Any update on L17 core / extended core changes?

by **bahamut** » Sun Sep 17, 2017 1:14 pm

HERE WE GO! The solution!

I started this last night(If i recall correctly..?!) been working on it, I need feedback!

: ;) I'm a newbie at this! WIP; teaser.jpg (98.03 KiB) Viewed 1320 times

by **Leto_II** » Sun Sep 17, 2017 2:02 pm

Please bear with this long post, as I have a lot to say. I will try to be concise.

First, I would like to apologize if I came across as angry during the Mumble discussion last night. I did (and do) have grievances regarding the overall direction and policies regarding core (some of which I believe are shared by a good number of pilots), and wanted to express and defend this position. When faced with what seemed to be a monolithic opposition by the administration and a few others - well, it's kind of hard for things to not get heated (and I don't like arguing in the first place). But it was (and will continue to be) an extremely good, and necessary, discussion, and I think we all appreciate the admins showing some transparency and community involvement.

Regarding Rie2mann, if the community wills it to be in extended core through open, transparent discussion (given its status on the margin earlier, which I think warrants such a discussion in the first place), then this is absolutely legitimate and welcomed. I am glad that such an open discussion may now occur.

However, let me reiterate what I hope was made clear in Mumble: The main issue at hand isn't about Rie2mann, but about the stagnant, limited nature of core in the first place. Let's take a step back, and try to take a look at what core has become: A very limited set of levels by one author, with a narrow band of styles. A good number of pilots have issue with this, yet the response from the administration is the same as when this issue was brought up by Vainiac a little while back: Organize among yourselves, and try to prevent this from happening - but within the confines of our existing system. It seems as though little consideration is given into re-evaluating this system in the first place - and yet, these main issues with core have worsened since Vainiac posted about it. Finally, another, smaller, but related issue is that the default stance of the administration seems to be exclusion, rather than inclusion.

I feel very strongly about preventing stagnation and in games, because I have seen it in other communities. I hope we all can make decisions to avoid this here.

Now, regarding specifics of Rie2mann: Thank you all for the commentary, suggestions, and discussion. I always liked those beige textures, but if visibility is an issue, then that certainly trumps any ideas of aesthetics. I am completely open to changing it (especially given suggestions of what texture to change it to - it could be a one-to-one change of beige to something else).

And finally, on the technical side, to Lothar and Drakona: I don't think your "unique pairs" metric measures what you want it to in the first place. Rie2mann had 30 "unique pairs," yet only 6 games did not have me in them? Shouldn't your "unique pairs" metric read something much smaller, then? I think, if you were to treat games and players of a level as a mathematical graph (games = edges, players = vertices), then "unique pairs" would measure a maximum number of edges that do not share a common vertex. This is the idea of a "maximum matching," and such a maximal number of edges is called a "matching number."

An algorithm for computing it (I think it's O(V^2 E), where V is number of vertices (players) and E is number of edges (games). This isn't bad, considering how small these graphs are for our ~20 member DCL):

https://en.wikipedia.org/wiki/Blossom_algorithm

Obviously, using this "matching number" for "unique pairs" would worsen Rie2mann's score by a lot. I estimate a score of around 3 - 4, so this mathematical diversion doesn't even "help" my level when compared to your existing metric (but, as I said above, the main idea of my above argument is about core in general, and not about Rie2mann). But, I think this method of counting "unique pairs" actually measures what you want it to. Given a fixed number n of people in the DCL, a complete graph of people playing everyone else in a level (i.e. the complete graph on n vertices) would yield maximal "matching number."

by **melvin** » Sun Sep 17, 2017 2:05 pm

one level to rule them all (baha's new level)

by **Mark392** » Sun Sep 17, 2017 2:07 pm

The only question I have is about the supposed limitedness of DKh levels? I see successful employment of every style and tactic in all of the levels listed as core, and we all know I dislike missiles, so what are these supposed limited bands of styles? I hear this referred to often, but never very well explained.

by **Leto_II** » Sun Sep 17, 2017 2:09 pm

I would say that they are all notably flat.

by **Mark392** » Sun Sep 17, 2017 2:11 pm

Good specific, thank you, Leto! Flat compared to what? They seem to be as tall as other popular non-DKH levels? Note: I LOVE vertical play l and employ it readily in union with horizontal play.

by **Leto_II** » Sun Sep 17, 2017 2:20 pm

"I don't know," (Blarget) "I don't care," (Lee) "ethylene," (Swarthy) "anon," (Big Rat) "Rainman," "June Bug," (not sure who made these last two, maybe someone else can fill it in) alongside with perhaps my "Rie2mann" and "Fourier," are examples of levels that feel much less flat than the DKH levels. Some might have similar "heights" in raw number of cubes, but I think we all know that capability for interesting, vertical play is more about connectivity and other variables.

by **Leto_II** » Sun Sep 17, 2017 2:21 pm

Also, baha, I'd like to fly around in it, as I can't really get a good idea from the picture. I always like new levels, though, so please continue!

by **bahamut** » Sun Sep 17, 2017 3:27 pm

I've just finished, Uploading to dmdb after i submit this post!

by **Code** » Sun Sep 17, 2017 3:55 pm

very nice map! i played a game in it!