 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I am currently in the process of raytracing an 8000x6000 image for an A3
poster - the problem is it has been going for two and a half days and
has become terminally slow! I predicted that it would take about three
and a quarter days, but I'm not sure now - as you can imaging, this is a
great inconvenience for anyone else who wants to use the computer!
Because I was bored, I was looking at the POV console output (I'm using
POV 3.1e.msvc for Windows, if it makes any difference), and I saw
something about "continued trace". So I looked this up in the POV-Win
help and saw that there was a "Continue Raytrace" option in the render
window :^)
However, I am rendering my scene with radiosity, and I have noticed that
if you render a certain section with radiosity it is inconstant with
another section which is ever-so slightly different, and very
inconsistent with a full image render. So, does anyone out there know
whether it is safe to just stop the job and use continue when you are
rendering with radiosity - I don't want to see any inconstancies which
will spoil the image and waste three or so days!
Can anyone help?
Chris Harrison
http://www.ChrisHarrison.co.uk/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Bob Hughes wrote:
>
> Think it's been said that a continued (+C) radiosity render is now fixed.
> But don't go trying it on my say so please. See if someone else speaks up
> first who has actually done it using the latest version. I take it you
> meant the inconsistencies were while using 3.1a or some other.
I read this a few hours ago and decided not to answer with possibly the
wrong advice. I think Bob made one very rational statement. Don't go trying
it on my say so. With as much time that has been invested in the rendering
of the scene till now this is not the time to experiment. Let this one
finish then verify on scenes that only take a few minutes to render weather
this is a potential problem. I too believe I had heard that there was a fix
in place for this problem but I wouldn't bet my 2 days render time on it
to find out.
--
Ken Tyler
mailto://tylereng@pacbell.net
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thanks :^)
Chris Harrison
http://www.ChrisHarrison.co.uk/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Good advice Ken. I don't think it's fixed yet.
GrimDude
vos### [at] arkansas net
Chris Harrison wrote in message <37357424.3BD20147@btinternet.com>...
>Thanks :^)
>
>Chris Harrison
>http://www.ChrisHarrison.co.uk/
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I don't think it is either. It is caused by the initial pass (the mosiac
preview step) which creates the initial radiosity database. This will be
different if you render different sections of the scene. I think somebody
fixed the feature that lets you save the database and re-load it later...
Then you could generate the database for the whole scene and re-load it
as you distribute it. I'm not sure who fixed it, though.
I want to do something similar for photon mapping, but I'm not sure of the
details yet.
-Nathan
GrimDude wrote:
>
> Good advice Ken. I don't think it's fixed yet.
>
> GrimDude
> vos### [at] arkansasnet
> Chris Harrison wrote in message <37357424.3BD20147@btinternet.com>...
> >Thanks :^)
> >
> >Chris Harrison
> >http://www.ChrisHarrison.co.uk/
> >
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
