Driftwood GOP1 AQ2 'AQUAMOTION v2' did give me some errors.
So i put back the Driftwood GOP1 AQ2 'AQUAMOTION' Experimental patch
And it did what it must do, working without any error.
Iso 160 to 4000.
Shutter 50 to 320, and lens from 14 to 150 and tele2.

Nothing special in the clip, just a lot of movement..

What I think about noise and ISO. There are no true ISOs. There is the native sensor signal. Then before the analog signal gets converted to our 8 bit sample, a "pre-amp" boosts the signal based on the selected ISO. Anyone familiar with a transistor, a nominal signal is applied to the transistor, and a voltage is added to boost the signal on the output. My guess the amount of signal gain is controlled by Pulse Width Modulation PWM. And depending on the quality of the input crystal and the PWM, some "frequencies" of pulses to control the gain are probably "cleaner."

Hey guys! Heres a video I shot with the Aquamotion V2's 720 60p at 66 and 55 mbps. I am so very satisfied with Aquamotion V2 720 but it fails sometimes too. Sometimes it run perfect, sometimes i get the card write speed limitations. anywho, watch it for some slow mo rugby! :) enjoy!

Teaser Trailer for "What Lies Between Us." Shot on Quantum v2.

After seeing v9b, which came out the day we started shooting, I wish I has used 9b. These clips came in at 88Mb/s. We've done one P/U shot for the film so far with 9b...and it looks amazing. And the 9b clips are coming in at 149Mb/s. Driftwood is my own personal Jesus Christ.

that looks real good, and the music is Wow, all those things together.. good job

@shian wow now that looks great! i read you said that sampling white balance create noise, what about auto white balance is it cool to use it or it also creates noise?

@mozes Yeah, our composer, Christian, is pretty badass.

@Swiss_Boy @Alex I always seem to get in trouble for "shameless self-promotion" when I cross post - even when its relevant, so the details of the shoot are in the ColorGHear thread, where I posted them last week http://www.personal-view.com/talks/discussion/1817/colorghear-toolkit-color-grading-system-for-ae#Item_348 But the smooth skin tones is a special application of ColorGHear's GHrain Killer. There are some before and after stills in the thread.

There was a really nice grain on the v2 footage, not as nice as 9b, and I only treated it in spots for effect.

And Swiss_Boy, I only encountered noise when trying to sample for WB, so I've gone back to dialing it in the way I've done from day one. I see no need to sample, or auto WB ever again. It's not very hard to figure out what color temp your lights are and adjust for it (and/or gel them to get them where you want), and then save it as a preset. I haven't done any kind of scientific test on the subject because it hasn't been a big enough problem for me, and one easily avoided.

Just like the ISO bug. I never use the 3rd stop ISOs only the full stop ones 200, 400, 800, 1600 and 3200, I do all my in between exposure adjustments with scrims, dimmers, gels, and moving lights closer or farther away if necessary.

@shian Hey thanks for the fast & instructive response. Yea i know you can save 4 presets of white balance but i'm not sure that if you can also save the one where you set the colors vector + K number the last one on the right so i will have to check that. And I think i will also start to use the full stop Iso but i find this video where it shows that iso 640 is kinda good

@driftwood

Did the most strenuous test I could this morning before work:

Panny 14mm, autofocus continuous left on, filming television full frame with the worst card I have (ADATA class 10 32 gig). Did two tests, one at ISO 160 one at 640 -- the main two iso use. Both failed to span. By way of reference, Aquamotions span fine on this card. Although weirdly the files show they get up over 4 gigs before failing which makes me wonder if there is something there that can be recovered.

Ran tests again with Sandisk 30mbps (extreme video cards). Spanned no problem at both ISOs.

Mac based so sorry, no stream parser.

Thanks again for the continual improvement of these. Nice to know all my 30 mbps sandisk cards won't be gathering dust!

(as an aside, can confirm the minor difference in record times with different isos -- 8 1/2 minutes for 4 gigs for iso160, compared to 6 1/2 for 4 gigs for iso640).

@ignatius -- can you edit your post to include what card you used?

Thanks man! No worries -- was just curious :-)

@driftwood

Initial impressions on Quantum 9c...

Clips spanned at 24h and 720SH on a Sandisk 64gig 95Mbps card. Three times at 24H and twice at 720. I stopped the recordings both times because I need to leave soon. Ran these tests using the "kit lens" 14-42mm, image stabilization on, sweeping motion all over the place with very detailed scenes.

Image quality looks very similar to the Quantum100 patch -- to me. It doesn't kick me in the face like 9b, but it certainly looks good enough for serious applications.

Windows 7 reports "total bitrate" of 78927 kbps on the 24H and 57709 kbps on the 720SH.

Nope. Nothing to do with that. This patch is purposefully left on 20/20/20.

Here, I am using AVC-Intra 50/100 spec bitrates, logical memory usage video buffer control, frame buffers from AVC-Intra spec , fb2, and frame limits all combined for stabilising AVC INTRA rate control. This is not the final version of this setting - purely experimental until the new ptools comes out.

I have been playing with special Intra matrices and a few other bitrates / coded picture sizes borrowed from AVC Intra. Of course, we can never replicate AVC-Intra 100 in its purety as that is a 10 bit 4:2:2 implemtation, the GH2 is only 4:2:0 8 bit. But we run it bloody close, and its looks good with tighter Q and the ripped AVC Intra 100 matrix to provide the closest colour implementation we can get to the Panasonic spec.

This means, all being good, that the final Quantum AVC Intra patch (when its released next week) will mix well with camcorders using the P2/AVC Intra specced implementation.

It won't be 100% but its close - my own findings have shown close resemblance to the AVC-Intra 100 footage but don't forget 10bit (1024 colour space) offers 4 times the colour space of 8 bit - 256 shades in the range (8 bit is responsible for the poor colour gradations seen in particularly in blue skies, etc...)

Later tonight a test patch of AVC-Intra 50 settings will be released in Quantum 9d (Quantum 50 v2). Remember, these are test patches for you. Don't treat them as final BUT give me feedback.

I am withholding my Q settings until the next ptools comes out sometime end of next week.

is quantum v9b much improved over seaquake?

all the talk is about quantum these days!