 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I noticed some errors in my week-long render, after stopping and then
restarting the render. You can see in the image that in a few spots
pov-ray did not completely trace the image. I'm not sure what the final
image will look like. Will it have blank holes? Will the holes be filled
in with the coarse pre-trace stuff? I dunno.
Mike
Post a reply to this message
Attachments:
Download 'continue_render_error_01.png' (1336 KB)
Preview of image 'continue_render_error_01.png'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 2/2/2016 11:27 PM, Mike Horvath wrote:
> I noticed some errors in my week-long render, after stopping and then
> restarting the render. You can see in the image that in a few spots
> pov-ray did not completely trace the image. I'm not sure what the final
> image will look like. Will it have blank holes? Will the holes be filled
> in with the coarse pre-trace stuff? I dunno.
>
>
> Mike
Oops. I forgot that the error on the far left was my mistake and not due
to a bug.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 03.02.2016 um 05:27 schrieb Mike Horvath:
> I noticed some errors in my week-long render, after stopping and then
> restarting the render. You can see in the image that in a few spots
> pov-ray did not completely trace the image. I'm not sure what the final
> image will look like. Will it have blank holes? Will the holes be filled
> in with the coarse pre-trace stuff? I dunno.
Well, you /could/ wait and see...
I need to ask once again what version of POV-Ray you are using; I recall
that there had been problems with individual blocks not being rendered
properly after continuing a render, and while I can't recall the
details, to all current knowledge these have been sorted out by now.
What you see might just be a case of a block taking significantly longer
time than the others, exaggerated by the abort and continue (because
incomplete blocks in an aborted render are started from scratch in a
continued render).
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 2/3/2016 1:29 PM, clipka wrote:
> Am 03.02.2016 um 05:27 schrieb Mike Horvath:
>> I noticed some errors in my week-long render, after stopping and then
>> restarting the render. You can see in the image that in a few spots
>> pov-ray did not completely trace the image. I'm not sure what the final
>> image will look like. Will it have blank holes? Will the holes be filled
>> in with the coarse pre-trace stuff? I dunno.
>
> Well, you /could/ wait and see...
>
> I need to ask once again what version of POV-Ray you are using; I recall
> that there had been problems with individual blocks not being rendered
> properly after continuing a render, and while I can't recall the
> details, to all current knowledge these have been sorted out by now.
>
> What you see might just be a case of a block taking significantly longer
> time than the others, exaggerated by the abort and continue (because
> incomplete blocks in an aborted render are started from scratch in a
> continued render).
>
version 3.7.0.msvc10.win64
It may be a few days still before it is done and we can find out for sure.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 03.02.2016 um 20:41 schrieb Mike Horvath:
> version 3.7.0.msvc10.win64
>
> It may be a few days still before it is done and we can find out for sure.
Having dug though the old records, I'm now convinced that there is
nothing to worry about - even if those squares will never show properly
in the preview window. There is a known (but apparently mostly
forgotten) problem where a render abort-continue will mess up the
preview, but measures have been implemented to ensure that the resulting
image will be ok.
See http://bugs.povray.org/task/270 for a bit more details.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 04.02.2016 um 00:32 schrieb clipka:
> Am 03.02.2016 um 20:41 schrieb Mike Horvath:
>> version 3.7.0.msvc10.win64
>>
>> It may be a few days still before it is done and we can find out for sure.
>
> Having dug though the old records, I'm now convinced that there is
> nothing to worry about - even if those squares will never show properly
> in the preview window. There is a known (but apparently mostly
> forgotten) problem where a render abort-continue will mess up the
> preview, but measures have been implemented to ensure that the resulting
> image will be ok.
>
> See http://bugs.povray.org/task/270 for a bit more details.
I have to correct myself: Your resulting image may /not/ be fine if you
use the official 3.7.0 release version.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Yes yes, these continued updates from CLipka are the best to prevent
panic and fear... :-D
I can imagine, to develop such a complex program like POV-Ray is a hard
task. Lots of difficult programming code. CLipka and the others do a
great job. If only I could develop in C++, I would join them.
I think, if the final image has some errors, I would - just as fix -
re-render only those areas afterwards and use a graphic program to merge
them into the final image.
On 03.02.2016 19:29, clipka wrote:
> Am 04.02.2016 um 00:32 schrieb clipka:
>> ...but measures have been implemented to ensure that the resulting
>> image will be ok.
>
> I have to correct myself: Your resulting image may /not/ be fine if you
> use the official 3.7.0 release version.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I finished the render, and the errors exist in the completed image as well.
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
