Tyler Purcell

Everything. Narrative, documentary, commercials, industrials, BTS, trailers. The only thing I don't do is model/fashion work. There is no other explanation. "Mistakes" are corrected when you turn on the histogram and realize how far off you are. "Mistakes" are corrected when you look at the video monitor and see how off it is. "Mistakes" are corrected when your DIT watches the material (in this case with Red Cine X) and tells you something is wrong. I believe the shoot was multiple weeks, but on the same sound stage.

David, I bet your accurate. It's 10 bit in 4k and 12 bit in 2k. Most likely a limitation of the processor.

Yes, it's actually very unusual. Most editors stay away from coloring. I only do it because it interests me. Sometimes what I do is final, other times I will export my DaVinci project for someone else to finish. Needless to say, I enjoy the "finishing" aspect of projects as much as editing, so I attempt to secure that part of the project whenever I can. Not accidentally, absolutely intentional. I don't have any footage online, but I will gladly post some samples someday. I'm going to reboot that project in October/November and I'll have things online once more. I was professionally trained as a youth and used it as a backup job for years.

One little detail... I think the imager processor is 10 bit. They seem to put that in the specs very clearly.

Here is the difference. You and I can't afford an Alexa, but if we had the money, we'd probably own one or at least a RED of some flavor. We're kinda stuck in a sub $5k camera world. I'm not a blackmagic devotee at all. In fact, outside of the pocket and Ursa Mini 4.6k PL, I think the rest of the cameras are seriously flawed. I haven't bounced on the URSA Mini yet because I think they'll do a major update soon. Still, I've never been just a shooter. If I shoot something, I'm most likely editing it. So things like codec's take a higher precedence then things like imager size, resolution or even built in audio capabilities. This is because, time is money and for me, the quicker I can edit something, the quicker the turn around, the more money I make. Most of the time I'm looking for "acceptable" rather then "perfect". I also like run and gun, so having a lot of dynamic range, especially in the highlights, is important for me. Maybe not so important for a big movie shoot where you can control everything. Hence the reason people get away with using crappy cameras on bigger shows. What I personally look for is camera system that fits perfectly with a native post production workflow. I'm also looking for something cinematic, something that doesn't have highlight clipping issues, that doesn't flatline in situations with too much brightness and isn't overly noisy.

But look at the price! HOLY poop! Canon is trying to break into the higher end market. Problem is, that market is pretty flooded with very good cameras.

My job is to make what other people shoot look good. So when handed a shot, it doesn't matter what format it's in, I have to make it look good. Lets say your signal to noise ratio is 50db. When you record a LOG curve, it requires translating that curve during coloring, which means raising the highlights and lowering the mids and blacks. Since you've baked in particular noise level, you're increasing the noise when you do this coloring work. I'm not talking imager noise, but "codec" noise. It's only hogwash because people like you, refuse to learn. I'm not making anything up, I work with top colorists all the time, on bigger jobs that I simply can't do. They always balk when I send the MPEG anything, they just don't want to deal with it because in their eyes, it's substandard. Few months ago, thanks to another thread where people said MPEG worked fine... I produced a video that destroyed the XDCAM 50 codec. Now that I have access to an F5, I will gladly produce a video that destroys the XAVC codec as well. We both know that's going to be a tough endeavor, but I know it's weaknesses and I will flaunt them. Of course... my "comparison" camera like in the last video, will be a $1000 blackmagic pocket camera in Pro Res HQ. I will use the same ISO, same stop, even the same glass just to keep that out of the equation. When I have time, I will make this, just to have a laugh. That's not what I said, I said it's unfathomable that the myriad of cinematographers I've cut and colored, are bad cinematographers. It also humors me when those same people work with a RAW or Pro Res camera it looks fine. So is it the cinematographer, or is it the camera? The feature I'm currently editing has a baked in look. That's because the cinematographer did a test with log and found the signal to noise ratio on the MPEG file to be a problem. Also I doubt anyone shooting with an F65 uses XAVC codec. Again, with Pro Res or RAW codec, the higher end Sony cameras look fine. Then it's down to all the other Sony bugaboo's, which the list goes on and on. Thank god the F5 and F55 have most options right on the side screen.

Robin, please back off. You need to stop being so close and tight to your Sony cameras. I have my reasons for not liking Sony cameras, especially the current fleet. You don't have to constantly defend your purchase decision with me. I don't own one, but then again I would never recommend one to anyone, ever. I have enough 4k proof right here for those people to see and understand. It's hard to transmit 10 bit 444 4k over the internet and for you to see it on a calibrated monitor on your side. So lets just say, since my job is to make what you shoot look good, that perhaps I have more insight to what each of these cameras does. Again, in perfect lighting, without any true dynamics, what you shoot with is down to glass more then anything else. When you get out of that world, when you truly run and gun, that's when these cameras show their true colors. I wrote a diatribe on the other thread, maybe you'll understand more where I come from. I also never said the material I work with is unusable. I said it requires a lot of re-working and the net result is noisy/grainy all the time. I think it's unfathomable that all the cinematographers, guys who have shot for 30+ years, don't know what they're doing. This is why I tend to blame the camera, especially since it's almost ALWAYS Sony cameras that are the problem. Since I'm an editor and colorist for hire, I've worked with a lot of cameras.

I'm discussing literally dozens of projects, shot over at least 6 years, none of which I've shot. Yes, this year I've had more experiences using modern 4k MPEG cameras, but I stay away from them based on the following experiences: Unlike your job, MY job is to make the show. I take the footage you shoot and put together what the audience see's and hears. Most of the time I work in Avid and color with Symphony/DaVinci. My workflows are industry standard, based on my editorial experiences working at various top post houses in Hollywood. So what I see is what I know. I'm not looking at a viewfinder and saying the shot is good. I'm not looking at a monitor on set and saying the shot is good. I'm not looking at a meter and saying shot is good. I'm looking at the final output camera file and having to match that file to other material, which in most cases is shot by the same cinematographer. This requires me to "tweak" the shot material in order for it to be a match. What I've learned over the years is based on my experience in post, more then it is a perfect setting on set, where you aren't seeing the actual file matched against maybe 600 shots that will compile the final project. All you know is what you're shooting at that very moment, you don't know how it will look in the back end, but I do. Armed with that knowledge and frustration when coloring, trying to meet clients expectations, I've formed my opinion on this matter. Mind you, I work A LOT with RED and Alexa cameras as well. I also shoot quite a bit with Pro Res cameras, like the one's I own. I use the same workflow with the Pro Res material and I'm happy to say, I rarely have the same issues. Yes there are times when poop happens, like the massive documentary project I took on earlier this year, which was under exposed by at least 3 stops and shot with a Red Epic. It took me days to come up with a workable solution that could be applied to all shots and even then, I will have to re-color once the client is done editing. I also don't like swiss army knife products at all. I'd say the only thing I own that's like that is my iPhone, it does a lot. All of my cameras have one function. All of my lenses do as well. This is probably why I dislike most digital cameras. My personal digital camera, the Blackmagic Pocket camera, looks fantastic for a $1000 body with $600 piece of glass. In the color suite, I find coloring my pocket cameras to be a piece of cake. I rarely have problems and when I do, it's because I was running and gunning on location and didn't have the time to make it better. I've pushed my pocket cameras where the MPEG cameras fall apart and far greater. Not saying the pocket camera has a better imager per say, just saying when it comes to noise levels and how much luminance and chrominance is in each frame. This goes for the Alexa as well, it's just an amazing looking camera right out of the box, that doesn't require much tweaking to make look awesome. Well anything looks good streaming 8 bit 4:2:0 50Mbps MPEG. Ohh wait, that's ALL forms of media delivered to the home; internet, satellite, cable, broadcast. So again, how do YOU know what it actually looks like? I've seen many documentaries and narratives shot with XAVC cameras and they always look like ENG cameras from the 90's. Lots of motion blur, lots of clipped highlights and plenty of noise. Most people use cameras like the FS7 due to it's high ISO setting, so that doesn't help either. I could care less if the audience can see in the dark, I'd rather have less noise on screen. I proved a few months ago how bad the 50Mbps Long GOP codec is... when I'm done with these two films, I will gladly post raw material in 4k for you to see how noisy and how much of a pain it's been to work with the FS7.

I never said XAVC is longGOP. I'd be glad to show you samples of how poor the XAVC codec is when you push it. I just can't do that right now... too busy editing.

Yes, I know and I never mentioned XDCAM anywhere. I'm specifically talking and mentioned many times iFrame, which as you well know is XAVC. Dude, I shot and serviced Sony for over a decade. I can say anything I want about them.

I have never worked with a good cinematographer. Maybe I'm just too picky.

Right, but that is a worthless codec that doesn't work for anyone but 1080p customers watching on a small television. I ignore it's existence because it's not 2005 anymore. That codec should have died off when the XDCAM optical format died in 2010 ish.

I said iFrame many times.

XAVC IS MPEG!!!!!

XAVC is MPEG!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! HELLO!!!!!!!!!!!!!!!!!!!!!! I don't have problems with stuff I've shot. I have problems with stuff OTHER people have shot. 3rd party companies are forced to make special stuff for Sony cameras.

Sorry, I was referencing many different cameras at the same time.

The cameras I've been talking about record out of the box in iFrame MPEG. Yes, the FS7 and F5 both have hardware updates to allow additional codec's. I've only seen people equip their cameras with those upgrades on rare occasions. I just used an FS7 with the Pro Res card, but only because I was like "HOLY poop" it was a unicorn in the wild! When I'm shooting, I work at a the lowest ISO I can and I will light more to compensate. However, I will still underexpose a tiny bit, to protect highlights. With film, I also work at the lowest ISO I can, but I over expose to that particular ISO, but not by much.

I don't use a meter when shooting digitally, I use a histogram and zebra set at 70. So I'm just use to protecting my highlights, probably a bit more then most people. "LOG" doesn't make enough of a difference. It actually makes the MPEG noise much stronger because it decreases the signal to noise ratio. I've worked quite a bit with raw recently, just to keep myself up to date on the coloring aspects of it. Even with RAW, I'd rather use a lower ISO and under expose a tiny bit, just when the 70% zebra's are starting to show up, that's the cap of my highlights. So I ignore the cameras native ISO rating and shoot like film. This gives less grain/noise, even when you punch it up. This trick works flawlessly on every camera I've worked with, outside of Sony. The Sony cameras don't like it, but then again, I despise Sony. I'll say this much, the C300MKII stuff I shot, looks perfect, with zero grain/artifacts. Yet the stuff our "cinematographer" shot has been noisy, he shot 2000 ISO outside in broad daylight with 1/64 ND. With film, I work just the opposite, I use a meter and I saturate the living crap out of it because I know there will be data in the highlights no matter what. I generally run a full stop over exposed on film, sometimes even setting my meter to compensate. Obviously there are occasions when you don't want that look, but generally I prefer the "pop" look it delivers.

I used Cinelab for years and never once had an issue. Love'em! :)

Yep, they sure are! You do have to push them a tiny bit in order to reduce the noise, that's 100% accurate. The only problem is when you do that, you're limiting how much excess range you have. This is why a camera that isn't noisy, where you CAN underexpose a tiny bit to retain better highlight levels, will be a lot better. This is for sure one the reasons I discount Sony cameras... that AND the cost v features... they tend to be expensive for what they are, thanks to the proprietary accessories.

I'll say this much, I think seeing a lens wide open is far more important then a lens closed down. It really tells you how good the glass is. As David points out, glass becomes more similar when stopped down. The need for good/high speed glass is really only warranted when near to all the way open. Whenever I do a shoot, first thing I do is test lenses all the way open and figure out how much they can be pushed. Looking forward to seeing the test! Thanks for sharing. :)

I'm coloring a show right now that's mixed FS7 and A7SMKII. We shot SLog2 with both cameras and I made a home-made LUT for the A7SMKII because the cinematographer's exposure was all over the map. I'm pretty happy with the results, but its been a lot of work and it's just talking heads. I can't imagine dealing with the sun or anything outside of a closed environment with a larger dynamic range camera is necessary. The FS7 material looks MUCH better and the F5 in my opinion is TWICE the quality of the FS7. We did some tests with an external recorder and one of the shots in the final project is Pro Res origination. Honestly, it was the worst looking of the shots, don't know why, but it needed the most work and has the most noise. Not sure if the HDMI output is noisy or not, but perhaps it is. Honestly it really depends on how you shoot it, that's going to be the key. I've done quite a bit of A7SMKII work recently for one of my clients and in certain limited dynamic range situations, the camera looks OK. Where it fails in my view is it's lack of dynamic range and it's just not very colorful. Almost all of my corrections include heavy saturation gains, even on top of the LUT. The color science of the F5 is MUCH better, it's a very good looking camera. I've colored lots of F5 material, most of it recorded in Pro Res and even when you underexpose, you can get something out of it. Over exposing on ANY Sony camera is death to your shot, they have very little tolerance compared to the more "cinema" specific cameras like the Alexa SXS and Red Dragon. I personally don't think you can match'em very well, even if you use the same glass.

More and more, people are reverting back to separate system sound, thanks to the advent of lower-cost external recorders. In fact, every single show I've been on this year has been double system sound, first time in my entire career. I'm a huge advocate of single system sound, but it appears the market is changing again. When I refer to "run and gun" I'm using using it as a general term that refers to a single camera unit, which is easy for one person to use. I also like to use the term "ENG" because originally the "all in one" camera was designed for that purpose and to this day, the external basic design hasn't varied much. Yes, manufacturers put their own personal touches on the cameras, but anyone who picks up a different brand's camera, should be able to make it work quick, unlike today's cinema cameras, which are so heavily menu driven. I also do think, if you're spending money on a real production, you'd better make it for a 70ft screen because that's where things are headed, that's what netflix delivery is already requesting. They already have higher standards then our cinemas!

It is when you add the extension box to the back, which I feel is critical to making that camera look decent, thanks to the addition of "proper" codec's. I mean lets face it, the only people buying camera's with XLR inputs, built in filter wheel and extremely limited MPEG recording, are people doing run and gun productions, which is in the same world as ENG in my opinion. The FS7 works fine with static talking heads and running around in a war zone, neither one of which is being blown up to a 70+ ft wide screen. "Cinema" cameras are an entirely different animal. Some cameras like the F5 and F55, are a hybrid between ENG and Cinema. However, when you shoot "cinema" you really need RAW or extremely high quality, full bandwidth recording, like Pro Res XQ as mostly all products made for cinema are heavily manipulated in post and need that bandwidth. Coloring the FS7 and C300MKII stuff (with native codecs) is a lesson in futility. Might as well be wearing a straight jacket and coloring with my nose.