Boothand

I have some demos I want to capture while controlling parameters such as timescale with binds on the go. It doesn't seem to register these clicks. In any case, if it did, it would still be hard to control, because when I start capturing (capture avi 60 fileName), it captures it at something like 1.5x the speed, which makes it hard to control. The movie looks fine though, but I would like to choose the non-gradual way of doing timescale.

Haven't done any more to the UVs (which is what will determine whether it looks good or not), but guess I'll have to know if it's likely to be used, otherwise I may not find time for it ^^. In any case, it's probably just the head and not a full character, unless I stumble upon some massive inspiration gust. Seems like @@Mandalorian is working on a Bossk project though, but I didn't quite understand what was modified and what was re-made on that thing. If this head model (while disregarding the texture I rendered) is favorable, it could be used there (the vertex count on the neck should match), otherwise I'm not likely to do anything to it in a while.

No rend2 features in itself require you to recompile (it's only the renderer), but I don't think it's quite finished when it comes to dynamic lights yet. To really achieve dynamic lighting that looks good, I found that you could recompile the map and bypass the q3map2 lightmap (_cs 0 and _rs 0 on worldspawn). This way, the pre-compiled shadows won't exist and won't interfere with rend2's dynamic lighting. However, it would obviously be rendered incompatible with people running the vanilla renderer, until Xycaleth potentially finds a way to bypass the lightmap in rend2 itself. The other side effect is that cubemaps won't yet work properly without a lightmap present. I'm not sure if more has been done to the parallax mapping (or if it's at all possible to fix what bothered me), but at the time I played around with it, everything kind of broke at the seams. It looked good, but two different surfaces (in a map) that didn't line up alignment-wise and angle-wise, would break the illusion and you could see into the parallax from each surface, *under* the other. If it's just an unfortunate consequence, it still looks very nice when used in the right place.

If you don't wanna use light emitting shaders, you'd probably have to initiate the position of the pallets to somewhere with even/optimal lighting and let them compile there, then set their position where they belong once the map is run. It might be a bit of a hassle though, depending on your setup.

Have you tried using Psyk0's autolinker script?

Donnie Darko should be good for Halloween, though it's not really a horror movie.

Thanks for those links, @@Didz! Very useful. What annoys me the most is things like, it seems to be a whole lot of trouble to get the accurate mouse position, and if you change the layout of things in the HTML for example, the mouse will be offset once again, after using ugly code to manually align it by eye-measure. Same goes for general behavior of objects when using a smaller window (which in my case also resizes the canvas to fit). As a result, I've had to divide everything by the width of the potentially varying canvas to get the same behavior on different monitors or if people happen to use a resized browser window. I guess somewhere along the way, that went wrong though, and now the game goes crazy if I'm playing it in a resized browser window

It's expected that everyone's gonna have different results. For me, the ball always lags through the opponent at higher speeds. A friend of mine says it never bugged through, and he even had trouble winning.. Yeah, it's not very pleasant (mouse coordinates... brrr!). Using Javascript because we were supposed to, and we've just started the programming course. Hmm. What browser are you in? If you have Chrome, that should work... Yeah, that's intended. It does an idle thing while waiting, and when it *sees* the ball half way back, it has a max speed that it tries to stop the ball with. To avoid it running out of bounce, use boostfields©™ to make it go faster (also the only way to beat the computer). Edit: Haha, even though I've tried supporting different window sizes, playing the game without the browser in full size makes the AI move wayyy faster and messes up.. stuff. That's also a major facepalm when it comes to games in JavaScript.

Try it out! Do note, the AI is a very bad loser. http://tiny.cc/winandforswinn In any case, has anyone in here ever done Javascript? The game is kind of broken by the bad collision detection once the ball speeds up. If you look at my code (right click the page and 'View page source'), do you see anything that could be done to avoid it skipping past the collision? Edit: Woops, forgot all the //comments are in norwegian, if someone tries to navigate the code..

Would be nice to have multiple desktops on Windows*

I think he looks a bit like a mixture between Alice Cooper and Libby from Lost. That's not bad though. Love the overall cheesiness, like the enthusiastic laugh mixed perfectly into "Great!". Anyway, would be very nice to have multiple desktops.

Cool drawings! But I would personally prefer a bit more contrast between the buttons and the background.

I did make a time machine as you requested, but it's a bit buggy. Ended up in dinosaur land. Can't get back. Think I forgot an if-statement. Need @@Xycaleth to fix the code. BRB Pterodactyl.

For how long have you had it? That's quite close to 2 months of game time!

Well, you could make a caulk brush into a func_door and give that entity a model to display in its stead (model2), like with func_statics, if you want to use a model. And I think you might need an md3 conversion for it to move (or to display while moving to be accurate), but I haven't tried with doors, so I'm not sure.

Well, an .ase model won't be solid, so you'd have to make custom collision. Like I showed you once, you can use the method where you make a static caulk brush and specify a model to display in its stead, if this is what you mean (in which case, you'd need md3 for a moving object IIRC) I think this is a stunning example of detail in GtkRadiant: http://jkhub.org/topic/1120-moonbase-labs/?p=26440 I suppose it was made in a large scale, then made to .ase and shrinked, otherwise correct me! @@Szico VII

@@DT85, you could try to make the offending areas into func_groups, and add a _lightmapscale key with a value lower than 1 (the default) to them. Also, have you tried different settings on the light entity casting the shadows?

Hmm, can recall a few times it has happened when I only compiled a -meta without removing the old one compiled with -vis and -light etc. I'm pretty sure it has replaced them as well before though, so not sure what's causing it...

The ceilings look kind of overly lit, from these screenshots, primarily the two first ones. I wouldn't expect bounce light to be that strong. Where is the light source(s) in screenshot #3?

In worst case scenario you could probably just as well rip sounds from Master Betty from Kung Pow Sounds just about right.

Hmm, wouldn't the lightmap being baked into the level prevent shadows from objects from being cast onto the model?

Here's what I have so far, but not looking all as good as I had hoped, at this moment. The texture has some way to go still. Might change some things around in the UVs too. Haven't implemented that much scaliness yet, but didn't see so many in the face on most Bossk references. The geometry should be fine-ish though. The vertex count of the neck matches the original, and total 1,108 polygons after triangulation (561 vertices). Sketchfab link. Texture link. Reference: http://www.loresdelsith.net/universo/cr/bin/bossk02.jpg Edit: I realize I should make the face take up a lot more space in the UVs. Also for some reason I almost always end up with some unsmooth texture. The colors don't fade well into each other. I think it's because of Zbrush somehow.

One of my projects here is based directly on Bossk. http://jkhub.org/topic/4493-trandoshan-variant/?hl=trandoshan If there's interest, maybe I'll do some more on it this week.

I didn't technically have a thirdparty antivirus when I caught the flu. Uninstalled avast some time ago after getting really annoyed at it consistently blocking an important dll file during installing Maya. But have installed it again now, done a long booted scan and also used the Malwarebytes link and got rid of something at least, thanks Circa! So it *feels* like it's gone at least.