[QUOTE=FalconKrunch;47446824]Yeah! Height based fog, awesome!
[t]http://i.imgur.com/gNDxTuv.png[/t]
Too bad it's a really cheap and shitty way of doing it
[t]http://i.imgur.com/InlYBGw.jpg[/t][/QUOTE]
If you want to make it work while in or below the fog layer, you can integrate over the fog column (integrate the fog density formula by hand, then plug that into the shader) to easily find the effective fog between two heights by subtracting the results. Then you multiply with the slope coefficient towards where you're looking at to get the total fog stack, which is the exponent of the exponential falloff function giving you the transparency factor.
This should give you (not quite but relatively) correct results from any possible view. It doesn't account at all for varied scattering and attenuation of sunlight though, which would have to be added with additional integrals. (The latter definitely shares a lot of calculations with the default fog gradient though, so you can make it really cheap.)
If you post your "fog density by height" I'll give you an example of the process for fog transparency.
[editline]3rd April 2015[/editline]
The only issue is that this is undefined/really imprecise if you look horizontally at something... You'd probably have to branch or smudge there if you want to do it with this method.
[I]In theory[/I] it should be possible to remove that issue though.
(I'm not good with discrete maths so I'm not sure how to do it.
Physics is more or less continuous everywhere so you can just "fill in" discontinuities like that, and if it's not it's usually because the model isn't good enough or the examined problem isn't physical in the first place.)
[QUOTE=Sombrero;47447184]Watcom finally compiled my mixed code into a correct flat binary.
Let's try it ou-
-vid-
:v:
Oh well, at least it somewhat functions. I'm not the best with pointers.[/QUOTE]
Hey, still better than this
[thumb]https://raw.githubusercontent.com/cartman300/CarpOS/master/screenshots/garbage.png[/thumb]
Also why are you nude in the video. :v:
[QUOTE=cartman300;47447327]Hey, still better than this
[thumb]https://raw.githubusercontent.com/cartman300/CarpOS/master/screenshots/garbage.png[/thumb]
Also why are you nude in the video. :v:[/QUOTE]
Because it's hot af in here.
the question is why not be nude in all of your videos? have you seen the ones i post? deceptively, i never wear pants when i do those
#themoreyouknow
[QUOTE=Sombrero;47447184]Watcom finally compiled my mixed code into a correct flat binary.
Let's try it ou-
[video=youtube;LoZMgHQ7N8A]http://www.youtube.com/watch?v=LoZMgHQ7N8A[/video]
:v:
Oh well, at least it somewhat functions. I'm not the best with pointers.[/QUOTE]
oh gosh. the fact that you were in the nude and the way you said "oops" reminded me of those alantutorial videos.
[video=youtube;KH4NrUxcsYs]http://www.youtube.com/watch?v=KH4NrUxcsYs[/video]
[QUOTE=Tamschi;47446354]Which ones, precisely?[/QUOTE]
Yeah I can't remember it could've been because of an unrelated bug. It [i]shouldn't[/i] matter because when the Activity is destroyed the broadcastReceiver is unregistered from the BroadcastManager anyways so you should never have an OnReceive call where the BroadcastReceiver's reference to the Activity is null.
I'm trying it again tonight since I need to be able to send the track title back to the UI.
[editline]3rd April 2015[/editline]
The metrics you get on the Google Play developer console is pretty cool. It would be great if there was a way to tell how many users are [i]actively[/i] using the app, I'd like to use Google Analytics but I think that'll put me over Xamarin's app size limit pretty quickly.
That said, I think this is a decent indicator, especially since I had to add a wake-lock (for WiFi) permission which requires someone to explicitly accept the app update:
[img]http://i.imgur.com/YL3Y5hK.png[/img]
jesus christ those final alantutorial videos
[QUOTE=andrewmcwatters;47448542]jesus christ those final alantutorial videos[/QUOTE]
Kinda reminds me of How To Basic...
[img]http://i.imgur.com/tHb8ajA.png[/img]
send help
[QUOTE=Berkin;47448710][IMG]http://i.imgur.com/tHb8ajA.png[/IMG]
send help[/QUOTE]
That's nothing, I once got over 900, plus another 400 or so lines that the compiler couldn't parse due to the error count.
Today I finally managed to finish loading my custom model's file type in, but this time it now supports loading custom animations and keyframes on a node-per node basis.
Now all I have to do is make a proper handler to manage playback, and load it into my animation maker widget.
[QUOTE=Berkin;47448710][img]http://i.imgur.com/tHb8ajA.png[/img]
send help[/QUOTE]
Perhaps it's just me, but does anyone else immediately consult the Output window rather than the Error List? Since there's a tendency for a single mistake to cause a massive chain of errors, I find it's quicker to look at the build Output log and fix the error at the top first.
[QUOTE=WillKirkby;47449047]Perhaps it's just me, but does anyone else immediately consult the Output window rather than the Error List? Since there's a tendency for a single mistake to cause a massive chain of errors, I find it's quicker to look at the build Output log and fix the error at the top first.[/QUOTE]
You don't have to build (explicitly) to get the errors.
[QUOTE=Berkin;47448710][img]http://i.imgur.com/tHb8ajA.png[/img]
send help[/QUOTE]
I'm sorry for [URL="https://github.com/TheBerkin/Rant/issues/47"]reporting[/URL] that issue.
lol jk pls fix it
[QUOTE=Simspelaaja;47449060]I'm sorry for [URL="https://github.com/TheBerkin/Rant/issues/47"]reporting[/URL] that issue.
lol jk pls fix it[/QUOTE]
[img]http://i.imgur.com/gKiDGMb.png[/img]
More projects need this
[QUOTE=Tamschi;47447246]If you want to make it work while in or below the fog layer, you can integrate over the fog column (integrate the fog density formula by hand, then plug that into the shader) to easily find the effective fog between two heights by subtracting the results. Then you multiply with the slope coefficient towards where you're looking at to get the total fog stack, which is the exponent of the exponential falloff function giving you the transparency factor.
This should give you (not quite but relatively) correct results from any possible view. It doesn't account at all for varied scattering and attenuation of sunlight though, which would have to be added with additional integrals. (The latter definitely shares a lot of calculations with the default fog gradient though, so you can make it really cheap.)
If you post your "fog density by height" I'll give you an example of the process for fog transparency.
[editline]3rd April 2015[/editline]
The only issue is that this is undefined/really imprecise if you look horizontally at something... You'd probably have to branch or smudge there if you want to do it with this method.
[I]In theory[/I] it should be possible to remove that issue though.
(I'm not good with discrete maths so I'm not sure how to do it.
Physics is more or less continuous everywhere so you can just "fill in" discontinuities like that, and if it's not it's usually because the model isn't good enough or the examined problem isn't physical in the first place.)[/QUOTE]
My "fog density by height" is really fucking terrible
[url]http://pastebin.com/twDMgvLU[/url]
I just lerp between my finalcolor and the fog, based on the height of the fragment
I made an icon for Pushjet. Is it any good or should I change it?
[img]http://a.pomf.se/ebchfw.svg[/img]
[QUOTE=Mega1mpact;47449195]I made an icon for Pushjet. Is it any good or should I change it?
[img]http://a.pomf.se/ebchfw.svg[/img][/QUOTE]
Downscale it to something like 64x64 and look at it from that perspective
[QUOTE=Fourier;47449219]Downscale it to something like 64x64 and look at it from that perspective[/QUOTE]
I've done that whilest I was doing the gradient and it's currently the only reason why I'm doubting using this icon. I could make the nose a bit thicker or something.
[img]http://i.imgur.com/Wf8wwFf.png[/img]
[QUOTE=ruarai;47449256]If you turn it a little it kind of looks like an arrow. That's a gimmick you must exploit.[/QUOTE]
It's an svg so I can animate it. Make it look like the jet is flying away out of frame and appearing back into frame a second later.
I'm gonna make that now. Sounds too cool not to do.
[QUOTE=Karmah;47448771]That's nothing, I once got over 900, plus another 400 or so lines that the compiler couldn't parse due to the error count.
[/QUOTE]
Yeah. Just remove a single brace in a header and you will be swimming in tons of errors.
[QUOTE=Mega1mpact;47449229]I've done that whilest I was doing the gradient and it's currently the only reason why I'm doubting using this icon. I could make the nose a bit thicker or something.
[img]http://i.imgur.com/Wf8wwFf.png[/img][/QUOTE]
i think the tailfins or whatever could be a bit larger so it's more obvious what it is
[editline]3rd April 2015[/editline]
also that thruster block thingy isn't visible at all so make it larger too or remove it
[QUOTE=DarKSunrise;47449330]i think the tailfins or whatever could be a bit larger so it's more obvious what it is
[editline]3rd April 2015[/editline]
also that thruster block thingy isn't visible at all so make it larger too or remove it[/QUOTE]
I'm thinking of hiding the thruster if it's smaller then 80px or something (if you can do that with SVG). I kind of like the thruster. It adds a little more detail that show it's a jet.
[img]http://i.imgur.com/SNiu6pS.png[/img]
[t]http://a.pomf.se/paddbd.svg[/t]
How long have you done pixel art by the way? I'm so jealous.
im only a programmer, second game's art is [URL]http://facepunch.com/member.php?u=249719[/URL] except for the akihiko portrait which is from persona
Oh, someone else does the art? I always figured it was you.
[editline]3rd April 2015[/editline]
Right, cool.
[QUOTE=Mega1mpact;47449367]I'm thinking of hiding the thruster if it's smaller then 80px or something (if you can do that with SVG). I kind of like the thruster. It adds a little more detail that show it's a jet.
[img]http://i.imgur.com/SNiu6pS.png[/img]
[t]http://a.pomf.se/paddbd.svg[/t][/QUOTE]
only critique is that its very weighted, if i were doing it i would create a mood board of possible colours, arrangements etc to and decide from there instead of trying to make a one off final icon
I'd maybe increase the padding a little bit.
hmm now i use a logarithmic depth and i output the depth to a new framebuffer's red channel
now for SSAO i need to make it linear again :(
but on the good side: i have no z fighting for sub cm sized objects on a galaxy sized distance xD
and on the bad side: i lost fast-z which i could compensate somehow but that one isn't a big performance impact anyways...
[QUOTE=Mega1mpact;47449169][img]http://i.imgur.com/gKiDGMb.png[/img]
More projects need this[/QUOTE]
That's not even the best one:
[img]http://i.imgur.com/ogpZa5b.png[/img]
It's just a matter of waiting for the right issue to use them on.
Sorry, you need to Log In to post a reply to this thread.
