Personal View site logo
Make sure to join PV on Telegram or Facebook! Perfect to keep up with community on your smartphone.
Please, support PV!
It allows to keep PV going, with more focus towards AI, but keeping be one of the few truly independent places.
Driftwood Cluster X series 3:│moon T7│ЅріzZ T6│Nebula T7│Drewnet T9│
  • A 'cluster' of different GOPs (Group of Pictures) forms the basis of the Cluster X series - to boldly see how far we can push each GOP factor in terms of superb looking artifact free Intra & prediction based Longer GOP (> 3) out of the panasonic cameras - where a natural scale of compression and consumed bitrate begins to appear according to its GOP size.

    What's so good about all these new Cluster X settings? In simple terms, after looking at the rate control, quantisation and scaling matrices found in the GH3, Nikon D800 and Canon 5DMKIII I decided upon a radical approach to find out if the GH2 was capable of getting near their very low QP values (which range from predominantly 1 to 30 depending on the detail and luminance of the subject).

    What I found was (and this had me concerned before) was that Panasonic's implementation / or ptools labelling!) of rate control using the GH2's quantisation parameter isn't too clear. I needed a way of having a minumum Quant and a desired maximum quant. Strangely with Initial Quantizer = 30, Quantizer for 1080 modes = 3 (+-2 can offer a QP of 1 !) and Quantizer for 720 modes = 30 I could sortof achieve this!!!!

    With 720 using B frames (switched on in the 720 encoder 1 settings as = 3) Quantizer for 720 modes seems to work as a highest Quantisation setting (high is worse, low quant is best) together with Initial Quant of 30 (don't worry about such a high starting value here - after the first frame it settles straight back to lowest achievable (best) for the scene. Subsequently, Quantizer for 1080 modes = 3 gives me my lowest desired (best quant) for each opt 24p, 1080i, HBR and 720 modes.

    This quantizer rate control now seems to work as in newer High Profile Level 5 (5.1) cameras such as the 5DMKIII and Nikon D800 (and seemingly the GH3) and when matched with a new softer looking matrix - as analysed on the Canon 5DMKIII (with adjustments) - brings a degree of softening to the over sharpe Pany lenses.

    I'm also sure that 1080i/FSH?HBR modes work better with half the coded picture buffer size of 720p modes (indeed I switched off 720 Bottom setting (the buffer amount as in automatically works it out) whilst offering a fairly low buffer in bitrate settings in 'patches for testers'.

    In summary, the range of QP will be (certianly in 1080p24 modes) from 1 to 30 - often staying very low (less than 10) all round. I believe (although only tested by myself) that the result offers extremely good looking 720p modes (probably the best ever), certainly superior 24p with Intra & Long GOP, and great looking HBR/FSH and iA mode control (though further strenuous tests may well be required). High dynamic functions work it seems too.

    All the new Cluster X series settings for each GOP attached below have been tested on the notoriously sharp Pany OIS lenses. Give me feedback.

    Its as simple as A, B, C, D, E, F ! Each .ini setting in ptools for Cluster X Series will equate to setA ('DREWnet' 12/15 GOP), setB ('Nebula' 6 GOP), setC ('Spizz' 3 GOP), setD ('Moon' 1 GOP/3 GOP). Any updates will stay on these sets.

    "Cluster X Series isn't for pussies!" Driftwood 2013

    If you enjoy these settings please don't forget to donate to personal-view by clicking on the link above. Gratitude and thanks to Vitaliy Kiselev.

  • 1811 Replies sorted by
  • 1 GOP Intra 'moon' T7 - Top Grading - Best Motion - Best Setting Ever? (You decide)

    Update News (updated 21/07/2013) Two months after the experimental moon T6 and lots of new research I present you 'moon' T7.

    UPDATE: Both variations of moon T7 (24p & 23.976p) are included in the one zip below. Both also include the slightly adjusted 720p60 for stability in iA modes (thanks @flaschus ).

    • INTRA 24p minimum 130-147Mbps - best looking ever with extreme Quant values - period.

    • INTRA 25p/FSH50i at 80Mbps.

    • 3 GOP 30pHBR/FSH60i at 72Mbps+.

    • 720p modes minimum 50-70Mbps max - best ever 720 setting? ;-)

    • new softer matrix - targeted at Pany lenses.

    • iA modes working it appears.

    • iDynamic mode on 24p best on Low.

    • Great sound from per (Per Lichtman of Pasadena Pulse AKA thepalalias!!!!) ;-)

    • Includes the long-time best looking mjpeg setting and greatest Quant.

    • Choose between setd 24p, 25p/50p, 30p & 60p or sete 23.976p, 25p/50p, 29.97p & 59.94p frame rates.

    • Please don't forget to donate to personal-view and help this site continue.

    An early version of moon T7 is on the other thread, just going over a few tiny little changes to improve iA mode stability - released here shortly.

    Here's Cluster X 'moon T5' and 'Nebula' in action courtesy of the boys at 'DeployAerial' ( @DeployAerial );-

    Latest moon T7 variations inc. setd 24p and sete
  • 3 GOP 'Spizz' - Hi-Quality - Pro Motion

    Update News (19/08/2013): Spizz Trial 6 is released!

    3 GOP expert Katsu (aka @bkmcwd) and r&d PAL specialist @towi put together some ongoing ideas of theirs and mine in this new setting of the experimental Spizz series. We've chucked out so many variations before deciding to release this that I was beginning to think T6 will never be realised! Anyways, what we've got is the best looking 3 GOP in areas of quantisation, bitrate spread and measured performance for all 1080 modes. 720 is 6 GOP.

    There's some unusual values in the Top bitrate settings - an experiment based on a Texas Instruments formula of 'maxbitrate is 1.5 times that of actual bitrate' under the current rate control - but everything is allocated in AVCHD Compression accordingly and close to what you can expect.

    For the first time we see the 1080i matrix adapted more towards the left side (not just top left) as an interlaced slice is scanned differently to progressive slices which I believe improves the spread of QP values and DC coefficient range in the low frequencies. Certainly the results on Elecard look good.

    Ive tried hard to make this work with iA modes or iDynamic and though it passes my tests some of you may find that it may / or may not pass yours. So trial away.

    Spizz T6 on average offers similar bitrate consumption to previous Spizz releases and settings like Flow Motion but is much more constant - qualified by the higher QP values without any lineage breakup in the overall picture.

    • 24p 100Mbps average can go much higher if pushed. The best QP around (below QP10 for most pics).

    • 720p modes on 6 GOP and refined to work with iDynamic/iA modes - 55Mbps avg

    • 1080i/HBR modes around 68Mbps avg finely measured matrix and quantisation. (QPs below 10!)

    • Great sound from Per Lichtman of Pasadena Pulse AKA @thepalalias

    • The best mjpeg mode from Driftwood (anamorphic users should experiment with their own frame sizes)

    Don't forget to donate to Vitaliy's link above if you use this setting.

    GOP3 Spizz T6 versus Flowmotion v202

    Cluster X - Spizz T6 Driftwood-Katsu-Towi
  • 6 GOP - Middle Earth 'Nebula'

    'Nebula T7' (Released 24/07/2013) Including all recent findings in moon, even lower quantisation and improved GOP related stuff.

    • 1080p24 at 86+Mbps avg

    • 1080iFSH50/60/HBR25/30 at 50+Mbps

    • 720p50/60 superior setting at 60Mbps - this could even beat moon's 720 modes for quality.

    • new softer matrix - targeted at Pany lenses.

    • iA modes working it appears.

    • iDynamic mode on 24p best on Low.

    • Great sound from Per Lichtman of Pasadena Pulse AKA @thepalalias

    • Includes the long-time best looking mjpeg setting and greatest Quant.

    Here's an early 'Nebular' 6 GOP video courtesy of Ernesto Mántaras ( @charlie_orozco ) - Has anyone ever shot a video on Cluster X Nebular/ not many of you like GOP 6 it seems...? ;-

    Nebula T7 6 GOP mid bitrate superior quality
  • 12/15 GOP 'DREWnet' T9 - Traditional Long GOP 12 (PAL) / 15 (NTSC)

    Update News (Monday 22nd July) DREWnet T9 released today. Including the findings of moon T7 (read above) You'll get a QP of 1 throughout many less busy pictures but most pictures will resolve sub QP of 10. Death charts code at a quant of around 15-16 average!

    • Superior Quantisation

    • 80Mbps constant in 24p modes for exceptional quality.

    • 1080i/FSH Maybe only around 40Mbps but the quantisation and PSNR is brilliant!

    • 720p modes Just the best there is.

    • iA and iDynamic modes simply work!

    • MJPEG mode - best available (like moon)

    • Per Lichtman stunning audio resolution.

    To download Click on the file icon link below labeled 'DREWnet'

    Here's Cluster X 'DREWnet T8' in action courtesy of Matteo Sacher ( @microfourthirduser ):-

    Driftwood - DREWnet T9 12-15 classic with 6-12 GOP 720p modes -
  • Latest News Here soon.




  • Continuing with some of the "motion blur" talk from series 2, I've found this is as well in my first tests of Moon 7. I always test first with static, well lit and properly exposed shots of me talking, and I'm about as animated as most humans but I do "talk with my hands" as they say, and the blur is noticeable as I'm hammering home a real earth shattering point. I transcoded to Cineform, and then rendered out to MP4 with Vegas Pro, avg 25 MB, peaking 50. Shooting w/ Lumix 20 and 14. But then again, I've also become accustomed to shooting at 1/40 with GH2, and I'm using the same 45 MB/s cards as danielortega is, which could be part of the problem. (?)

    Apart from that, the setting is really nice. Viewing on my JVC HD projector at 120 inches I wouldn't say it's too soft. I'm testing out various settings for a project I have coming up, and honestly I'm still quite a fan of the original Intravenus with the '444 Soft', and I like stuff in focus, John Ford/William Wyler depth of field etc, and it holds up pretty darn nice. Anyway, if my tests weren't me talking like an idiot about the Spring Breakers Blu Ray I'd put it up, but that'd be too embarrassing, so I'll keep testing out some stuff, including 1/50 shutter, and the 'B' and 'C' trials, etc. Again this isn't much of a complaint, these settings are terrific and the work is appreciated big time.

  • @driftwood Here is a quick edit of the full moon over London filmed with Moon T7

  • Yay! Thankyou @WarLightUK Love it. :-)

  • ok Mr. driftwood, now create some magic for poor mans SD cards, i have SDXC 64GB Transcend....have i become annoying :\

  • @warlightUK what lens? and did you mess with the colors at all to get the moon to look more white? or was the moon actually that color temp?

    @driftwood 7c and 1/50 looks a little better, I see a few clips that for some reason appear to having a "slow motion" playback. it appears less actual "motion" blur (probably thanks to 1/50) but it's more looks like a crazy frame rate/slow motion? and my pc shouldn't be lagging in playback via VLC, 32gig of ram and a geforce gtx 690 overall from a-c was huge improvements, I'm going to grade some stuff and render it out to a more natural format and see if the playback looks a little cleaner.

    moon 7c overall is probably the best setting since sedna, intravenous, and moon t3&5. ALL of the settings are really nice and tons of hard work was put into each, but those 4-5 settings I just mentioned just have something extra special about them : )

  • @dado023 Did you try REDshift?

  • @GravitateMediaGroup The lens is Lumix 100-300mm... Didn't do any colour correction but the WB was set to 5200K

  • Yeah like GMG said the "blur" is kind of like step-printed slo mo, not motion blur per se.

  • @jeffharriger

    the more I watch the footage the more I think it's fine now. I really think this setting doesn't play well with 1/40, and either whatever @Driftwood did made a huge difference, or 1/50. Only way to really know is to shoot some 7a footage at 1/50 and compare.

    @matt_gh2 cinema, and I believe -2

    currently uploading some 7c footage

  • @driftwood to be honest i didn't, is it stable :) ?

  • Its as stable as you stand.

  • @driftwood but what is the real difference in video quality if both REDshift v1b and v1 have the same bitrate, i mean it has QP10 VS QP16, but what that gives in practical usage? disregarding incamera playback.

  • I shot my first tests of Moon Trial 7 today. I'm absolutely stunned!

    This is without a doubt the best, smoothest, cleanest, most organic looking footage I've ever seen come out of my camera. Nick, you are a god!

    Even at 1600 ISO, this is footage I could use without apology.

    My only complaint is the motion cadence seems to be a bit...odd. Even at 1/50 shutter. (And in my work, I do often need to go down to 1/25, so this is something that concerns me.) It's not quite as 'flowing' as I'd expect, there's almost a 'stuttery' look to the footage, like a frame is being skipped every now and then, or like the shutter was actually set to 1/1000. Frame by frame playback reveals no anomalies. It might possibly be my system, so I'm burning a Blu-ray, but Moon T5 footage showed no such cadence weirdness on the same system, so we'll see.

  • I think I might have found the cadence weirdness. Moon Trial 7 seems to be recording at exactly 24 fps, not the expected 23.976. Both Premiere Pro and KMPlayer report this 24 fps frame rate. When I added that to a 23.976 sequence, I got the weirdness. When I watched the clip in KMPlayer, it looked fine. When I created a 24 fps sequence, it also looked fine in PP.

    So the question: Is this exact 24 fps by design? I believe Blu-ray can handle it, and film is usually at exactly 24, so this might be a good thing.

  • @Jim_Simon I just tested moon trial 7c, and i'll have to agree with you. Quality is the best I"be seen out of the moon series. 1600iso looks really good. I'll make some better tests on the weekend. Thank you @driftwood Will be donating again soon!

  • @Jim_Simon When you shoot with other settings and then Moon Trial 7 right after without changing any other variables, do some of them get reported differently than others? Or could they be getting interpreted that way across the board? Ditto for framerates for each in KMPlayer (or VLC).

  • @driftwood whatever "motion" issues I was having I really don't see anymore, original clip or YouTube, both look pretty much the same, now all I get is just NORMAL 1/50 24p motion blur as expected when shooting handheld ; )

  • @thepalalias I went back and checked my other footage, including some recent Moon T5, and they all report 23.976. Moon T7 is the first I've ever seen report as exactly 24 in three different programs (now including VLC).

  • @jim_simon is it 24 in 7a b and c? maybe it's one of the changes nick made to fix the "motion" issue

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