Brett Munoz
-
Posts
35 -
Joined
-
Last visited
Content Type
Profiles
Forums
Articles
Posts posted by Brett Munoz
-
-
You're essentially getting a 4K image scaled to 1080 with the C100ii. So unless you're wanting to crop in more than 50% it's not really an issue. Now 120fps would be nice
Funny, I spoke to a Canon rep and he stated that the C100MKII does NOT scale down which I found weird since I too read somewhere it does.
-
I shoot lots of weddings and also music videos/shorts. I really like the Canon C100MKII look but hate the lack of 4k and 120fps. The c300mkii is awesome but out of my budget. I really like the FS5 but do notice it has a "slightly" more videoish look when compared to the c100mkii (I think cause it is so sharp). Also, I might have a hard time matching colors with my 5diii and gh4. Any feedback is greatly appreciated!
-
Lmao. What a POS! Will sell 10 cameras.
-
Well no offence taken but I never shot my kids in a backyard. Will I don't have backyard. Aside from joking, I shot 8bit Log for national HDTV, for high-end commercials (Toyota, Ford, High-end make up/fashion clients), and 2 national-wide cinema release feature films. All 8bit. Strong grades were made by high-end post-production houses, and phenomenal image quality was produced. The funny thing is I never want to shoot 8bit, it's just that every project I direct I go to location and they have a C300, it's just the de-facto standard.
All I am saying is that 8bit is not on itself the main devil. For example, the 8bit H.264 image of the C300 can be pushed further than the BMP4K at 12bit raw! Why? because the moment I try to touch the BM file I get FPN and horrible shadows, due to a sensor defect. While the C300 has a good sensor and a modest codec giving a more robust image for grading. I was shocked grading both side by side, both having the same sensor size, the same lens, one at 12bit raw less gradeable than one with a compressed 8bit H.264 one.
The point is that it's SO many variables that make up high image quality, so many. Sensor size, lowlight performance, shadow DR, highlight roll-off, aliasing, moire, rolling shutter, lens resolution, colour science, image processor, downsampling algorithm, and of course codec, which in itself has many variables aside from just bit depth, including chroma subsampling, bit-rate, and most importantly the compression algorithm/quality of the codec, log-gop, all-i, motion based, VBR, CBR, etc
All variables being constant, a 10bit image will show less banding than 8bit on solid gradients (which therefore gives more room to push, especially with a sutble gradation in the scene), and 12bit will give more than 10bit, and so on. It is a fact. And that's where you're coming from. I agree.
But in real world, variables are never constant. Never be tricked into thinking an image on a certain camera is good/bad/gradeable/brittle based on the bit depth on the spec sheet. Real world examples of many cameras prove the opposite can and does happen more often than you'd think.
I believe in 8bit. I believe in 8bit Log. But only when the camera manufacturer gives me a good sensor and processor and an adequate Log gamma for 8bit.
A recent disaster was the GH4 with V-Log at 8big, I sold the camera for that. The image quality of the camera simply could not handle the post-production required to correct the Log images. On the a7s it's somewhere in-between, yes while the image is robust enough in terms of de-logging artefacts, the colours end up brittle for a reason that I don't know if it's the weak codec for shooting LOG or just a bad processor (colour science/matrix). Canon and Nikon Log makes me believe in 8bit Log and how great it can be in retaining huge dynamic range and great gradeability with small file sizes.
Don't get me wrong, I don't think it matters and most people won't be able to tell the difference. Content is king! However, on a budgeted professional shoot, I have only seen Reds or Alexas on raw or prores but I am sure there are the exceptions. As for reality TV, I've seen it all. A buddy of mine works on Daly show and is being shot on the 5d Mark III at H.264.
-
Log does work very very well with 8bit. 8bit is not the issue, weak codecs are.
Example, Canon C300, 8bit 50mpbs, Canon Log works a treat, even with the 8bit 24mbps C100, works beautifully. Sony A7s, works great (no banding or codec break up when while de-logging - some poeple don't like the a7s colour science but that's irrelevant).
Log is simply a logarithmic gamma curve to get a flat tonal range easier to grade and protect both ends of the spectrum it will work with any codec that has the ability to be strongly contrast-adjusted without introducing artefacts.
GH4 4K 100mbps codec is one of the strongest 8bit codecs, as is the 200mbps 1080p, not to mention the ability to record 10bit 600mbps 1080p ProRes HQ through HDMI (You can get that for 300$ Ninja Star or 500$ Ninja2 that gives you an extra 5" field monitor)
A strong Log curve coupled with tweaked colour science (which is what V-log is, it's also a little colour magic) is a perfect companion for the GH4.
I don't mean to offend but maybe it's good enough for grading video of your kids in the backyard but for professional environments where you know you will be color grading you will roll with 10bits if not raw because its not worth the risk. I rather bake in 8bits if I know I will only be making small adjustments.
-
How does color in the new a7s ii compare to Canon?
-
Log on compressed 8bit video files? Did I miss something? Wouldn't grading such files cause banding and other such horrifics on files from the Gh4?
-
Does anyone know of any metabones alternatives? The speed booster is too pricey for me. I don't need control of the aperture. I currently have the lens turbo however, it seems to suffer from a blue dot in the center.
Sony FS5 - why I bought one
In: Cameras
Posted
regardless, still looks as detailed as a scaled down 4k image.