Normal distribution is great, but look at beta-binomial distribution. You can control the distribution of values better then just by shifting normal distribution.
Thanks for the suggestion, I was actually planning on remaking this video and implementing normal distributions in a different and more appealing way, but I’ll look into that too, and maybe add that into the video.
first B) e: also was confused throughout the video but still fun to watch. any way you could make a video about OpenCL/parallel GPU computing on non-nvidia GPUs? xD
Thanks for watching! So the main use for this method is to make things in your games less uniform. If we are doing heights and want to model it to be realistic, we can make it follow a normal distribution to have a majority of the characters close to a certain height and have less be small or tall. This is a lot more realistic than having the same chance of any height in a range of values but is definitely more confusing lol. Think of the mean as the target value and standard deviation as how much the numbers can differ from the mean. And regarding making a video on parallel gpu computing on non-nvidia gpus, the answer is probably no. I made a video a while back on compute shaders which is kinda like parallel gpu computing, I think... I may revisit that stuff in the future but when I was doing it, I had it in my mind that I was just leaning about this thing and not that I was going to spend lots of time on that when I could be learning other aspects of game dev. I also have an nvidia gpu and would have no clue how to make things work on other gpus if they worked differently in the first place. If I were to make another video remotely related to this it would probably be about threading, but I have no actual plans to do it.
this method is cool when u want values to be pretty similar but with little changes here and there but u still need the Random Function made by Unity to get the input probability, will try to make a Random function (pseudoRandom) using a seed for example ? (cuz tbh when i clicked in this vid i thought u we're gonna do that since i know very little about normal distributions)
Thanks for watching! My rationale is why would you need to make your own random number generator in the first place when there is unity’s random number and system.random for not unity and both work great.
@@WillHessGameDev Exploring how things work has always fascinated me (and ik am not the only one), since u can change things around to suit ur needs even more when u know how they work
“Unique method of something I learned in school that I can apply to game dev that I realized wasn’t used by many other game devs and has no tutorial on TH-cam”, what it actually is, 1) does not fit in the title and 2) would not be good for views lol. Sorry if you felt clickbaited with the “new” but it’s a method that I came up with myself (to the best of my knowledge) and literally can improve your games in certain circumstances.
Normal distribution is great, but look at beta-binomial distribution. You can control the distribution of values better then just by shifting normal distribution.
Thanks for the suggestion, I was actually planning on remaking this video and implementing normal distributions in a different and more appealing way, but I’ll look into that too, and maybe add that into the video.
first B)
e: also was confused throughout the video but still fun to watch. any way you could make a video about OpenCL/parallel GPU computing on non-nvidia GPUs? xD
Thanks for watching!
So the main use for this method is to make things in your games less uniform. If we are doing heights and want to model it to be realistic, we can make it follow a normal distribution to have a majority of the characters close to a certain height and have less be small or tall. This is a lot more realistic than having the same chance of any height in a range of values but is definitely more confusing lol. Think of the mean as the target value and standard deviation as how much the numbers can differ from the mean.
And regarding making a video on parallel gpu computing on non-nvidia gpus, the answer is probably no. I made a video a while back on compute shaders which is kinda like parallel gpu computing, I think... I may revisit that stuff in the future but when I was doing it, I had it in my mind that I was just leaning about this thing and not that I was going to spend lots of time on that when I could be learning other aspects of game dev. I also have an nvidia gpu and would have no clue how to make things work on other gpus if they worked differently in the first place.
If I were to make another video remotely related to this it would probably be about threading, but I have no actual plans to do it.
this method is cool when u want values to be pretty similar but with little changes here and there
but u still need the Random Function made by Unity to get the input probability, will try to make a Random function (pseudoRandom) using a seed for example ?
(cuz tbh when i clicked in this vid i thought u we're gonna do that since i know very little about normal distributions)
Thanks for watching! My rationale is why would you need to make your own random number generator in the first place when there is unity’s random number and system.random for not unity and both work great.
@@WillHessGameDev Exploring how things work has always fascinated me (and ik am not the only one), since u can change things around to suit ur needs even more when u know how they work
Very true
new?
sum two dice-rolls, you get normal distribution
one dice roll is sample from uniform distribution
knowledge is unnecessary weight to carry, to come out of the mouth after, yes it a face fart
intuitive is so much better, you have skill, you are not a robot, or should not be
“Unique method of something I learned in school that I can apply to game dev that I realized wasn’t used by many other game devs and has no tutorial on TH-cam”, what it actually is, 1) does not fit in the title and 2) would not be good for views lol. Sorry if you felt clickbaited with the “new” but it’s a method that I came up with myself (to the best of my knowledge) and literally can improve your games in certain circumstances.