What I Did Last Week #1 (1/16/2011-1/22/2011)

[PyTom]

Welcome to the first "What I Did Last Week" thread. Over the last couple of years, I've noticed that there's a lot of posting, and especially concrete posting, towards the start and end of projects --- but not much in the middle. The hope is that this series of threads will fix this, by providing a place for people to talk about what they've done over the course of time. To help foster a sense of community, I want to try time-based threads, rather than project-based threads.

The challenge here is simply to post, once a week, a synopsis of what you did in the last week; and to do so at a level of detail that will let people follow along and make suggestions.

A couple of rules:

1. Anyone can make a new What I Did Last Week thread, if the old one is out of date. The goal is to post at least once weekly, with weeks running Sunday-Saturday. If you don't see a thread, make a new one. (If more than one thread for a week is created, we'll merge them.)

2. Be specific. Don't just post, "I wrote 100 words, made 2 sprites, and made 1 background." Show us - give us a synopsis or excerpt of the writing, or post a picture or two. The idea is to give people something to comment on, rather than going "yeah, that's nice." Use spoiler tags for large, game-ruining spoilers - but maybe not small ones, as people should expect some spoilers here.

3. Collaborative Criticism is welcome. This should be considered a "Critique Please" thread. That means that if you have suggestions for how a project can be made better, post them. If there's something you especially like or dislike, post about it.

4. Don't post lack-of-progress reports. If you didn't do anything this week (and it happens to everyone), don't post about it here.

This is an experiment; I hope it leads to more discussion of in progress games, and I hope that discussion leads to more and better games.

[PyTom]

While I hope to (after 6.12 is out the door) start collaborating more on games, this week I put my spare time to Ren'Py development.

Ren'Py

- I triaged my TODO list, so I now know what's going to be in 6.12, and what didn't make it. I've decided that there will be another feature release relatively quickly after 6.12, so I felt more comfortable about slipping some features.

- The image prediction support has become quite a bit more intelligent, especially when screens are involved. Ren'Py will constantly predict the first screen of the game menu, so a right-click should bring the player instantly there. Whlle at the game menu, Ren'Py will predict all of the reachable screens, again improving performance. This only works with screens, rather than the more traditional mechanisms. Screens are also more predictable in the menu and say statements - for example, 6.12 will be able to predict a side_image, or a dialogue window with a custom background.

- The tutorial game has been redone to use screens. This both gives it the benefits of screens, and helps me to find screen problems. The definition of the "choice" and "nvl" screens changed, to make writing screens easier. The screens.rpy file from the tutorial game is also present in the template game, so new projects will be using screens by default.

- Implemented a system of screen variants. You can now define one set of screens, and then a second "touch" variant that will be used on touch-based devices, like Android phones.

- Rewrote Solid and Frame to use GPU scaling, after noticing that doing them on an ARM was taking an absurd amount of time. This also tends to make things use less image cache space, which helps free space up for more advanced reloading.

- Added a new mousearea widget. This defines an area of the screen which has its hovered and unhovered properties called when the mouse enters and leaves, respectively. This is different from buttons in that, while only one button can be focussed at a time, having an mousearea overlapping a button works fine.

- Random cleanups. There were lots of little checks and fixes that were in my TODO list as having to be done before a release. I got a lot of these taken care of, moving Ren'Py that much closer to being releasable.

[jack_norton]

Hehe nice, that's what I did:
- implemented an aggro system for Planet Stronghold based on a suggestion from a user in my forums
- played the first tech demo sent by the coder for my 3d boxing game made in Unity3d
- wrote Rebecca and Philipp subplots
- wrote storyboard for a superhero game and commissioned some sprites
- spoke with Scout about ending/important event CGs for Love & Order and commissioned them
- did lot of testing on Planet Stronghold and some on Love & Order

may have forgot something but that were the main events

[PyTom]

Not a huge amount of detail, there - what were the subplots about?

[jack_norton]

Well you get to know that Rebecca had a twin ssister who was a medic and got killed by some alien, and you can help her with her revenge and trigger romance CG+ending later in the game. As for Philipp, you discover he was a treasure hunter which unlocks three special battles to get very powerful pieces of equipment.

Ah, forgot that I also added a check to prevent getting more than a kiss/romance CG on same playthrough

[KimiYoriBaka]

sounds interesting. I think I'll try too. I'm not good at keeping track of the passing days, but this should have been one weeks worth:

-finished my prototype for an ace attorney clone in screen language using characters from the touhou universe and handed it to some of friends. the case I used was too simple and didn't really allow much freedom as far as how to prove the defendant's innocence (my goal was to have multiple solutions for everything) and there were still places where I had forgotten to fix things, including a section of the trial where I forgot to include anything for providing evidence. I will not be posting this, as i'd rather not have anyone else see it. (the graphics and story are embarrassingly bad.)

-I finally figured out how to get allegro to work on my laptop (i'm not very experienced in programming outside of renpy). In addition, I got started on a potential aerial shooter that, if my ideas actually work, will have very low level graphics and efficiency as a cost to have a custom level option. I've programmed most of the classes I'll need, but they'll all need to be adjusted for various reasons, and I'm particularly concerned about my boss class, which I found out would take 200 lines of code to use even for just a simple 20 sec shot pattern.

I'm wondering if I should stick to renpy...

-I finished the second out of three base script routes for my story "castle of arhannia", and handed it to my friend that is most likely to actually read it and give me input. it came out to 4747 screens of dialogue at 14.1 words per screen, for a total of 66,775 words. It covers the second of two optional love interests, tazha, as well as all the scenes the mc needs to go through to find out the true nature of the mentally-ill princess, aiya. as a result of tazha being a lot less talkative than the other one, hima, this route came out to 1000 words less than the first one, despite having more scenes.

-even though I was going to put all the artwork on hold till the script was done, I decided to cancel that because I just haven't been drawing enough lately. Being primarily an artist, this can't be allowed, so I started on my first cg attempt. because I'm mainly a traditional artist, I intend to do as much as possible by hand, so I'm going to draw each background and each cg on four sheets of printer paper held together by painter's tape. I'm hoping this will allow enough space for detail, as well as help with the blending of colors, as they should blend together when I resize the images on the computer. I also plan to do the character sprites on a single sheet so they'll come out sharper than the bgs after resizing.

the cg shows the mc holding hima softly while hima cries.

I'm a bit nervous as to whether the art style I'm using fits the characters, but I figure that something I'll only find out for sure upon release, as I've dropped the idea of releasing a demo

[papillon]

The third dungeon-exam for Magical Diary, which involves finding the exit from a randomly-generated dungeon (the earlier dungeons are fixed) but the code work on that comes from poor Spiky, I only have to design the challenge and the text to encounter during it.

There is no individual spell that will solve the puzzle in one go unless you're very lucky. There are several spells you can cast which will tell you if the exit is in the dungeon square you're currently standing in. Some are cheap manawise and some are very expensive. There is a spell which will always point in the direction of the exit, but you'll have to recast it every time you move. There are two spells that will let you follow a trail which has been laid down leading to the exit, and again you'll have to recast the spells as you move. There's also a spell that will reveal all squares adjacent to yours (and uncover the exit if it's in one of them), which is of midrange cost. So part of the puzzle is in managing your mana. There are many solutions but some are not practical... unless you've been pushing up your mana stat. And of course depending on what you've been studying you may not have access to all the spells.

Otherwise, recent MD plot fragments have included returning from Thanksgiving break and finding out what you and your roommates were doing during that time period. Also, a secret society holds inductions during the first week of December and certain characters may or may not be invited. I'm about to start on yet another detention scene, this one for characters who manage to earn punishment during Exam 3. Every detention is unique, so poor Professor Grabiner and I have to keep inventing more ways to torment the PC!

Outside of Magical Diary, I have decided to look into the feasibility of doing a short pure visual novel based on something I mentioned in the Ideas Dump thread over a year ago and have always wanted to play with.

Remember Trap!Princess?

If I do it, it will not be the huge epic some people might have wanted, because I have enough epics on my plate. Current design looks to have about three or four endings only. But because it's smaller than my typical fare, I expect it would also be sold at a cheaper price.

[PyTom]

KimiYoriBaka:

Depending on how storytelling-oriented your shooter is, you might be interested in the new SpriteManager that's part of Ren'Py 6.12. It's fast enough for a lot of shooter games, although it might be marginal for the bullet-hell type.

I think it's great that you're having a friend proofread your game. You might also want to consider a "hostile proofreader" - I'd worry that friendship could get in the way of horribly frank feedback. (Do other people have comments on this?)

Papillon:

How robust is the game to "mana screw"? If the user knows that he could solve puzzle with more mana, will he be able to go back and train that stat, or is he stuck having to either load from a save, or figure out another way?

[papillon]

The exams occur at set points along the school schedule, so you need to work with what you've trained by that point in time. Everything has a lot of solutions, and if nothing else, you can always "give up" in an exam and say that you just can't solve it. You'll be penalised grade points, but the game moves on. You're not just stuck there.

And depending on what your goal is, you might want to fail an exam, in order to get demerits and see how that changes the plot.

I love my heavily-branching plots. Important things are much harder to miss than minor things, but very minor things can come back and affect dialog months later.

[Samu-kun]

While I hope to (after 6.12 is out the door) start collaborating more on games, this week I put my spare time to Ren'Py development.

Ren'Py

- I triaged my TODO list, so I now know what's going to be in 6.12, and what didn't make it. I've decided that there will be another feature release relatively quickly after 6.12, so I felt more comfortable about slipping some features.

Mmm... When can we expect 6.12 to come out, and once it does, how can we go about getting official help on our games? If it comes out some time next month, that'd be just about the time Homeward will be in a fully playable state and ready for extensive programming additions...

For what I've done this week:

-Programmed roughly 80% of the school festival arc. Progress is somewhat slower than expected, since there are still a lot of incidental character sprites that need to be made. I made the rather regrettable decision that every single tiny variation in emotion needed to be captured by a character sprite, so I'm stuck making a new expression for almost every ten lines of dialogue. Not sure if it'll pay off in the end, but I'm already committed to it, so it'd be too weird to quit half way.
-Met with professor in charge of the (real life) school side of the project. Played the game up to the school festival arc. Took her the whole meeting (1 hour) to get through the

Seattle arc

, sports festival arc, and beach episode arc. The whole general arc might take about two hours? Ran out of time, couldn't talk about other stuff.
-Ran into conundrum. I was never on the rooftop of a school in Japan, so I don't have any photographs of it. The

confession

scene at the end of the school festival arc takes place on the roof of their school. Will have to make the background by hand using a photograph composite. Ugh. Early progress.

-More problems. The sprites made at the beginning of the project are now getting obsolete. The character sprites made later into the development into the game look better. Example:

Problem: Bizarre looking hands

Now somewhat fixed.
The only problem is that there are about 120 Haruka sprites that need to be fixed with the new hands. Don't want to waste who knows how many hours fixing a minor error. Debating whether to recruit someone to fix it for me. Will truly be a thankless and boring job though. In addition, there are too many other errors that I want to fix that'll just bog up the production schedule if I start revisiting things I've already done. (Also, the hands are partially obscured by the message box anyways, so it's kinda... >_>;)
-Sketched out CG

[kinougames]

What we did last week:

---About 9 new item pictures.
---8 new CG scenes.
---Worked with the English voice actors.
---Worked on finishing up final tiles for the tilesets.
---Testing the stat system.
---Started programming the second minigame and fixing bugs in the tester.
---Worked on plans for distribution in Japan.
---Worked on monster designs to be sprited!
---Worked on GUI portions.

[Aleema]

OH snap. Cool thread idea. Shame I'm not programming a lot of cool stuff to show off this week. I mostly post in my game's forum with this stuff, but if it's cool here too, then I'll ramble away. Since last Sunday, I ...

- Commissioned Lorey again, thanks to donations, and she sent some sketches for new sprites, including the alternate Dallas.
- Scripted some events that need music, expression changes, and sound effects, specifically a video game scene where I had a lot of fun recreating what a generic fighter game (i.e. Mortal Combat) would sound in the background. I love messing around with sound effects. ^^; Basically, Robin was button mashing with her fighter to win. The girl voice doesn't come up too well though.
- Decided I should finally upgrade my game from Ren'Py 10.2 to 11.2, which was fine, until I realized I had to rebuild some of my screens because they broke, which turned into the most frustrating experience I have ever had with Ren'Py, ever. Besides the window having to actually close and restart for every Shift+R refresh (don't like that), things are working better (like my jukebox) and I immediately made use of the Notify() feature for Quick Save/Load.
- Wrote some Emmitt and Dallas scenes, which is a big deal for me, because writing is suuuch a choooore.
- Argued with the writing staff because I'm a bitch. Luckily they're not so they'll say I wasn't.
- Decided that my mini-games are too complex and I should seriously, seriously consider making them stupid easy. Memory games have been suggested. If I can get a super simple Match-3 game to work (godhfgfdghdohsd) then that will be the main one. There are a lot of pygame examples of Match 3, but I barely understand pygame, so I'll just be coding it from scratch with my knowledge, meaning it will suck. BUT, I love coding so it's not like I won't have fun doing it.
- Put all my variables in one RPY file. I feel so ... organized.

- Set up all the persistent data for the 60+ achievements you can earn in the game, and successfully debugged the achievements screen (I wan't calling it from the menu properly) and got the pop-up notification working when you unlock an achievement. I'm currently working to make it disappear more efficiently, because right now it lingers in the BG until you make an interaction, and I think if I follow Notify()'s example ... SL can make that not the case. Won't know until I try.
- Added a lot of new posts to the fake forum in the game, thanks to all the contributers, like selenity136 and Londonivy.
- Probably a lot of other misc balance stuff with the code. I'll know to remember now.

LAST WEEK'S POLL RESULTS:
Which mode do you usually play these kinds of games?
Windowed: 72.5%
Fullscreen: 27.5%

Should there be a confirmation prompt before Quick Save/Load?
Yes, accidents happen and I could lose progress!: 57.1%
No, quick saves should be QUICK!: 33.3%
Don't have quick save/load.: 9.5%
(From this poll, I decided to make only Q.Load have a confirmation prompt.)

CURRENT POLL: How do you like to advance the text?

~~~
On a side note, is the date in the title's thread suppose to mean we post about what we did between now and next Sunday, or that's when we're allowed to post here? Because I just listed things I did last week ... >_<

[SusanTheCat]

- Put all my variables in one RPY file. I feel so ... organized.

YAY! *clap* *clap* *clap* *clap* *clap* *clap* *clap* *clap*

On a side note, is the date in the title's thread suppose to mean we post about what we did between now and next Sunday, or that's when we're allowed to post here? Because I just listed things I did last week ... >_<

Hey! That was my question.

Slow week in terms of Visual Noveling. Brain overload week at work.

I got the outline done for the Ren'Py course and Lesson One being proofread and validated.

I came up with a zillion ideas while walking my dog, none of which I wrote down.

Susan

[Radhreni]

KimiYoriBaka:

I think it's great that you're having a friend proofread your game. You might also want to consider a "hostile proofreader" - I'd worry that friendship could get in the way of horribly frank feedback. (Do other people have comments on this?)

WARNING: Herd of teal deer approaching!

Friends and family rarely make good proofreaders -- they are too close to you, and can't look at your work impartially. (Teachers are generally not a good idea, either.) This is not to say to avoid them entirely; sometimes you need that extra bit of confidence that no, your writing does not suck. Oftentimes they can help in partial ways, such as spelling, grammar, or getting stuck on details. (In my case, any sentence I write that snags my husband's dyslexia filters gets rewritten. It's a red flag.) And sometimes you're lucky enough to have someone who can thoroughly dissect a story and will tell you what you need to hear, even if it hurts for a little while. And it will, at first. Give your story a week to rest. By then you'll have bounced back and had time to digest those crits, so you can start on the next draft.

A good proofreader will do more than make sure your spelling and grammar are sound. They will analyze every element of your story. Plot, pacing, characterization, voice... If necessary, they won't shy from breaking your story apart -- but said breaking will give you the parts to reassemble a better constructed story.

And you want more than one proofreader. Multiple opinions are helpful. (For one thing, if more than one proofreader stumbles onto the same problem...there's a problem. If all of them point out different little things, without consensus, you're in pretty good shape.) Three or four is a good number.

But a "hostile proofreader?" I don't like that term. If it refers to someone who isn't afraid to tear a story apart to analyze the vitals and help you improve them, that's one thing (and that's not hostile, that's constructive crit at an advanced level). Hostility toward the story...that's a recipe for disaster. (I'm sure there are cases to the contrary, but odds are it'll be a trainwreck if you try it.)

Teal deer version (abridged in no particular order):
Get more than one proofreader. Don't rely on friends and family for impartial critiques. Proofreader with a good sense of story craftsmanship = HUGE plus. Don't seek out antagonism.

[Aleema]

That's funny, because I know my sister will tell me "that's ugly don't use it" and when I proof her school work, I'll "This paragraph is just you BSing, scratch it" ... but we also support each other's projects because we know it makes each other happy. I'll never tell her making flowers out of duct tape is stupid, because she made something with it, is proud, and I should be happy for her. So I half agree.