remove : function(i) {
internalArray = internalArray.splice(i, 1);
},
Riven had it right the first time. Array.splice modifies the content of the array being called returning the deleted elements.
Not to be confused with Array.slice which doesn’t modify the original array.
Sorry, my bad, I posted my reply without noticing you had already pointed it out :emo:
I took another sunset photo on a beach tonight.
Note: the ball in the sky is actually the moon, not the sun.
Today during my 6 hours train travel from work back home I read kinda everything about point-in-polygon tests, including Eric Haines’ article about “Point In Polygon Strategies” - http://erich.realtimerendering.com/ptinpoly/ and all the strategies mentioned in it.
After that I emailed him at acm.org asking whether actually no one on earth ever considered the most intuitive and natural way of accelerating the “Crossings Test” (i.e. even/odd algorithm or “raycast algorithm”) by using a freakin’ dead simple binary tree?? Or am I just super unable to Google things up?
I mean what the hell?
Okay, granted, that article is from 1994, but that article talks about 1,000 vertices being representative for “large” polygons??? And the so far best algorithm taking 485 milliseconds to solve a point test with such a polygon? the hell? I can solve a point query for a 524,288 vertices polygon in under 0.6 microseconds.
Would someone knowledgable in this topic/field please tell me that there are such algorithms already out there?
After Eric’s kind response and saying he would not know of such an algorithm publication he encouraged me to write one for the JCGT.
Those people only like C code, not Java 
, so translated to C: https://raw.githubusercontent.com/httpdigest/pointInPoly/master/intervalTreePolyPoint.c
Doing proper memory management in C is… tricky, but one can squeeze a tiny bit more performance out of it.
Here is a video of a small demo program showing interactive polygon drawing and mouse cursor inside testing: http://i.imgur.com/hPTRIW7.gifv
I’ve been fiddling around with voxels a bit. Not going to do a Minecraft clone or anything, just experimenting with 3D.
Some of you may remember all of the posts I have made trying to figure out why my audio locks up or doesn’t work.
I’m happy to say after many months I have finally narrowed down the issue My speakers (the device that OpenAL used) were causing the issues, because when I finally plugged a pair of nice headphones into my computer sound worked right away
Hell yes, SATs were delayed because of snow. Hoping to get a 2000+ this go got a 1890 something last time.
Good luck!
When I took the SAT (long ago) there were only two tests, English and math, 800 pts possible each. I “studied” for the English by reading lots of Huxley and looking up every word I didn’t know. Only managed 700 on the English, but strangely it was at a higher percentile (of UCB applicants) than the 750 I got on the math. It got me into UC Berkeley, my goal. (But the lack of study skills–why study if you can get ok grades via cramming?–made getting through college very difficult.)
Wow! Congrats on getting in there, that’s one of the colleges I’m looking at. Right now my top picks would be Duke, MIT, Berkeley, UVA, and Cal Tech. I have such a small chance of getting into all of those I have a lot of backups, but I’m hoping that I can get into one of them! Been working hard to get my GPA above a 4.0 before my junior year ends, and I think it’s gonna happen.
As a sidenote, solved this monster today, as well as got two 9 second solves:
Wow, that’s awesome. I just got into the 2 minutes range with my Dayan Zhanchi 3x3x3 cube, where my best is 1:42. You’ll be my inspiration.
Hey everyone starts somewhere! I’d suggest a bit of an upgrade. The zhanchi was good when it was released (really good) but now there are much better cubes out there. I personally am a fan of the Moyu Aolong and Congs Design Meiying. I’ve tried a dozen or so cubes and those three are the ones that have really stood out to me. If you’re looking for a bit of a cheaper alternative, the QiYi Thunderclap is really good for $9. It almost became my main. Keep practicing and you’ll be sub 20 in no time
It’s been mentioned before, but I finally started working on it:
So far I’ve only gotten through 15 problems in about 6 hours (on and off), but it really pushes your thinking skills!
Distance field fonts!
Features:
Technical:
EDIT: Getting 3500 FPS with the above text, and it’s actually CPU limited. The above screenshot was taken with MSAA to anti-alias the bounds frame. >___> MSAA doesn’t affect the actual font quality at all.
Returned from the dead, which is to say my new computer finally arrived.
You really only notice how many programs actually play a part in your daily routine when you have to reinstall them all…
It very impressive, rendering text in openGL is a much bigger pain than it should be. Well done for making it as easy as java2D’s drawString.
I hate to be a nit-picker, but is there a thin vertical white artifact (|) between some letters? Such as before the first i in ‘|ipsum’, and on the second line before the second i in ‘suscip|it’. Seems to only affect i’s as far as I can tell.
Heh, you’re right. It’s due to texture bleeding from insufficient padding. I’ll have to reconvert the font. xd It happens more frequently for smaller fonts, but I thought this was big enough…
Hey! Me too. One of my RAID drives decided to explode so I took the opportunity to just replace everything. Getting a new machine these days is tough. Now on day 5 of setting the damn thing up. (sigh)
@theagentd: one distance per sample? What’s interesting to see is close in at (what should be) sharp angles and very small…assuming either will happen.