Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Simon

Pages: 1 ... 214 215 [216] 217 218 ... 277
3226
Lix Main / Re: Feature wish: un-Stretched widescreen
« on: June 17, 2015, 03:18:28 AM »
ccx has explained it perfectly. Thanks. 8-)

Right now, choose Integer scaling, or play windowed.

Support for all screen resolutions will come in the D port using Allegro 5. The support has been a development goal from day 0 there. Screenshot with my desktop resolution of 1280x1024, which is 5:4.

-- Simon

3227
Edit Simon 2015-06-16 18:00 UTC: namida has reuploaded the archive, now everything is playable.

Both geoo's and my level are chopped off in the middle of the files.

Reupload would be nice. :-)

(What am I experiencing: geoo's level is chopped off in line 322, my level is chopped off in line 105. My level should have 107 lines instead. There seem to be chopped off exactly as many bytes as there are linefeeds in the original file. Downloaded the archive twice, didn't fix the problem. Unpacked with two different programs, didn't fix the problem.)

(What others would experience: Lix refuses to start the level and displays icons for missing terrain.)


-- Simon

3228
General Discussion / Re: Best quotes from IRC and Mumble
« on: June 16, 2015, 10:49:18 AM »
[12:28] <geoo> l.get_good returns true if the contained level is valid, and false if it's not or doesn't exist, right?
[12:29] <SimonN> right, not exist implies not good
[12:29] <geoo> so basically, when I hold the key I want the same to happen as in the !get_good case, right?
[12:29] <SimonN> right
[12:30] <geoo> except that it doesn't work


Later:

[12:34] <SimonN> the gameplay will again decide which level to load, what a mess
[12:36] <geoo> so it must be possible for levels to be both good and not good...oh wait nvm you're overwriting l
[12:37] <SimonN> the magic doesn't happen in this file
[12:37] <SimonN> gameplay/gameplay.cpp:106, determine_filename
[12:38] <geoo> haha, so is the function even used?
[12:38] <SimonN> which function?
[12:38] <geoo> ReplayBrowser::on_file_select
[12:38] <SimonN> it is used, but only to check whether the level is good
[12:38] <SimonN> if it's good, we go into the game, and the game again decides which level to use
[12:39] <SimonN> it so happens to come to the same conclusion as the browser
[12:39] <geoo> that's terrible redundancy
[12:39] <SimonN> this is extremely dumb code
[12:39] <SimonN> yes
[12:39] <SimonN> also, gameplay.cpp:109 the comment is wrong, but the code is right
[12:40] <SimonN> the comment should be "always use pointed-to level"
[12:41] <geoo> ok it works now


-- Simon

3229
Forum Games / Re: Family Feud 2015
« on: June 15, 2015, 02:50:40 AM »
How is the next host decided anyways?

If you would like to host one, then it's you!

(wait, how did I got here? ??? ;P)

-- Simon

3230
Lix Main / Re: Discoverable UI (scrolling, dir select)
« on: June 14, 2015, 09:18:31 PM »
Thanks. I'll try to put it in the tooltips for now.

You're making a reasonable case against the buttons. Then, tutorial + tooltips seems to be the best we can do, other than fat in-your-face advertisements that obscure the main play area.

I will make two options: one to show tooltips, one to show skills on hotkeys. This was demanded a couple years ago already, I should have done it earlier.

-- Simon

3231
Contests / Re: Official Level Design Contest #5
« on: June 13, 2015, 10:06:58 PM »
increased the "a gimmick of your choice in NeoLemmix" to two gimmicks



-- Simon

3232
Forum Games / Re: Family Feud 2015
« on: June 11, 2015, 04:05:50 PM »
Excellent ideas.

Quote
Goals: Each team scores 10 x Chasers x (Beaters + 1) x 2 if at least one Keeper.

How does the 'if' connect to the equation? I assume this means "stuff x2 if at least one keeper, stuff x1 if no keeper."

-- Simon

3233
Forum Games / Re: Family Feud 2015
« on: June 10, 2015, 08:05:04 PM »
why namida turned out to be the most popular answer to Q10? He explicitely announced that he would not be the host this time!

Nobody else underlines. :-[

And namida liked the second-answers round, and would have liked more participation on that.

Thanks for hosting!

-- Simon

3234
General Discussion / Re: Ichotolot visiting Simon
« on: June 10, 2015, 12:43:44 PM »
In late July/early August 2015, geoo will be back in Germany. We're aiming for a meetup in Göttingen, with Ichotolot and me.

Nepster, you're also from Germany. Do you live reasonably close to us? 8-)

Are there any other users who'd like to come around? We've already tried to persuade Akseli, with limited success:

<Simon> I'm not too sure whether Akseli would like to visit us, his first gut reaction to me meeting Ichotolot was "how can you be sure the guy won't attack you?" :D
<geoo> We could turn your cellar into a sauna and lock him up in there for 7 days
<Simon> or he could learn C++
<geoo> yes, let's offer a free C++ or D course to Akseli, he'll definitely bait and then we banish him into the sauna
<geoo> damn should have said that off the record :P


You do not have to be a programmer! We can teach you will have a great time no matter what!

<Akseli> good stuff in the logs of the morning :D
<geoo> haha Akseli, are you coming over?
<Akseli> wait, I'll finish my porridge and answer after that
<geoo> that's a yes! :D welcome, it's gonna be fun
<Akseli> haha :D
<Akseli> I haven't been travelling for ages


-- Simon

3235
Forum Games / Re: Zendo, play by forum
« on: June 09, 2015, 12:14:50 AM »
Koan list only.

White koans:
AAA(A*)   DOGSSS    GAMING  MIRE      SPAM
ABC       DUG       GGG     QQQQ      SSS
AKALAZAM  FIQQ      GITSMH  QQQQFIRE  SSSSSS
AMEN      FIRE      GNIGMA  QQQQQQQQ  STR
AYZ       FIREQQQQ  GOD     Q{16}     S{9}
CLAM      FREAK     GYMHIT  REQQ      YAZ
DDD       GAM       MEAN    RIFE      YIU
DIG       GAME      MEGA    SITE      ZYA
DOG       GAMERS    MIGHTS  SMH
DOGAAA    GAMESS    MIGHTY  SMIGHT


Black koans:
A          DOGCC      FIREQQQ      ING                 OA        SITES
AA         DOGE       FIREQUISITE  KLZM                OL        SOMETHINGDDD
BAC        DOGENIGHT  FIRER        LIX                 POOL      SPAMGOD
BBB        DOGF       FIRES        LOOP                Q         SS
BBBB       DOGI       FIREY        MEANT               QQ        SSSS
BBBBB      DOGRST     FIRST        MIGHT               QQQ       SSSSS
BBBBBB     DOGS       FREAKKK      MIGHTYCC            QQQQQ     TE
BCA        DOGSS      FREAKY       MIGHTYDDD           QQQQQQ    THMIG
BCD        DOGSSSS    FRQQQQ       MIGHTYDOGGY         Q{12}     TTT
BLACK      DOGT       GAMER        MIGHTYFREAKY        Q{65536}  TTTT
BLACKPOOL  DOGTOR     GAMERST      MIGHTYGGGGGG        Q{65537}  TTTTT
CCC        EEE        GAMES        MIGHTYMIGHTY        RK        TW
DDDD       EERIE      GAMESSS      MIGHTYMIGHTYMIGHTY  RMEGA     WHITE
DDDDD      ET         GIT          MIGHTYQQQQ          RS        Y
DDDDDD     FFF        HHH          MIGHTYS             RST       ZIX
DOGA       FIGHT      HOT          MIGHTYSSS           RTS
DOGAA      FIR        HOTLIX       NIGHT               S
DOGC       FIREQ      HPT          NS                  SERIF


(x)* is a repetition of (x) 0 or more times.
(x){n} is a repetition of (x) exactly n times.

-- Simon

3236
Forum Games / Re: Zendo, play by forum
« on: June 08, 2015, 09:21:42 AM »
The general rules of the game disallow the empty koan. It's not part of the koanspace.

The reason is pragmatic: In the tabletop game, a marked empty koan looks like a stray marking stone.

Allowing the empty koan is a minor variant, and it doesn't affect the game. If you want to test the empty koan, then do it. Secret rules can be evaluated on the empty koan in the normal fashion. ("The X is Y" fails if there are no X, or if there are more than one X, or if there is exactly one X that is not Y.)

-- Simon

3237
Forum Games / Re: Zendo, play by forum
« on: June 08, 2015, 03:18:28 AM »
Right. Edited.

My instinct was to avoid a weird mix of regular expressions and math, but it's probably the only concise notation. >_> Let c be a letter, n >= 3. Then c{n}, which denotes n times the token c, is white iff n divides id0(c). This is true even for A, B, C, because no number >= 3 divides 1 or 2, and all numbers divide 0.

Having this for same-letter koans, there is obviously some magic happening at 3, and some method to mangle non-same-letter koans. The given example koans for anagram-changes-buddha were 3 letters long.

New koans:

SOMETHINGDDD
MIGHTYDDD
MIGHTYCC
ABC
BAC
AYZ
ZYA
YAZ
BCD

-- Simon

3238
Forum Games / Re: Zendo, play by forum
« on: June 08, 2015, 01:02:06 AM »
No?

Let id0 be the function on single letters that maps A to 0, B to 1, C to 2, ..., Z to 25.

Then Proxima's half-rule states: For each letter c that is not A, a koan that contains exactly n >= 3 copies of c [and nothing else, EDIT Simon after namida's reply] is buddha iff id0(c) divides n. Equivalently, that koan is budda iff there exists a natural number m with id0(c) * m = n.

The marked koans violate this, because the koan BBB is black. (Everything of the form B{n} with n >= 3 appers to be black.) Also, namida mentioned there exist two letters, such that koans consisting only of copies of that letter are always black.


Noes, I can't read properly. n divides id0(c) is the claim. That would make B and C the two letters that make only black same-letter koans.

-- Simon

3239
Lix Main / Re: Discoverable UI (scrolling, dir select)
« on: June 06, 2015, 02:55:48 AM »
Hi, yes, you're welcome to make many posts in a row, each is valuable and appreciated much. 8-) Thanks.

Ideas for directonal-select buttons: Right, it's what you've suggested. I imagine the button to take one click, then either be active until it's deactivated, or be good for only one assignment. For more fine-grained and fast assignments, the user should learn the hotkeys, as usual. The dead-space problem you're suggesting is very real. (To compare, skill buttons never become dead space, since they show information.)

Thinking about removing modifiable spawin interval: The idea isn't as far-fetched as it sounds. L1 and ONML have Flow Control, which is a single level out of 220 to be designed entirelly around it, and the harder WAFD need it too. Killing only one level is acceptable for the greater good.

The critical opinion is how player-modifiable spawn interval helps execution, and people have reported 5 to 10 levels from the 240-level community pack where it's integral to the solution. When culture has grown over the design, the culture locks the design in place.

(To compare this with one-way arrows: I don't have one-way arrows either, because I deem them unnatural and weird to draw. That doesn't stir up opinions as much as culling the spawn interval. A few level ideas won't be possible, but the level design space remains wide open still. It's harder to take them out than not introducing them in the first place.)

Exact setting of spawn interval: I've taken this as a feature request. The spawn interval in-game buttons could be replaced by a single button. This pauses the game and opens a modal dialog (hehe) where the user can precisely set the spawn interval. Or something. But precise control is a must, if we want to keep the user-changable spawn interval at all.

Tutorials to introduce directional select: We have this in the batter tutorial, and we want to make a dedicated tutorial level at some time. The longer the tutorial session becomes, the fewer people will play through it all. OTOH, even if it's discoverable from the main UI, people might disregard it as hard-to-use via buttons, and not bother to learn it either.

Tutorial for right-click scrolling: This is new even to players who have played other Lemmings-like games. Those won't play through tutorials, but would need to know it still. The only alternative to tutorials seems to be aggressive advertizing for RMB-scrolling whenever the user scrolls a lot with the screen edges.

Button for right-click scrolling: Very interesting, could be doable. Let's see what ideas this spawns.

As last resort, tinker with edge scrolling: RMB scrolling and egde scrolling can be selected independently in the options menu. Per default, both are active. I want the user's instinct to lead them to an implemented scrolling method. I could make the edge-scrolling speed option faster by default, so RMB-scrolling becomes less of an improvement. Alternatively, I could disable edge scrolling by default, but I'd rather not do that.

Mantra about GUI design: Correct, hiding non-critical stuff in menus is normal, and the functions discussed in this thread are too time-critical for a menu. The hotkey/RMB is the correct trigger. The button would only be for discoverability. Of course, this leads to the question: If discoverability is the problem, not having the button per se, there is maybe a different solution. Excellent push into an important, fresh direction.

Anyway, thanks for your valuable input, also for the IRC discussion. I have had guests over in real life, I'd have participated more otherwise!

-- Simon

3240
Lix Main / Re: Discoverable UI (scrolling, dir select)
« on: June 05, 2015, 04:36:56 PM »
Thanks. Yeah, I feel too that something crispy is missing for the RMB scrolling.

With the release rate (spawn interval), I'm most concerned about whether it eases execution, or even is flat-out required for the solution. If you consider it useful here, then I'd rather leave it in.

If its only use were saving time, turbo-fast-forward >>> is often enough, even without minimal spawn interval = fastest release rate.

-- Simon

Pages: 1 ... 214 215 [216] 217 218 ... 277