Blaise Douros
Forum Replies Created
-
There are various gamma settings available in most of Sony’s professional cameras, including settings for the knee and blacks. The FS7 allows you to load LUTs that you can either view only in the viewfinder or bake into the footage–part of the reason I got it. I will likely shoot with a base Rec.709 LUT on the Log footage for a lot of my day-to-day corporate talking head stuff, but when I have a more documentary or narrative project, that’s where shooting flat will be a bonus for me.
Todd, I agree with you that it’s not ALL about specs…but things like 10-bit 4:2:2 vs 8-bit can get important really fast if you need the ability to create looks in post.
But yes, a camera isn’t good unless it has a good driver!
-
The Ursa Mini gets expensive faster than you think, what with the CFast cards, V-mount batts, and expensive OEM accessories which are not included. Plus it is totally unproven…
If you want the dynamic range, consider the FS7. Comes fully loaded, and works with Canon lenses with a Metabones adapter. Ping me in two weeks, I’ll tell you how I like it 🙂
-
I may seem naive, but I believe the XF300 images rivals the Alexa for a saturated image.
o_O
I can’t agree with you there. Maybe in saturation, but not in dynamic range, which is what really matters for getting that coveted cinema look.
I wouldn’t say the XL1 “revolutionized” anything. I mean, it was a nice prosumer DV cam. If anything, the Panasonic AG-DVX100AP was the camera that made everyone think you could shoot cinematic video (first readily accessible video camera to do 23.976 via 3:2 pulldown, cine-like gamma).
I think the Panny’s in general do a pretty good job of rolling off their highlights, even on their cameras with lower dynamic ranges. And the latest Sonys, with 14 stops in the log profile, give you a ton of latitude. I mean, if you’re not getting enough saturation out of the Sonys, the codecs can be pushed! Hell, the Sony X70 has a 10-bit 4:2:2 codec, recording internally, for $2k! I just don’t think Canon is up there with them when it comes to sensor and codec technology. The best Canon’s got is 8-bit 4:2:2. Sure, you can go off-board to an Atomos or whatever…
Which brings me to another point. If you love that XF300 so much, why not pick up an offboard recorder, and get your footage in ProRes 4:4:4 so you have additional latitude from the camera? Might breathe a few more years of life into it, and give you additional options.
-
Each manufacturer has its advantages. Panny’s lenses on their all-in-one camcorders aren’t stellar, but damn, they get their gamma right, and their color rendition is pretty nice, too. Canon is great at color rendition, and Sony is always pushing the codec envelope.
Having been a big fan of Canon for a long time, I just today bit the bullet and ordered a Sony FS7 kit.
I love the Canon glass and color rendition, but hate that they are two damn years behind on their feature sets on their camera bodies, which, BTW, are extremely expensive. The FS7 is going to get me 10 bit HD 4:2:2, true log gamma profiles, 180fps (!!!) overcrank, and 4K options…plus, I can adapt whatever glass I want. For half the price of a C300 mkII.
When it comes to the Big Three (Panny, Canon, Sony), I’m at the point where I just get the camera that serves me best, and to hell with brand loyalty. But I don’t buy JVC. I fully admit I am prejudiced where they’re concerned.
-
Could be that he’s using different lenses from you. There’s no particular noise in your images. You are overexposing your shots slightly, and haven’t color-corrected or post-sharpened them, so that’s probably not helping. The comparison video was also shot in highly-contrasting light, while yours was shot on what appeared to be a cloudy day–naturally less contrasty.
-
I’ll add my voice to the CC supporters. I was skeptical as well, but no more. The regular updates and access to the entire Adobe Suite you get is absolutely unparalleled. I effing love it.
I’ve been using it for two straight years, and never once, never, not a single time (did I mention never?) had an issue related to licensing. The only major issues I’ve seen are some easily work-aroundable video-card related glitches in Yosemite, and of course, the inevitable faulty-keyboard-to-screen-meat-to-hardware errors (i.e., user error).
The people who are holding out on CS6 are kidding themselves. Adobe is NEVER going to go back to the old model. Why? They get a regular infusion of dollars that can support a development team that is devoted to literally doing one thing: supporting and improving that software. The users get faster iterations on the software, and constant bug fixes, updates, and improvements.
Is it better to spend $1200 to upgrade to a new version of the partial CS suite every two years, or better to pay $600 a year to have access to ALL of Adobe’s apps? Seriously, you probably pay more for your cell phone and cable bill, INDIVIDUALLY, than you will pay for Adobe CC every month.
-
You’ll just have to slow down the audio in post. You could record it at a higher sample rate, but unless you are actually performing a resampling operation on it versus a slowdown, it won’t make a difference. Audio isn’t recorded in FPS, so there won’t be a setting on your recorder that will perfectly match.
The key is that when your clips are synced in your timeline, you need to link them so that the slowdown is performed on both at once. You don’t want to try to manually slow the audio down separately from the video–that’s a recipe for a major headache. Get it synced, and let your NLE do the rest.
-
Yeah, looking at the video, it’s definitely not out of focus. I think you’re running into the typical issues with DSLRs–Ryan did a good job of summarizing them, so I won’t rehash. I do disagree with him on the lens bit–my experience with that Rokinon lens is that it is soft (although I have also heard that there is some variance between different copies)–so stopping up a little won’t hurt your overall sharpness.
The 6D can produce great images when processed right–I often use 6D and 5DIII footage together. But, I ALWAYS sharpen it a bit in post–there is lots of detail that can be recovered. Here’s what I would do:
First, your exposure on this shot looks a bit hot, which can make things appear slightly soft. Throw the Fast Color Corrector on, and drop the highlight level slightly. You’ll recover some of your sky, faces, and the whole thing will appear to saturate slightly more.
PP (at least, PP CC, which is what I’m using) has a great software sharpener–throw it onto the clip and set it to the 30-40 range, whatever looks best to you. That will sharpen it up, but not to the point that it looks like garbage.
These two effects, in this order, are what I use to dress up 5DIII and 6D footage. You’ll be amazed at what a little software sharpening can do to footage from these cameras–I certainly was.
It also occurs to me–you said in your original post that you’re keeping a constant aperture, and varying the shutter speed. You should keep your shutter speed at 2x frame rate (so at 60 fps, 1/125 of a second shutter) and vary your aperture. You always want to keep that 1/2x ratio, unless the situation calls for a specific slow-shutter or fast-shutter (think Gladiator’s fight scenes) effect. Stopping up that Roki will only improve it, anyway.
-
The Rokinon 14mm is certainly no Zeiss prime. It is pretty soft. But it’s not going to cause pixelation. I mean, 720P is inherently softer than 1080P footage in general, and especially when you’re downscaling it on a DSLR sensor. Does it respond at all to software sharpening?
Without seeing an example clip, it’s hard to say. Can you throw one up?
-
I’ve scaled the footage up to 1080p using MPEG Streamclip…
There’s your problem.
When you do this, you’re stepping on the quality of your footage not once, but twice. Transcoding it, and upscaling it, are both things that can and will add visible pixelation to your image.
So what you’re seeing likely isn’t grain, but compression artifacts and/or visible pixels from upscaling 720P to 1080P.
What are you transcoding to? Can Premiere CS6 not take the native files from the camera? You want your transcoded files to be as close to the original settings as possible. Make sure, too, that the framerate is the same.
I would suggest transcoding to a 720P file, and letting Premiere handle the upscaling, or better yet, work in a 720P sequence and downscale your 1080P footage.