Wednesday, April 18, 2012

More UI for transitions

Found my bug with NGui, the atlas for images also needs to include fonts. Although NGui works with fonts in a separate atlas using two atlases causes Z level drawing problems.

Working on UI helped me realize a better way handling transitions, that will be flexible enough for rest of prototyping I think, hopefully even into the product. I played with the base orientation along xyz axis.(which is why everything is upside down) This will take a bit more work. I want NGui, user expectations and rotating through angles along multiple dimension to all work together. The user expectation is most flexible because I can swap axis letters when displaying co-ordinates to users. If I have to I may be able to swap co-ordinates for NGui but then I won't be compatible for upgrades so need to do it in extension, could be pretty messy. When rotating an object in 3D rotating about X then Y is not the same as Y then X . I want user to be able to twist an object on screen and only change one axis angle value. I think if I lay my object out "flat" correctly this will work, but need to test this. I see in the demo that the way the 4th panel moves has changed when I changed global orientation. How to let the user control this will be a UI problem.

Green and red frames represent start and end of transition. Moving them changes panel numbers, but this can only be done in editor. i have to write the drag, resize and twist code for run time still.

Almost finished refactoring transitions.


V0_0_5

UI for transition panel

UI for transition panel is almost done for first draft. Tested in browser instead of editor and found a display problem fixed it, did some more work on panel and it broke again. I though I had found the cause but more testing of NGui today.

Last night I went to an entrepreneur talk and networking event. Talked a bit about the project with people and realized this morning I am over geeking the design. I do not plan on removing any of the features I planned but I will need a new dead simple setting automatically does transitions with user only placing art work in final position. The simplest one will let users give a single size and import artwork as all whole pages and produce the manga. Probably 80%~90% of artists will use the two simplest setting. The manga produced by the simple settings will be same as by advance setting so artist can use simple settings and then tweak with advanced settings.

v_0_04

Sunday, April 15, 2012

Seperating UI for touch and mouse.

Have played a little with the UI with fingers and mouse. I think the multi coloured gizmos will work good on computer with mouse. Fine control and no twisting make them a good fit. although for 2D placement drag and drop instead of gizmo will be my first option. For touch the multi finger twist and pinch means the gizmos are not needed. so that means using #ifdef compiler directives to separate the two interfaces, and "Finger gestures" packge by fatalfrog.com on unity.

The FPS counter showed 30 FPS on iPad and iPhone 4 which is max frame rate on those devices. on a iPhone 3G with iOS 4.x the frame rate dropped to 20 FPS for the mesh deformation in panel 3. That would be too slow for general performance but for worst case on 3 year old phone that is fine. I am not going to worry until frame rate gets down to 25 on iPhone 4.

I think I know how I want to do the popup panes for control position and transitions of panels, from UI point of view, but first I have to change the animation system. Currently there is no central control of animations so I can not stop them all anytime. I need to do this when reseting or moving all frames to final position in composer. This is exactly what JobManager.cs what meant for.

Started on pane for editing transitions but need to do some more work in gimp first making some graphics.

v0_0_3

Saturday, April 14, 2012

MangaPad First Draft UI and internals datastructures.

I don't really know how much I want MangaPad to be capable of in the first release version. So there will be the usual tension between wanting to make my software to general and worry I'll make a lot work for myself later if don't make general enough. Since like most programmers I tend to make things to general I will be trying hard to keep it minimal.

I am writing this in Unity3d so there will only be a single source for both browser and iPad.I don't know what it should do or how the UI should look. So first off I will try and get some minimal UI up that works on browser and maybe iPad. I started by refactoring the data structures to be cleaner. The third panel has a fancy transformation of the panel from a sphere to flat, I will leave that in but not refactor it yet. First I will just work simple transformations of location, rotation and scale. I found a library on the Unity Asset store for $15 that makes a UI for those. It is in JavaScript, and has mouse clicks hard coded in (so probably won't work on iPad) and does not work with an orthogonal camera, but only $15. So I have have included it will use it for testing the feel of its UI. If it feels good I can recode to c# . I also changed my camera to perspective but moved it far away and reduced its field of view to 5º so it looks orthogonal if I keep everything in the same plan.

I am also using Prime31's opensource UItoolkit code heavy modified for the panel transitions. I think I will use the opensource JobManger code too. I plan to use NGUI for the Composer GUI. The fancy transitions are done with Mega-Fiers which is great mesh transformation code. I hope it is not to heavy to use lots of it at runtime on an iPhone.

Getting some artifacts on the screen when I rotate or scale the panels. I wonder if shaders (main becuase that is the code I understand the least, so I always suspect it). Turns out not to be shaders but a second GUI camera I have does not have the right culling masks.

Added a FPS count (frames per second) so I can also track performance on iPhones.

version 0.0.2






MangaPad new sideline project

OK the plan is blog a a project from start to finish. I will also post links to prototype of the software. The most current version will be found at Ryuuguu.com

I first though about a manga program 2 years ago but it was only one of many ideas. Then I looked at some of the tools I had seen coming out recently and thought I could do a good manga composition program with these and it would run fine on an iPad or browser.

First I looked around for what manga iPad apps were new. I had looked a couple years ago. I did not see anything like what I wanted to make but I did see one that had a similar end product. The idea behind Manga pad is that it will allow artists who already have art work in digital format to compose panels in to scenes (like pages) then publish this to iPad or browsers. Each panel would also have a transition. Such as sliding in from the right. Looking at my prototype will give you an idea of what I mean. Press the gold button at the bottom to go to next frame. The original manga KinyuRival is by a friend. The app I saw was by CyberManga and is also called CyberManaga (app). It also has voice actors and is multilingual something I had not considered. Also I saw on their website that they are interested in making a manga composer for PCs aimed at independent artists. There were no updates about the composer software so I contacted them and got a friendly enthusiast reply from Kohei Nishino. He is a professor of managa at Kyoto Seika University. Since April is the start of the school and business year in Japan it is a very busy time. Hopefully we will meet when things have gotten less busy.

Next post I will discuss some of my first draft UI design ideas.

Wednesday, January 6, 2010

Learning Japanese on your iPhone

I have found 3 programmers are useful for learning Japanese on my iPhone. The first two are dictionaries based Jim Breens jmdic and KANJIDIC2. I am not sure about the third. For s straight Japanese to English Dictionary "Japanese" is what I use it also includes some JLPT level divide flash card practice. The second dictionary is Shinkanji. Its strength is it accepts handwritten input for Kanji you don't know the pronunciation of, also it good for browsing from Kanji to Kanji by radicals. Both show stroke orders and lets you keep lists.


As study tool I use my iPhone the most not with a app but a website ReadTheKanji. It is a Kanji flash card site but with two differences. First it shows you words like "行く" or "説明" to learn instead of single Kanji with all the possible pronunciations and meanings. I find it much better memorize one thing at a time than 5 things at once. ReadTheKanji reduces it 2 things, a reading and an English meaning. It also give a sentence for each word.  Its layout is not good for the iPhone but I still find myself spending hours reviewing and learning new words/Kanji with it.  The other frustrating point is that the site is often slow.  An iPhone app version would solve both these problems. Hopefully I will be writing that app with them in the coming months. It is an app I would use a lot so I want it done right and as they say - If you want something done right do it yourself.  

Monday, December 21, 2009

Startup Weekend Tokyo

Spent the weekend coding, networking and making a pitch at Startup Weekend Tokyo. It was a good time and I learned a lot. Started Saturday with breakfast and networking for an hour or so. Then we sat a different table by our primary expertise. I was at the developer table, there were also marketing, design and idea tables. Dave McClure with Geeks On A Plane got the startup activities going with a session of  half baked dot com . My team cam up with FreshYopparai.com (FreshDrunk.com in English).  Then it was on to the first pitches of the session. any one who wanted to could get up and give a two minute pitch and then the best 5 would be chosen by vote and those 5 would be developed over the next day. The pitches were translated between English and Japanese by Samuel Baron. He did a great job. I am amazed by the ability to remember a a minute of pitch and smoothly translate it to a another language. My pitch did not get many votes but I did get some interest and question if I was going to do follow through and try the startup or if the pitch ws just for the weekend.  (I don't know the answer to that yet.)

I ended up joining the iPhone Yoga app team. Yoichi Shirakata was the idea man. In marketing was Yoshiko Yamamoto and Loren Fykes. Design - Gohsuke Takama. Programing Aaron Franco, Takashi Kaneko and I.  The Saturday afternoon was spent discussing what we should make and why. Market size, competition and product features. When we had agreed we had a feasible product for a business, I started coding. From here my perceptions of what was was going around me become quite limited. I went into head down mode and only had one ear on the surrounding conversations. Which was mainly in Japanese so I would have needed to concentrate to follow it. The app design did not include any 3D features but Unity3D was still great for prototyping work. I had some basic features running on the iPhone within an hour so.  Having something on a phone that people could see and feel (it used the accelerometer)  help my (and others I think) energy level. The ended around 21:00 and people went home or off to the bar. 

The second day started with breakfast and a little hanging out, but people were pretty focused on getting to work on their project. We were down to 4 people in  our project. This was not a complete surprise some of our team had told us at the beginning had told us they would not be able to come the second day.  I spent the morning on twitter stuff for the app, getting it to tweet the right information in the right format,with some help from Aaron. Aaron was working on the website while I did app work. Gohsuke and Yoichi were working on the business plan and presentation. We got a call for two a sentence description for the panels summary sheet. I came up from coding to join that discussion a little. We then got a quick lunch in our room. 

Time for the presentations, there was a panel 6 people I think. I dozed off a few times until it was time for our presentation. When we went up to the front and started Gohsuke going through the power point I realized I had not really seen all the slides and not in order.  I found myself thinking "Wow this is a real business plan". I gave the live demo of using Wubble (our product) to track your balancing time. We took some questions from the panel and later I saw a nice tweet from Dave McClure "Wubble presented compelling idea for suite of mobile fitness apps for jp women".  Wubble took first place among the entries as best startup business plan/pitch.

We then had a Japanese TechCrunch interview and things wrapped up.

Here are some videos. Our presentation is at about 1:30 in the first presentation(1/2) video and the the Awards presentation video.