version 3: very slow preview on multi CPU/core macs?

forgot to add some details: when calculating the actual effect, version 2 takes 38 seconds, version 4 needs 40 seconds.

please see also a screenshot which i took during calculation, which shows the CPU activity: it looks like every 2nd core is not used (probably no hyperthreading). even if the rest is distributed evenly across the cores, the speed remains at a very slow level. can this be further optimized?

Quote
3dworks wrote: is it my configuration only that version 2 is actually performing much slower than version 2?

uhm, read this again.

Quote
3dworks wrote: forgot to add some details: when calculating the actual effect, version 2 takes 38 seconds, version 4 needs 40 seconds.

uhm, there is no version 4.

you are not the only one that have noticed and tested that FF 3.0 could be slower than 2.0, but I have found after that is not always and it depends on the configuration you choose and the filter choosen to make the test.

Here are some threads that probably you could find useful and interesting for this

See this thread here --> Comparison between FF2 and FF3 render speed test

From this thread is interesting this comment from GMM

Quote
GMM wrote: Spaceray, you should get the same results with FF2 and FF3 (or close to the same) if you set Progressive previews to Legacy and turn on preview cache preservation. These are the only two settings that make FF3 different from FF2. (note that I mean preview cache preservation and not the double-precision thingy on the Rendering tab – leave that off).

See this thread here -->What is important and priority for FF render speed ? NOT HARDWARE !

In this thread you can see that is not the same to test the speed with one filter or another as there are many variables and conditions to meet to test, and many different speeds depending the filter construction and the effect it does, size resolution, etc.

See this thread here --> FF Perfomance difference between Quad Core AND a Intel HT Quad Core ?

SIMPLE BENCHMARK TEST

Here I have made a benchmark test with FF 2.011 using the Creepy filter by Vladimir Golovin (I have choosen this filter as it is very CPU intensive and it does NOT depend on the source image as other, as this effect is generated by FF) and using default 2.0 options.

Intel i7 2600K 3.4 Ghz Quad Core with 4 HT cores (4+4)
16 GB 14900 DDR3 Special RAM
SSD SATA 3.0 6GB/s
Nvidia Geforce 9600 GT with latest drivers
Windows 7 Home premium 64 bit

Time is in minutes.

FF2 1000x1000 1.20

FF2 1200x1200 2.11

FF2 1400x1400 3.03

FF2 1600x1600 4.10

FF2 1800x1800 5:12

FF2 2000x2000 6:08

FF2 2600x2600 9.38

FF2 3000x3000 13:10

FF2 3600x3600 17.27

FF2 4000x4000 23:42

FF2 4600x4600 30.04

FF2 5000x5000 36.38

FF2 5600x5600 45.54

FF2 6000x6000 55.09

FF2 6600x6600 71.05