This is my first test run of a macro I wrote to generate "blobscapes."
This one is a single blob composed of 763200 spheres. 

Peak memory used: 247 Meg
Parse time: 40 minutes 39 seconds
Trace time: 4 hours 38 minutes 2 seconds
(640x480 no aa no radiosity)

I think I may have succeeded in creating the most innofficient,
memory-and-time-consuming macro ever.



-- 
- Doug Eichenberg
  http://www.getinfo.net/douge
  dou### [at] nlsnet

Post a reply to this message

Attachments:
Download 'iso-8859-1' (2 KB)

Hmm neat looking image but if you don't mind me asking, why blobscapes?
There are many, more accurate and faster methods to produces landscapes,
HF's and isosurfaces for example.

Doug Eichenberg wrote:

> This is my first test run of a macro I wrote to generate
> "blobscapes."This one is a single blob composed of 763200
> spheres. Peak memory used: 247 MegParse time: 40 minutes 39
> secondsTrace time: 4 hours 38 minutes 2 seconds(640x480 no aa no
> radiosity) I think I may have succeeded in creating the most
> innofficient,memory-and-time-consuming macro ever.
> --
> - Doug Eichenberg
>   http://www.getinfo.net/douge
>   dou### [at] nlsnet

Post a reply to this message

> Doug Eichenberg wrote:
> 
> This is my first test run of a macro I wrote to generate "blobscapes."
> This one is a single blob composed of 763200 spheres.
> 
> Peak memory used: 247 Meg
> Parse time: 40 minutes 39 seconds
> Trace time: 4 hours 38 minutes 2 seconds
> (640x480 no aa no radiosity)
> 
> I think I may have succeeded in creating the most innofficient,
> memory-and-time-consuming macro ever.
> 

Doesn't look bad at all, but isosurfaces are probably more effective for such a
thing.  It would not be slower in many cases, and has nearly no parsing time and
memory requirements.  

How did you place the blob components? eval_pigment?

BTW, please don't post html next time.

Christoph

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

In article <39bb4495@news.povray.org>, "Doug Eichenberg" 
<dou### [at] nlsnet> wrote:

> This is my first test run of a macro I wrote to generate "blobscapes."
> This one is a single blob composed of 763200 spheres. 

It has an interesting look to it, like some kind of strange geological 
formation in a dry seabed. It might be interesting how an isosurface 
using this in a blob pattern would work...
It would probably be faster rendering than a blob object, at least, if 
you use a blob object as input instead of entering the components 
directly into the pattern(the blob object has internal bounding, the 
blob pattern does not...but it does have the ability to use blob 
objects, at least, it does in MegaPOVPlus).


> I think I may have succeeded in creating the most innofficient,
> memory-and-time-consuming macro ever.

In order to do that, it would have to beat my POV-Script liquid 
simulator, complete with inter-particle forces. I eventually abandoned 
developing it after a 3 hour parse.

-- 
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/

<><

Post a reply to this message

--
- Doug Eichenberg
  http://www.getinfo.net/douge
  dou### [at] nlsnet
"Thomas Lake" <tla### [at] homecom> wrote in message
news:39BB4826.178D1611@home.com...
> Hmm neat looking image but if you don't mind me asking, why blobscapes?
> There are many, more accurate and faster methods to produces landscapes,
> HF's and isosurfaces for example.

I think it just has a neat look to it.  I plan on adding random rock
formations, kind of
like Jaimes stalagmites/stalactites in the wilderness round.

Post a reply to this message

--
- Doug Eichenberg
  http://www.getinfo.net/douge
  dou### [at] nlsnet
> Doesn't look bad at all, but isosurfaces are probably more effective for
such a
> thing.  It would not be slower in many cases, and has nearly no parsing
time and
> memory requirements.

I haven't really messed with isosurfaces.  I don't think they would be able
to produce a random landscape like this.  This one takes a seed value and
creates a new landscape each time, and it allows you to control maximum
height, rate of ascent/descent in height, etc.

> How did you place the blob components? eval_pigment?

I came up with a formula where the height of each spherical component is
based on the height of the points around it, and has a random chance of
moving up or down by a percentage of the average of those heights.  The
height of each component is stored in an array.

> BTW, please don't post html next time.

Ooops... sorry!

Post a reply to this message

--
- Doug Eichenberg
  http://www.getinfo.net/douge
  dou### [at] nlsnet

> It has an interesting look to it, like some kind of strange geological
> formation in a dry seabed. It might be interesting how an isosurface
> using this in a blob pattern would work...
> It would probably be faster rendering than a blob object, at least, if
> you use a blob object as input instead of entering the components
> directly into the pattern(the blob object has internal bounding, the
> blob pattern does not...but it does have the ability to use blob
> objects, at least, it does in MegaPOVPlus).

I really haven't messed around with iso's yet.

> In order to do that, it would have to beat my POV-Script liquid
> simulator, complete with inter-particle forces. I eventually abandoned
> developing it after a 3 hour parse.

Ouch.

Post a reply to this message

Doug Eichenberg wrote:
> 
> I haven't really messed with isosurfaces.  I don't think they would be able
> to produce a random landscape like this.  This one takes a seed value and
> creates a new landscape each time, and it allows you to control maximum
> height, rate of ascent/descent in height, etc.
> 

Of course it would be difficult to achieve those "sphere structures" with an
isosurfaces, but in general they are much more flexible.  Control of maximum
height and slope is not difficult at all.

Christoph 

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Neat! a landscape made with blobs; I think you should call it a
blobscape :-)

Post a reply to this message

Wow
 An HTML post and Netscape can't read it at all.. must have an
improperly closed table or something, but it shows up in teh reply
window... very strange.

I like the image, and it has possibilites, but I can see how lots of
components show up to low to be seen. You might be able to improve the
algorithm some by focusing it on the surface... then again, you may be
doing that already.

Keep up the good work. I'd like to see this go places. I tried the same
thing a while ago to make water dropelts, and with some settings it made
a nice little villiage as welll.

Josh


Doug Eichenberg wrote:

> This is my first test run of a macro I wrote to generate
> "blobscapes."This one is a single blob composed of 763200
> spheres. Peak memory used: 247 MegParse time: 40 minutes 39
> secondsTrace time: 4 hours 38 minutes 2 seconds(640x480 no aa no
> radiosity) I think I may have succeeded in creating the most
> innofficient,memory-and-time-consuming macro ever.
> --
> - Doug Eichenberg
>   http://www.getinfo.net/douge
>   dou### [at] nlsnet

--
Josh English -- Lexiphanic Lethomaniac
eng### [at] spiritonecom
The POV-Ray Cyclopedia http://www.spiritone.com/~english/cyclopedia/

Post a reply to this message