|
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 5/13/21 3:49 PM, jr wrote:
> hi,
>
> William F Pokorny <ano### [at] anonymousorg> wrote:
>> ...
>> // This a povr branch only example!
>> ...
>> #declare Nrml00 = normal {
>> average
>> normal_map {
>> [0 bevy type 1 bump_size 0.5] // unbiased wrinkles alt
...
> File 'bevr.pov' line 28:
> Parse Error:
> No normal pattern type specified. For example, bumps.
> Fatal error in parser: Cannot parse input.
> Render failed
>
The bevy keyword (as normal pattern collection, pattern) is new. I've
not yet released code that supports it. I should've added a note. And I
should tweak that error message as 'type' has become misleading.
Bill P.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
hi,
William F Pokorny <ano### [at] anonymousorg> wrote:
> On 5/13/21 3:49 PM, jr wrote:
> > William F Pokorny <ano### [at] anonymousorg> wrote:
> >> ...
> The bevy keyword (as normal pattern collection, pattern) is new. I've
> not yet released code that supports it. ...
ah, ok. can you clue me/us in wrt planned update/release? (roughly)
regards, jr.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 5/14/21 5:38 AM, jr wrote:
> hi,
>
...
>> The bevy keyword (as normal pattern collection, pattern) is new. I've
>> not yet released code that supports it. ...
>
> ah, ok. can you clue me/us in wrt planned update/release? (roughly)
>
I'm working on some commits today, so I might be able to put something
out fairly soon.
I know the internals of the bevy collection itself is a work in
progress, but probably I can throw a construction tarp over what is not
done. I'm not done with all the include self testing either, but quite
lot fixed... We'll see.
Bill P.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> hi,
>
> William F Pokorny <ano### [at] anonymousorg> wrote:
>> ...
>> // This a povr branch only example!
>> ...
>> #declare Nrml00 = normal {
>> average
>> normal_map {
>> [0 bevy type 1 bump_size 0.5] // unbiased wrinkles alt
>
> the line above (28) gives me trouble.
>
>
>> [1 bevy type 4 bump_size 0.5] // Lean already wrinkled at +y
>> }
>> }
>
In an average, the control values determine the weight of that part. So,
that [0 bevy type 1 bump_size 0.5] mean zero weight, zero contribution.
This mean that it could be removed without affecting the final result.
That zero weight may be the cause of your problem.
Try changing it to :
// effectively the same as the above code
#declare Nrm100 = normal{bevy type 4 bump_size 0.5}
or
// Equal weight
#declare Nrm100 = normal{
average
normal_map{
[1 bevy type 1 bump_size 0.5]
[1 bevy type 4 bump_size 0.5]
}
}
or
//very little contribution for type 1
#declare Nrm100 = normal{
average
normal_map{
[0.0001 bevy type 1 bump_size 0.5]
[1 bevy type 4 bump_size 0.5]
}
}
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Just went through povr installation with jr.
Got the biscuit render, so looked good.
Went into qtpovray38 distribution directory to test a .pov file.
bald@Alienware ~/povray38/scenes/objects $ povr blob.pov
/home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: 5:
/home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: Cannot fork
bald@Alienware ~/povray38/scenes/objects $ povr chess.pov
/home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: 5:
/home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: Cannot fork
Any ideas?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 5/14/21 10:20 AM, Alain Martel wrote:
>> hi,
>>
>> William F Pokorny <ano### [at] anonymousorg> wrote:
>>> ...
>>> // This a povr branch only example!
>>> ...
>>> #declare Nrml00 = normal {
>>
>> the line above (28) gives me trouble.
>>
>>
>>> }
>>
>
> In an average, the control values determine the weight of that part. So,
> that [0 bevy type 1 bump_size 0.5] mean zero weight, zero contribution.
> This mean that it could be removed without affecting the final result.
> That zero weight may be the cause of your problem.
>
With povr the zero weight is part of the trick to pass the results of
this map entries normal perturbation to bevy type 4 so what you get is a
chain of perturbations ending in one result and not some average on
perturbation results.
That said, your suggestion should still work, but will result in a
normal both averaged and chained... Not something I've actually tried.
I'd guess it would come in as a mix of the lower two images.
All of the bevy stuff is still sitting only on my machine. I haven't
released it.
Bill P.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 5/14/21 3:59 PM, Bald Eagle wrote:
>
> Just went through povr installation with jr.
> Got the biscuit render, so looked good.
>
> Went into qtpovray38 distribution directory to test a .pov file.
>
> bald@Alienware ~/povray38/scenes/objects $ povr blob.pov
> /home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: 5:
> /home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: Cannot fork
>
> bald@Alienware ~/povray38/scenes/objects $ povr chess.pov
> /home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: 5:
> /home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: Cannot fork
>
> Any ideas?
>
Just guesses.
The 'cannot fork' stuff usually happens when a unix/linux system has run
out of resources to start another process. Lack of memory/swap space,
/tmp, /var - maybe. There are sometimes limits on the number of
processes in place, but this in my experience is not all that common.
For starters you could try running the 'df' command and look at disk
usage. The command 'free -m' can be useful too.
Somewhere you posted about your machine... Let me try and find that and
I'll post back here if I see anything obvious.
---
Note! I expect quite a few existing scenes will NOT render correctly
with the povr branch without modification.
Let me try quickly to run a v3.8 blob scene here with my povr wrapper
script. It runs OK for me, but my povr is a little different.
I don't see a chess.pov, but chess2.pov doesn't run because it tries to
include color.inc and that file is not in povr's core set of includes.
In any case I'm not getting 'cannot fork' errors...
Bill P.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 5/14/21 5:27 PM, William F Pokorny wrote:
> On 5/14/21 3:59 PM, Bald Eagle wrote:
>>
>> Just went through povr installation with jr.
>> Got the biscuit render, so looked good.
>>
>> Went into qtpovray38 distribution directory to test a .pov file.
>>
>> bald@Alienware ~/povray38/scenes/objects $ povr blob.pov
>> /home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: 5:
>> /home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: Cannot fork
>>
>> bald@Alienware ~/povray38/scenes/objects $ povr chess.pov
>> /home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: 5:
>> /home/povray/povray-3.8.0-x.povr_b0bcf6c0/bin/povr: Cannot fork
>>
>> Any ideas?
>>
> Just guesses.
>
> The 'cannot fork' stuff usually happens when a unix/linux system has run
> out of resources to start another process. Lack of memory/swap space,
> /tmp, /var - maybe. There are sometimes limits on the number of
> processes in place, but this in my experience is not all that common.
>
> For starters you could try running the 'df' command and look at disk
> usage. The command 'free -m' can be useful too.
>
> Somewhere you posted about your machine... Let me try and find that and
> I'll post back here if I see anything obvious.
>
Machine stats you posted look good.
---
Had a few other thoughts meanwhile.
- What was your entire command line with flags & ini ?
- If just say the blob scene you can probably point to a normal povray
v3.7/v3.8 install and see if you get the same problem.
- You could run the povray command directly. One executable would be in
the unix directory below where you issued make:
<theMakeDir>/unix/povray blob.pov
- You could copy the blob.pov scene to where you ran the biscuit scene
and try and run it there.
- You could try (a) to cut the resources required by using the '+wt1'
and '-d' flags. I guess in that vein you could just run 'povr --help'
while in the 'problem' directory.
- You could try and start another process than some form of POV-Ray -
say gimp - from within the povr script when you are seeing the error.
- While in the directory with blob.pov, if you run 'ls -l blob.pov' does
it show the read bits set on the left. Something with 'r's like:
'-rw-rw-r--'
... Yeah, guessing. It's been many years since I've seen that error.
Bill P.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
William F Pokorny <ano### [at] anonymousorg> wrote:
> > The 'cannot fork' stuff usually happens when a unix/linux system has run
> > out of resources to start another process. Lack of memory/swap space,
> > /tmp, /var - maybe. There are sometimes limits on the number of
> > processes in place, but this in my experience is not all that common.
> >
> > For starters you could try running the 'df' command and look at disk
> > usage. The command 'free -m' can be useful too.
> - What was your entire command line with flags & ini ?
Er, for what?
bald@Alienware ~/povray38/scenes/objects $ povray bezier.pov
povray: cannot open the user configuration file
/home/bald/.povray/3.8/povray.conf: No such file or directory
Persistence of Vision(tm) Ray Tracer Version 3.8.0-alpha.10064268.unofficial
(g++ -std=gnu++11 5.4.0 @ x86_64-pc-linux-gnu)
This is an unofficial version compiled by:
BaldEagle <cre### [at] netscapenet>
The POV-Ray Team is not responsible for supporting this version.
<snip....>
----------------------------------------------------------------------------
Parser Time
Parse Time: 0 hours 0 minutes 0 seconds (0.087 seconds)
using 1 thread(s) with 0.012 CPU-seconds total
Bounding Time: 0 hours 0 minutes 0 seconds (0.000 seconds)
using 1 thread(s) with 0.000 CPU-seconds total
Cannot create render state output file.
Cannot create render state output file.
Render failed
> - You could copy the blob.pov scene to where you ran the biscuit scene
> and try and run it there.
The biscuit scene automatically rendered as part of the install process.
> - You could try (a) to cut the resources required by using the '+wt1'
> - While in the directory with blob.pov, if you run 'ls -l blob.pov' does
> it show the read bits set on the left. Something with 'r's like:
> '-rw-rw-r--'
bald@Alienware ~/povray38/scenes/objects $ ls -a -l
total 288
drwxr-xr-x 4 root root 4096 Jun 21 2020 .
drwxr-xr-x 20 root root 4096 Jun 21 2020 ..
-rw-r--r-- 1 root root 1463 Jun 21 2020 bezier.pov
-rw-r--r-- 1 root root 955 Jun 21 2020 blob1a.pov
-rw-r--r-- 1 root root 1246 Jun 21 2020 blob1b.pov
-rw-r--r-- 1 root root 1077 Jun 21 2020 blob1c.pov
-rw-r--r-- 1 root root 1023 Jun 21 2020 blob.pov
-rw-r--r-- 1 root root 1105 Jun 21 2020 chesmsh.pov
-rw-r--r-- 1 root root 115136 Jun 21 2020 chess.inc
drwxr-xr-x 2 root root 4096 Jun 21 2020 csg
-rw-r--r-- 1 root root 1522 Jun 21 2020 fractal1.pov
-rw-r--r-- 1 root root 2349 Jun 21 2020 fractal2.pov
-rw-r--r-- 1 root root 1405 Jun 21 2020 fractal3.pov
-rw-r--r-- 1 root root 1277 Jun 21 2020 fractal4.pov
-rw-r--r-- 1 root root 5082 Jun 21 2020 isosurfaces.pov
-rw-r--r-- 1 root root 1499 Jun 21 2020 lathe1a.pov
-rw-r--r-- 1 root root 1518 Jun 21 2020 lathe1b.pov
-rw-r--r-- 1 root root 1555 Jun 21 2020 lathe1c.pov
-rw-r--r-- 1 root root 3198 Jun 21 2020 lathe2.pov
-rw-r--r-- 1 root root 5357 Jun 21 2020 mesh2.pov
-rw-r--r-- 1 root root 1257 Jun 21 2020 polygon.pov
-rw-r--r-- 1 root root 4104 Jun 21 2020 primitiv.pov
-rw-r--r-- 1 root root 5344 Jun 21 2020 prism1.pov
-rw-r--r-- 1 root root 5388 Jun 21 2020 prism2.pov
-rw-r--r-- 1 root root 1354 Jun 21 2020 prism3a.pov
-rw-r--r-- 1 root root 1810 Jun 21 2020 prism3b.pov
-rw-r--r-- 1 root root 1597 Jun 21 2020 prism3c.pov
-rw-r--r-- 1 root root 571 Jun 21 2020 prism3.inc
drwxr-xr-x 2 root root 4096 Jun 21 2020 quartic
-rw-r--r-- 1 root root 3561 Jun 21 2020 sor1.pov
-rw-r--r-- 1 root root 2617 Jun 21 2020 superel1.pov
-rw-r--r-- 1 root root 2629 Jun 21 2020 superel2.pov
-rw-r--r-- 1 root root 4247 Jun 21 2020 superel3.pov
-rw-r--r-- 1 root root 4963 Jun 21 2020 torus1.pov
-rw-r--r-- 1 root root 1759 Jun 21 2020 torus2.pov
-rw-r--r-- 1 root root 1342 Jun 21 2020 ttf1.pov
-rw-r--r-- 1 root root 1821 Jun 21 2020 wtorus.pov
I have qtpovray running on this machine, and it works just fine - rendering a
scene file. But not THOSE scene files. Couldn't make the state file.
Just did chown -R bald:bald povray38 on that whole parent directory, and
qtpovray renders those, but povr now gives me --- no output whatsoever.
bald@Alienware ~/povray38/scenes/objects $ povr blob.pov
bald@Alienware ~/povray38/scenes/objects $
no new image file in that directory after silent command
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 5/14/21 6:31 PM, Bald Eagle wrote:
> William F Pokorny <ano### [at] anonymousorg> wrote:
>
>>> The 'cannot fork' stuff usually happens when a unix/linux system has run
>>> out of resources to start another process. Lack of memory/swap space,
>>> /tmp, /var - maybe. There are sometimes limits on the number of
>>> processes in place, but this in my experience is not all that common.
>>>
>>> For starters you could try running the 'df' command and look at disk
>>> usage. The command 'free -m' can be useful too.
>
>> - What was your entire command line with flags & ini ?
>
> Er, for what?
No idea. I'm fishing for cause. :-)
The 'no output at all' you now see makes as little sense to me - as a
likely problem - as did the unable to fork message.
The cannot create state file is what I got too - with 'povr' - when
trying reproduce your fork issue last night and not having permissions
to write to the current directory.
As I expect you know by now, directories installed system wide as
packages or with 'make install' / 'sudo make install' almost always have
an owner of root. Further they are read only except for the owner. Those
permissions should be left as is to prevent accidental corruption.
Normal users - even those able to play as 'root' should work elsewhere
in directories and with files they do own.
The povr script ships in the root build directory but must be modified
and installed somewhere in your PATH. If you type the command:
'which povr'
what do you see?
I see: /home/pokorny/bin/povr because I copied the povr script to my
/home/pokorny/bin directory - which is part of my PATH environment
variable by default. Plus, I updated the contents of 'povr' to point to
the build location directory. You can point the 'povr' script to an
install directory too, if you configured with
--prefix=<eventual_install_dir> before building and after a successful
'make install'.
'echo $PATH'
shows the contents of the search path used for executable commands.
You can execute commands with the full path assuming the permissions are
set for an executable. For example, no matter the current directory
permissions, I can type:
'/home/pokorny/bin/povr --version'
and see in the output:
...
POV-Ray (povr) 3.8.0-x.povr_19812865.unofficial
This is an unofficial version compiled by: wfp
...
Further, such a full path command should run in any directory no matter
permissions so long as the permission bits are set correctly. When I use
the command:
ls -lt /home/pokorny/bin/povr
I see:
-rwxr-xr-x 1 pokorny pokorny 806 May 14 17:53 /home/pokorny/bin/pov
Let's start the debugging with the commands above.
>
...
>
Aside: As a general aim I'm thinking most unix/linux users should
install packages maintained for their OS distributions.
For those wanting to play with code branches, a goal of the povr branch
is to NOT use or need 'root' or 'admin' authority at all. Further, to be
able to run as many versions side by side as they wish in local,
smallish, directories without cross interactions - something long an
issue. There are tentative code changes in the povr branch to support
this aim. For more see, for example:
https://github.com/POV-Ray/povray/issues/374
The povr script as shipped is a template, and for users in the playpen,
like me, I have many such scripts pointing to different versions. Where
'povr' as a command is how I execute my most current code and compile.
Bill P.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |