Personal View site logo
Cluster X: Series 1 - moon - ЅріzZ - nebula - drewnet - Slipstream
  • 1008 Replies sorted by
  • This is veri interesting, was made with Moon and Angenieux 12-120mm C Mount .

  • Yesterday I took a quick test with Moon trial 4. Everything works perfectly except the Ex-tele mode. There are flickers in the clouds in video with Ex-tele mode.

  • @driftwood Moon trial 4 rocks!!! 24p H/L - The quality is great! And it runs even with 30mbs(8gb) ultra SanDisk! I tried even with all the auto stuff on and iso 1600 still didn't stop. ex tele worked well too. The image is actually quite sharp with the the panny 20mm at f1.7

  • Hey Guys,

    Been following this thread for a while now and am excited by the prospect of 'Moon'. I installed the 'Trial 4' patch the other day and have got a couple of questions (That may have been answered elsewhere so please bare with).

    • Haven't tested other modes yet, but in HBR mode there is no in camera playback unless you switch the camera on and off. Is that normal with this patch? If it is normal, is there a way of fixing it?

    • There 'seems' to be a slight increase in rolling shutter. I say 'seems' based on same lens (Nikon 50mm ais), different patch comparison. It's probably just me being a little too shaky, but was wondering if anybody else experienced this?

    • Also, for a UK PAL user (HBR), which of the @driftwood patches --> Cluster X│moon | ЅріzZ│nebula │drewnet would you guys recommend for best quality & reliability?

    Many thanks.

  • @driftwood any update on a moon trial 5 release, or will it be considered "official" I'm thinking about doing a bmcc ProRes vs GH2 type of video but want to wait for whatever is considered the most official moon.

  • still fiddling with best quality 720p modes... there's something strangely wrong with the way 720 modes work.

    Pic: Here's the problem with Panasonic Zoom lenses such as the 14-140 lens when you attempt to zoom. As you can see everything as far as the bitrate/GOP/buffer is trundling along nicely until you zoom in - see the spike and change in frame sizes. 3 GOP B frames and even with P frames as seen below (p frame 720 is what the encoder was setup to use as default) forces the encoder to go skewith then fallback.

    gh2-basic-720pGOP3 with p frames - zoom problem on pany lenses.png
    771 x 291 - 28K
  • @pinger007

    Point taken, but hard lights are well for wides, not for close ups in interiors - like what burnet said, it's too harsh for skintone and details.

    it's not that hard to use 1/4 frost. i like harsh lighting for daylight interiors but tend to difuse those aswell!

  • @driftwood. I reduced the high end bitrate of the Moon (3,4) 1080i/60 and 720p60 from 93MB to 88MB and increased the low end from 48MB to 58MB with necessary adjustments to the top/bottom bitrate settings. Now FSH and SH modes are working fine with no errors even in full zooms with great quality. The FH/H are also giving me higher stable bitrates as backup if necessary.

  • Thanks for your report @Zaven13 Im attempting perfect 3 GOP 720p60 - painful but getting success at lower bitrates around the 72-77 mbps. By all means pm me your ini and lets take a look.

  • @driftwood
    Out of the following settings which allow spanning?

    24P H = yes or no
    HBR = yes or no

    24P H = yes or no
    HBR = yes or no

    24P H = yes or no
    HBR = yes or no

    24P H = yes or no
    HBR = yes or no

    PS = I'm using sandisk 64gb 95mb/s

  • Cheers Nick @driftwood.

    The spike on differential frames make sense because temporal redundancy is broken with sudden movement (and non linear on top of it) so it will take an extra encoding effort. Have you tried looking at frame info to see if this extra bitrate is texture or MVs?

    Now about those Intra frames... is it fallback? I don't know! Does it ever recover later in the footage? If not, and if E-PSNR isn't fluctuating, I would place some money on lens softness! Especially if this is the 45-200mm. More so if you were using MF, as you could be sending the lens OOF if you zoom a lot (DOF can be decreasing wildly, and so does content contrast).

  • I have settled it with opt2 tables and buffer balancing. moon Trial 5 is about to appear. It appears that GOP Related is neccesary to infinitely control the weighting against frame limit. Lifting frame limit (unchecked)and raising GOP Related weighting basically does the same thing.

  • I’m very happy with the QP in Moon Trial 4. It would be great if you could leave it at the same value, Nick!

  • My wee brother (13) has bugged me to let me use my gh2 for a year now and i finally let him borrow it. i loaded moon and this is what he came back with. super shaky footage but not bad for a first attempt. He edited the footage in videopad as he's not aloud on my pc.

    @driftwood just had a wee look at the videopad footage and there was lots of weird CC plugins. I took everything off and looks much better now! Think he might have run into a couple of traps such as iso bug....

    @mrbrill yeah i haven't seen much glasgow vids on here

  • *** NEWS **** Cluster X Series: 'moon' trial 5 Released today with stability improvements to 720p modes and FSH/HBR modes - and a balance of the best of Trial 3 & Trial 4 on 1080p24 - Go check it out on page 1. :-) BTW 720p60 is now fast motion 3 GOP like 720p50. This is worth the upgrade.

    @Mr_Moore You won't notice the change - its almost the same - promise.

    @Diffusion33 Good attempt by your bro' - However, even in 1080p mode on YouTube you can tell the rendering from Videopad / conversion to Youtube is breaking up - somewhere something has ruined the original footage.

  • Hey pal, thx so much for this incredible new setting 'm o o n' t5 ! Have a shooting tomorrow - will try that out.

  • @driftwood yeah the footage looks baked, i'll try and get hold of the MTS and see about editing them myself and re-posting.

  • @Diffusion33 - nice to see Glesga on the forums!

  • Cluster X Series: 'moon' trial 5 spanning tests

    All tests pointing at a TV screen with plenty of moving footage using a Pany 35-100 lens! I suspect as usual that datarate plays a big part in spanning.

    Test 1. 1080p24 H A unformatted sandisk 64gb 95mbps card with 40Gbs spare recorded at 1 hour and 22 mins and 13 secs and stopped on 'memory card full msg' - creating 8 x 4.29Gb files and 1 x 1.82Gb file before stopping - avg datarate of 56Mbps for the scene. Result: YES! Spans See clip53 in FCPX (which adds all the recorded clips together into one clip on import) test1 pic below

    Test 2. HBR25p A freshly formatted sandisk 16Gb 30mbps card at 10 mins and 2 seconds at 54,5Mbps average datarate and stopped at 4.29Gb - Result: No Span! I think I have a idea to make this span - maybe next trial (note: 30p HBR does span)

    Test 3. HBR25p A freshly formatted sandisk 64gb 95mbps card at 11 mins at 54Mpbs and stopped at 4Gb. Result: As test2, No Span! I think I have a idea to make this span - maybe next trial (note: 30p HBR does span)

    Test 4. 720p50 A freshly formatted sandisk 64gb 95mbps card at over an hour and 5 4Gb files + before battery ran out and correctly saved file and shut down. Result: YES! Spans.

    Test 5. HBR30p A freshly formatted sandisk 16Gb 30mbps card at 31 mins and 10 seconds at 2 x 4.29Gb files written. Result: YES! Spans to two files.

    Test 6. 720p60 A freshly formatted sandisk 64gb 95mbps card at 10mins 48 secs to one 4.29Gb file at avg 50.5Mbps datarate. Result: No Span

    test1 - Final Cut Pro X showing the 1 hour 22 mins footage in one file.png
    1673 x 1174 - 1M
  • moon trial 5 NO GRADE, very windy here today so my tripod got a few wiggles in a shot or 2

    lens - rokinon 85mm

  • @GravitateMediaGroup How do you achieve such flat image? Gh2 always gives me very contrasty footage... it´s because it was a cloudy day? Nice test... lots of detail and movement. Thanks!

  • I did notice when I shot a couple clips inside I got a "little" bit of noise in the shadows. I'm not sure if was more or less than T3 or 4, but it's nothing neat video cant handle if needed.

    is trial 5 pulling the same amount of detail as previous builds or more?

  • @danielvy what picture style are you using? I use standard, most folks are still caught up on smooth & nature

  • @GravitateMediaGroup Standard is my personal fav and the one I always recommend. Shouldn't be any more noise than normal - could well be the scene changing light/different settings etc... Best to check in an unchanging scenario and let me know which is best, TRIAL 3, 4 OR 5.

  • @GravitateMediaGroup I use Cinema -2, -2, 0, 0, ill try standard... Also. Tomorrow ill be testing Moon V5 in daylight conditions. Ill be posting some video.

This topic is closed.
← All Discussions
Start New Topic

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In with Google Sign In with OpenID

Sign In Register as New User

Tags in Topic

Top Posters