|
pokeyman posted:I like this one better, though I think you've inverted the segmented control labeled 'single/double/stats'. I should be pushing the current mode in, at which point the mode I'm leaving pops back up. Think a cassette tape player: when you hit stop, the play button pops back up. Thanks for the positive feedback, guys. I decided to go with the blue theme and incorporated the changes you suggested. The main reason for ignoring wagers is that this app is meant to keep a record of your Coryat scores and judge your chances of making it on the show (see http://www.pisspoor.com/jep.html). Daily doubles and final jeopardy don't factor into that. Basically, if you're getting 3 or more stars in this app on a consistent basis, you have a very good shot.
|
# ? Sep 14, 2011 20:28 |
|
|
# ? May 9, 2024 09:39 |
|
parkov posted:The main reason for ignoring wagers is that this app is meant to keep a record of your Coryat scores and judge your chances of making it on the show (see http://www.pisspoor.com/jep.html). Daily doubles and final jeopardy don't factor into that. Basically, if you're getting 3 or more stars in this app on a consistent basis, you have a very good shot. Oh, right on! Definitely leave them out then. Hopefully I'm the only one who was confused when I looked for the 'final jeopardy' button.
|
# ? Sep 14, 2011 20:32 |
|
I need a terrain engine in a couple years, so I thought I'd make one now. It's nicely modular; all Terrain knows how to do is to work a heightmap and maintain a quadtree, with a shader where the unpacking magic (geomipmaps with skirts) goes in one function that can be easily copied to another shader. Normal map generation and maintenance, frustum clipping, distance LOD blending, occlusion query LOD offseting, and anything else are in modules that can be freely attached and detached. The TerrainEditor class is for working with the height or colour data. Here we see it applying erosion using Fast Hydraulic Erosion Simulation and Visaulization on GPU with the thermal weathering from The Synthesis and Rendering of Eroded Fractal Terrains. The red stain is sediment getting pulled down the slopes by rainfall. I'm pretty sure From Dust just implemented Interactive Terrain Modeling Using Hydraulic Erosion, which primarily adds layering where each layer can behave in a distinct manner. I don't think I want layering, but I'm going to see about having sediment pickup take on the colour of the material it's grabbing or cause it to be stained. vvv I don't need the simulation for anything planned, it's just a tool during editing that was easy to implement. But it is realtime, so making it an aspect of gameplay as From Dust does wouldn't be hard. The biggest problem is that XNA's RenderTarget2Ds are not preserved after a device lost event, and you can't copy from a RenderTarget2D to a Texture2D; you can only download it, which is worthless for realtime. So if I were to use it for anything on Windows during gameplay, I would have to use some other framework. It wouldn't matter for the XBox because there is no device lost event there. The Wizard of Oz fucked around with this message at 00:43 on Sep 18, 2011 |
# ? Sep 18, 2011 00:13 |
|
The Wizard of Oz posted:I need a terrain engine in a couple years, so I thought I'd make one now. Great another minecraft clone . But seriously that's pretty cool and didn't realize there's papers on this stuff. What will you need terrain simulation later on for? Looks like there can be some civil engineer applications. I really should stop being a big baby and making the leap into 3d but matrix math scares me (even with a minor in math).
|
# ? Sep 18, 2011 00:31 |
|
Finally finished writing a fancy MARC Record (bibliography information for books) editor. I wrote about a MARC templating system I made a while back to replace some COBOL we still have running. This new system is to actually build the original record before we apply templates to it. It also supports the fancy new RDA version of MARC21. Then we have to be able to take those MARC record files and turn them into an actual bibliography card for our customers who, for some reason, still use actual catalog cards. Sadly I wasn't able to escape the horrors that were COBOL. This part of the app writes out a flat text file that's a bastardized version of the already bastardized MARC21 format. Then runs it through a COBOL app that converts it to a .dat and a .idx file (some kind of magic binary database? gently caress if I know!), which then get passed into another COBOL app which writes out roughly 1500 characters that need aligned and arranged on a 66x24 grid while ignoring a small subset of them. It's really an elegant fix for what would otherwise be a coding horror. I'd love to take out all the COBOL crap but, apparently, no one knows how it works. It was outsourced to some other company in 1982 (based on the COBOL source file) and not enough customers use these drat cards to make it worth figuring it out enough to rewrite. It's still a coding horror, but dammit the code is pretty good all things considered.
|
# ? Sep 22, 2011 18:46 |
|
Frozen-Solid posted:Finally finished writing a fancy MARC Record (bibliography information for books) editor. I've heard horror stories of library and bibliographic software from my wife, who is herself a librarian. Well done for actually creating something that looks good and appears functional
|
# ? Sep 22, 2011 19:01 |
|
csammis posted:I've heard horror stories of library and bibliographic software from my wife, who is herself a librarian. Well done for actually creating something that looks good and appears functional Thanks. One of these days I'm going to try and take everything I've written and make a public version of it. Right now it's extremely tied into our own company database structures and internal processes. I'm making progress though! At least the class structures I wrote that actually read/write/modify MARC21 are open sourced. My boss is awesome and let me release it
|
# ? Sep 22, 2011 19:15 |
|
What's the significance of the colors in the boxes across the middle? At first I thought it was red for bad, green for good. But I'd think Missing RDA = 0 would be good.
|
# ? Sep 23, 2011 08:02 |
|
I'm trying to make an embedded managed runtime with a very thin C++ integration layer and much better performance than interpreted dynamic languages like Lua. It's also thread-safe and lets you serialize pretty much anything (including threads). The compiler supports nearly all of the language features and template types, the low-level IL translator was just finished (pictured), x86 native code output's next up. I can almost taste the pre-alpha release! (Yes I know it optimizes like poo poo, but compared to globally-blocking interpreted languages that hammer the allocator it'll be great) OneEightHundred fucked around with this message at 05:14 on Sep 25, 2011 |
# ? Sep 25, 2011 05:05 |
|
Frozen-Solid posted:At least the class structures I wrote that actually read/write/modify MARC21 are open sourced. My boss is awesome and let me release it
|
# ? Oct 4, 2011 11:42 |
|
DoctorScurvy posted:I am very interested in this, since MARC21 is a complete dog's breakfast, where is it available? http://csharpmarc.net I haven't had time to do much with sample code, but the source download has a bit. Feel free to email or pm me if you need help with it. Frozen Peach fucked around with this message at 13:47 on Oct 4, 2011 |
# ? Oct 4, 2011 12:43 |
|
code:
Opinion Haver fucked around with this message at 00:29 on Oct 5, 2011 |
# ? Oct 5, 2011 00:24 |
|
Dealing with gender and sexuality issues always seems to create a poo poo storm, but I felt like trying to create a new way to gather data about these issues. Specifically I wanted to remove coarse labels and allow for individuals to fit themselves into areas mapped onto these reuleaux triangles. You can play around with the demo here. The next step is running some experiments with this method compared to other more commonly used methods of collecting such data. Though, I'm kinda interested if looking at gender/sexuality in such a way causes people to self reflect more than just select from this drop down.
|
# ? Oct 6, 2011 06:22 |
|
clockwork automaton posted:Dealing with gender and sexuality issues always seems to create a poo poo storm, but I felt like trying to create a new way to gather data about these issues. Specifically I wanted to remove coarse labels and allow for individuals to fit themselves into areas mapped onto these reuleaux triangles. I don't know if this is all the data you want to collect, but you could also ask about biological sex.
|
# ? Oct 6, 2011 08:09 |
|
clockwork automaton posted:You can play around with the demo here. Good idea to let them "decide" what they fall into, but I hit your demo, and found it pretty hard to get to the complete corners. If I am man, I don't want my doctors thinking I am 99.5% man and .5% woman.
|
# ? Oct 6, 2011 09:48 |
|
LOLLERZ posted:I don't know if this is all the data you want to collect, but you could also ask about biological sex. Yeah, these would probably be used in addition to selecting biological sex which is a solved problem. Moey posted:Good idea to let them "decide" what they fall into, but I hit your demo, and found it pretty hard to get to the complete corners. The way that the space gets transformed as long as you are close to this edge you won't be considered part woman. If you inspect the elements on the page as you move the points around you'll notice that some values start getting negative on the extremes.
|
# ? Oct 6, 2011 15:38 |
|
yaoi prophet posted:Who says Haskell can't be used in the real world? There's a bit of boilerplate code and the register form template (and like 20 lines of imports), but other than that that's the entire thing. Made it with digestive-functors-snap-heist and the latest builds on the 0.6 branch of snap, snap-core, snap-server, and heist. Man that's... that's something all right. I love seeing these illogical scenarios, it's like someone building a working car out of lego. Sure it's probably not the best car, but it sure took some ingenuity.
|
# ? Oct 6, 2011 21:19 |
|
clockwork automaton posted:Dealing with gender and sexuality issues always seems to create a poo poo storm. Some people don't like to compartmentalise things and no matter what you do their ~snowflake~ way will never be expressed. besides shouldn't it be an x-y graph with male vs female - your 'all/none' thing doesn't really give a sense of proportion, and then people could put multiple points in because they are special etc etc maybe you should just have a tickybox [ ] heteronormative [ ] other
|
# ? Oct 6, 2011 23:54 |
|
You would not believe the trouble I had refactoring my code to work on binaryqueer platforms.
|
# ? Oct 7, 2011 01:02 |
|
What if I'm binary-curious?
|
# ? Oct 7, 2011 01:21 |
|
tef posted:Some people don't like to compartmentalise things and no matter what you do their ~snowflake~ way will never be expressed. besides shouldn't it be an x-y graph with male vs female - your 'all/none' thing doesn't really give a sense of proportion, and then people could put multiple points in because they are special etc etc Sexuality and gender are phenomenally complicated subjects. No person is or could be all "male", all "female", all "straight", or all "gay", as those words include thousands of highly local cultural concepts that are both contradictory and don't add up to a homogenous body. Those cultural issues will skew even questions like this, as many people would be inclined to polarise their answers even when within their own culture they might enjoy things that would have them labeled effeminate or masculine. Since you can break this down with other questions, you can identify possible skews and explore what produced it.
|
# ? Oct 7, 2011 01:40 |
|
tef posted:
..or consider using the excellent 21-dimensional klein sexual orientation grid! 3 timeframes: Past (entire life up until a year ago) Present (last 12 months) Ideal (what would you like?) Seven indicators: A) Sexual Attraction: To whom are you sexually attracted? B) Sexual Behaviour: With whom have you actually had sex? C) Sexual Fantasies: About whom are your sexual fantasies? D) Emotional preference: Who do you feel more drawn to or close to emotionally? E) Social preference: Which gender do you socialize with? F) Lifestyle preference: In which community do you like to spend your time? In which do you feel most comfortable? G) Self-identification: How do you label or identify yourself? And seven levels, which you could make continuous or w/e: Each of the 21 boxes should contain a value from 1 to 7, categorizing the individual's answers to the questions. For variables A to E the possible answers are: 1 = Other sex only, 2 = Other sex mostly, 3 = Other sex somewhat more, 4 = Both sexes, 5 = Same sex somewhat more, 6 = Same sex mostly, and 7 = Same sex only. For variables F and G these range from 1 = Heterosexual Only to 7 = Homosexual Only. http://en.wikipedia.org/wiki/Klein_Sexual_Orientation_Grid
|
# ? Oct 7, 2011 02:39 |
|
Scaramouche posted:Man that's... that's something all right. I love seeing these illogical scenarios, it's like someone building a working car out of lego. Sure it's probably not the best car, but it sure took some ingenuity. Honestly, as weird as Haskell is, there are some things I sorely wish other languages had, like pattern matching and algebraic data types. But that's for another thread.
|
# ? Oct 7, 2011 02:48 |
|
clockwork automaton posted:Yeah, these would probably be used in addition to selecting biological sex which is a solved problem.
|
# ? Oct 7, 2011 02:52 |
|
in my city we have "art walks" where the arts district hosts a big party on the last friday of every month I'm going to setup a projector screen 10x10ft with this gigantic gently caress off eye that tracks people with a kinect and tells them to stop buying so much poo poo
|
# ? Oct 7, 2011 05:02 |
|
the eye is generated with n-layers of textures generated in glsl shaders the system that generates the iris features is a 3 or 5 state totalistic cellular automata calculated on the GPU It runs live, it's pretty chilling to watch the pupil dilate when it has a good focus on you...
|
# ? Oct 7, 2011 05:03 |
|
it's about consumerism
|
# ? Oct 7, 2011 05:04 |
|
openframeworks++
|
# ? Oct 7, 2011 05:06 |
|
Got a video?
|
# ? Oct 7, 2011 05:49 |
|
taqueso posted:Got a video? Yeah this sounds cool but we need a video to judge its true merit!
|
# ? Oct 7, 2011 09:20 |
|
I'll cut something together this weekend.
|
# ? Oct 7, 2011 13:03 |
|
I'm absolutely terrified of whatever the hell I'm looking at.
|
# ? Oct 7, 2011 20:49 |
|
I want it as my screensaver.
|
# ? Oct 7, 2011 21:09 |
|
Doh004 posted:I'm absolutely terrified of whatever the hell I'm looking at.
|
# ? Oct 8, 2011 00:34 |
|
Weekend project: A little Nancy extension which exposes your methods in C# as async calls in Javascript. I really like what Now.js is doing for RPC on Node and want to clone it, so this weekend's task was a fun proof of concept. All you need to do is subclass my thing and write some public methods: Then in your page, include 'nanio.js' which is emitted by my lib (and JQuery, currently the sole external dependency since I'm not currently interested in rewriting ajax handling) And then your calls work! Yay! Todo: rip off the rest of Now.js's features
|
# ? Oct 11, 2011 00:05 |
|
Dolex posted:As long as you do not shop at Walmart you have nothing to fear from EYE. You need to implement non-human pupils. Like goat's pupils with their rounded rectangle, or cat's pupils that narrow to a slit. That would make the whole experience significantly more otherworldly.
|
# ? Oct 11, 2011 01:40 |
|
Factor Mystic posted:Todo: rip off the rest of Now.js's features
|
# ? Oct 11, 2011 10:18 |
|
Doesn't ASP.NET's WebMethod feature already do that?
|
# ? Oct 11, 2011 18:29 |
|
OneEightHundred posted:Doesn't ASP.NET's WebMethod feature already do that?
|
# ? Oct 12, 2011 12:14 |
|
|
# ? May 9, 2024 09:39 |
|
Sagacity posted:That's limited to ASP.NET Xml Web Services or something and is only designed to handle stuff like an Ajax call Guess it doesn't do push though.
|
# ? Oct 12, 2011 15:25 |