@driftwood - "Perhaps youre being late to the ball and getting your arse in gear is hurting and your arrogance has led you to hijack the thread?"

Alright, since you insist on hijacking the thread into a personal dispute, let's talk about arrogance. You seem to assume that since Chris approached you with an idea that you put some work into, you now own both it and this thread, and resent my participation. If that's the case, I'd suggest you start your own blog, like EOSHD, and ban me from it.

With respect to Chris, he vetted the 3x3 matrix idea with me as well, and disclosed the details back in July. That's why I was already familiar with it, and was able to analyze this patch in technical detail and distinguish his work from yours. I actually appreciated the fact that you followed through with it, and was interested to see how well it would work in practice. But when you suddenly released the Sharp2 matrix, I had to reverse-engineer the different matrices just to figure out which variant I should be testing.

Now, as for your gratuitous insults. Your paranoid accusation that my independent technical review of Apocalypse Now was some kind of "moment of glory" for me is simply pathetic. Anyone who's read through my patch threads would know that I've created plenty of my own opportunities to claim as much glory as I could want. Sorry, but this isn't about you or any competitive thing at all. I'm interested only in how well Chris' matrix works, and you happened to provide a convenient way to test it out.

And on the topic of mutha'fuckin' testing, I have a bug report. The Driftwood Cluster v7 'Apocalypse Now' - 6 GOP Nebula '444 Soft' variant froze up twice in a row while I was shooting some daffodils, tripod-mounted in 720p60 SH mode, after about 7 seconds. Manual-exposure movie mode with a manual focus lens, ISO 400, f5.6, 1/60 sec, 30MB/sec Class 10 Sandisk Extreme. You might want to take another look at that.

@driftwood: Please don't let LPowell highjack your unique thread, maybe he is just jealous of your outstanding, groundbreaking performance here.

I bought the GH2, tested all the settings, tested LPowells "Flowmo", tested bkmcwd settings, tested all your settings. The only thing i have to say is: From my point of view "Drewnet 444" and "Nebula 444" are the best ever rockstable settings you can get for the GH2 !!! It's just unbelievible what this little consumer camera can serve using your settings !

Keep on going your "mutha'fuckin" hard work - we all luv it !!!

Thank you all involved in Cluster development. For a change there is clutter 7 gop6 soft. I should not say that but Sorry for colors- I found out that I have a big problem with calibrating my U2410 so CC its done intuitive, Hope its acceptable and it looks similar on your screen to mine.

@bkmcwd

I would like to try your settings, but I am a little confused.

Could you please tell me the difference between your "Valkyrie 444 TYPE-ZERO 'Apocalypse Now' with cbrandin 'Original 444 Matrix' " and your "Valkyrie 'Apocalypse Now' with cbrandin '444 Soft Matrix"?

In which cases should I use one or the other?

Thanks

First post on the boards, but I've been following the GH2 hacks for a while now, and thought I'd get stuck in and upload some test footage with the new settings.

I used the 3GOP Valkyrie Soft setting with two different manual lenses. Interestingly, the Miranda 28mm lens I use for a couple of shots I always find to be soft, but the details look fine to me, even with the soft variant. The colours on the flowers are brilliant I thought. No grading, only tweaked the contrast, so I hope this is useful.

Great work with these settings! Thanks so much.

@itimjim Great to see someone else using that Tokina! I have the old version from Japan and love it. Love the lens flares you got with it.

I was a huge fan of reAQuainted and Sedna Q20 and really like the sharp2 Apocalypse. I use old Russian and German lenses mostly and they are getting me close to what i would ultimately like. GOP1 to 3 is better for an overall crispier look in gradations and shadow. I tend to shoot as flat as possible everything to -2 and then adjust in post to get the ultimate look. Probably a low contrast filter on top of all would even be better. Paradoxically the lower contrast reaches the sensor, the more detail gets preserved. For after effects: use (multiple times) the unsharp mask. For FCPX there is a free plugin effect "unsharp mask" that is also very useful to bring back the detail. Get the contrast back with Curves effect in After effects or with Nattress (Graeme Nattress wrote software for RED) Curves, Luma Curves and Chroma Curves in FCPX (the default S-curve already gets nice results). Really worth it. Try it and see how great an image you can get with this workflow.

Is it busy while I am away for one day?

@bozfx

"Is the "TYPE-ZERO" Valkyrie matrix patch considered to be the 'sharp' variant?"

No, it is the version which adopted the first edition of the 444 matrix of Chris and Nick. I am going to make the sharp matrix version of Nick after this. Rather than "Chris's original matrix was soft", by this original matrix, since the balance of the I frame and the B frame is not good for GOP3, I consider that a 444 Soft matrix version is better, IMHO. Then, it is because this setting is a starting point of "Valkyrie" as to why I released TYPE-ZERO.

@Mirrorkisser

Many thanks for understanding my situation poor at English. :-)

@jebsly

Thanks for testing Valkyrie! :-)

And all praise is to Chris and Nick!

Here's something I threw together yesterday where I live. It's MGMs old backlot #3. Tech info at start of video. It's Cluster v7 'Apocalypse Now - DREWnet' 12/15 GOP Sharp2 Seems like there might be a couple glitches in the upload. I have one going up on Vimeo right now to see if it fares any better.

@cbrandin Thanks for posting frame grabs of your matrix resolution tests. I think they clearly illustrate the inherent difficulty of choosing a number that represents the camera's "lines of resolution". In both charts, the individual vertical lines blur into gray in a ragged manner, and exactly where you choose to assign a numerical cutoff is debatable. Beyond that, I'm actually seeing more noticeable aliasing in the Soft 444 matrix, in the form of bright vertical ringing effects around the outer edges of the flute and the zero in the number '10'. This is often evidence of a digital filter with too sharp of a cutoff, which may indicate that the 255 components in the Soft 444 matrix are a little to harsh for optimal results.

In general, there's really no objective way to measure the resolution of any digital camera, it's a subjective judgment call. At high spatial resolutions, the image sensor's output is contaminated with ultra-fine aliasing artifacts that are literally folded down into the upper resolution range. As a result, the genuine resolution of the sensor is submerged within spurious alias products and there's no way to separate them. With digital cameras, "sharpness" is a marriage of refined technology and aesthetic preference.

Sure - that's because the gradation rendering is much, much better. Also the way detail is rendered is a little softer and less rectilinear - compensating for the "sharp" video look.

People complain about the "harsh" video look - well... that's significantly related to detail sharpness. I think it's funny when people "want" maximum sharpness, then turn around and use soft lenses - I don't get it!

@lpowell - The criteria I chose was the first instance where the wrong number of lines are shown (8 versus 9). But your right - that decision is ultimately subjective. Part of the reason they look different is because they are not scaled to relative resolution. If the lower res is scaled, it looks more similar. These are JPEG screen grabs, so we have layers of compression artifacts happening. Also, I didn't optimize any other settings and that results in cruder rendering overall with the soft 444 matrix because it needs more bandwidth. I was just trying to determine relative resolution and nothing else.

@cbrandin wrote: "I think it's funny when people "want" maximum sharpness, then turn around and use soft lenses - I don't get it!"

Pretty dirt simple explanation on that. If you're experienced with lenses -- MFT lenses are rather mediocre on a range of levels. I couldn’t' bash these newest Olympus primes for long enough.

On the other hand -- combine the excessive sharpness of the GH2 with what's being called "soft" film lenses -- and the result is a winning combination.

@LPowell Aliasing? Battery drops? You've clearly got some problems. I assume next in line is "it bricked my camera."

I've had no battery drops, no corrupt files, and have certainly not been plagued by aliasing and/or moire on any of these AN variations.