Oh RIT
all4pocketlint: good afternoon sir
DarkLordSprite: Hello
DarkLordSprite: how are you?
all4pocketlint: im quite fine. and yourself?
DarkLordSprite: doing good, just organizing my diffEq notes, numbering the pages and making a table of contents
all4pocketlint: sounds riviting
DarkLordSprite: i average just under 5 pages per class of DiffEq
DarkLordSprite: thats just over 2 pages an hour
DarkLordSprite: :P
DarkLordSprite: some days like today are like 9 or 10 pages
all4pocketlint: heh i take about two typed pages in this class im in now every class
all4pocketlint: so what exactlyt is DiffEq?
DarkLordSprite: Differential Equations
DarkLordSprite: it is kind of like calculus, except that it isn't
DarkLordSprite: it is pretty much teaching us even though it isn't saying it, that everything can be defined by a solvable equation
all4pocketlint: hm.
all4pocketlint: and hows that
DarkLordSprite: because you can have equations of different derivatives, with coefficients of those equations being equations, all in an equation
DarkLordSprite: and there is a theorem that says if there is a differential equation there is a solution to it
all4pocketlint: right
all4pocketlint: that makes a bit of sense. although to be honest i had to read it 3 times
DarkLordSprite: lol
DarkLordSprite: well you have an equation called y
DarkLordSprite: and you signify what derivative of y it is by tick marks in the superscript, like first derivative is y' second is y''
DarkLordSprite: so a second order differential equation is like, ay'' + by' + cy = d, where a, b, c and d are constants
all4pocketlint: mhm
DarkLordSprite: but you can also have y'' + p( x )y' + q( x )y = g( x ) where p( x ), q( x ) and g( x ) are functions
DarkLordSprite: and you can have more than just two derivatives
DarkLordSprite: so you can pretty much make a function to describe anything
all4pocketlint: anything?
all4pocketlint: can a function be made to describe how great pie is?
DarkLordSprite: sure
DarkLordSprite: if you have the right inputs
all4pocketlint: hahaha
all4pocketlint: :-)
DarkLordSprite: i mean, it would be more variant with process variation
DarkLordSprite: but i could use statistics to come up with a model equation
DarkLordSprite: it won't be a perfect predictor on how good the pie is, but it will be close
DarkLordSprite: if taste is a factor, and we put a number to taste, then run experiments using a DOE table to test all of the different factors that go into making pie
DarkLordSprite: i can find out which factors are insignificant to the taste and which are significant and create a model equation to predict if you have this factors inputted, how good the pie will taste
all4pocketlint: really now
DarkLordSprite: yeah
DarkLordSprite: there are a lot of factors, it would be a large experiment, probably multiple experiments
all4pocketlint: how would one conduct an experiment like that?
DarkLordSprite: Well i'd have to know the inputs
DarkLordSprite: like, baking time
DarkLordSprite: amount of different ingredients
DarkLordSprite: and then i would set up a DOE table, which is an outline of your experiment, that tells you which factors to change and which to hold the same
DarkLordSprite: i would run a Fractional Factorial design becuase there are so many inputs
DarkLordSprite: then using statistical software i would get parameter estimates, and create my equation based on those
all4pocketlint: mhm
DarkLordSprite: needless to say, there would be a hell of a lot of pie to eat
all4pocketlint: : )
DarkLordSprite: it would be pricey
DarkLordSprite: but through the experiment, we could safely say that we could make the best pie ever
DarkLordSprite: because we optimize our process
all4pocketlint: indeed.
all4pocketlint: one could always sell the unneeded test pies.
DarkLordSprite: they wouldn't be unneeded
DarkLordSprite: we need to taste them
DarkLordSprite: because we need responses
all4pocketlint: but would we taste the entire pie?
DarkLordSprite: we could have some random people eat the pies
DarkLordSprite: and have them rate it
DarkLordSprite: on a continuous scale from 0-10
DarkLordSprite: or rather
DarkLordSprite: any real number lol
all4pocketlint: hahahahaaaa
DarkLordSprite: so they could say that this pie ranks pi
all4pocketlint: hehe
DarkLordSprite: there are a lot of factors
DarkLordSprite: i was just thinking how this could be done
DarkLordSprite: and i decided that it couldn't be done on my budget
all4pocketlint: of course nto
all4pocketlint: not*
all4pocketlint: it would be an expensive thing to do
DarkLordSprite: it would be posible to do the experiment if we hold everything else constant and only change 2 or 3 factors
DarkLordSprite: 2 factors we'd make 4 pies, 3 factors we'd make 8 pies
DarkLordSprite: 4 = 16,
DarkLordSprite: 5 = 32
DarkLordSprite: see how fast it goes up?
DarkLordSprite: lol
DarkLordSprite: how many ingrediants go into pies?
DarkLordSprite: we could also narrow it down to just the filling
DarkLordSprite: and then just the crust
DarkLordSprite: and if we have more than 3 factors we cna do a fractional factorial design
DarkLordSprite: so 4 factors could be done with only 16 pies
DarkLordSprite: but we'll need 3 center point pies
DarkLordSprite: so we'll actually need 3 more pies for each of those numbers
all4pocketlint: right
DarkLordSprite: so 2 = 7 pies, 3 = 11 pies, 4 = 11 pies, 5 = 19 pies
DarkLordSprite: if we run fractional factorials for over 4 factors
DarkLordSprite: and we won't want to do experiments with more than 4 or 5 factors at a time
DarkLordSprite: as you can see the numbers quickly get out of hand
DarkLordSprite: 7 pies is a bit much
all4pocketlint: right. so how many pies are a minumum?
DarkLordSprite: 7
DarkLordSprite: well, it could be done with 4 pies, buuut we won't get good information
DarkLordSprite: and we'll only learn about two factors
all4pocketlint: right and that wont suffice for the best pie ever
DarkLordSprite: we would have to do multiple experiments with overlap so we can see the interaction between factors
DarkLordSprite: because with the 7 pies we'd learn about the factors and their interactions with each other
DarkLordSprite: we need interactions, because they are normally significant
DarkLordSprite: 3 effect interactions normally aren't significant
DarkLordSprite: but 2 effect interactions are
all4pocketlint: i didnt know that
all4pocketlint: why?
DarkLordSprite: because they can effect each other
DarkLordSprite: but 3 effect interactions normally aren't significant
all4pocketlint: hm
DarkLordSprite: it just statistically isn't significant normally
DarkLordSprite: sometimes they are
DarkLordSprite: but not often
DarkLordSprite: and higher number effect interactions are even less likely
DarkLordSprite: Hello
DarkLordSprite: how are you?
all4pocketlint: im quite fine. and yourself?
DarkLordSprite: doing good, just organizing my diffEq notes, numbering the pages and making a table of contents
all4pocketlint: sounds riviting
DarkLordSprite: i average just under 5 pages per class of DiffEq
DarkLordSprite: thats just over 2 pages an hour
DarkLordSprite: :P
DarkLordSprite: some days like today are like 9 or 10 pages
all4pocketlint: heh i take about two typed pages in this class im in now every class
all4pocketlint: so what exactlyt is DiffEq?
DarkLordSprite: Differential Equations
DarkLordSprite: it is kind of like calculus, except that it isn't
DarkLordSprite: it is pretty much teaching us even though it isn't saying it, that everything can be defined by a solvable equation
all4pocketlint: hm.
all4pocketlint: and hows that
DarkLordSprite: because you can have equations of different derivatives, with coefficients of those equations being equations, all in an equation
DarkLordSprite: and there is a theorem that says if there is a differential equation there is a solution to it
all4pocketlint: right
all4pocketlint: that makes a bit of sense. although to be honest i had to read it 3 times
DarkLordSprite: lol
DarkLordSprite: well you have an equation called y
DarkLordSprite: and you signify what derivative of y it is by tick marks in the superscript, like first derivative is y' second is y''
DarkLordSprite: so a second order differential equation is like, ay'' + by' + cy = d, where a, b, c and d are constants
all4pocketlint: mhm
DarkLordSprite: but you can also have y'' + p( x )y' + q( x )y = g( x ) where p( x ), q( x ) and g( x ) are functions
DarkLordSprite: and you can have more than just two derivatives
DarkLordSprite: so you can pretty much make a function to describe anything
all4pocketlint: anything?
all4pocketlint: can a function be made to describe how great pie is?
DarkLordSprite: sure
DarkLordSprite: if you have the right inputs
all4pocketlint: hahaha
all4pocketlint: :-)
DarkLordSprite: i mean, it would be more variant with process variation
DarkLordSprite: but i could use statistics to come up with a model equation
DarkLordSprite: it won't be a perfect predictor on how good the pie is, but it will be close
DarkLordSprite: if taste is a factor, and we put a number to taste, then run experiments using a DOE table to test all of the different factors that go into making pie
DarkLordSprite: i can find out which factors are insignificant to the taste and which are significant and create a model equation to predict if you have this factors inputted, how good the pie will taste
all4pocketlint: really now
DarkLordSprite: yeah
DarkLordSprite: there are a lot of factors, it would be a large experiment, probably multiple experiments
all4pocketlint: how would one conduct an experiment like that?
DarkLordSprite: Well i'd have to know the inputs
DarkLordSprite: like, baking time
DarkLordSprite: amount of different ingredients
DarkLordSprite: and then i would set up a DOE table, which is an outline of your experiment, that tells you which factors to change and which to hold the same
DarkLordSprite: i would run a Fractional Factorial design becuase there are so many inputs
DarkLordSprite: then using statistical software i would get parameter estimates, and create my equation based on those
all4pocketlint: mhm
DarkLordSprite: needless to say, there would be a hell of a lot of pie to eat
all4pocketlint: : )
DarkLordSprite: it would be pricey
DarkLordSprite: but through the experiment, we could safely say that we could make the best pie ever
DarkLordSprite: because we optimize our process
all4pocketlint: indeed.
all4pocketlint: one could always sell the unneeded test pies.
DarkLordSprite: they wouldn't be unneeded
DarkLordSprite: we need to taste them
DarkLordSprite: because we need responses
all4pocketlint: but would we taste the entire pie?
DarkLordSprite: we could have some random people eat the pies
DarkLordSprite: and have them rate it
DarkLordSprite: on a continuous scale from 0-10
DarkLordSprite: or rather
DarkLordSprite: any real number lol
all4pocketlint: hahahahaaaa
DarkLordSprite: so they could say that this pie ranks pi
all4pocketlint: hehe
DarkLordSprite: there are a lot of factors
DarkLordSprite: i was just thinking how this could be done
DarkLordSprite: and i decided that it couldn't be done on my budget
all4pocketlint: of course nto
all4pocketlint: not*
all4pocketlint: it would be an expensive thing to do
DarkLordSprite: it would be posible to do the experiment if we hold everything else constant and only change 2 or 3 factors
DarkLordSprite: 2 factors we'd make 4 pies, 3 factors we'd make 8 pies
DarkLordSprite: 4 = 16,
DarkLordSprite: 5 = 32
DarkLordSprite: see how fast it goes up?
DarkLordSprite: lol
DarkLordSprite: how many ingrediants go into pies?
DarkLordSprite: we could also narrow it down to just the filling
DarkLordSprite: and then just the crust
DarkLordSprite: and if we have more than 3 factors we cna do a fractional factorial design
DarkLordSprite: so 4 factors could be done with only 16 pies
DarkLordSprite: but we'll need 3 center point pies
DarkLordSprite: so we'll actually need 3 more pies for each of those numbers
all4pocketlint: right
DarkLordSprite: so 2 = 7 pies, 3 = 11 pies, 4 = 11 pies, 5 = 19 pies
DarkLordSprite: if we run fractional factorials for over 4 factors
DarkLordSprite: and we won't want to do experiments with more than 4 or 5 factors at a time
DarkLordSprite: as you can see the numbers quickly get out of hand
DarkLordSprite: 7 pies is a bit much
all4pocketlint: right. so how many pies are a minumum?
DarkLordSprite: 7
DarkLordSprite: well, it could be done with 4 pies, buuut we won't get good information
DarkLordSprite: and we'll only learn about two factors
all4pocketlint: right and that wont suffice for the best pie ever
DarkLordSprite: we would have to do multiple experiments with overlap so we can see the interaction between factors
DarkLordSprite: because with the 7 pies we'd learn about the factors and their interactions with each other
DarkLordSprite: we need interactions, because they are normally significant
DarkLordSprite: 3 effect interactions normally aren't significant
DarkLordSprite: but 2 effect interactions are
all4pocketlint: i didnt know that
all4pocketlint: why?
DarkLordSprite: because they can effect each other
DarkLordSprite: but 3 effect interactions normally aren't significant
all4pocketlint: hm
DarkLordSprite: it just statistically isn't significant normally
DarkLordSprite: sometimes they are
DarkLordSprite: but not often
DarkLordSprite: and higher number effect interactions are even less likely