow up to #4613.

         Test job prior to a main submission(s). 

Particularly for a job requiring a lot of integration time it might be beneficial to have a reasonable idea of optimal exposure times. In the absence of anything that could assist in determining those, it might be a good idea to pre-submit a short test job. These ideally will only ever be used in exceptional circumstances and if the results are needed asap.

Prior to any such test it's strongly advisable to mention it here on the forum.  Someone may well be able to assist and if not, at least members will have been informed ahead of time. This is important otherwise the job may appear to uninformed members to be jumping the queue. Such a job should be kept fairly short in terms of total integration time owing to the effect it will have on existing ones in the queue.

When submitting such a job it needs to be made clear in the notes by using a pre-agreed description recognisable to the roboscopes team. Something like "exposure test job". The team will then endeavour to have that prioritised. As previously mentioned these should be used sparingly as I'm certain they don't want to have to keep tweaking individual jobs too often. 

Finally, Steve has added in his earlier reply advice on entering jobs for targets yet to rise, something that needs to be regularly monitored by us so the pier is kept operating. 

Cheers, 

Ray 

Regarding #4615: I am in complete agreement!

Good morning/afternoon? Manuel,

That's really appreciated. :)

We just need to give it a couple of weeks to allow others the time to look all through this and post their suggestions and comments. After discussing these then this can be turned into a list of recommendations.

Best wishes,

Ray 

Regarding #4613:

Thanks for the reply Ray ... let's see what is possible regarding job prioritisation. ... For Steve to say.

Interactive discussions:

1. Indeed, availability and time zones are a barrier. One way to address this might be to have a fortnightly or monthly fixed point that people can join or not.

2. I am in Switzerland, i.e. GMT+1. email is manuelcp.astro@gmail.com and I can be reached by WhatsApp/Telegram on +4179 200 9148. Best is between 6pm and 8pm my time on weekdays ... if you want to chat.

Cheers

Manuel

Manuel, 

Thanks for the suggestion let's see what Steve has to say. 

Still living very much in the dark ages I'm afraid, no Facebook, Twitter, WhatsApp etc. and being frank but not wishing to appear rude, there is little interaction on the forum anyway, so I'm not sure you'll get the majority to join in. As a not so good alternative we could use email. If some others agree to doing this as an alternative to just using the forum, then I'll post my address.

Really appreciate your keenness, 

Best wishes, 

Ray (UK) 

Regarding #4613:

Thanks for the reply Ray ... let's see what is possible regarding job prioritisation. ... For Steve to say.

Interactive discussions:

1. Indeed, availability and time zones are a barrier. One way to address this might be to have a fortnightly or monthly fixed point that people can join or not.

2. I am in Switzerland, i.e. GMT+1. email is manuelcp.astro@gmail.com and I can be reached by WhatsApp/Telegram on +4179 200 9148. Best is between 6pm and 8pm my time on weekdays ... if you want to chat.

Cheers

Manuel

 

Morning everyone (Just about)

As you all know prioritisation is on a scale of 1-4

0 = HA
1 = SII OIII
2 = RGB
3 = Luminance
4 = Comets (proposed test jobs)

My suggestion would be to add a small test data set to Priority-4, this would make it the highest priority in the pier queue and moon avoidance parameters allowing it would do this before all other jobs.

I do have a couple of caveats to make however

1. When a tester job is placed it must be stated in the notes section with the order that its a tester and state Priority-4
2. test jobs should be short  (2 hours or under)
3. All priority jobs are discussed with other members prior to input - No exceptions
4. Please please try to forward plan as it will really help with implementing this feature and save a backlog
5. No moaning at us if priority jobs delay other work being captured although if it is used sensibly it should have very little impact on daily running of the pier in my opinion

Apart from that it should work fine and have no issues implementing this feature if you all want it, it may also work with re-inputs (bad subs) but I will have to have a think about that etc :)

Ps

My apologies for number 5 being so blunt but I have just spent over a month every clear evening behind the scenes tweaking our automated software to get it to do what I want (not easy) which was getting most jobs on p-14 finished in between a poor run of weather because a lot were in the same part of the sky and I don't really want to be doing that stress again any time soon.

Steve

Regarding #4613:

Thanks for the reply Ray ... let's see what is possible regarding job prioritisation. ... For Steve to say.

Interactive discussions:

1. Indeed, availability and time zones are a barrier. One way to address this might be to have a fortnightly or monthly fixed point that people can join or not.

2. I am in Switzerland, i.e. GMT+1. email is manuelcp.astro@gmail.com and I can be reached by WhatsApp/Telegram on +4179 200 9148. Best is between 6pm and 8pm my time on weekdays ... if you want to chat.

Cheers

Manuel

 

Morning everyone (Just about)

As you all know prioritisation is on a scale of 1-4

0 = HA
1 = SII OIII
2 = RGB
3 = Luminance
4 = Comets (proposed test jobs)

My suggestion would be to add a small test data set to Priority-4, this would make it the highest priority in the pier queue and moon avoidance parameters allowing it would do this before all other jobs.

I do have a couple of caveats to make however

1. When a tester job is placed it must be stated in the notes section with the order that its a tester and state Priority-4
2. test jobs should be short  (2 hours or under)
3. All priority jobs are discussed with other members prior to input - No exceptions
4. Please please try to forward plan as it will really help with implementing this feature and save a backlog
5. No moaning at us if priority jobs delay other work being captured although if it is used sensibly it should have very little impact on daily running of the pier in my opinion

Apart from that it should work fine and have no issues implementing this feature if you all want it, it may also work with re-inputs (bad subs) but I will have to have a think about that etc :)

Ps

My apologies for number 5 being so blunt but I have just spent over a month every clear evening behind the scenes tweaking our automated software to get it to do what I want (not easy) which was getting most jobs on p-14 finished in between a poor run of weather because a lot were in the same part of the sky and I don't really want to be doing that stress again any time soon.

Steve

 

 

 All good for me! 

Good morning/afternoon? Manuel,

That's really appreciated. :)

We just need to give it a couple of weeks to allow others the time to look all through this and post their suggestions and comments. After discussing these then this can be turned into a list of recommendations.

Best wishes,

Ray 

Of course, let's see what others are saying and suggesting ...

Best wishes

Manuel

Manuel, 

Thanks for the suggestion let's see what Steve has to say. 

Still living very much in the dark ages I'm afraid, no Facebook, Twitter, WhatsApp etc. and being frank but not wishing to appear rude, there is little interaction on the forum anyway, so I'm not sure you'll get the majority to join in. As a not so good alternative we could use email. If some others agree to doing this as an alternative to just using the forum, then I'll post my address.

Really appreciate your keenness, 

Best wishes, 

Ray (UK) 

Manuel, 

Thanks for the suggestion let's see what Steve has to say. 

Still living very much in the dark ages I'm afraid, no Facebook, Twitter, WhatsApp etc. and being frank but not wishing to appear rude, there is little interaction on the forum anyway, so I'm not sure you'll get the majority to join in. As a not so good alternative we could use email. If some others agree to doing this as an alternative to just using the forum, then I'll post my address.

Really appreciate your keenness, 

Best wishes, 

Ray (UK) 

Hmmm ... maybe the new modern rather than the dark ages! 

Indeed, I noticed that ... a pitty I think, really. So, yes, lets go through email for now. 

BTW, I am just one hour "ahead" in Switzerland ... so easy.

All the best

Manuel

Manuel, 

Thanks for the suggestion let's see what Steve has to say. 

Still living very much in the dark ages I'm afraid, no Facebook, Twitter, WhatsApp etc. and being frank but not wishing to appear rude, there is little interaction on the forum anyway, so I'm not sure you'll get the majority to join in. As a not so good alternative we could use email. If some others agree to doing this as an alternative to just using the forum, then I'll post my address.

Really appreciate your keenness, 

Best wishes, 

Ray (UK) 

Manuel, 

Thanks for the suggestion let's see what Steve has to say. 

Still living very much in the dark ages I'm afraid, no Facebook, Twitter, WhatsApp etc. and being frank but not wishing to appear rude, there is little interaction on the forum anyway, so I'm not sure you'll get the majority to join in. As a not so good alternative we could use email. If some others agree to doing this as an alternative to just using the forum, then I'll post my address.

Really appreciate your keenness, 

Best wishes, 

Ray (UK) 

Hmmm ... maybe the new modern rather than the dark ages! 

Indeed, I noticed that ... a pitty I think, really. So, yes, lets go through email for now. 

BTW, I am just one hour "ahead" in Switzerland ... so easy.

All the best

Manuel

Just a thought :)

I could ask Phil about turning on the PM DM (direct message) in the forum so members can email behind the scenes. It may make life easier for certain things ?

Steve

Just a thought :)

I could ask Phil about turning on the PM DM (direct message) in the forum so members can email behind the scenes. It may make life easier for certain things ?

Steve

Good morning Steve,

Yes, that would certainly help as well, many thanks!

One question: Is there a way to link e.g. an Excel file to a DM or to a forum post? The reason I am asking is that I am working on the plan and would wish others to review, verify and of course suggest changes. This would also allow the final plan to be available easily to anyone in the syndicate.

All the best

Manuel

Good morning Manuel, 

Looking forward to seeing what you have come up with and appreciate the effort you are putting in on this.

Just a suggestion. Maybe it would be a good idea when you share your plans to create a new post, titled something along the lines of 'Let's discuss mosaics'. At that time I may create one called 'Let's discuss targets' and re-post my original short list of single targets together with anything suggested in the meantime. I'm sure Mr Asbury :) has some good ideas.

I think re-posting my original post will help to tidy things up and make it quicker and easier to navigate. Together with your one on mosaics there ought to be plenty for us to discuss ahead of the completion of the pier. 

Please do not take this seriously, although it does have one merit. 

Perhaps a mega mosaic, once submitted we could take most of the year off and come back when it finishes. Suggestion appears below. 

Just a thought :)

I could ask Phil about turning on the PM DM (direct message) in the forum so members can email behind the scenes. It may make life easier for certain things ?

Steve

member to member messages have always been turned on.

In the top menu theres a bit for 'conversations'

Hi Phil, 

Thick as 2 short planks here. 

Does this apply only to me. Apparently I'm not allowed to post private messages. I expect everyone else is allowed to. :)

Cheers,

Ray 

Hi Phil, 

Thick as 2 short planks here. 

Does this apply only to me. Apparently I'm not allowed to post private messages. I expect everyone else is allowed to. :)

 

Cheers,

Ray 

What does it say Ray?

Screenshot below. 

Screenshot below. 

Fixed.