Post, discuss, or show off your MLP fan art - check tags for commissions

Search /art/ threads

Password  (for post and file deletion)

File 136631210444.png - (2.24MB , 2000x1918 , shipping chart easter 2012.png )
122613 No. 122613
The time has come to update the shipping chart.

I've been gathering the data for these for two years, and it's been a year since the last edition. Since I'm good with numbers but have no artistic talents whatsoever, I do like I did last time and ask /art/ for help.

The last chart, made by CiscoQL, is shown as the picture. He did a good job, but it's hard to fit in 72 characters and a crazy amount of ships on a single image. Since then, the number of characters has gone from 72 to 127 and the number of ships has gotten even crazier. I'm not sure if the new data can be made to fit in a single chart and not be a total clusterfuck.

If somepony feels they can make this chart and have it look good, then that's great. But either way I wanted to compile the data into more useful graphs. I'm a somewhat competent programmer, so I made a program to analyse the data and put it in fun tables - like listing the most common ships of the gayest characters.

But right now this is only a drab text document with a few tables. Nopony wants to read that, and that's where I'm asking for help. I'm asking for two things:
(1) Advice on what tables of information could be extracted from my data (from anypony).
(2) To make a chart and/or make the document visually interesting (preferably no more than one person at one thing).

Who feels up to the challenge?


Unspoiler all text  • Expand all images  • Reveal spoilers
>> No. 122614
File 136631247524.png - (1.26MB , 1500x1500 , 132120912876.png )
For comparison, the chart from a year and a half ago.
>> No. 122633
I've made a little java thing which helped determine all the connections that they are and if there were any individuals which could be put by themselves.

I think I can continue working on this to make it better somehow. Discrete mathematics, graph theory and programming ftw~

Here's the source code: you'll need to download this jar file in order to compile and run:

I've determined that Cranky, Mathilda, Chowder, Colton, Daisy, Sir Lintsalot and Tootsie Flute can all be in a separate graph since they don't interconnect with the others. Also, it will probably be best to start with those who have the lowest amount of connections and work inward, towards the main 6.
>> No. 122636
File 136640443006.gif - (58.43KB , 360x360 , 132619104822.gif )
Nice! Considering how long I've been in the shipping chars business, I should read up on graph theory. Yes if the chart is going be easy to survey, structuring it properly is key. I guess the optimal solution would have no overlapping lines. I have a feeling graph theory could show if that's possible (though it probably isn't).

I can't figure out how your Shipping.xls file looks, I've tried several variations and I keep getting NumberFormatExceptions. I doesn't help that I have no experience with exel. For my program I just put the data in a notepad file and split the Strings by tabulars and made a String[][]. Could you tell me how you structured the file, or show me the results directly?

Excuse me for having a bad memory for internet names (and names in general), but aren't you the same guy that did the chart the last time? I think I remember your livestream channel being named Midnight Star.
>> No. 122687
Oops I forgot I removed that part of the code. If you look at the same pastebin, it should be fixed now. All you need to do is put the .xls in the same folder you are compiling/running the program.

Yes I'm the same guy.

The output looks like this:

The important bits being the degree of each individual and the bottom part which specifies which ponies can be separated.
>> No. 122707
I read up a bit on graph theory. Your degree list is the same as my Versatility Awards list, except I trunked mine to the top 20 in the document. I can see how this i useful for drawing (characters with higher degrees closer to the centre).

Considering crossing lines, I have come to the conclusion that the graph is not planar and definitely has a high crossing number, hard to calculate how high. No help there.

I looked at graph drawing as well, and there is a whole science to it so it's not very simple. I had an idea about how to divide the characters into groups so that everyone is in the same group as their most shipped character. It's far from perfect (the Rainbow Dash group is huge, for one thing) but it might be an indication of what characters should be close together on the graph. Here are my results:
>> No. 122768
So, is anyone interested in helping by either drawing the chart, makig other tables/charts or just coming up with ideas?
>> No. 122792
Remaking in /collab/:
>> No. 127294
Pffft. The characters I like to ship aren't on that list. Yet Winona is....Who ships Winona with a pony? I mean seriously.
[Return] [Entire Thread] [Last 50 posts] [First 100 posts]

Delete post []
Report post