[Sidefx-houdini-list] tool efficiency benchmarks

Craig Zerouni craig at sidefx.com
Thu Nov 2 20:03:52 EST 2006


I think about this stuff a lot - in fact, it's part of what I get paid 
for. And yet, the problems with it are obvious.

One interesting data point though - when a company I won't name was 
setting up to do animation with software product A, which they had never 
used before, they based their manpower requirements on their 
considerable experience with software product B. However, as they got 
into the film, which had > 700 shots by far, they discovered that they 
had roughly 25% more staff in the animation dept than they needed for 
the schedule they had.

Tentative conclusion: software product A was more productive than 
software product B.

Is that ironclad evidence though? Never, because every film is 
different. But if you could do this over large enough numbers of shots, 
with large enough numbers of artists, you would be able to make 
reasonable conclusions. But nobody is ever going to spend the money, on 
purpose anyway, to conduct those kinds of experiments.

CZ

Andrew D Lyons wrote:
> True. Method and Morale are important considerations.
>
> I might add that if people want to send examples of film or studio
> specific case data to me off list that would be great.
>
> Cheers
>
>
>
> On 02/11/06, Ivan DeWolf <ivan at martian-labs.com> wrote:
>>
>> part of the problem in quantifying this is there are wide differences 
>> in artists
>> speed and techniques.
>>
>> I know some artists that plod along but use shortcuts and simple 
>> hacks to get
>> things out the door quickly, and other artists that scurry around 
>> like mad
>> generating thousands of sops or lines of code over months to create a 
>> simple
>> effect. It's never as simple as "packageA vs. package B"; I would 
>> argue that
>> the other outside influances carry far more weight in an efficiency 
>> equation
>> than choice of software.
>>
>> even the same artist might have different levels of efficiency 
>> depending on
>> other influences in their lives.  This week he's slow, but last month 
>> he was
>> crankin'...
>>
>> Quoting Andrew D Lyons <tstexture at gmail.com>:
>>
>> > I've often wondered whether there was any scientific way of
>> > establishing whether "tool A" was better for certain tasks than "tool
>> > B". It seems a really difficult thing to quantify. You can say that
>> > "house A" used "tool A" and made 800 shots in 12 months with a crew of
>> > 12 artists and 3 TD's. "House B" on the other hand used "tool B" and
>> > also made 800 shots in 12 months but needed 24 artists and 12 TD's.
>> > People might argue that "Film B" had more challenges, or the  Director
>> > was a vision-less nut-case etc. How do you then quantify and take into
>> > consideration "challenging", or the effect of poor direction?
>> >
>> > I don't want to start a software flame war. I'm looking more for
>> > people's ideas on scientific methods that could be used to benchmarks
>> > tools - more than applications of those methods to the tools that are
>> > out there.
>> >
>> > Having said that I'd also be really interested to see any raw data and
>> > case examples that people might have to share - where the task
>> > involves shot counts > 700 shots.
>> >
>> > Thanks!
>> >
>> > Cheers
>> >
>> > --
>> > =======================================
>> > Andrew D Lyons | Digital Artist | http://www.tstex.com
>> > =======================================
>> > _______________________________________________
>> > Sidefx-houdini-list mailing list
>> > Sidefx-houdini-list at sidefx.com
>> > https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list
>> >
>>
>>
>> -Ivan
>>
>> ---------------------------------------------------------
>> "Perfection is achieved when nothing else can be removed"
>> -Yvon Chouinard
>> _______________________________________________
>> Sidefx-houdini-list mailing list
>> Sidefx-houdini-list at sidefx.com
>> https://lists.sidefx.com:443/mailman/listinfo/sidefx-houdini-list
>>
>
>



More information about the Sidefx-houdini-list mailing list