Driftwood Quantum X Settings, Series 2: Sedna A, B, C

Stray

@mozes yep thats a cadence problem, also, I don't see any useful info from your test, as you have said that you've merged the driftwood settings with your own. So its telling us nothing about the new matrix at all, as you have a completely different setting to any of us. Also, testing at ISO 4000 is fine for stress testing I guess, but I figure we're all interested in how it looks in practice, and none of us are going to shoot at ISO 4000 sensibly. Also the youtube encoding is awful and also tells us nothing, we would need to download the original source file.

rsquires

@driftwood @frame I confirm that pre hack Clipwrap 2.4.7 on OSX will convert the .mts to a .mov as a re-wrap not transcoding to Pro res with perfect playback etc. This is an HBR 25p file recorded with firmware 1.1. After hack to Orion Clipwrap will convert the file but you get choppy playback. If you view the original .MTS file in Movist and VLC yes it plays back fine, but Quicktime and Clipwrap have a problem. I spoke with the developers of Clipwrap and they say the wrapper that is there is not standard H264 so they can't do anything about it. 5D2RGB does a good job of transcode to Pro res, but it would be nice for space sake to have a .mov version of the h264 that played back ok.

mozes

just loaded Quantum X (v4c Orion) CBR with 'Dark Matter v3
And wow, why am i playing around when settings this good are here just for download.
i have done the same clip with the orion DMv3 it it looks so much better...

onionbrain

@Driftwood

My very initial, very preliminary reaction to Dark Matter v3 -- is that it's the best thing I've seen since 9b, and possibly the best thing I've ever seen on a GH2. That's very initial, very preliminary -- and I reserve the right to come back at you in a few days with a scathing reaction that concludes in praise of 9b.

I walked around a dimly lit room with a fast lens at ISO 500 and filmed something in every shadow I could find. A seven minute clip (24H) spanned perfectly on a Sandisk 64gig, 95 Mbps card and imported as one clip into Premiere Pro CS5.5. Some things were gorgeous, other things weren't. Neat Video didn't play nicely with the clip -- but it's too early for anything credible on that.

The stream was in the 143 Mbps range.

In the next few days I'll put this Dark Matter V3 through hell on my familiar test range. After that I can respond with greater accuracy.

rsquires

@onionbrain Like your style. Neat Video is the messiah of noise reduction in my opinion, it really is phenomenal. In what way didn't it like it?

rockroadpix

Tried HBR, 24H, & MJPG on V1. I liked the results. There's some nice smoothing going on, Loading up V3. Wine is open on the computer and on the desk!

Is it me or do these version ups take longer with these?

Oedipax

The firmware flashes definitely feel like they're getting longer! I haven't measured it though.

Loaded up Dark Matter v3, but I won't get to test it out until tomorrow. Thanks again Nick.

rockroadpix

Just tried V3.

It's official, I'm leaving my wife and kids and heading to some seedy area of Paris with V3 to live out my days Jim Morrison style! Very nice grain qual, @driftwood .

Hats off to you and Vitaliy. SUPERB.

onionbrain

@rsquires

Thanks!

It's still way too early for anything credible on that. This system has been on for two weeks and it's low on disk space. So, the problem could be on my end.

That said - I twice got an error I've never seen before and when it did run on two separate sub-clips the results were indistinguishable from the footage without Neat Video applied.

Again, it's too early for anything credible on that, and in retrospect I probably shouldn't have mentioned it.

thepalalias

Quick question:

I've just shot comparison exterior night footage (high ISO) for Orion 4c Dark Matter v1,2,3 and am about to shoot some Orion 4b. I would like to shoot Quantum v5b and v9b as well as SeAQuake V2 for comparison as well.

In order to best compare these older settings, which of the following are required: - Using PTool v3.63d instead of 3.64d (which I used for the newer ones)? - Using Firmware v1.0 instead of 1.1?

thepalalias

Haven't heard back on my last question so I'm testing Pictoris settings instead.

Also shot Factory defaults with Firmware 1.1 and Rocket V3b so far.

euggie2000

Pictoris v5c VBR occasionally produces gray frames. Transcend 64gb class 10. 24p Gonna try CBR version

thepalalias

I may have time to shoot more settings after I finish Pictoris.

So far: - Orion (all 4 settings) - Factory 1.1 - SpanMBU v2 - Pictoris CBR Hybrid

All shot with PAL/European settings in the following modes.

24H & EX24H
SH & EXSH
HBR & EXHBR

80% mode may also have been included for a given mode.

I would like to be as comprehensive as possible if time allows tonight. Can someone answer my question about FW1.0 vs 1.1 and PTool 3.63 vs 3.64d in regards to the older settings?

I prefer not to go back and forth between FW1.1 and FW1.0 because in one direction it causes all my preferences to be lost, but I will do it if needed to properly test.

terry2 · Screen Shot 2012-03-06 at 2.22.42 AM.png

@Driftwood Dark Matter V3... (MAC) Shutter 50, iso 400, open 2, Manual lens (zeiss cp2), -2,0,0,-2 Shows VBR on media info but on QT bitrate doesn't change. Fixed on 127.

terry2 · Screen Shot 2012-03-06 at 2.30.41 AM.png

@Driftwood Dark Matter V3... (MAC) Shutter 60, iso 400, open 2, Manual lens (zeiss cp2), -2,-2,-2,-2 Shows VBR on media info but on QT bitrate doesn't change. Fixed on 123.

peaceonearth

Hi folks, I am new to this an don't understand this matrix thing. What's the difference of these matrices? Would anybody be so kind to explain it on a short hand to a non-tec-guy? Thanks!

terry2

@filthy Thank you, I've tried but no difference. All Videospec, Media Inspector, Mediainfo etc still shows VBR and not CBR. I wonder if they just read what's written in the metadata and not "actually" analyze the file. I'm on Mac.

Meierhans

I guess the easiest way to explain is: The video codec inside GH2 looks onto the world as combination of frequencies rather than single pixels. These encoding matrixes tell the codec how much bandwidth it should allocate for which frequency. So the matrix affects the way how small details are weightened vs. larger ones. A too steep matrix will result in oversharp look and banding, a too flat one in muddy picture without enough details.Guess there is even a little more to it ;)

Hope this helps, tell me if I am wrong somewhere.

peaceonearth

@Meierhans thanks! If I understood right the original matrix is to steep therefore you have to lower sharpness? Orion is more flat? Is it a matter of taste or is there a general recommendation? thanks again

kingmixer

@thepalalias That's awesome that you're in the midst of a comparison test! I think I speak for more than myself when I say it'll be exciting to see how they compare and I appreciate the effort!

Kihlian

@Bueller I find it interesting, could you post a short video?

JPB1138

Orion 4vb... just some static shot test (mostly) in my backyard. Color graded in Color. Hopefully something a little more interesting next time. Overall happy with Orion and GH2. Some weird green noise, but mostly because i pushed the grade.

Looking forward to 4vC as I like night shooting. enjoy.

Frame

@driftwood, @rsquires I can now confirm as well, unhacked v1.1 Firmware HBR plays fine, transcodes fine, no problems at all. But why does Quicktime play any bitrate/hack @24p fine, but fails on 50i? Why does VLC play all (and probably Movist) ?

A short recap: HBR Hacks play fine in VLC, with some small glitches, maybe one or two in a 1:30 recording. In Quicktime Player, choppy playback of HBR-Hacks with Perian and AVCCAM-Importer installed Theclitches glitches will make Quicktime Playback hang/loop the last second. No transcode/rewrap of HBR to .mov possible using 5DtoRGB or Clipwrap when Perian and AVCCAM-Importer are installed.

When Perian and AVCCAM-Importer are UNINSTALLED: Choppy Playback in Quicktime, but transcode/rewrap possible, again resulting in choppy playback of the rewrapped rewrapped or transcoded file

24P recordings are just fine. HBR recordings with unhacked Firmware are also fine. Hacked CBR-HBR recordings also play fine but transcode/rewrap. (Have not tested CBR recordings yet with Perian/AVCCAM uninstalled) With unhacked HBR or hacked 24P at any Datarate, Perian or AVCCAM-Importer will cause no problems with Clipwrap or any other tool. All transcodes/rewraps are correctly shown as 25FPS in Quicktime, MediaInfo etc.

A non-stisfying workaround: Convert HBR-VRB to mkv using "Remux". Quicktime will take some time loading the rewrapped file, but now plays it back like like VLC again with the small glitches (wich seem to be on the recording) But Quicktime will now show the file as 50FPS. All tested with the Sandisk 64bG UDXC Claas 10 cards

Does any other Mac user experience the same? I am on 10.6.8. Are there any Windows based workarounds? We could still transcode/rewrap using Bootcamp... I will store my dailies on a Windows based portable anyway...

@thepalalias Thanks for your reminder about acg settings. 448 bitrate works fine. On AGG 203, I miss some db when recording full scale, analog 0dB sine from my field mixer which provides -34dBu output level. I am around-12dB in the recording, would like to get to the standard -9dB headroom. All the best, Frame

Rammstein

v5b Hybrid Matrix seth - HBR freeze my gh2 after 3 min. or sometimes 5 min.(unhappy) - FSH stop recording after 7 min.(unhappy) + 24h works very well continuously over 40 minutes. uninterrupted.(happy) I think 70 mb is too much for my GH2 in FSH or HBR. I need FSH and HBR.

rsquires

@frame I assume Remux is available for mac. Does FCPX play it (mkv files that is)?

Howdy, Stranger!

Categories

Tags in Topic

Top Posters