I'm trying to fork the github project so I can use their OAuth code
Edited:
This link describes how to do it but I lack the cognitive abilities to do it
I'm trying to fork the github project so I can use their OAuth code
Edited:
This link describes how to do it but I lack the cognitive abilities to do it
I'm scrotum deep in this shit...
..and I'm adding text shadows/outlines in aero glass mode to mintty as a side-project.
Did some more code cleanup, i felt silly to have the map generating in the map rendering class.
Writing this reminded me that every tilemap engine i have made, generates the map in the same class that draws it. It just feels so stupid, like having a group of architects planning a building in the building they're planning the building on
I also just realized moving the player is also done in the camera class. Well that can wait until tomorrow, need sleep.
Its not a game,the stars are just a skybox im using.
Edited:
I really need to begin on dividing my maps in chunks, but i have no idea how
I am totally original.
LibreOffice is OpenOffice, but "more open".
What language are you using for it?
*watches it till the end of the video and see's C++*...
Edited:
And better math formula support
How are you handling the math for this? I ask because if you're doing this as an N-body it's incredibly stable and I'd love to know how you did it. A few years ago I did something similar:
For whatever reason (probably boredom) I went all-out on the math behind it and made it as realistic as I possibly could, but in doing so it became incredibly difficult to create stable systems. I guess that's how it should be.
Are you only counting local gravity from whatever something is directly orbiting?
It looks like the bodies are just too massive in your simulation.
They are, I didn't have any videos saved so I just ran it and grabbed one real quick that clearly demonstrated the issue, but it still exists inherently no matter what your masses are. It's the 3-body problem and it's unavoidable really, you can stabilize it but you can't make it go away. In my case that's actually really really stable, I'm using a 6th order symplectic integrator with my time step. What kills it visually is my orbit calculations which obviously fluctuate hilariously with every step because I don't do any smoothing over time.
All of that said, just because I'm using a lot of math and I refined this like a dozen times doesn't mean there's nothing left to learn, and if he's got something in his that I don't I'd at least be interested in knowing what it is :)
And has a bit more updates (around 12 IIRC).
Last updated
OpenOffice.org 3.3.0 27-01-2011
LibreOffice 3.5.2 05-04-2012 (Can't find the date of 3.5.3)
IcedBox would be correct in that the orbits are fixed paths. I'm using a polar coordinate system with the sun as a focus of the ellipse. (r, theta) where r = p/(1 + Ecos(theta)); p = pow(b, 2) / a; E = sqrt(1 - (pow(b, 2) / pow(a, 2))).
We went over orbits in my engineering physics class last week so I decided to make a simple model of the inner solar system during my Python class. I'd like to someday create what you posted.
That's somewhat along the lines of what I thought you were doing. That's how mine started ;) In truth while my math is strong my physics is arguably weak, I knew very little short of F= ma and gravitational forces like you said you went over recently in class when you started. My project started exactly like your's, then I posted a question to the physics subforum of Gamedev.net and had a couple of really smart guys take a lot of interest in helping me work through the complexities of the whole thing. I might be biased because it was my thread, but I think it's a pretty interesting discussion (albeit one-sided, like I said I knew nothing of this stuff at the time). Take a look if you're interested in improving your's, I stumbled on just about every problem you can come across while doing this stuff and they taught me a lot.
I got points marked off for using parts from Lua. Seriously.
Oh yeah! (OpenTK C#)
Pretty much hello world with OpenGL, but this time with VBOs
Hooray!
http://filesmelt.com/dl/Phyzicle1.apk
New test build of Phyzicle for Android
This time it should work better on the Galaxy S and the Android 2.2 devices, so if you guys could test it, that'd be awesome :)
Once again, I'd like:
1) Device
2) OS
3) Result of running the app
There are many other fixes (softkey support, better UI sizing, and so on) but I'm mainly concerned about the device compatibility. If you have a device that worked in the previous test, feel free to try it again, just to make sure I didn't fuck something up!
Will this work on 4.0+?
I tried it on Android x86, but it said that there was a problem parsing the package :(
A couple of things about your app:
A) Please have your app hide the bottom bar on devices with no hardware keys.
B) Your polygon tool is fucked.
C) The UI is a bit confusing, I'm not really sure where to start with the app.
D) Something feels 'off' about scrolling and zooming etc.
Edited:
Android 4.0.1 - Kindle Fire Rooted & Flashed with Linux 3.0 kernel
That space stuff has kind of given me an idea for a game. After sitting down and letting my dead heart spew its creativeness, or lack of, on to a page, I think I have a decent and simple idea. Just want to say that I'm not stealing the planet orbiting idea, as that won't be a major mechanic.
Yes
Edited:
It's an ARM package
Edited:
Why? How would you exit it?
What's fucked about it?
I'm not sure how much simpler the UI can get than "tap the square button, draw squares"
Hard to figure out what you mean.
This is progressing quite nicely:
I hope this doesn't get covered in app reports for null :(
Well the bar thing handles itself, when you tap the very bottom of the screen it comes back up. It gives a whole solid cm of extra viewing space which is a big deal on 7" tabs.
Draw circles with your n sided polygon tool, watch as it produces crazyness. You should probably find some way to prevent that, or something.
I for instance couldn't figure out how to freeze things, and starting the simulator was non-obvious. You'd expect the start button to be big and in an obvious place.
The fourth might not be your fault, but take a middle-end android device on any version prior to Android 4.0, that's how your scrolling feels.
I don't think there's much of an idea left to steal from planet orbiting anyway. N-body sims have been done to death and there's more than a handful of games based around this concept.
Good luck tho!
Holy fuck is it really 1cm on a 4.0 ROM? I'm running 2.3.3 and the bar is about 5 mm, which feels perfect on the Kindle Fire.
The triangulation will fail if you draw self-intersecting shapes. It's not worth the effort to prevent it, but I am investigating how to properly triangulate self-intersecting shapes.
Jeez, that's the first item in the FAQDid you even bother open the side menu and tapping the big question mark? As for the start button, it feels pretty obvious to me, no one else has brought that up. But thanks for the feedback, I'll take it into consideration.
The Kindle Fire is known to have extremely laggy touch input, if that's what you mean. It does feel slightly less responsive than other tablets.
I spent this entire afternoon and evening trying to get a Java CAS (Computer Algebra System) to compile correctly. It would only compile after I gave Eclipse 3 GB of heap memory and it wanted to recompile itself every time I changed a single line in my own application.
Now you would expect me to say something about a stupidly simple fix, e.g. some compile checkbox in the Eclipse settings. Unfortunately I found out something rather depressing: it's a known bug with the Android SDK.
Now, the Java compiler does actually know about which files change and which don't and it only rebuilds new .class files for those files that change. Unfortunately there's another step in the pipeline that translates Java bytecode to bytecode suitable for Android, the dex tool. This application performs that process on every single file (library or otherwise) every time I want to run my app to test something new.
This process takes up to a minute, so now I pretty much have to resort to using a dummy library that imitates the CAS just so I can stay sane while developing this app.
Oh well, at least I learned something new.
Well the touch input is actually TOO sensitive with a 4.0 rom, and it's pretty accurate your app and netflix are the only touch issues I get.
Can you please describe the issue a bit better? What doesn't work exactly? When you tap a button, it doesn't really get pressed? Or something else? The Kindle Fire's screen is horribly sensitive (even compared to the PlayBook, which already had a horribly sensitive screen compared to an iPhone or iPod), so there's a "touch radius" which can be configured. Tap 5 times, type in "touchpx 20", and tell me if that helps.
I don't think it's so much physical as maybe like the 'tactile feel' of it. it's a bit floaty, and there's no reference frame since the background is static.
That command didn't really fix it, but I think the I nailed where it comes from down. Place your finger on the screen, scroll, stop moving your finger and stay as still as possible, then lift. It "snaps" a bit.
Uh. Oracle has rights to the OpenOffice name, so new versions of OO are called LibreOffice.
OpenOffice LibreOffice = new OpenOffice();
Edit: Whelp, looks like I'm horribly late..
Edited:
It looks like you're trying to compile mono under cygwin. I thought it compiled fine under Windows natively?
one of the many reasons why android is a ghetto
Oh man, that filesmelt background made me think my phone's screen was fucked at the bottom
ICS on a nexus s.
Works exactly the same as before.
A few things I noticed this time around:
- The box tool is selected by default. I think this should be the move tool instead. The first thing I did was try to pan, which made me draw a box and get rather confused.
- Menu items can be "selected" while you're trying to scroll, which is confusing. You should only show the tick on release, or you think you've made a mistake. Maybe highlight the background, box or text instead if you want to keep some kind of feedback.
- The app takes a while to restore when the phone is unlocked. Not much you can do about that I guess, but for example anomaly warzone earth doesn't take any time at all.
- It wants a permission entitled "phone calls". This is ominous, but I guess you need something from there or you would have removed it.
I also felt that the background could have a texture that scrolls with the view, to give it a better feel than just stuff floating around.
I get a parse error on my ZTE Blade with Android 2.2 as soon as I try to open the apk file.
Download it again?
I think you might have accidentally pressed it because it's in the corner. I just tried it and it wasn't selected at startup (and there's no reason for it to be selected at startup)
Good point.
Really? That's odd. I never noticed that.
Those are all of the permissions it asks for. It comes as a template with Marmalade so I haven't messed much with it (I did remove a lot of unused ones). Maybe it's READ_PHONE_STATE? I have to check out what that's for.Code:<uses-permission android:name="android.permission.READ_PHONE_STATE"/> <uses-permission android:name="android.permission.SET_ORIENTATION"/> <uses-permission android:name="android.permission.INTERNET"/> <uses-permission android:name="android.permission.ACCESS_NETWORK_STATE"/> <uses-permission android:name="com.android.vending.BILLING" />
A lot of people have mentioned this and I should probably try it, thanks for the tip.
Yeah I just tried it again and you're right, my bad.
The phone state thing I think has to do with the device ID or something.
It turns out I downloaded the Filesmelt download page. Derp.
Edited:
Alright, installed it but the colors are all wonky and it instantly freezes as soon as I try to do anything. Can't expect much from a Blade, though.
1) Motorola Droid (The very first one.)
2) Cyanogen 7.1.0 (Android 2.3.7)
3) Nothing happens. Black screen.
Why don't you support 3 year old phones. 0/10.
Edited:
it is a joke.