@karl Thanks for testing my setting! :-)

"Using this opportunity, I did some IQ comparison shots, I just posted them in the FlowMotion 2 thread."

I am looking forward to seeing the result very much.

"Regarding beta6: Are you sure that a higher bitrate for 720p60 is worth the risk to have the same problem that FM 2.01 still has? I mean: Can you actually see a better IQ for 720p60 in beta6 than in beta5?"

60p of FM2.01 was immediately frozen also in my test. BETA5 and BETA6 of me do not stop in the same test. For the moment in MY view, I think that around 53M is the limit of 60p. The efforts for raising IQ in the limit are continued. Although I have already begun the next improvement with another approach, I think that BETA6 is better than BETA5.

@LPowell

60p of FM2.01 was immediately frozen also in my test with pany 14-45mm F5.6 SS1/60 ISO1600. Why does only 60p of GH2 have so low tolerance?

@bkmcwd GH2 Memory. Not enough for 60 frames per second. That is why.

As soon as you go over Max GOP Size of 1,000,000 Bytes youre in troubled waters.

You have to try and divide up the frames in the GOP accordingly - and that means allocating x to i frames, x to P/b frames - its like dividing up the pie. Once youre happy with whatever frame limit/ FBs, etc... keep tweaking 'GOP Related' together with increments/decrements in quantisation and encoder setting 720p ... but notice when B frames look great, i frames begin to get lower!

Fun game eh?! ;-)

@LPowell Film mode was CINEMA[+2.-2.+1.-2] when it froze. Not with test chart but real fine textiles under the halogen bulb.

Although the frozen direct cause is not known, I think whether the bit rate was too high.

@driftwood

I realized now that the second half of your comment is added...

Although I have tried before the setting in which the B frame becomes large relatively, I am never pleased with I frame size becoming small too much. By such a reason, I am looking for the setting that the size of the I frame does not fall as much as possible, now.

Thanks again! :-)

@karl

Many thanks for detailed comparison test in FM2 thread! :-)

I think that the result may change by the difference in Deblocking Tables. I am just still going to improve further more. Please tell me the point which you noticed anything, thanks.

@kndkndknd

Thanks for testing my settings! :-)

i just returned from japan. I used your setting golgop3 natural for a bunch of stuff there so that I could conserve space compared to the higher bitrate driftwood settings. Your patch did very well. Thank you.

@ Silent users of my settings :-)

I release new "GOLGOP3-13-MAX_SednaAQ1C_Standard_BETA1" setting.

I think that recording will stop shortly after not making especially 720/60p moderate so far. Since the I frame became quite small with 6GOP in 720/60p, I returned to 9GOP.

Stability (don't stop during recording) is thought as important especially in my settings.

*Deblocking tables are substituted by your liking.

Although the performance under lowlight is not different from a normal version, the image quality in the scene which requires high stress improves in this MAX version. However, with MAX version, playback in camera may not often be available in 24p. This is because the maximum frame size which can be played back by GH2 may be exceeded. Of course, the file which cannot be played back in camera can also be satisfactorily played in PC. Supposing you consider that the playback in a camera is important, I will not recommend you this setting.　Please use normal version. However, IQ of MAX version is better than normal versions especially in high stressed scene, IMO.

Special thanks to @driftwood. :-)

Please try it!

*Of course, already combined with excellent audio tune @thepalalias's "Pasadena_Pulse_Audio_V2_Beta2"

NOTE:

*Although bit rate setups are quite high, the bit rate which actually appears are lower than these. Since the actual bit rate changes by the object to shoot, please confirm by yourself.

*24H and 24L are high quality 3GOP setting with super mega size I-frames.

*24p80%(30p) is available in 24L only.

*In PAL Mode, I especially recommend you use with SanDisk 95MB/s 64GB SDXC card strongly.

*In camera playback is sometimes not available in 24p. In other mode, Sometimes available after reboot the camera.(Not tested full yet, but probably)

*EX Tele works.(Not tested yet, but probably)

*Spanning in 24H, 24L with SanDisk 95MB/s 64GB SDXC Card. In other mode, need test.

*Spanning in other cards needs to be tested.

Please download original raw data from vimeo and check actual IQ.

@safari

Thanks for using GOP3ZILLA! :-)

@sohus

I recommend you GOP3ZILLA_standard and natural_standard in my settings. :-) http://personal-view.com/talks/discussion/comment/57882#Comment_57882

After a test I did with "cluster v3" today, I installed GOP3ZILLA-MAX_standard_V2_BETA7. I experienced a serious problem when recording 24H (at ~100MBit/s) using it: Within the same ~ 15 minute recording, there are many sequences of several seconds each (not just single GOPs) that have damaged image data in the lower part of the screen - see attached screen shot file.

I have not seen this particular phenomenon before, only isolated glitches within one single GOP.

(720p60 recording worked fine at ~50MBit/s.)

@karl

Thanks for testing my setting always! :-)

I think that your problem is a problem of the player or something in PC. Even if you encode data at a trial to the format which is different with editing software, does a problem occur?

@karl

OK, how about returning the Deblocking tables to the stocks at a trial? However, I cannot think solving, since such a phenomenon has not been experienced until now. :-(

Or is it over the frame size which can be played in your player? Although a problem is not in the data itself, the frame size of MAX versions becomes as large as it cannot often play back in GH2. The same situation may be also in your player.

I appreciate the continuing feedback.

@Karl ~ A StreamParser analysis of a section of the file (and posted to this thread) would help shed some light as to where the errors are occurring...