Well well well..
I've been plotting a bit on this image. I
It has two trees from a well known macro :-) nothing fancy there.
The chess-pices are from the examples.
The water is a box, scaled in the same size as the world.

The world is a 50x50 grid(limitation in the file creator in windows.. dunno why,
stack overflow bug.) since I wanted the use of a premade inc file.

The parse time wasn't too bad because of theis (8 sec, I think).

I'd really have liked to add some more pices here, but I got too bored with it..

Everything is scaled down to fit the 1,1,1 size, then scaled up to fit the
world. This makes the pices easy to place, but adds a few seconds in parsing.

Here are the macro calls :
noise(0.0005) 
//Center hole
change_lin(<1/2, 0, 1/2>, -0.50) 

//A hole at the top right corner.
change_lin(<1/1, 0, 1/1>, -0.25)

//A hole close to center and towards the viewer
change_log(<2/5, 0, 1/5>, -0.25 ) 


//three tops, with two holes between them.
change_log(<08/32, 0, 23/32>, 0.50) 
change_lin(<11/32, 0, 26/32>, -0.5) 
change_log(<14/32, 0, 28/32>, 0.40) 
change_log(<04/32, 0, 19/32>, 0.40) 
change_lin(<07/32, 0, 22/32>, -0.5)

//fit to specifications.
finalize() 

And then a camera placement and so on..
To place the items, I picked a square 
(every squares center is accessable with point(<1/squareX,0,1/squareY>) )
Then used the #debug v2str( <1/sqX, point(<1/sqX,0,1/sqY>),1/sqY> ) 
to get the position..

This made it fairly easy to insert in the file..
I felt it was too tedious to wait for the parse all the time, but If you have
the patience, or a faster computer than my p200, you can skip the extra include
step.

Comments, ideas?




-- 
//Spider
        [ spi### [at] bahnhofse ]-[ http://www.bahnhof.se/~spider/ ]
What I can do and what I could do, I just don't know anymore
                "Marian"
        By: "Sisters Of Mercy"

Post a reply to this message

Attachments:
Download 'macroland-demo1.jpg' (252 KB)

Preview of image 'macroland-demo1.jpg'

Darn, I forgot the reflective sphere :-)

Post a reply to this message

Is that a one-eared horse (knight chess piece) I see? Yes indeed, it
does appear to be. I think that thing only renders right in POV 2.0 or
some such. Maybe it's POV-Ray 1.0...
That's a pretty think misty air there, makes more trouble for my eyes
than I have already :) If it dropped off faster upwards I think it would
have looked better, imho.
I like seeing that "landscape" of yours down inside and up close like
that, even if the squares have been enlarged. Bet you are working the
supersampling overtime.


Spider wrote:
> 
> Well well well..
> I've been plotting a bit on this image. I
> It has two trees from a well known macro :-) nothing fancy there.
> The chess-pices are from the examples.
> The water is a box, scaled in the same size as the world.
> 
> The world is a 50x50 grid(limitation in the file creator in windows.. dunno why,
> stack overflow bug.) since I wanted the use of a premade inc file.
> 
> The parse time wasn't too bad because of theis (8 sec, I think).
> 
> I'd really have liked to add some more pices here, but I got too bored with it..
> 
> Everything is scaled down to fit the 1,1,1 size, then scaled up to fit the
> world. This makes the pices easy to place, but adds a few seconds in parsing.
> 
> Here are the macro calls :
> noise(0.0005)
> //Center hole
> change_lin(<1/2, 0, 1/2>, -0.50)
> 
> //A hole at the top right corner.
> change_lin(<1/1, 0, 1/1>, -0.25)
> 
> //A hole close to center and towards the viewer
> change_log(<2/5, 0, 1/5>, -0.25 )
> 
> //three tops, with two holes between them.
> change_log(<08/32, 0, 23/32>, 0.50)
> change_lin(<11/32, 0, 26/32>, -0.5)
> change_log(<14/32, 0, 28/32>, 0.40)
> change_log(<04/32, 0, 19/32>, 0.40)
> change_lin(<07/32, 0, 22/32>, -0.5)
> 
> //fit to specifications.
> finalize()
> 
> And then a camera placement and so on..
> To place the items, I picked a square
> (every squares center is accessable with point(<1/squareX,0,1/squareY>) )
> Then used the #debug v2str( <1/sqX, point(<1/sqX,0,1/sqY>),1/sqY> )
> to get the position..
> 
> This made it fairly easy to insert in the file..
> I felt it was too tedious to wait for the parse all the time, but If you have
> the patience, or a faster computer than my p200, you can skip the extra include
> step.
> 
> Comments, ideas?
> 
> --
> //Spider
>         [ spi### [at] bahnhofse ]-[ http://www.bahnhof.se/~spider/ ]
> What I can do and what I could do, I just don't know anymore
>                 "Marian"
>         By: "Sisters Of Mercy"
> 
>   ------------------------------------------------------------------------
>  [Image]

-- 
 omniVERSE: beyond the universe
  http://members.aol.com/inversez/homepage.htm
 mailto:inv### [at] aolcom?Subject=PoV-News

Post a reply to this message

Bob Hughes wrote:
> 
> Is that a one-eared horse (knight chess piece) I see? Yes indeed, it
> does appear to be. I think that thing only renders right in POV 2.0 or
> some such. Maybe it's POV-Ray 1.0...
Heheheh, Yup, it's the one-ear allright :-)


> That's a pretty think misty air there, makes more trouble for my eyes
> than I have already :) If it dropped off faster upwards I think it would
> have looked better, imho.
Well, It should be easy, But then... Hmm, it'd force me to open the file and
edit it... too much trouble ;-)
Nah, I'll see to it.

> I like seeing that "landscape" of yours down inside and up close like
> that, even if the squares have been enlarged. 
Well, they are large, I know. but then, there are two reasons.
	1) 	Object placing. (If it becomes smaller I'd have to remove the
		objects, they'd look awkward from this POV)
	2)	ParsingTime. It'd be easy to increase the amount of squares and 
		thus have smaller ones. but i'd parse slower.
and, in this image, you only see a very small part of the whole landscape.
> Bet you are working the supersampling overtime.
hehe, it took some AA at least :-)

> Spider wrote:
> >
> > Well well well..
> > I've been plotting a bit on this image. I
> > It has two trees from a well known macro :-) nothing fancy there.
> > The chess-pices are from the examples.
> > The water is a box, scaled in the same size as the world.
> >
> > The world is a 50x50 grid(limitation in the file creator in windows.. dunno why,
> > stack overflow bug.) since I wanted the use of a premade inc file.
> >
> > The parse time wasn't too bad because of theis (8 sec, I think).
> >
> > I'd really have liked to add some more pices here, but I got too bored with it..
> >
> > Everything is scaled down to fit the 1,1,1 size, then scaled up to fit the
> > world. This makes the pices easy to place, but adds a few seconds in parsing.
> >
> > Here are the macro calls :
> > noise(0.0005)
> > //Center hole
> > change_lin(<1/2, 0, 1/2>, -0.50)
> >
> > //A hole at the top right corner.
> > change_lin(<1/1, 0, 1/1>, -0.25)
> >
> > //A hole close to center and towards the viewer
> > change_log(<2/5, 0, 1/5>, -0.25 )
> >
> > //three tops, with two holes between them.
> > change_log(<08/32, 0, 23/32>, 0.50)
> > change_lin(<11/32, 0, 26/32>, -0.5)
> > change_log(<14/32, 0, 28/32>, 0.40)
> > change_log(<04/32, 0, 19/32>, 0.40)
> > change_lin(<07/32, 0, 22/32>, -0.5)
> >
> > //fit to specifications.
> > finalize()
> >
> > And then a camera placement and so on..
> > To place the items, I picked a square
> > (every squares center is accessable with point(<1/squareX,0,1/squareY>) )
> > Then used the #debug v2str( <1/sqX, point(<1/sqX,0,1/sqY>),1/sqY> )
> > to get the position..
> >
> > This made it fairly easy to insert in the file..
> > I felt it was too tedious to wait for the parse all the time, but If you have
> > the patience, or a faster computer than my p200, you can skip the extra include
> > step.
> >
> > Comments, ideas?
> >
> > --
> > //Spider
> >         [ spi### [at] bahnhofse ]-[ http://www.bahnhof.se/~spider/ ]
> > What I can do and what I could do, I just don't know anymore
> >                 "Marian"
> >         By: "Sisters Of Mercy"
> >
> >   ------------------------------------------------------------------------
> >  [Image]
> 
> --
>  omniVERSE: beyond the universe
>   http://members.aol.com/inversez/homepage.htm
>  mailto:inv### [at] aolcom?Subject=PoV-News

-- 
//Spider
        [ spi### [at] bahnhofse ]-[ http://www.bahnhof.se/~spider/ ]
What I can do and what I could do, I just don't know anymore
                "Marian"
        By: "Sisters Of Mercy"

Post a reply to this message

Here you go Bob. A new one, 
no chess-pieces in this, and the bush ended up underwater...
Sorry, but I don't feel like doing another render at the moment. 
it took 1h 20m  to parse it, and even more to render. 

I really hope you see the difference now :-)
Actually, its the same landscape, but with approx 100 000 points calculated, 
instead of 2500. (316x316 versus 50x50 grids)

and the number of boxes equals the number of points...

While parsing(The output takes memory), my swap file grew to 115 Mb, together
with my 64Mb ram, that's telling.
:-)

Next time, I'll skip the normal on the water, and the reflection. 
it takes too much processing time.

Hmm, for the vfaq(if it's not there) 
speed-issues : never have a surface both reflecting and filtering/transmmitting.
*hehe*

out, or sumthing.

stats :
1024x768 @AA0.3 (should have used more AA)
Mem : 92 637 325 (Probably allocated more, measuring my runtime-tools)
Parse : 1h 20m 39s
Trace : 2h 31m 05s
total : 3h 51m 44s



-- 
//Spider
        [ spi### [at] bahnhofse ]-[ http://www.bahnhof.se/~spider/ ]
What I can do and what I could do, I just don't know anymore
                "Marian"
        By: "Sisters Of Mercy"

Post a reply to this message

Attachments:
Download 'demo3-landscape.jpg' (226 KB)

Preview of image 'demo3-landscape.jpg'

Spider wrote:
> 
> Here you go Bob. A new one,
> no chess-pieces in this, and the bush ended up underwater...
> Sorry, but I don't feel like doing another render at the moment.
> it took 1h 20m  to parse it, and even more to render.
> 
> I really hope you see the difference now :-)
> Actually, its the same landscape, but with approx 100 000 points calculated,
> instead of 2500. (316x316 versus 50x50 grids)
> 
> and the number of boxes equals the number of points...
> 
> While parsing(The output takes memory), my swap file grew to 115 Mb, together
> with my 64Mb ram, that's telling.
> :-)

Greetings Spider Man,
 
  Pretty cool looking stuff dude. I haven't looked at the source yet but
I am curious as to wether you are storing your data inside of arrays.
I understand this can help a lot where memory effeciency is an issue.
Also concerning your problems with the stack overhead. I was reading
through the section in the docs concerning adc bail out, max trace levels,
and max intersections. In each area of discussion there was mention of
the stacks use in the program and may give you some insight as to how
you might change these global settings or program your code around their
limitations (or process attributes depending on your point of view).

Cheers,

-- 
Ken Tyler

mailto://tylereng@pacbell.net

Post a reply to this message

Uncle Ken transmitted : 
> Greetings Spider Man,
Salutations to you Uncle Ken.

>   Pretty cool looking stuff dude.
Thanks :-)
> I haven't looked at the source yet but
> I am curious as to wether you are storing your data inside of arrays.
Hmm, I store the actual data in arrays, yes. 
but the problem is when writing out data.... :-/

> I understand this can help a lot where memory effeciency is an issue.
yes, since it would allocate all the memory at once and don't change it in
between, yes.

> Also concerning your problems with the stack overhead. I was reading
> through the section in the docs concerning adc bail out, max trace levels,
> and max intersections. In each area of discussion there was mention of
> the stacks use in the program and may give you some insight as to how
> you might change these global settings or program your code around their
> limitations (or process attributes depending on your point of view).
I'll look into it, but I don't think they will solve this problem, the problem
isn't when handling the data, but when writing to a file, and I don't know why.

> Cheers,
The same :-)

-- 
//Spider
        [ spi### [at] bahnhofse ]-[ http://www.bahnhof.se/~spider/ ]
What I can do and what I could do, I just don't know anymore
                "Marian"
        By: "Sisters Of Mercy"

Post a reply to this message

For little ol' me? Oh thankyou Spider!
Okay, I'm over it. That tree... what happened to it?! The leaves are
lost in a grean and gray murkyness. And your poor water got baffled by
all those tiny squares as well.
You may have stumbled upon something good though, no "jaggies" could be
found in all that black and white "pixelation" of both the water AND the
land too.
You really shouldn't have re-rendered at the cost of such a massive
parse. And heck, I thought you had more RAM than that (128 at least),
64MB is what I'm working with too.
What a parse to render ratio. Wow.


Spider wrote:
> 
> Here you go Bob. A new one,
> no chess-pieces in this, and the bush ended up underwater...
> Sorry, but I don't feel like doing another render at the moment.
> it took 1h 20m  to parse it, and even more to render.
> 
> I really hope you see the difference now :-)
> Actually, its the same landscape, but with approx 100 000 points calculated,
> instead of 2500. (316x316 versus 50x50 grids)
> 
> and the number of boxes equals the number of points...
> 
> While parsing(The output takes memory), my swap file grew to 115 Mb, together
> with my 64Mb ram, that's telling.
> :-)
> 
> Next time, I'll skip the normal on the water, and the reflection.
> it takes too much processing time.
> 
> Hmm, for the vfaq(if it's not there)
> speed-issues : never have a surface both reflecting and filtering/transmmitting.
> *hehe*
> 
> out, or sumthing.
> 
> stats :
> 1024x768 @AA0.3 (should have used more AA)
> Mem : 92 637 325 (Probably allocated more, measuring my runtime-tools)
> Parse : 1h 20m 39s
> Trace : 2h 31m 05s
> total : 3h 51m 44s
> 
> --
> //Spider
>         [ spi### [at] bahnhofse ]-[ http://www.bahnhof.se/~spider/ ]
> What I can do and what I could do, I just don't know anymore
>                 "Marian"
>         By: "Sisters Of Mercy"
> 
>   ------------------------------------------------------------------------
>  [Image]

-- 
 omniVERSE: beyond the universe
  http://members.aol.com/inversez/homepage.htm
 mailto:inv### [at] aolcom?Subject=PoV-News

Post a reply to this message

Bob Hughes wrote:
> 
> For little ol' me? Oh thankyou Spider!
*gracious bow*
> Okay, I'm over it. That tree... what happened to it?! The leaves are
> lost in a grean and gray murkyness.
Well, I did as adviced and lowered the fog_alt, but in the test renders (12
boxes, no trees) It turned out to thin so I had to decrease the distance... that
is what happened to it.. :-/

> And your poor water got baffled by all those tiny squares as well.
Yeah... It really was a hell to render, I'd guess(I was by the Tv :-)

> You may have stumbled upon something good though, no "jaggies" could be
> found in all that black and white "pixelation" of both the water AND the
> land too.
Now here is something you'd have to explain more. I don't understand what you
mean.

> You really shouldn't have re-rendered at the cost of such a massive
> parse.
hehe, I had never done this before, so I thought, Whatta heck and did. and once
it had passed an hour, I thought I had done so much already, so I could let it
finish ...

> And heck, I thought you had more RAM than that (128 at least),
> 64MB is what I'm working with too.
Yeah, I had 96 a while ago.... bbut I had to give the 32 back ... :-/

> What a parse to render ratio. Wow.
yea... hehe. I think you understand why I'm pretty fed up with the fwrite
crashes .-)

 
> Spider wrote:
> >
> > Here you go Bob. A new one,
> > no chess-pieces in this, and the bush ended up underwater...
> > Sorry, but I don't feel like doing another render at the moment.
> > it took 1h 20m  to parse it, and even more to render.
> >
> > I really hope you see the difference now :-)
> > Actually, its the same landscape, but with approx 100 000 points calculated,
> > instead of 2500. (316x316 versus 50x50 grids)
> >
> > and the number of boxes equals the number of points...
> >
> > While parsing(The output takes memory), my swap file grew to 115 Mb, together
> > with my 64Mb ram, that's telling.
> > :-)
> >
> > Next time, I'll skip the normal on the water, and the reflection.
> > it takes too much processing time.
> >
> > Hmm, for the vfaq(if it's not there)
> > speed-issues : never have a surface both reflecting and filtering/transmmitting.
> > *hehe*
> >
> > out, or sumthing.
> >
> > stats :
> > 1024x768 @AA0.3 (should have used more AA)
> > Mem : 92 637 325 (Probably allocated more, measuring my runtime-tools)
> > Parse : 1h 20m 39s
> > Trace : 2h 31m 05s
> > total : 3h 51m 44s
> >
> > --
> > //Spider
> >         [ spi### [at] bahnhofse ]-[ http://www.bahnhof.se/~spider/ ]
> > What I can do and what I could do, I just don't know anymore
> >                 "Marian"
> >         By: "Sisters Of Mercy"
> >
> >   ------------------------------------------------------------------------
> >  [Image]
> 
> --
>  omniVERSE: beyond the universe
>   http://members.aol.com/inversez/homepage.htm
>  mailto:inv### [at] aolcom?Subject=PoV-News

-- 
//Spider
        [ spi### [at] bahnhofse ]-[ http://www.bahnhof.se/~spider/ ]
What I can do and what I could do, I just don't know anymore
                "Marian"
        By: "Sisters Of Mercy"

Post a reply to this message

I have little tolerance for parse. Something go wrong only by render
time is the most aggravating. Hey, thought here, parse optimizations:
most spurious stuff parsed first, least last ;)
Didn't mean to confuse about the pixelization comment. I should have
added the word "illusion" and a smilie :)
From what I could see, if jaggies were present or not, I would not have
noticed much judging by all the sharp contrasting small scale blacks and
whites in there (land as well as water part).

Spider wrote:
> 
> > What a parse to render ratio. Wow.
> yea... hehe. I think you understand why I'm pretty fed up with the fwrite
> crashes .-)
> 

-- 
 omniVERSE: beyond the universe
  http://members.aol.com/inversez/homepage.htm
 mailto:inv### [at] aolcom?Subject=PoV-News

Post a reply to this message