Personal View site logo
100Mbps Flow Motion v1.11 Failsafe Patch with HBR 25p & 50p modes
  • 574 Replies sorted by
  • @LongJohnSilver With a Sandisk 30Mb/s card, Flow Motion will definitely span in 24L mode, and will span on moderate-bitrate scenes in 24H mode. With a 95Mb/s card, it will span reliably in both 24L and 24H modes.

    The Golden Gate bridge clip above, however, probably requires a high bitrate due to the active ocean waves. Flowing water is one of the most demanding challenges for the AVCHD encoder, as there are countless independently moving details in each frame.

  • Thank you Lee, Your work is of great value! I'm in the middle of something and I will try it as the hack and patch will be ready for the 25p (aka 1.1)

  • Would using a 95Mb/s card decrease the crashing in SH modes significantly? I've been experiencing crashes in SH mode 120FPS with 14-140 kit lens at f9.0 on a SanDisk Extreme 45MB/s outdoors pretty easily.

  • @jshzr I've only found 95Mb/s cards to make a difference in 4GB file-spanning. Stay tuned, however, I'll be releasing an updated verison of Flow Motion soon, with much more robust support for high bitrates in SH 720p modes.

  • @jshr absolutely not. 64mb 95sec helps with spanning. I've not used any setting that crashed with 30mbsec sandisk that didn't crash with 64mb 95sec sandisk.

    I wish it was different. But it's not. Just spanning. Sometimes.

  • @LPowell & @Chauncy Thanks for the info. Saved me some money :)

  • Hi, will this work with the new v3.64 PTool or does it require modification?

  • Flow Motion v1.0 is designed for Panasonic firmware v1.0 and can be used with either PTool v3.63 or PTool v3.64 when you load GH2__V10.bin into PTool.

    When you load Panasonic firmware v1.1 into PTool v3.64, a different set of patches is activated to support the new firmware features. Some of the v1.0 patches used in Flow Motion v1.0 are not available for use with the v1.1 firmware, which makes it incompatible for use with v1.1.

    I'll soon be releasing a Flow Motion v1.1 update which will make full use of Panasonic v1.1 firmware.

  • Great, looking forward to your update!

  • Ouch. I flashed FloMo v1.0 with PTool v3.64 and Panasonic firmware v1.1. No adverse effects so far, but then again I don’t know what I’m missing.

    Also looking forward to Flow Motion v1.1!

  • I did the same thing, except I loaded the Quantum patch. Then reflashed back to V11 stock. I hope I didn't screw up my camera.

  • No need to worry about using the v1.0 Flow Motion patch. While it's not optimized for v1.1 firmware, it won't do any harm to the camera.

  • Flow Motion v1.1 Patch updated with HBR 25p support

    See the first post of this thread for info and downloadable INI file.

    http://www.personal-view.com./talks/discussion/2099/100mbps-flow-motion-v1.1-update-with-hbr-25p-support/p1

  • Thank you for this sir .. this thread has explained more to me about the workings of my cameras than any other. Thanks again

  • @LPowell - been very happy running with your 75mbs patch on my GH13's for a while - now I have my GH2 and am about to hack with Flow Motion for 1.1, should I upgrade my Transcend Class 10's? Thanks in advance.

  • Thanks for update @LPowell, but with your new patch for FW1.1 my bitrate stays allways below 22 mbit in HBR mode and below 30 mbit in 24H Mode. Is that normal. I tested in normal conditions, filming my children, no deathchart.

  • Great news! Thanks a lot for this update! Flow Motion is my favorite patch for FW 1.0, so it's great that is works on FW 1.1 now :-)

  • Testing Panasonic GH2 modified with PTool 3.64 and LPowell Flow Motion v1.1 settings in HBR Lens used Lumix X VARIO PZ f3,5-5.6/14-42 Shutter speed 1/50 Mode: Smoth -2,-2,-1,-2. Bitrate around 37mbits. Thanks to @LPowell
    Original .mts file

  • @mrbill I've used a pair of 16MB Transcend Class 6 cards on both GH1 and GH2 for well over a year and found them to be second only the the Sandisk Class 10 cards. I'd definitely recommend trying the Transcend Class 10 with Flow Motion, they should be able to work in all modes. For assured file-spanning, I recommend using 24L, FH, and H modes.

  • Great - got loads of Transcend class 10 cards

  • Tomaso:

    What bread of dog is that? They are awesome!

  • @David_Cole - Nova Scotia Duck Tolling Retreiver, "Toller" the smalest retreiver. Yes they are very nice and very willing to work. Look att videos in www.kopppartrollets.se galleri:-)

  • LPowell, a while ago I told you that the B-frame quality was not matching the I-frame quality. I think I know what is going on with that now. The GH2 encoder encodes I-frame macroblocks that are dark with a lower quantization parameter than the encoder's base setting - up to 2 lower. Mids use the QP setting, and bright areas are up to 2 higher in the QP. So just shoot something that's dark and noisy (ISO 800 or 1600), and you'll see cleaner noise in the I-frames than in P or B frames. I can upload a Flow Motion sample showing this, if you like.

    Do you know a way around this problem? Nothing I've tried has worked.

    Can you explain your 24L trick mode? What does that do for spanning? I see that you have the 24L bit rate set below top, which is set below bottom. I could never figure out what the bottom setting does.

  • @balazer In I-frames, the GH2's AVCHD encoder typically assigns macroblock QP values in a +/-2 range around the base QP value. In P and B-frames, it most often applies the average QP value to the entire frame. However, I have seen it vary the QP value when encoding Intra macroblocks in P and B-frames. This happens most often in cases of rapid movement between frames.

    The algorithm it uses to determine macroblock QP appears to be rather complex. I'm convinced that it's not solely based on illumination, as I have seen brightly illuminated macroblocks encoded at a coarser QP than darker macroblocks which used a finer QP. I believe that image detail is a factor, as the encoder seems to use finer QP to encode low-contrast areas than it does with high-contrast details!

    At present, there's no way to alter the encoder's QP range algorithm, and it would no doubt require a new type of patch to do so. The visible results of this behavior are quite subtle and are in most cases too elusive to spot, since the average QP value is the same for I, P, and B-frames. Flow Motion minimizes these variations with its custom Scaling Tables, so it's less likely to be as much of an issue as in earlier patches.

  • @tomaso Thanks for uploading your Flow Motion sample video, your dogs are beautiful! This is a fine example of fast motion against a high-contrast, brightly illuminated background. These conditions often produce distracting flicker effects with camera movement and I'm very pleased with how clean and smoothly detailed the video looks. I recommend downloading the original MTS file to see its original recording quality,