 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Christian Froeschlin <chr### [at] chrfr de> wrote:
> For examples, rendering the sample advanced/abyss.pov with
> 640x480 AA 0.3 took about 3 minutes with 3.6 on my system,
> but 9 minutes with 3.7 in the SSE2 version using thread
> count 1.
Have you tried with a scene which doesn't use media? For example if
I render biscuit.pov at 1024x768 without antialiasing I get 19 seconds
with POV-Ray 3.6 and 16 seconds with 3.7.
IIRC there was some development still going on with the media code.
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Did anyone try to reproduce the effect that 3.7 seems to be
> significantly slower (read: more than any reasonable overhead)
> on some single core systems (in my case: P4 2.6 WinXP)? This is
> of course not urgent for beta26, but if the final version has
> the same behavior, it will seriously deter some people from
> using the new version. I realize the real benefit of 3.7 is
> for multi-core systems, but it suppose it is also in your
> interest that all users work with the newest version.
>
> For examples, rendering the sample advanced/abyss.pov with
> 640x480 AA 0.3 took about 3 minutes with 3.6 on my system,
> but 9 minutes with 3.7 in the SSE2 version using thread
> count 1.
In some features, 3.7 *is* slower than 3.6, for example due to global
caches that were used before, but aren't thread-safe. It's not that big
of a slowdown, though. Also, the tiled rendering used for multithreading
causes extra antialiasing samples (on the tile edges). Is there a way to
change the tile size yet?
Of course there were also optimizations to other features.
In any case, 3x slower doesn't look like just "overhead", that's a lot!
If you see any important slowdown on a scene *without* media (see Warp's
post), *that* would be a problem worth reporting.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Nicolas Alvarez <nic### [at] gmailisthebestcom> wrote:
> > 1) Lost transformations under certain circumstances. Usually happens
> > when unioned, clipped_by objects are transformed with either transform{}
> > or Point_At_Trans(). Other conditions produce bad results as well. I can
> > make a more detailed post about this if necessary.
>
> Test cases is what they surely need :)
Provided in
http://news.povray.org/povray.beta-test/thread/%3C47a84bb1%241%40news.povray.org%3E/
Possible fix in
http://news.povray.org/povray.beta-test/thread/%3C47a1714a%40news.povray.org%3E/
Tom
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp wrote:
> Have you tried with a scene which doesn't use media? For example if
> I render biscuit.pov at 1024x768 without antialiasing I get 19 seconds
> with POV-Ray 3.6 and 16 seconds with 3.7.
Yeah, biscuit.pov seems fine here too. However, for grenadine.pov,
which also does not use media, I got 1:59 on 3.6 and 2:24 for 3.7
rendering 640x480 AA 0.3. Not so bad, although still noticable.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Christian Froeschlin <chr### [at] chrfrde> wrote:
> Yeah, biscuit.pov seems fine here too. However, for grenadine.pov,
> which also does not use media, I got 1:59 on 3.6 and 2:24 for 3.7
> rendering 640x480 AA 0.3. Not so bad, although still noticable.
Did you look if the difference is in the photon shooting time or in
the tracing time? Here I get slower photon shooting but faster tracing
with 3.7 than with 3.6.
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp wrote:
> Did you look if the difference is in the photon shooting time or in
> the tracing time? Here I get slower photon shooting but faster tracing
> with 3.7 than with 3.6.
No distinct effect, both phases were slower, and the factor
was similar. BTW, the render window in 3.7 now opens before
the photon shooting is done, was that intentional?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Cason wrote:
> I'll be working on getting a beta 26 out sometime in the next few days, no
> guarantees since - as always - real life comes first, but hopefully it'll
> be up by early next week.
Unfortunately I missed that target. I've had a out-of-hours project bugging
me for the past few months which chews up all my time; I had hoped to have
it done last weekend. Sigh.
On the up side, when it is done, I'll have about 15 or so hours per week
free that I can put back into POV-Ray, so some things will start to get
done again.
Thanks for the replies to this thread: comments have been taken into
consideration. I don't expect to be making any large changes or fixes in
the next beta; mainly it is the smaller bugs that I'd like to reduce the
number of in order to get us closer to correct functionality. In
particular, radiosity isn't getting touched until everything else is done
(at least not by me anyhow), and photon issues I'll defer to Nathan on.
The other speed issues (i.e. ones not related to photons or radiosity)
bother me a lot and I will look into them as my first priority once I have
the next beta out.
thanks for your patience,
-- Chris
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Cason wrote:
> The other speed issues (i.e. ones not related to photons or
> radiosity) bother me a lot and I will look into them as my first
> priority once I have the next beta out.
only media seemed to be really bad, without that it was at
most 25% slower, which isn't too terrible considering the
same scene was faster on other systems, and will in any
case have a significant speedup on multiple CPUs.
Also, the README in the beta sources mentioned you used
the Intel compiler for 3.6 but not for 3.7, which makes
it hard to compare the speed with respect to actual code
changes. I'd expect icl to give better results than cl,
especially on Intel CPUs like mine ;)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
FYI beta 26 for Windows will be uploaded in the next day or so.
-- Chris
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Cason wrote:
> FYI beta 26 for Windows will be uploaded in the next day or so.
This is now available, with the exception of the 64-bit version, which is
delayed due to a system crash.
The Linux version ought to follow soon. Once that's up, I'll update the
source code archive.
-- Chris
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
