Andross

Like is there a list of most expensive to least expensive objects or something?

is-there-a-way-to-know-the-cost-of-certain-objects-on-the-processor-before-running-the-patch

I have maked in illustrator paths that i want to import and use in 

Jitter. So i know that i have to convert this illustrator doc .ai 

But i do not manage to do that properly. Which 3D program do i have 

When i have managed to do that > Jitter question :

Is it possible to use this .obj 3d path in Jitter like a 3d path to 

Is there some help or example patch to do that or to give me a hint.

I have maked in illustrator paths that i want to import and use in  
Jitter. So i know that i have to convert this illustrator doc .ai  
to .obj and import that in Jitter.
But i do not manage to do that properly. Which 3D program do i have  
to use?

When i have managed to do that > Jitter question :
Is it possible to use this .obj 3d path in Jitter like a 3d path to  
move other 3d objects along that path?

>I have maked in illustrator paths that i want to import and use in 

>Jitter. So i know that i have to convert this illustrator doc .ai to 

>But i do not manage to do that properly. Which 3D program do i have to use?

>When i have managed to do that > Jitter question :

>Is it possible to use this .obj 3d path in Jitter like a 3d path to 

>Is there some help or example patch to do that or to give me a hint.

maybe convert it to a text file ?
then load it into a coll

>Hi,
>
>I have maked in illustrator paths that i want to import and use in 
>Jitter. So i know that i have to convert this illustrator doc .ai to 
>.obj and import that in Jitter.
>But i do not manage to do that properly. Which 3D program do i have to use?
>
>
>When i have managed to do that > Jitter question :
>Is it possible to use this .obj 3d path in Jitter like a 3d path to 
>move other 3d objects along that path?
>
>
>Is there some help or example patch to do that or to give me a hint.
>
>
>
>
>Regards,
>Bart
>


not as far as I've ever seen, although that would make a great data 

analysis/display assignment, if someone is looking for a school 

On Jul 31, 2006, at 12:25 PM, Robert wrote:

> Like is there a list of most expensive to least expensive objects 

not as far as I've ever seen, although that would make a great data  
analysis/display assignment, if someone is looking for a school  
project...

>
> Like is there a list of most expensive to least expensive objects  
> or something?


I have tried that for other projects also but it takes to much time 

Is there a programme or way to do that fast and properly? So that i 

do not have to clean 1000 code lines so that jitter can understand it 

>> I have maked in illustrator paths that i want to import and use in 

>> Jitter. So i know that i have to convert this illustrator doc .ai 

>> But i do not manage to do that properly. Which 3D program do i 

>> When i have managed to do that > Jitter question :

>> Is it possible to use this .obj 3d path in Jitter like a 3d path 

>> to move other 3d objects along that path?

>> Is there some help or example patch to do that or to give me a hint.

Yes, but how do you do that properly?
I have tried that for other projects also but it takes to much time  
and gives a lot of problems.
Is there a programme or way to do that fast and properly? So that i  
do not have to clean 1000 code lines so that jitter can understand it  
properly.

> maybe convert it to a text file ?
> then load it into a coll
>
> -mj
>
>
>> Hi,
>>
>> I have maked in illustrator paths that i want to import and use in  
>> Jitter. So i know that i have to convert this illustrator doc .ai  
>> to .obj and import that in Jitter.
>> But i do not manage to do that properly. Which 3D program do i  
>> have to use?
>>
>>
>> When i have managed to do that > Jitter question :
>> Is it possible to use this .obj 3d path in Jitter like a 3d path  
>> to move other 3d objects along that path?
>>
>>
>> Is there some help or example patch to do that or to give me a hint.
>>
>>
>>
>>
>> Regards,
>> Bart
>>
>

Quote: Andross wrote on Mon, 31 July 2006 05:25

----------------------------------------------------

> Like is there a list of most expensive to least expensive objects or something?

no, because the "CPU cost" of an object depends on the 

patch i.e. how often it is been instatiated and with 

what other objects it is been used together.

Quote: Andross wrote on Mon, 31 July 2006 05:25
----------------------------------------------------
> Like is there a list of most expensive to least expensive objects or something?
----------------------------------------------------

no, because the "CPU cost" of an object depends on the 
patch i.e. how often it is been instatiated and with 
what other objects it is been used together.

make 10 copies of [zerox~].
look at the CPU % window in max.
then try 20, 30, 40, and 100.

yeah, but, you can instantiate 10 copies of 'method a', (method a 

being either an object or abstraction to processing something) and 10 

copies of 'method b', and see which is most efficient. Granted, the 

context of the parent patch obviously matters, but I still feel like 

there could be a better way to write cleaner/faster code, yes?

On Aug 9, 2006, at 10:22 AM, Roman Thilenius wrote:

> Quote: Andross wrote on Mon, 31 July 2006 05:25

> ----------------------------------------------------

>> Like is there a list of most expensive to least expensive objects 

> no, because the "CPU cost" of an object depends on the

> patch i.e. how often it is been instatiated and with

> what other objects it is been used together.

yeah, but, you can instantiate 10 copies of 'method a', (method a  
being either an object or abstraction to processing something) and 10  
copies of 'method b', and see which is most efficient. Granted, the  
context of the parent patch obviously matters, but I still feel like  
there could be a better way to write cleaner/faster code, yes?

>
> Quote: Andross wrote on Mon, 31 July 2006 05:25
> ----------------------------------------------------
>> Like is there a list of most expensive to least expensive objects  
>> or something?
> ----------------------------------------------------
>
>
> no, because the "CPU cost" of an object depends on the
> patch i.e. how often it is been instatiated and with
> what other objects it is been used together.
>
> try this:
>
> make 10 copies of [zerox~].
> look at the CPU % window in max.
> then try 20, 30, 40, and 100.
>
>
> dont try -110

> yeah, but, you can instantiate 10 copies of 'method a', (method a 

> being either an object or abstraction to processing something) and 

> 10 copies of 'method b', and see which is most efficient. Granted, 

> the context of the parent patch obviously matters, but I still feel 

> like there could be a better way to write cleaner/faster code, yes?

Experience helps a lot. Not only experience in Max/MSP/Jitter, but if 

you've worked with lower-level languages and/or studied some DSP, 

then you've got added experience along the lines of "Fourier 

Transforms are expensive, table-based waveshaping is cheap" and you 

can extrapolate that to specific objects. (The last example is over- 

simplified, but hopefully you get the idea.)

A table of objects sorted by relative expense is a nice idea but how 

do you deal with objects that are cheap most of the time but have to 

do heavy computation occasionally? What if the heavy computations 

come up at regular intervals (FFT stuff, the lp.pvvv~ & friends 

objects) or erratically, say when parameters change? Then some 

objects pay a higher price if you use signals in control inlets, 

others don't notice much difference. On top of that, some of the 

relative expense may change simply by using a different compiler, 

different OS, or (gasp!) a different processor.

I'm not saying that the idea isn't a good one. I'm just trying to 

explain some of the difficulties involved. My suggestion to Robert is 

to start a personal list. No matter what comes of the list, you'll 

learn a lot in the process. And that's a good thing.

On 9-Aug-2006, at 16:22, Roman Thilenius wrote:

For tracking CPU usage more precisely (which one presumably would 

want to do in a project comparing CPU usage of different objects), 

you might want to try the Running CPU Usage example distributed with 

the free Litter Power Starter Pack. URI below.

Peter Castine +--> Litter Power & Litter Bundle for Jitter

Interface Building for |home | chez nous|

Max/MSP Extremely cool |bei uns | i nostri|

On 9-Aug-2006, at 21:56, vade wrote:
> yeah, but, you can instantiate 10 copies of 'method a', (method a  
> being either an object or abstraction to processing something) and  
> 10 copies of 'method b', and see which is most efficient. Granted,  
> the context of the parent patch obviously matters, but I still feel  
> like there could be a better way to write cleaner/faster code, yes?

Experience helps a lot. Not only experience in Max/MSP/Jitter, but if  
you've worked with lower-level languages and/or studied some DSP,  
then you've got added experience along the lines of "Fourier  
Transforms are expensive, table-based waveshaping is cheap" and you  
can extrapolate that to specific objects. (The last example is over- 
simplified, but hopefully you get the idea.)

A table of objects sorted by relative expense is a nice idea but how  
do you deal with objects that are cheap most of the time but have to  
do heavy computation occasionally? What if the heavy computations  
come up at regular intervals (FFT stuff, the lp.pvvv~ & friends  
objects) or erratically, say when parameters change? Then some  
objects pay a higher price if you use signals in control inlets,  
others don't notice much difference. On top of that, some of the  
relative expense may change simply by using a different compiler,  
different OS, or (gasp!) a different processor.

I'm not saying that the idea isn't a good one. I'm just trying to  
explain some of the difficulties involved. My suggestion to Robert is  
to start a personal list. No matter what comes of the list, you'll  
learn a lot in the process. And that's a good thing.

On 9-Aug-2006, at 16:22, Roman Thilenius wrote:
> look at the CPU % window in max.

For tracking CPU usage more precisely (which one presumably would  
want to do in a project comparing CPU usage of different objects),  
you might want to try the Running CPU Usage example distributed with  
the free Litter Power Starter Pack. URI below.

--------------    http://www.bek.no/~pcastine/Litter/    -------------
Peter Castine             +--> Litter Power & Litter Bundle for Jitter
                                Universal Binaries on the way
iCE:  Sequencing,  Recording &
       Interface  Building  for                   |home    | chez nous|
       Max/MSP   Extremely cool                   |bei uns |  i nostri|
       http://www.dspaudio.com/                   http://www.castine.de

For Max objects you can also take a look at the help file for the 

cpuclock object. I have several times banged objects ad ridiculous and 

used this approach to see how long that takes. One such example can be 

For Max objects you can also take a look at the help file for the 
cpuclock object. I have several times banged objects ad ridiculous and 
used this approach to see how long that takes. One such example can be 
found here:

Is there a way to know the cost of certain objects on the processor before running the patch?