Shake

On 3/30/07, Deke Kincaid wrote:
>
> http://www.jnack.com/adobe/photoshop/fountain/
> http://av.adobe.com/russellbrown/CS3StackModeSM.mov
>
> -deke

---------- Original Message ----------------------------------
From: "Randy Little" Reply-To:
Date: Fri, 30 Mar 2007 14:17:38 -0500

>that rocks
>
>On 3/30/07, Deke Kincaid wrote:
>>
>> http://www.jnack.com/adobe/photoshop/fountain/
>> http://av.adobe.com/russellbrown/CS3StackModeSM.mov
>>
>> -deke

Deke Kincaid wrote:
> http://www.jnack.com/adobe/photoshop/fountain/
> http://av.adobe.com/russellbrown/CS3StackModeSM.mov

On Mar 30, 2007, at 2:02 PM, Deke Kincaid wrote:

> http://www.jnack.com/adobe/photoshop/fountain/
> http://av.adobe.com/russellbrown/CS3StackModeSM.mov
>
> -deke

On Mar 30, 2007, at 12:37 PM, Lewis Saunders wrote:

> Deke Kincaid wrote:
>> http://www.jnack.com/adobe/photoshop/fountain/
>> http://av.adobe.com/russellbrown/CS3StackModeSM.mov
>
> Neat. So does anyone have a plugin which does a median across several
> image inputs? Not possible in a macro, right?
>
> L

One fairly straightforward approach might be to (auto)align the plates,
difference matte each plate from the others keeping only what's common with
most (maybe add your resulting difference mattes for each plate together and
then threshold the resulting matte to generate the alpha for that layer),
and layer them with the other plates. You'll want at least one plate that
you're arbitrarily declaring to be the 'background' to fill in for things
like leaf flutter (or the water in this example) that's not going to be the
same between any of the plates.

It's definitely something you'd want to build into a re-usable,
re-purposable piece of script rather than a make-it-everytime kind of
thing. On a case-by-case basis, it's probably easier just to paint yourself
a cleanplate.

The biggest limitation in Shake is going to be the lack of an autoalignment
tool that'll accept a half dozen or more inputs - and the lack of
auto-expanding inputs on macros a la the "Layer" node.

I'm interested in seeing a demo of the photoshop stack mode with more
divergent plates: how does it handle a series where the camera moved more
than in the example, for instance.

On 3/30/07, Lewis Saunders wrote:
>
> Deke Kincaid wrote:
> > http://www.jnack.com/adobe/photoshop/fountain/
> > http://av.adobe.com/russellbrown/CS3StackModeSM.mov
>
> Neat. So does anyone have a plugin which does a median across several
> image inputs? Not possible in a macro, right?
>
> L

---------- Original Message ----------------------------------
From: Phil Pham Reply-To:
Date: Fri, 30 Mar 2007 14:27:44 -0700

>I've been kinda doing this with a bunch of Mix nodes across inputs.
>But I always get a little it of ghosting. I wonder how they are able
>to do that.
>
>
>On Mar 30, 2007, at 12:37 PM, Lewis Saunders wrote:
>
>> Deke Kincaid wrote:
>>> http://www.jnack.com/adobe/photoshop/fountain/
>>> http://av.adobe.com/russellbrown/CS3StackModeSM.mov
>>
>> Neat. So does anyone have a plugin which does a median across several
>> image inputs? Not possible in a macro, right?
>>
>> L

On Mar 30, 2007, at 2:31 PM, Eddie Offermann wrote:

> Can't recall if I've seen a publicly-available macro for this on
> fxshare or
> whatever, but it's doable without a plugin. The variable number of
> inputs
> is the tricky part, but it can be worked around.
> One fairly straightforward approach might be to (auto)align the
> plates,
> difference matte each plate from the others keeping only what's
> common with
> most (maybe add your resulting difference mattes for each plate
> together and
> then threshold the resulting matte to generate the alpha for that
> layer),
> and layer them with the other plates. You'll want at least one
> plate that
> you're arbitrarily declaring to be the 'background' to fill in for
> things
> like leaf flutter (or the water in this example) that's not going
> to be the
> same between any of the plates.
>
> It's definitely something you'd want to build into a re-usable,
> re-purposable piece of script rather than a make-it-everytime kind of
> thing. On a case-by-case basis, it's probably easier just to paint
> yourself
> a cleanplate.
>
> The biggest limitation in Shake is going to be the lack of an
> autoalignment
> tool that'll accept a half dozen or more inputs - and the lack of
> auto-expanding inputs on macros a la the "Layer" node.
>
> I'm interested in seeing a demo of the photoshop stack mode with more
> divergent plates: how does it handle a series where the camera
> moved more
> than in the example, for instance.
>
> On 3/30/07, Lewis Saunders wrote:
>>
>> Deke Kincaid wrote:
>>> http://www.jnack.com/adobe/photoshop/fountain/
>>> http://av.adobe.com/russellbrown/CS3StackModeSM.mov
>>
>> Neat. So does anyone have a plugin which does a median across
>> several
>> image inputs? Not possible in a macro, right?
>>
>> L

>
> you lost me on the file open part. this shit is getting too hard.

On 3/31/07, Sam Cole wrote:
>
> >
> > you lost me on the file open part. this shit is getting too hard.
> no, it's
>
> because
>
> MR adobe man
>
> ......talks.......
>
> LIKE! a
>
> retard
>
> WOW
>
> ./sam
>
>
> p.s.
>
> So has anyone figured out how to do a temporal median in shake logic?

On 3/31/07, Randy Little wrote:
>
> Ok Mr. Adobe Man is one of the best designers alive. Russel Brown. He is
> a bit crazy and has been making those demo's longer then most people have
> been using computers. I used to work with crazy russel back in the mid
> 90's. Well I was at Macromedia and he was at Adobe.
> http://www.russellbrown.com/ Russel got is style because most people
> doing demos back then would put you to sleep. Russel being a very very very
> good designer decided when he was asked to do these that he would be more
> Mr. Crazy Russel. (he is a little wacky) But he is also the SR. Creative
> Director for Adobe.
>
> R.
>
>
>
>
> On 3/31/07, Sam Cole wrote:
> >
> > >
> > > you lost me on the file open part. this shit is getting too hard.
> > no, it's
> >
> > because
> >
> > MR adobe man
> >
> > ......talks.......
> >
> > LIKE! a
> >
> > retard
> >
> > WOW
> >
> > ./sam
> >
> >
> > p.s.
> >
> > So has anyone figured out how to do a temporal median in shake logic?

one exception is Deke McClelland, his photoshop tutorials are very good
and advanced, and often you pick up very useful tricks in a side remark.
he also has a good pace and is mildly funny ;)
http://movielibrary.lynda.com/authors/author/?aid=98

i wish somebody would do something similair for shake..
++ chris



at 31.03.2007 Sam Cole wrote:

> no, it's
>
> because
>
> MR adobe man
>
> .......talks.......
>
> LIKE! a
>
> retard
>
> WOW
>
> ../sam
>
>
> p.s.
>
> So has anyone figured out how to do a temporal median in shake logic?

On 3/31/07, chris wrote:
>
>
> lol, i thought i'm the only one that feels that way..
>
> frankly, i dont care if he's a good designer or not or if he's a bit
> crazy (well, maybe if i'd known that before hand i wouldnt feel outright
> insulted). point is i cant watch that stuff and the learning effect is
> usually rather small... most tutorials i've seen seem to be made for
> people who hardly know how send an email attachment, and cover stuff
> that i could just as well find by looking up the manual.
>
> one exception is Deke McClelland, his photoshop tutorials are very good
> and advanced, and often you pick up very useful tricks in a side remark.
> he also has a good pace and is mildly funny ;)
> http://movielibrary.lynda.com/authors/author/?aid=98
>
> i wish somebody would do something similair for shake..
> ++ chris
>
>
>
> at 31.03.2007 Sam Cole wrote:
>
> > no, it's
> >
> > because
> >
> > MR adobe man
> >
> > .......talks.......
> >
> > LIKE! a
> >
> > retard
> >
> > WOW
> >
> > ../sam
> >
> >
> > p.s.
> >
> > So has anyone figured out how to do a temporal median in shake logic?

On 3/31/07, Stewart Pomeroy wrote:
>
> Aren't those side snarky remarks in the manuals amusing?
>
>
> S.
>
> >> i wish somebody would do something similair for shake..
> >> ++ chris
>
>
>
> --
> //stewart pomeroy
> //cg artist
> //

On Mar 30, 2007, at 9:17 PM, Randy Little wrote:

> that rocks
>
> On 3/30/07, Deke Kincaid wrote:
>>
>> http://www.jnack.com/adobe/photoshop/fountain/
>> http://av.adobe.com/russellbrown/CS3StackModeSM.mov
>>
>> -deke

> This kind of averaging has exited in Flame for 10 years under the
> name of compound - hardly revolutionary.

On 3/31/07, Randy Little wrote:

On Apr 2, 2007, at 1:35 PM, Sam Cole wrote:

>> This kind of averaging has exited in Flame for 10 years under the
>> name of compound - hardly revolutionary.
>
> A median is different from a mean.
>
> ./sam

On 2 Apr 2007, at 15:14, Christopher Noellert wrote:

> ... it's a weighted average - but it's still a form of an average,
> which is why the trick works. Mean was also one of the choices in
> the menu by the by.
>
> Regardless, the point is still valid. Still curious about the
> entropy setting...
>
> Best,
> Chris
>
> --
> Chris Noellert
> Senior Flame / Digital Post Technical Director / Prima donna
>
> Nordiskfilm Post Production Stockholm
> (Formerly Filmteknik/Frithiof Film to Video. AB)
> Tulvaktsvagen 2
> 115 40 Stockholm
>
> Tel: +46 8 450 450 00
> Fax: +46 8 450 450 01
> Dir: +46 8 450 450 17
> Mob: +46 7 024 616 31
> AIM: cmnoellert
>
> Reel: http://se.nordiskfilm-postproduction.com/movieviewer.aspx?
> movie=Video_250018.mov
> Web: http://www.nordiskfilm-postproduction.com
>
>
> On Apr 2, 2007, at 1:35 PM, Sam Cole wrote:
>
>>> This kind of averaging has exited in Flame for 10 years under the
>>> name of compound - hardly revolutionary.
>>
>> A median is different from a mean.
>>
>> ./sam

On 2 Apr 2007, at 15:59, howard Jones wrote:

> Just to stick my head up my arse and be totally anal,
>
> a median isn't a form of an average, it IS by definition an average -
> taking the central point of a distribution rather than the average of
> the distribution or
> (and here comes the point) as the mode does - the most common.
>
> In the example given in CS3 surely this is a mode average rather than
> a mean or median (ie the most common colour is
> used not the central or average)?
>
> and yes this would be very handy.
>
> I would check this theory against the movie but I've lost the link.
>
> Right head out of my arse and time to enjoy the sunshine.
> Howard
>
>
>
> On 2 Apr 2007, at 15:14, Christopher Noellert wrote:
>
>> ... it's a weighted average - but it's still a form of an average,
>> which is why the trick works. Mean was also one of the choices in
>> the menu by the by.
>>
>> Regardless, the point is still valid. Still curious about the
>> entropy setting...
>>
>> Best,
>> Chris
>>
>> --
>> Chris Noellert
>> Senior Flame / Digital Post Technical Director / Prima donna
>>
>> Nordiskfilm Post Production Stockholm
>> (Formerly Filmteknik/Frithiof Film to Video. AB)
>> Tulvaktsvagen 2
>> 115 40 Stockholm
>>
>> Tel: +46 8 450 450 00
>> Fax: +46 8 450 450 01
>> Dir: +46 8 450 450 17
>> Mob: +46 7 024 616 31
>> AIM: cmnoellert
>>
>> Reel: http://se.nordiskfilm-postproduction.com/movieviewer.aspx?
>> movie=Video_250018.mov
>> Web: http://www.nordiskfilm-postproduction.com
>>
>>
>> On Apr 2, 2007, at 1:35 PM, Sam Cole wrote:
>>
>>>> This kind of averaging has exited in Flame for 10 years under the
>>>> name of compound - hardly revolutionary.
>>>
>>> A median is different from a mean.
>>>
>>> ./sam

On 2 Apr 2007, at 16:10, howard Jones wrote:

> although if it was median used I suspect thats why its on a grey road.
>
> On 2 Apr 2007, at 15:59, howard Jones wrote:
>
>> Just to stick my head up my arse and be totally anal,
>>
>> a median isn't a form of an average, it IS by definition an average -
>> taking the central point of a distribution rather than the average of
>> the distribution or
>> (and here comes the point) as the mode does - the most common.
>>
>> In the example given in CS3 surely this is a mode average rather than
>> a mean or median (ie the most common colour is
>> used not the central or average)?
>>
>> and yes this would be very handy.
>>
>> I would check this theory against the movie but I've lost the link.
>>
>> Right head out of my arse and time to enjoy the sunshine.
>> Howard
>>
>>
>>
>> On 2 Apr 2007, at 15:14, Christopher Noellert wrote:
>>
>>> ... it's a weighted average - but it's still a form of an average,
>>> which is why the trick works. Mean was also one of the choices in
>>> the menu by the by.
>>>
>>> Regardless, the point is still valid. Still curious about the
>>> entropy setting...
>>>
>>> Best,
>>> Chris
>>>
>>> --
>>> Chris Noellert
>>> Senior Flame / Digital Post Technical Director / Prima donna
>>>
>>> Nordiskfilm Post Production Stockholm
>>> (Formerly Filmteknik/Frithiof Film to Video. AB)
>>> Tulvaktsvagen 2
>>> 115 40 Stockholm
>>>
>>> Tel: +46 8 450 450 00
>>> Fax: +46 8 450 450 01
>>> Dir: +46 8 450 450 17
>>> Mob: +46 7 024 616 31
>>> AIM: cmnoellert
>>>
>>> Reel: http://se.nordiskfilm-postproduction.com/movieviewer.aspx?
>>> movie=Video_250018.mov
>>> Web: http://www.nordiskfilm-postproduction.com
>>>
>>>
>>> On Apr 2, 2007, at 1:35 PM, Sam Cole wrote:
>>>
>>>>> This kind of averaging has exited in Flame for 10 years under the
>>>>> name of compound - hardly revolutionary.
>>>>
>>>> A median is different from a mean.
>>>>
>>>> ./sam

On Apr 3, 2007, at 6:06 PM, Christopher Noellert wrote:

> My turn for anal behavior:
> There are several different kinds of calculations for central
> tendency, the kind of calculation that should be used depends on the
> type of data (level of measurement) and purpose for which the central
> tendency is being calculated:
> Arithmetic mean