Ludum Dare 32 - Page 3

Ludum Dare 32

Moderator: Content Developer

gajop
Moderator
Posts: 3051
Joined: 05 Aug 2009, 20:42

Re: Ludum Dare 32

Post by gajop »

We picked a theme: Corruption, did some brainstorming and started coding it.
Concretely: https://github.com/SpringCabal/
For the brainstorming doc I'm sending you a link in PM since everyone with a link can edit it (other trusted individuals interested may ask to get it).
User avatar
PicassoCT
Journeywar Developer & Mapper
Posts: 10450
Joined: 24 Jan 2006, 21:12

Re: Ludum Dare 32

Post by PicassoCT »

Made some buildingparts for a quick and dirty random generated medieval city
Image
Attachments
buildingparts.jpg
(54.72 KiB) Not downloaded yet
gajop
Moderator
Posts: 3051
Joined: 05 Aug 2009, 20:42

Re: Ludum Dare 32

Post by gajop »

Prep day1: https://www.youtube.com/watch?v=uZ89ibS6YrQ
Some recent commit broke Pica's houses so they aren't rendered properly.
User avatar
PicassoCT
Journeywar Developer & Mapper
Posts: 10450
Joined: 24 Jan 2006, 21:12

Re: Ludum Dare 32

Post by PicassoCT »

still looks promising..
gajop
Moderator
Posts: 3051
Joined: 05 Aug 2009, 20:42

Re: Ludum Dare 32

Post by gajop »

Ludum Dare Theme voting now starts (5 phases of voting): http://ludumdare.com/compo/2015/04/11/l ... ng-begins/
The number of themes will be drastically reduced in each, and the final one will be announced on the last day.
PS: I'm going to predict now that "Companion" will win.
EDIT: Oh, I didn't realize it's a different set of themes each week! Still wouldn't mind if Companion would win :D
gajop
Moderator
Posts: 3051
Joined: 05 Aug 2009, 20:42

Re: Ludum Dare 32

Post by gajop »

gajop
Moderator
Posts: 3051
Joined: 05 Aug 2009, 20:42

Re: Ludum Dare 32

Post by gajop »

My conclusion is that while we failed to make a complete game for the prep, we'll probably manage to create something in the actual compo, although our expectations shouldn't be too high still.

To share with the rest of Spring, the lessons that we learned and need to improve for the compo:

- Checkpoints! every 3 hours (?), everybody is saying what they’re doing, and discuss
- Make an exact plan of who does what, before the event starts. Separate responsibility by capability and previous experience.
- Reenable mumble at checkpoints because people regress to text
- At the beginning, create placeholders for everything: models, unitdefs, weapondefs and scripts. Otherwise weird errors happen such as units not being removed when killed (lack of scripts). Override placeholders instead of creating new files.
- - Unit model pieces should be defined in placeholder models and be valid. Minimal pieces should be…?
center? weapon (origin of attack)?
- - Figure out what was the issue with aiming (lack of pieces? pieces in wrong positions?)
- - Figure out how to make custom commands that can be used as skills (spells): have limited range and have a cooldown between usability; combine them - - with lups/cegs to do effects; options that require the unit to turn to face its target (either completely or just by turning its turret)
- Do polish incrementally, do important things before polish (e.g. units before music)
- One of:
- - (table for what people are doing-thing)
- - Task priority table/list: a short list of immediate priorities with assignees and ETAs (the current table with a list of all tasks isn’t really suitable for this)
- Availability timetable for each member filled up before the compo and updated during (“available”, “unavailable”, “uncertain” are valid options)
- Get a better clean project before starting:
- - Spring ABC (was ok)
- - gajop’s chili (was ok, but remember that it’s a submodule)
- - OOP weapons/units with created files in BaseClasses (had some minor errors that were fixed)
- - fully working widgethandler (caused issues with removing the default UI)
- - working widget selector (wasn’t configured)
- - a working lups setup (wasn’t done)
- - CEGs (wasn’t done)
gajop
Moderator
Posts: 3051
Joined: 05 Aug 2009, 20:42

Re: Ludum Dare 32

Post by gajop »

Btw, Anarchid and I still want to submit the game to the warmup weekend page: http://ludumdare.com/compo/2015/04/09/w ... /#comments
We still need to do animations, add zombie models and some textures, fix targeting issues, setup a default camera zoom and use the widget I added to disable mouse wheel (otherwise it looks kinda meh), add lups/CEGs for leech life and maybe some basic "blood" effect for all attacks, add terrain rendering off map (like ZK uses for example), cleanup the UI (remove all the pointless commands and leave only leech life, add the corruption level, also remove mouse-over selection widget), fix leech life (buff it, make sure it respects unit facing, add a 5s cooldown and have range), make the player die when corruption is 100 (let's say leech life adds 50 corruption and decays 0.1/frame or sth), make zombies stun civilians, fix house collisions (units can pass through houses as it is) create player-death end-game screen, add intro image, and lastly, package it for windows and linux.

Now that I've written all this.. ouch :D that's at least one full day of work imo. (and I can't do much until Thursday..)
Still, let's do this so we don't find any nasty issues for the real compo.
hokomoko
Spring Developer
Posts: 593
Joined: 02 Jun 2014, 00:46

Re: Ludum Dare 32

Post by hokomoko »

One more thing that I think will help is pick something easier to do:
1) No humanoids. use vehicles, tanks or anything that's super easy to animate.
2) No melee attacks. I don't know why it didn't work the way we wanted, but I'm pretty sure ranged attacks would have worked well enough.
gajop
Moderator
Posts: 3051
Joined: 05 Aug 2009, 20:42

Re: Ludum Dare 32

Post by gajop »

hokomoko wrote:One more thing that I think will help is pick something easier to do:
1) No humanoids. use vehicles, tanks or anything that's super easy to animate.
2) No melee attacks. I don't know why it didn't work the way we wanted, but I'm pretty sure ranged attacks would have worked well enough.
1. I'm actually not sure how hard it is to animate humanoids (using Anarchid's script), but you're right that it's probably easier to do than tanks. Picasso started doing that and just vanished at one point. What we definitely should do is make sure animations are done before textures (I'll make a project development path sometime in the next two-three days).

2. I'm not sure it would be fixed. I tried changing "Melee" to "Cannon" (with similar range), and even setting turret=true, but the issue remained. I'm betting the issue was badly placed weapon pieces ("Empty"), that were on very low height, maybe even below ground (you can see how they look if you use "BeamLaser" type). This is one thing that we need to confirm before the jam (which is why I want to make sure this project gets to a workable state before).
User avatar
Anarchid
Posts: 1384
Joined: 30 Nov 2008, 04:31

Re: Ludum Dare 32

Post by Anarchid »

Bad pieces will definitely screw up your aiming. The Empty on most units is placed to the front of model in Blender, but some hidden voodoo evil could displace it below ground... if the weapon is affected by ground it wouldn't shoot if aimpiece is from ground.

Things i'm going to try to do today and tomorrow:
- textures for exorcist and zombies
- make and/or implement animations
- figure out the melee thing incl piece positions

I agree that tanks are easier; let's see however how hard actually animations are once i get to them.

One more thing that i'd like to try would be a different unwrap strategy, but that requires new models at this point. We'll see.
User avatar
Silentwings
Posts: 3720
Joined: 25 Oct 2008, 00:23

Re: Ludum Dare 32

Post by Silentwings »

I can remove the unwanted usual commands from the UI, add the corruption bar & associated game logic. I don't think that will take long.

I'm not clear on is wanted for "leech life"? If it's going to go into the command queue and then only act when in sufficient range, I can't see what about it prevents it just being implemented as a normal attack command from a normal weapon, with health transfer implemented in UnitPreDamaged. Also, are we going to have more than one such spell?

I don't know if its possible to properly disable zooming, can that be implemented nicely or does it need a hack?

Imo the things that will eat time here are unitscripts + textures + models. We could focus on making the humans ourselves and add some random innocent animals (suitable model+scripts already exist in a game but I forgot which). It could add some cute variety without need for much additional work.
User avatar
Anarchid
Posts: 1384
Joined: 30 Nov 2008, 04:31

Re: Ludum Dare 32

Post by Anarchid »

Models, animations, and art in general do seem to be our primary bottleneck.

It would be very good if someone talented like picasso was highly available and equipped with a microphone, during the event, because it would be good to have more art people.
gajop
Moderator
Posts: 3051
Joined: 05 Aug 2009, 20:42

Re: Ludum Dare 32

Post by gajop »

Silentwings wrote:I'm not clear on is wanted for "leech life"? If it's going to go into the command queue and then only act when in sufficient range, I can't see what about it prevents it just being implemented as a normal attack command from a normal weapon, with health transfer implemented in UnitPreDamaged. Also, are we going to have more than one such spell?
That's definitely one option! I think for the submission we should focus on only one such spell, although initially I wanted more, which might have different use (some might not damage, some might be targeted at friendlies or multiple targets, and so on -> basically think at least Warcraft 3 variety of skills).
I guess this can still be hacked by having weapons do 0.01 dmg, and then use it to do a different effect instead.
Commands kinda seem nicer since they're automatically added as buttons in many such widgets, and also allow additional condition handling, but still very limited as far as the range/cooldown/facing (as normal weapons would). :S Any ideas?
Silentwings wrote: I don't know if its possible to properly disable zooming, can that be implemented nicely or does it need a hack?
I've done that, it's trivial, just return true in mousewheel event: https://github.com/SpringCabal/Corrupti ... ol.lua#L28
What I haven't done yet is to set a default zoom + panning level (so camera is zoomed to something sensible).
In Scened I have a way to make sure camera always follows units (as can be seen on the video), so that's settled.
Silentwings wrote: Imo the things that will eat time here are unitscripts + textures + models.
Textures aren't that important in the early phases, and can always be made simplistic (even in the final game), a simple multiple-color textured unit (one color for clothes and one for skin) is enough for the final release too. Animations are more important, they add life to the game!
Silentwings wrote: We could focus on making the humans ourselves and add some random innocent animals (suitable model+scripts already exist in a game but I forgot which). It could add some cute variety without need for much additional work.
Check the rules! We won't be able to apply for the "Graphics" category if we used existing art. I really think all models, unit textures and animations should be made ourselves. Ground textures should be ours too (I can do the map with just two original ground textures). The only thing in question are icons, and I think we should do our own if we have the time, or minimize them so we can still apply for art with a disclaimer ("The 3 icons used were obtained from pre-existing resources").

Anarchid wrote:Models, animations, and art in general do seem to be our primary bottleneck.

It would be very good if someone talented like picasso was highly available and equipped with a microphone, during the event, because it would be good to have more art people.
Yep, although we still didn't do CEGS, lups, stuns, or proper skills on our side ;) You're usually supposed to spend time "finding the fun" in your game, to make sure you'd actually want to play this, and this is mostly the objective of the coding and should be done while art is making textures (which come last before polishing as far as art is concerned).

As written in the "lessons", I'd really like to know more about what people were doing as well as their availability. I felt my productivity was pretty low when I was struggling with the unitdefs/weapondefs until we did our first checkpoint, and later there were times where it wasn't completely clear who was doing what (there were a lot of switching around that happened). Project management is tough I guess.
User avatar
Silentwings
Posts: 3720
Joined: 25 Oct 2008, 00:23

Re: Ludum Dare 32

Post by Silentwings »

Ok, I'll handle the spells + UI stuff. And then I'll look at lups.
just return true in mousewheel event: https://github.com/SpringCabal/Corrupti ... ol.lua#L28
Can't I get around that by binding a key to some zoom related event, or change camera event, etc? Although I don't think we should bother with worrying about such things right now.

Iirc there is a default binding (maybe t) to make the camera track a units position.
gajop
Moderator
Posts: 3051
Joined: 05 Aug 2009, 20:42

Re: Ludum Dare 32

Post by gajop »

Silentwings wrote:Can't I just get around that by binding a key to some zoom related event, or change camera event, etc?
I guess you can, but no one would be going so far Ludum Dare probably :)
And it's mostly for a better perspective (getting the probably default icon-level distance is kinda weird for a RPG-type game).
gajop
Moderator
Posts: 3051
Joined: 05 Aug 2009, 20:42

Re: Ludum Dare 32

Post by gajop »

The reason why we should make all art ourselves (or nothing):
jam rules-emphasis mine wrote: You’re free to use any tools or libraries to create your game. You’re free to start with any base-code you may have.

You’re free to use 3rd party Artwork/Music/Audio assets, or assets you previously created, but we ask that you OPT-OUT of the respected voting categories (Graphics, Audio). You can opt-out of a voting category when you submit your game.
User avatar
PicassoCT
Journeywar Developer & Mapper
Posts: 10450
Joined: 24 Jan 2006, 21:12

Re: Ludum Dare 32

Post by PicassoCT »

Anarchid wrote:Models, animations, and art in general do seem to be our primary bottleneck.

It would be very good if someone talented like picasso was highly available and equipped with a microphone, during the event, because it would be good to have more art people.
How about resolving bottleneck by shoving evo algos into zombies or not putting too much pressure on the bottle in general.. and those guys who tryied to get cloning material, including that really sexy female assasin are dead ! Better arrange the alternatives..
User avatar
FLOZi
MC: Legacy & Spring 1944 Developer
Posts: 6240
Joined: 29 Apr 2005, 01:14

Re: Ludum Dare 32

Post by FLOZi »

gajop wrote: - - Figure out how to make custom commands that can be used as skills (spells): have limited range and have a cooldown between usability; combine them - - with lups/cegs to do effects; options that require the unit to turn to face its target (either completely or just by turning its turret)
May or may not be useful to you;

MCL Abilities (MASC (sprint, has a cooldown) & flush coolant) http://sourceforge.net/p/mwspring/code/ ... lities.lua
MCL Uplink Abilities (Orbital Arty strike, with cool down - could easily be range limited using this function) http://sourceforge.net/p/mwspring/code/ ... lities.lua
MCL Perks (one off upgrades paid for with xp) http://sourceforge.net/p/mwspring/code/ ... _perks.lua & config file http://sourceforge.net/p/mwspring/code/ ... k_defs.lua
- Get a better clean project before starting:
- - Spring ABC (was ok)
- - gajop’s chili (was ok, but remember that it’s a submodule)
Any improvements to ABC? I dislike that git discards empty folders, there are some ugly ways to circumvent this. Why not 'C' based Chilli? (outdated?)
- - OOP weapons/units with created files in BaseClasses (had some minor errors that were fixed)
These are the S44/ MCL ones? What errors? (so I might fix upstream) :)
- - a working lups setup (wasn’t done)
Can be depended on externally now iirc, package is on rapid (though for S44 i ended up just including it directly as it was less hassle)

Thanks for the feedback on this, I find it very interesting. :-)
hokomoko
Spring Developer
Posts: 593
Joined: 02 Jun 2014, 00:46

Re: Ludum Dare 32

Post by hokomoko »

These are the S44/ MCL ones? What errors? (so I might fix upstream) :)
S44 ones.
They were mostly our errors I think, so I don't think there's anything to fix.
Post Reply

Return to “Ludum Dare”