TriAtlas project (2nd. edition) - first proof
Posted 01 April 2008 - 06:47 AM
I am preparing a new edition of the TriAtlas, trying to solve all the detected problems reported up to now. I have prepared a set of three sample charts (Orion, Andromeda and Ursa Major) for discussing and improving the maps before generating the new A, B and C sets.
PDF file with three B-set charts (2nd edition)
There are some very nice improvements (I hope):
(1) Chart arrangement by decreasing R.A., similar to Uranometria 2000.0 2nd ed
(2) Charts are now in portrait format. This allows viewing a larger region of the sky, placing two consecutive charts oposite.
(3) Label overlap problem practically eliminated. The system is quite good although I think I can still do some more improvements
(4) Tips indicating the location of the problematic objects, which allow labelling even extremely cluttered regions (eg. Large Magellanic Cloud)
(5) Double stars labelled by name and indicated by horizontal tips, but best double stars are plotted like in the former release
(6) Larger fonts, much more readable and still good for the scale of the maps
(7) Legend at the top and neighbouring charts at the borders in black background
(8) New symbols, etc...
Well, the best is to have a look in the link above. I am very, very happy with the new charts and I hope they will help to enjoy the sky to everybody.
Posted 01 April 2008 - 11:37 AM
The area around the Andromeda Galaxy is much too crowded. Most of those globulars would demand a 16 inch telescope. No one with a large telescope would try and use such a crowded map to find any of those very faint and difficult objects.
Also, please label "Barnard's Loop" with that name. There are places where a common name is how that object is spoken about.
I will be interested to see how many galaxies you allow in these Abell Galaxy Clusters. Many of them are very faint and difficult in medium sized scopes.
Thank you for your time and effort;
Posted 01 April 2008 - 12:22 PM
First, thanks for the feedback, and congratulations for your books (I have a couple of them). The same for the excellent observations you have made available throughout the years. I started this project with the old SAC52 database.
Don't worry for the Andromeda globs. No one would dare to search them with a low power atlas. The atlas just indicates that "they are there". For finding them, obviously one needs more power. I could find seven of them with my 10", but I needed a photograph to spot them; perhaps the TriAtlas C (currently its faintest stars are around 12.25 mag) could serve for some of them, but the best is a picture or a full power computer map. The reason to plot them in the B-set was simply that I thought the B-set would include all known globs (after all, there are very few); in some cases, the C set is necessary, and in the case of M31 globs, a picture is the best alternative.
I still have to prepare a list with common names for stars, asterisms and named objects (such as the Barnard's Loop). This feature will be present in this edition, that is sure. What I have linked is only a first proof to get some feedback.
Again, only some "brightest" (if this could be said at all) Abell clusters are plotted in the atlas, but no galaxies (except outstanding cases such as NGC 6166 and so on). I decided to include all Hickson and some Abell clusters. The base program (CNebulaX) includes more than a million galaxies (Hyperleda), but the atlas is cropped, and even at full power, most Abell's includes few or none Hyperleda galaxies. The first release of the C set included around 37000 galaxies, and B-set (the set to which correspond the sample charts), around 7000. With my telescopes, I have not seen Abell galaxies except some cases (Perseus cluster, Hercules AGC 2199, and a few more); I was able to image at least the CrB GalCl, though.
Posted 01 April 2008 - 07:50 PM
Congrats - I really like these previews, even if some parts are quite crowded with objects (e.g. M31).
One thing I noticed: the stars do not appear round when they're not at the center of the page.
Posted 02 April 2008 - 12:50 AM
You may find this hard to believe but here in the US finding A4 paper is almost impossible. I understand that the US is going its own way with 8.5x11 inch paper but it's all we amateur astronomers have to work with.
Thank you for your dedication to mapping the skies.
Posted 02 April 2008 - 07:33 AM
I have got the common names and asterisms files from the Saguaro Astronomical Club already. In fact, the core of the A and B sets is the SAC database, that for me it is one of the best (if not the best) deep sky resources. I have to say that I am quite fanatic of the SAC databases since the 5.2 release; I made my first deep sky program for managing it. Thanks for the idea, and if you know more sources, please tell me.
Posted 02 April 2008 - 07:39 AM
I can prepare PDF files at any size. What is the most common paper size in the USA? Please tell me, and I will prepare a sample PDF.
About paper sizes, Adobe Acrobat (or Adobe Reader) print menu includes an option to scale up or down the print job to any paper size (the maps are vector-based: there is no loss in resolution). Did you try this feature? I don't see any problem in changing the size to whatever measurement (except more web space to store the files).
Posted 02 April 2008 - 07:46 AM
I also like the look of the new maps. Another sample, this time of the A-set:
Proof page of the A-Set (TriAtlas 2nd, ed.)
I didn't see any distorsion at all in the printed charts, but if you magnify the charts to very high levels it could happen. I suppose that it can be generated by the virtual printer code and can be supressed increasing the detail (and the file size...).
Posted 02 April 2008 - 08:01 AM
I am quite amazed by the problems with the paper size. Have you tried to scale the paper using the print menu options of Adobe Acrobat -or Adobe Reader-?. Using this menu, you can print the charts at any scale value you wish, and fit the charts in 8x15" pages (or whatever size) without losses in resolution.
Is the 8x15 inches the standard US paper size? Please, confirm it or tell me the most usual page size, and I will generate charts for that size. It's a pity that countries do not find an agreement in basic things like this. In Europe the A4 is the standard. All printers are prepared for it.
About madness of making charts... Well, what I really like is to see the deep sky, and the basic tools are the sky charts. It is only that.
Posted 02 April 2008 - 05:55 PM
The "A" sizes makes more sense to me, as they all seem to be half's of each other unit (ie: Double the width of 2 A4 = A3). No idea why the US didn't stay with the same sizes though, as A4 is used EVERYWHERE else in the world, and 8.5x11 is ONLY in the US...
Heres a link about all international paper sizes...
Posted 03 April 2008 - 06:53 AM
I have prepared a proof with the paper dimensions you have told me (11x8.5" = 27.9x21.6 cm = letter size). Please have a look and print it. If it looks right, I can prepare editions for this page size. This is the sample chart:
Sample page in 8.5x11" paper size
The charts are prepared for 600 dpi printing; I am using an HP laserjet 1022 printer. I suggest a good laser printer: ink printers do not work so well for small high resolution details.
Posted 03 April 2008 - 05:34 PM
I just printed out your sample, and chose NO SCALING. Seeing as you left very little margin, the bottom of the chart didn't print on my printer. The top, left, and right all printed fine, but are very close to the edge, within the normal margin.
If I scale it ("Fit to printable area" option), then all prints fine, within the margins...
Printer margins are .5 inches from each edge, as well as top and bottom. MOST of the time I can override these, except for the bottom margin. I can not go past the .5 on the bottom or I loose whatever was there.
But I digress... Like I said, with Scaling, all prints fine.
Posted 03 April 2008 - 05:48 PM
Jose, thanks for your hard work on these charts, and for taking the time to accommodate us weird U.S. folks, it's appreciated!
Posted 03 April 2008 - 06:18 PM
And just for another opinion, I tried printing with both "no scaling" and "fit to printer" and both worked fine for me; "no scaling" didn't cut anything off any of the margins on my printer.
Yeah, this will vary from printer to printer. I actually think it has nothing to do with margins in the printer, but rather how the printer feeds the paper. The head couldn't print any more beyond that .5 from the bottom probably because of how it feeds?!? I can get (with other prints I've done) within an 1/8 inch of the top, left and right...
Posted 04 April 2008 - 11:30 AM
It is easy to change the format and print in letter-size paper or whatever else, bigger or smaller. In that way you not need to scale down the maps (tiny labels), and the quality is better. The total number of pages may change, but I can solve this. There is, however, a problem: the three sets require around 600 Mb for storage, and now, with the first version and all the other stuff (CNebulaX and other files), I have exhausted all my web space (around 1 Gb); I can replace the 1st. version by the 2nd., but I have to space to store two 2 full second versions. Perhaps I could store sets A and B in two page sizes, but the C set is impossible.
I will link the A and B sets in both A4 and letter-size (8.5x11") -probably next week-, but when I finish the C-set, I will have to remove the letter-size edition.
Posted 04 April 2008 - 08:10 PM
Yeah, Scaling is fine, but you end up with TINY labels. My eyesight isn't getting any better.
Posted 04 April 2008 - 10:25 PM
Great work, as ever... too much information for me (as I've told you in other times) but it really looks very nice, besides, I think it will be very useful for a lot of serious astronomers around the world. Greetings from MÃ©xico.
Posted 05 April 2008 - 06:53 AM
Creating two versions seems to be too much work in my opinion. I had no trouble printing the first version on letter paper.
Posted 05 April 2008 - 10:24 PM
I had no trouble printing the first version on letter paper.
Again, as I stated above, it all depends on your printer...
For the size, and as tiny as some labels are, I would LOVE to bring it to Kinkos and have them run off a copy on 14" paper... Laminate and spiral bind that!
Hmmm... Wonder what that would cost...
Posted 06 April 2008 - 09:53 PM
I was very excited to see this thread in early April. You seem to be ahead of schedule, as your 2nd Ed. was not expected before summer (at the earliest). Does this mean that an updated CNebulaX is also coming soon (and will it include the Sky Atlas Generator utility)??? If so, any quibbles that anyone has with your provided charts can be changed by the user with their own settings at chart generation time.
It seems that some of the people commenting in this thread may not have been familiar with your earlier efforts. I think your approach (of an A set, B set, and C set of increasing detail) is ingenius. For those who think your sample provides too much detail, I recommend they look at your A set.
I think your improvements of changing the charts to vertical so the two page spread becomes, in effect, one chart is a GREAT idea.
I believe the label overlap problem was the biggest deficit with the earlier edition. This only affected certain parts of the sky, but if you have solved the problem: KUDOS!
As recommended at Jose's web site, the A and B sets (together) might make a nice Field Atlas. The C set, is essentially a Uranometria 2000 sized set that could be broken into two volumes.
Also, for those who wish a more professional binding of Jose's charts, you might check out services like lulu.com. In fact, you can choose A4 paper at lulu, if you like. They also offer bulk rates for printings of 25 or more, so perhaps a group of interested users could go together for a group order and save.
In short, Jose, with CNebulaX and the TriAtlas you are providing an absolutely stunning set of resources for todays amateurs. I applaud your efforts and your generosity in giving back to the amateur community!
Posted 06 April 2008 - 10:31 PM
It prints very well for me either with or without scaling.
Thanks for taking the time to provide us wacky American amateurs with the ability to print on our wacky paper.
Posted 07 April 2008 - 03:01 PM
I have finished the B-set, with 107 charts (I have just printed my working copy and I like it a lot... ). These are the provisional downloading links (save the PDF files to disk before opening, because they are huge >40 Mb files!!!):
TRIATLAS 2nd edition - Set B - Charts 1 to 36
TRIATLAS 2nd edition - Set B - Charts 37 to 72
TRIATLAS 2nd edition - Set B - Charts 73 to 107
...and the index of the B set:
Index to B-charts (Triatlas 2nd edition)
DO NOT PRINT YET!!!Give me some more days to correct mistakes and make the final edition. The whole sets A, B and C will likely be finished next monday.
1) I have detected one error: 47 Tucanae is wrongly labelled on M104 (47 Tuc = NGC 104; when I processed the list of common names I typed "104" instead on "NGC 104", and CNebulaX looked for "M 104" and gave me its location). I have revised the list and the remaining objects seem to be right.
2) the RA axis of charts 99-108 is not well printed (not important for norhern observers)
3) I will likely give more overlap to charts A, but if I do this, it will imply changing the numbering of the A charts given in the B set
More improvements to add to those that I commented in the first message:
(9) Constellations codes are overlaid within each chart in each set
(10) Framing charts indicated to zoom out easily. In the B-charts, the A chart in which each B-chart is framed is indicated above the chart number (bottom right)
(11) Common star names (Betelgeuse, Rigel...) and common names of deep sky objects (Minkowski Footprint, Intergalactic Tramp...) are now also printed. Around 200 objects are labelled.
(12) Line sizes now indicate brighter objects (e.g., a brighter galaxy has a thicker outline than a fainter one). Selecting the bright objects is now much easier!!!.
(13) Much more galaxies are oriented acording to their position angle
I would also like to show you some of the new C-set...
The Virgo Galaxy Cluster (the center):
Center of the Virgo Cluster in the new set C
Note the different outlines for indicating brightness. It is a very nice improvement, I think. BTW, I have just found an error in the RA axis for charts 99-107, that I will fix hopefully very soon.
As I told you, I can prepare letter-size pages. When I finish the A set, I will place both (A and B) in letter size. However, I will have to remove one of them (150 Mb) when I prepare the C-set, because I have no room for everything. I have placed my new printed copy in a folder with protecting plastic pages, in A4 format. I prefer something smaller to have at the telescope, flexible and protected of moisture and tearing. This is a very cheap solution that I would recommend to everybody.
Hi Jorge, nice to have news from you again. In spite of what you are telling, I hope that the A nd B set will be very useful for you, it is quite comfortable but powerful at the same time. BTW, did you got the last CNebulaX?
I was using the first edition for nearly a year. It was quite good, except in overcrowded areas. Now I hope to have fixed all the problems. It includes many thousands of objects and it is very difficult to avoid label colliding. I have removed most of overlaps, but the algorithm sometimes cannot get convergence. I very hope you will find this new release good.
The reason to go faster is... that I was impatient to get the new atlas. I needed a long holiday time (Easter) after having finished all the job tasks that in my case are concentrated in Fall and Winter. I have concentrated all my efforts in the Atlas program, that has grown quite a bit and it is a separated program from CNebulaX; perhaps I could link it again. The problem is that is not so easy to deal with. And not only managing it: when it is running it represents a so massive work for printer drivers that usually congests it and I can only get blocks of 30 charts (except when charts are sent individually to the printer). I agree with all what you are telling. My main atlas is the A nd B sets, and I use the C-charts when I need more power. One can have a very comfortable and powerful atlas with the A and B charts.
OK. I thought that you had not tried scaling. This allows changing the margings and adapt to a different paper sizes the printing job. However, I will link an A and B set in letter size; perhaps this could be more convenient because the maps will be better adapted to the paper.
OK, time to go home...
Posted 07 April 2008 - 05:24 PM
Thanks again, JosÃ©!
This is just awesome.