Personal View site logo
Make sure to join PV on Telegram or Facebook! Perfect to keep up with community on your smartphone.
GF3 Stable Settings topic
  • 147 Replies sorted by
  • @stonebat Oops, my bad, it was actually the 44Mbps AQ4 patch that worked properly on the GF3.

  • @LPowell , Lee I sometimes get the no playback issue, but for some reason a power off/on clears it and all clips then playback fine. This is with the PV dog settings, which have otherwise been very reliable and I'm happy with the IQ. I originally had cbrandins 44 but like PV dog a little better. With it's low res sensor, I don't think there is much to gain with GF models with different patches other than reliability. Heck, why an I telling you, you know More about this stuff than I do.

  • @Rambo I was able to get most patches recording properly with the GF3, but it just doesn't seem to be able to play back high bitrate videos in-camera. Since it lacks 24p mode and is at heart a compact point-and-shoot camera, I decided that convenience and reliability were more important to me than maximizing bitrate.

  • gf2.jpg
    1900 x 960 - 1M
    gf2-CM.jpg
    764 x 500 - 146K
  • @ssh Not sure what your conclusions are, (debunking Sedna?) but my priorities for the GF3 can be stated objectively:

    • Records without fail.
    • Plays back in-camera.
    • Higher bitrates than stock firmware.
  • Lee, that playback fail was quite rare so it never worried me as my main use for GF cameras is set and run long takes as a B/C cam. I hear you on the 24p though.

    ssh, the exposures seem different in your samples but the detail in the PV dog setting was what impressed me when I tried it so I've looked no further. Maybe those 80+ bitrates are too much for the GF.

  • I too have only occasional playback errors, most (even higher bitrate) settings are OK. @LPowell What's your SD card? I know that I know nothing, just curious.

    My previous post was meant to illustrate @Rambo statement:

    "With it's low res sensor, I don't think there is much to gain though with GF models with different patches other than reliability."

    Not only there is no gain, the quality seems to get worse in most cases, and yeah GH2 resolves as much at 720 as GF at 1080.

    Canis Majoris (all flavors) introduces weird artifacts that I have not seen from any other patch. Just wanted to share. Not trying to debunk anything, none of the @driftwood 's stuff was ever intended for GF2, and from what I understand he pays little to no attention to anything but 24p which looks ridiculously good on GH2. I wish @driftwood would throw GF2/3 users a bone and develop something specifically for GF as he said he might.

  • I agree with @lpowell. The 44Mbps setting (actually 32Mbps for 1080i60 mode) is quite enough.

    It works quite well with compact m43 lenses on GF3. 14mm 2.5, 25mm 1.4, and x14-42mm. Compact and portable and hacked.

    398240_3710767684409_1132125665_3302155_2126790315_n.jpg
    960 x 717 - 110K
  • i am trying this out for first time. GF3 is my main photo camera so i want it reliable & stable. what would be the best setting to start with? 44mbps/AQ4 ? also does "stock GOP" mean i leave the GOP settings menu untouched? and do i pick the same bitrate for both FSH/SH & FH/H ? i have no clue what the diff is.. thanks

  • deshwasi - you can usually get little to no help here if you are a beginner. Trust me, I have tried.. I have GF3 myself, have increased AVCHD bandwidth 2x and changed AQ to 2. File size went 1.5x and saw little to no increase in quality. Can't really say the 1.5x file size increase is worth it (whatever IT is). Unless am doing something wrong.

  • @Stonebat what kind of mod is that on your GF3? Btw: Brown is so stylish! I regret getting the black. :P

  • Deshwasi, there is no need to alter any parameters in the settings, just load and use the tested ones like those in this thread.

  • Guys, try to keep normal usage of settings and patches words.
    As it just confuse people.

  • Would those who have hacked the GF3 recommend it as THE backup for a GH2 for video? All the Panasonic models and the naming convention seems quite baffling as to what has the highest quality video beyond the GH2.

  • @JeffGibbsTC With a pancake lens, the GF3 is very compact and its IA+ mode makes it a convenient point-and-shoot camera. Its video features, however, cannot match the manual control of the GH2 and the GF3's limited in-camera playback ability restricts hacking the bitrate to more than about 44Mbps.

  • Thanks for the input I was considering the Olympus OMD but that could be a problem with the 30p and limited bandwidth. I take my GH2 into all sort of difficult conditions and just need a backup, and the stealth factor of the GF3 seems great. If the GH3 was out now that would solve my problem--the GH2 would be the back-up. There is a great deal on the GF3 barely more than the 20mm pancake lens alone I think I will snap one up and give the patches here a whirl. By the way people keep RAVING about my GH2 footage with Sedna and CM night... mind blowing what's been opened up here for filmmakers.

  • Does anyone know what's the reason to hack GF3?

    I have tested 2 hight bitrate hacks against unhacked GF3 and there is no difference in resolution. Anything else that can improve?

    You can see PNG crops from my test samples (the last one is picture jpeg straight from the camera resized to fullHD and cropped to see the available resolution).

    This example is not most demanding scene, but I have observed the same (no resolution change) in more complex scenes.

    My test was done with: GF3 + m42 Revuenon manual lens 35mm F8, no filter, profile: landscape 0,0,0,-2

    2_sedna_AQ1.png
    1237 x 459 - 2M
    1_mysteron_burst.png
    1234 x 458 - 2M
    3_unhacked.png
    1229 x 457 - 2M
    4_jpeg_resized.png
    1231 x 453 - 2M
    0_unhacked.PNG
    746 x 412 - 902K
    0_mysteron_burst.PNG
    746 x 412 - 902K
    0_sedna_AQ1.PNG
    746 x 412 - 902K
  • @opek Anything mildly wide with a lot of details looks like shit on GF. Resolution stays the same with hacks, you will see a little less blocking with low bitrate hacks. My fav's at the moment are EOSHD and GOP3ZILLA. High bitrate patches are counterproductive on GF, see my post above in this thread and http://personal-view.com/talks/discussion/comment/57529#Comment_57529

    After running a GF2 + PL 25mm test against iPhone 4s I'm ready to throw GF2 out of the window. iPhone resolves 3 times more, GF's dynamic range is slightly better (bigger sensor) but with that resolution it's just not worth it. I really wanted to love my GF2 as an always-in-my-pocket "baby GH2" but it turned out to be a total waste of time. The only thing that GF had going for it for me was the size. Well, iPhone is smaller and GH2 with pancakes is tiny too. I'm done with GFs (at least for now).

  • @ssh I wouldn't replace GF with IPhone :). Most important is not only resolution, but also DoF one can get, low light capabilities and interchangeable lenses.

    It's a pity that the hack doesn't improve much but I still like filming with GF.

    Somebody should say here that GF is just a regular low budget m4/3 camera with nothing special in it and the hack changes almost nothing. It shouldn't be compared with pro-level GH2 which is in completely different league (especially with hack which works for it).

    P.S.: My (not expensive) full-HD camcorder has far better resolution, but no DoF, no lenses.. it's just something different.

  • Most important is not only resolution, but also DoF one can get, low light capabilities and interchangeable lenses.

    Use GH2 for that. Why bother with GF?

  • @ssh

    I don't have GH2.

    If I knew that GF has nothing special when it comes to video compared to other regular m4/3 cameras I wouldn't buy it. (but i though that hack can give GF little advantage).

    GH3 should appear later this year, so I'll wait for it and stick with my GF for now - it's ok for what I need it now.

    I agree with You, that for now only GH2 is real video master. GF is just a regular low budget camera which is not even above-average in terms of video quality - and it looks like the hack changes nothing.

  • @ssh @opek

    Check topic title please.

  • This is the hack for me as it is stable.

    I used a hacked GF3 using pvdogs patch for the GF2. You can find the patch here http://personal-view.com/talks/discussion/2321/gf2-stable-settings/p6 it is located towards the bottom of that page.

    I wanted to test it to the max... So I recorded until the battery died to see if it would span and to see if it would save my footage in the end and when I recharged the battery and turned the camera on it had saved all of the footage. So... Not bad. I wouldn't recommend using the gf3 for very wide shots as it won't pick up all the fine details like the GH2 but other than that this GF3 will be great!

    This clip is just a small portion of the footage taken in that 2.5 hour recording... I should also note... I used a Sandisk 64gb Ultra SDXC... "30 mbps" though it says write time is slower... it worked fine.

    Normal record time with the 64GB with the unhacked GF3 read 8 hours on the screen and after the hack it read 1 hour 25 minutes... HOWEVER!!!!!

    After 2.5 hours of recording it still had 45 minutes left.

    I was using the 14mm f2.5. on manual focus. And my living room is poorly lit, but the footage looks good to me. Also it was about 7PM and it was getting dark outside. SO I hope this helps. The GF3 is a good investment if you don't need manual controls... but be sure to set the flicker reduction to 1/50 to make sure your shutter speed isn't too fast.

    The one negative spot was that when I tried taking a photo while recording video as you can see during the shot of the grass I got one quick green frame.

  • Cluster v2e for GF users.

    Cluster v2e -6 GOP - with in cam pback and 2k mjpeg - setb.zip
    1K
  • Ok... I am having some second thoughts.

    Does anybody really notice a difference from their hacked stable settings vs. unhacked footage??? I am beginning to think there is absolutely no difference... not even in editing.