General discussion

[<hk>]

Your quoting style is very hard to read.

Clipping is a mapping technique where you build invisible walls around objects to prevent the player from jumping on them. The same goes for patches that let the player pass on way and block him in the other direction. This is used by mappers who don't know the "delay"-flag of the start timer to prevent Timer Resets.

If we drop the qualification round and do the time adding up thing then there should only be one map for each skill. Not three strafe maps. That way strafer aren't preferred.

[khtti]

IMHO:

Personally, I have no problem letting mappers play their own map.
There is potential for abuse of course, but they could just alias around any rule,
either by submitting the map under an assumed identity or by playing under one.
It seems better to just allow this and avoid giving anyone a reason to be shady.
Besides, this can't work at all without willing mappers, best not to demotivate them.

Please reject any map which contains FPS killing features.
Also, I don't think torture maps are suitable. That might be a bit ambiguous,
since for some people pdm01_pea_impostor and similar are torturous,
but I refer mainly to the CS Kreedz style runs that consist of 450ups cjs on tiny platforms.

It may seem wrong to limit mappers in this way, and excluding features that may appeal to others seems a bit extreme,
but to me theres nothing impressive about using the best jumping physics ever to complete maps that are equally possible in halo, UT, etc.
I suppose this one is open to discussion.

Thinkmaps are fine by me, though

The custom engine fiasco of last year was a huge mess, and did very little to prevent cheating.
I was paranoid enough not to use it at all, beyond proving to myself how useless it was.
If you ask me, quake3.exe is valid and anything else isn't.
Don't get me wrong, dfengine is *awesome*, I just think DFWC should be as traditional as possible.

The discussion on allowed / disallowed obs was lol, too. Lets not complicate things without need.

An unrestricted category would be fun, as an exhibition.
Scripts, strafebots, random OB's, timer reset routes, etc would be fine as long as physics are normal, just no -128.
Hopefully this might serve to promote honest playing in the main competition.
Of course, there must be competitors for this to be worthwhile.
( Offtopic but related: !topbot plz(please)! )

[$$-albion]

Agree with most things, except this:

If you ask me, quake3.exe is valid and anything else isn't.
Don't get me wrong, dfengine is *awesome*, I just think DFWC should be as traditional as possible.

Problem is, quake3.exe crashes after a certain amount of time (mostly after 30s-1min) on my laptop (probably due to vista). I've tried several things, but the only thing that worked was playing with alternate binaries, aka dfengine & cq3 ^^ And I still don't see why dfengine shouldn't be used. But I'm sure bringing this up again will create another massive shitstorm... *sigh*

[MidiMightyMoe]

I was afraid nobody would actually run it, but it seems some people are motivated .

My sugestions would be :

• No direct elimination. I am in favor of a point system rather than summing up times, time difference between maps are not the same, +1.000s doesn't refer to the same skill difference on every map. A lot of sports use a simple point system and its working well. This way people of every playing style have their chance, not only people who are good at long combo map.
• Player drop or qualification is a good thing imo, it feels more comptitive, you have more pressure even if you're not aiming for top 3. However, I wouldn't drop player after every map, I think we can have a first phase (let's say three maps) of easy/mid maps and top20 (or topXX) moves to the final for another set of three maps, much harder this time.
• Final phase should be one strafe map , one full rocket/rocket heavy map, and a combo run, including plasma and strafe/rocket/nade tricks that were not covered in the other map (maybe focusing more on tech things than on speed). Goal here is to cover as much diffrent skills as possible.
• First phase could be either one map of each style (strafe/rocket/combo), in this case points have be higher (doubled ?) on the final maps, or more fun/creative map that are just used for qualification but doesn't count in the final result, this way if an unconventional map ends up having cuts/cheesy way (dfwc2010-04) it doesn't ruin the whole thing (good people would still qualify). Maps have to be relatively short and easy. Goal at this point is to have the maximum amount of player having fun and comparing their demos with the pros.
• Focus more on gameplay than on graphics, good looking map are nice, but I prefer good gameplay as long as the design is decent/doesn't hurt eyes.
• More beta test than last year (I don't blame beta testers... you know i love you morbus :p), checking on every possible abuse is #1 priority. This could be done by either have more beta tester or by extending the beta-test period.
• More communication ! Last year no one knew what was happening, who would be qualified after which round, how the winner would be determined, wich engines were suitable etc. People have to know what the staff is thinking and such event needs some advertising. This includes people actually speaking enlgish (not me ? ) and fast result post as well (even if some demos have to be eliminated afterward).
• Staff people who get to play the map before the comp (mappers beta tester) are not eligible for top10 ?

Also, uZu wrote a program that checks for pmove_fixed 1 or other cvar (without watching the demo in the game of course), I think sousix and uzu are already using it for demo validation in frcup, and they said this could be easily implemented on the upload module of a site. This could prove useful, you should ask him.

And if I can help in any way, let me know, I'll be glad to !

[-MdC-uZuskaya]

Yep, for demo validation (at least cvars and regular info found in a demo file), you can use the last version of qdemos; this tool outputs infos which can be easily checked against matching patterns for automation. You may see a sample output here.

About the maps, I'd suggest to have a common/coherent set of textures or some common stuff for mappers to share (like a font, a pattern, ...), which could lead to a nice overall map pool

[<hk>]

We wrote our own demo parser for mDd. It has some extra features.

[-MdC-uZuskaya]

yet another closed-source initiative I suppose... using an existing open-source project and providing a patch wasn't a valid option?

[<hk>]

Submitting a patch to a project that does something different is rather pointless. We don't do high scores. We have a generic parsing core that supports a number of modules like for drawing routes as seen on the forum or anti cheating. Sanity checking is just another module.

For the open source part: For now we released all source code we were asked for, including the server demo source code.

mDd is not Q3bers

[$$-albion]

Suggestion: maybe you could make some threads to "recruit" some ppl(people) to manage the whole dfwc, general thread for the mappers who are interested in making a map for dfwc. Maybe also a closed thread stating the agreed facts and stuff. Putting everything in one thread is making it uncomfortable to read :p

And to the ppl(people) making page-long posts, please make a tl;dr :p ^^

[<hk>]

Don't worry. That's why I made this a sub forum.