Noah Yuan-Vogel

From the published specs of the newer Kinefinity Terra cameras, it says they can shoot up to 5K/6K ProRes so that may no longer be the case.

Yes I was going to throw in a qualification about my statement only being about image quality vs file size. Certainly there are workflow concerns that may trump those. Then again I'd be skeptical about how much more 444 or XQ really gives you for 4k from a 4.6k bayer source especially considering the framerate limitations, but I'm sure some people will want to use it anyway.

So Arri is only relying on the lower gain ADC for the top two stops of DR to overcome the DR-limited ADC? No other special stuff going on? So in a way the combined ADC is effectively 14.8bit I guess?

My understanding is that the 4.6k can shoot UHD prores downscaled from the full 4.6k sensor. Keep in mind it's up to 60fps for UHD Prores HQ but the UHD Prores 444 and XQ max is 40fps. Also the 4.6k RAW 4:1 bitrate is pretty close to UHD Prores HQ so it may not make sense to shoot UHD at more than Prores HQ anyway.

I just saw it in 2D as well and often noted several scenes with high motion blur that could have been a 270deg or 360deg shutter angle. It was a little jarring to me, but I didn't quite seem like the whole movie was shot that way, maybe certain scenes were? I also noticed a bit more digital colored noise in the shadows in some scenes which surprised me, but overall I definitely enjoyed the movie anyway.

4:2:0 for avccam recorded internally, 4:2:2 out through HD-SDI to external recorder. You probably wouldnt want 4:2:2 at 24Mbps anyway, since that'd be packing 33% more data into an already rather minuscule compressed bitstream. ie. it'd look like 18Mbps 4:2:0 but with more color resolution.

We can't see your images, you seem to have 3 images linked (like "https://cpanel25.syra.net.au:2083/viewer/home%2fstrictly%2fpic/Shoot.jpg") but the links may be through your website's secure private cPanel back-end not through a publicly accessible website. Try copying and reposting the links from your website, not from your administration page.

Agreed, my understanding is that LEDs (especially ones designed for high-cri and specific color temp) are not likely to have much higher luminous efficiency than HMI and that manufacturer's claims about 40W fixtures being 500W equivalents are more related to LEDs having a higher efficiency non-adjustable lens and reflector on each lamp and on their standard 'flood' / 'spot' beams being narrower than the that of most fixtures to which they are compared. Plus they are likely comparing to tungsten softbox units which of course would be larger sources and have a very different quality of light from a 1x1 LED fixture. Comparing diffused light to very focused (LED) light would be quite difficult. For example, the photometrics published by litepanels on their 1x1 standard spot are pretty similar to that of lowel's tungsten rifa 500w softbox, but the softbox is a diffused single source with 4x the area and essentially 180degree beam angle versus a smaller LED multi-source fixture with a 30degree beam angle. A better comparison to LED might be a fresnel source, in which case the 40w 1x1 lightpanel photometrics are closer to a 150w Arri tungsten fresnel. So perhaps a 4x4 could be compared to 2k tungsten fresnel, where you'd need a bit of diffusion for the LEDs to match the evenness of the 2k and the 2k would need a bit of diffusion to match the size of the LEDs. I'd be interested to see how the 4x4 compares to a 575W HMI Par with varying degrees of 4x4 diffusion on each. Having not used a Litepanels 4x4 myself, I'd be interested to hear what someone who has thinks.

Hey Ari, Good luck! If you ever need a goofy no budget light kit in NYC, you know where to call. Let us know how it turns out.

hey Frank! Great to see you on here. Thanks to everyone for all the suggestions. I was actually worried it was too cutty rather than too slow. I have a lot of trouble getting proper final versions of my work to provide partly due to most things in my reel being finished and delivered in either 2-5 days or 1+ year. Ill write more when I'm back in the US in a few days.

I just updated my reel, and I would be interested in any criticisms and feedback on it since I update it periodically. You can see it here: http://noahyv.com/index.php?option=com_content&task=view&id=10&Itemid=19 Really I'm open to any criticisms, so please share your thoughts. Pretty much everything on there was shot entirely with my light kit and a Canon 7D or Sony EX1. A lot of it is actually from web comedy content I've shot, though shots from a dramatic feature and short I shot are in there too. I'd prefer to get more work in dramatic narrative so that is the aim of the reel, if that helps focus your critique. Thanks for watching!

Really nice concept and look. The anamorphic aspect ratio really feels right for this given the size of the shoe and the horizontal direction of all the action. Is it wider than 2.4:1? 16:9 with 1.5x anamorphic perhaps? (=2.66:1?) It's a great look and I'd be interested to know what kinds of things you did for the color, some of it almost looks HDR/tonemapped but never overly so, might just be the combination of low contrast grading and post sharpening. Only criticisms I have at the moment are I had to watch it several times to appreciate the backgrounds since the cuts are so fast, and it seems a bit long for that sort of concept. It is very interesting, but too repetitive for more than a :30. The lens looks really nice, but it sounds like it could be quite hard to use in a shot that does not have fixed focus and a static camera. I noticed the out of focus highlights in one ext night shot look a little funky, but i suppose that may be expected for an old lens, though they also dont look like your usual anamorphic bokeh. Any info you can give on this?

Really nice looking shots. I'd be interested to know how carefully you planned those shots and the lighting and what kind of lights you used. How long did it take to shoot the whole thing? Also, did you shoot the whole thing at 24p or did you do some 720/60p? My only critique is that I don't think the red/amber grad works well for the content, it sort of takes away from the beauty of the place, making it look like there is always a sort of surreal storm is overhead. I can see the effect you were going for, but maybe it just would have worked better if the effect were not as strong. The juxtaposition of colors is interesting, it just seems too intense.

From what I understand, isos 100,200,400,800,1600 are all native (hardware gain only), whereas 120,250,500,1000,2000 use the same hardware gain as the lower native step plus 1/3stop of software gain and 160,320,640,1250 use the same hardware gain as the higher native step plus -1/3 stop of software gain. Hardware gain increases are able to increase sensitivity with minor noise increases, whereas software gain can only appear to change exposure by shifting midpoints and throwing away shadow or highlight detail. The -1/3stop software gain levels appear to have the least noise because they simply throw away shadow/noise detail. for example isos 160-250 all have the same clip point with different black and mid points, so if a practical is blowing out at iso250 and you lower 2/3stops to iso160 to compensate, the practical should still be clipping but the rest of the image will appear darker and image noise will appear lower. Whereas if all steps were implemented with hardware gain, the practical that just clips at iso250 might not clip at isos 200 or 160 but the difference in noise level would be less dramatic. If someone has evidence otherwise I'd be interested to see it as I dont believe Canon has verified any of the theories out there.

True, but in practice sensor size will affect focal lengths used and circle of confusion which are factors in the equation. I think it is quite reasonable to talk about sensor size as having a direct and inversely proportional relationship to DOF since it will be very probable in most normal systems that do not have wildly unusual CoC or optics.