Jack Jin

the f900 is a contender, but ism't the viper from 2004? Also, where there is bokeh i don't see any signature rainbow bokeh from the 3ccd on the f900, so I think it's film? But no idea what telecine/scanner/colorgrading suite could be on, the lighting is super interesting though!

This is a japanese music video shot around 2000, but the colors look very interesting and almost print film like and switches to a cold and desaturated palette midway through, any idea how and what equipment they could have achieved this look on? Thank you very much.

What scanner was used? Because the end result looks fantastic with amazing 60s/70s color depth.

https://www.emotivecolor.com/ I think the tungsten version is kinda meh imo, so i would recommend the daylight version only

Highlight roll off on internal footage and external prores is actually quite nice, very similar to p4k with the highlight recovery on, but braw on the s5 does not have any highlight roll off in clipped areas, as all highlight roll off is built into the colorscience and processing of the camera, and braw does not use the camera's colorscience but rather blackmagic's V-log color science. I would highly recommend s1alex again, as it really does make grading so much easier, V-log is quite weird in it's hue and s1alex just makes an incredibly nice and filmic gamma curve since it converts the v-log to arri log c. The pocket 4k's rolling shutter and the s5's rolling shutter in 4k apsc crop are about the same, around 13ms, and the s5 in fullframe is at around 21-22ms. Which for reference, the a7m4 is at about 27-28ms and the pocket 6k is at about 19ms.

It seems like the image quality has a very lo-fi and early digital point and shoot look, but the exposure ramp reminds of things that I've seen on film, it also seems like the footage was shot on a slow framerate then speed up. But I don't know how they would make the film look like this, is there any telecine or scanner that degrades the film image like this? Thank you.

So as an owner of these two cameras, here are my thoughts: First, the bmpcc4k right now is shooting blackmagic raw, and unfortuntely blackmagic raw is not true raw, it's partially debayered in camera, and for some reason on the p4k and p6k camera's braw is especially soft and mushy, i would say that the hd cinema dng from the bmpcc original is comparable to the p4k in braw. You really isn't getting the full 4k resolution out of the blackmagic raw files, especially due to their high frequency detail filters that eliminate fine details such as hair, fabric textures, small texts, etc. This was done as a way to denoise the footage in camera, and was sharpened as well, creating a mushy image, with a green fringing commonly present in the prores and blackmagic files on the pocket 4k post firmware 6.2, I originally thought that it was my lens's fault, but turned out it was braw, as the cdng stills did not have those green fringing. Second, the bmpcc4k's colorscience is intresting, it's a lot easier to manually grade by hand in terms of getting a pleasing image, due to the fact that blackmagic seemed to have aimed for skintones instead of accurate primary colors. So even though the skintone is very healthy, the greens are very brown. And if you attempted to fix those greens in post, the afformentioned mushiness and green fringing becomes extremely obvious. I owned a lot of pocket 4k luts, like the p4k2alexa, p4kalex by emotive color, the buttery luts, and filmconvert nitrate. Which all aim to match the pocket to arri alexa or a cineon log film scan of different film stocks. But since the alexa has very vibrant greens, even though the color look great on the p4k2alexa and p4kalex, the mushiness and green fringing is exaggerated. Third, I would say that the lowlight performance of the p4k, is about 0.66 stop noisier then the s5, and the exposure lattuide to be a stop lower then the s5 when having highlight recovery turned on (the s5 does not have highlight recovery in blackmagic raw), due to the fact that s5 is better then any camera under 10000 dollars when it comes to usable dynamic range and latitude . Speaking of which I do own a blackmagic video assist 12g to record both prores and blackmagic raw on the lumix s5. The lumix s5's blackmagic raw is somehow quite a bit more detailed then the braw out of the p6k, and the 4k apsc crop blackmagic raw on the s5 definitely has more detail then the braw out of the p4k. But in terms of natural texture it still does not hold a candle to the normal prores 10bit footage recorded on the video assist. But the weird thing about the lumix s series of cameras is that their noise reduction seems to have some waffle shaped artifacts, which is apparent when you push the colorgrade in post at iso4000 and above, with a bit of banding present on bright areas that should not be there on normal prores 10bit as well. I have reported these issues to the panasonic team so hopefully they will get fixed soon, in which case I would whole heartedly suggest shooting prores, as it has much better performance and have smaller file size, but until then i will be sticking with the blackmagic raw on the s5. Fourth, almost all modern mirrorless and dslr have a very nice and strong IR cut filter built into their sensor, which means that even with high levels of ND filtration the skin tone, and blacks in the image stills normal. However, the blackmagic cameras have always had bad IR cut filters, it is better on the p4k then on the original pocket, but outside even without any ND black fabric tends to turn purplish, and skintone becomes quite pink as well. This is not completely fixable in post. Thus a external IR cut filter would be needed on the p4k. I first purchased hoya ir cut filter, which does filter ir light well, however the filter's coating is very soft. So after a bit of usage the filter was very scratched, and if you shoot any wider then 35mm in fullframe equivalent, then you will see green vigenetting on the edges. Which is a pain to correct in post, the only IR cut filter that I recommend is the heliopan IRUV cut, which doesn't have the green vignetting issue and has a very strong coating just like a normal lens filter. In terms of usability I prefer the p4k, the dials position just makes sense and since you don't need a video assist on top of the camera the balance is great. The motion on the p4k is also very smooth, smoother then the s5 in the same framerate and shutter speed. I would also recommend you to get s1alex for the lumix s5, because it converts The V-log from the s5 to arri log c, which means that it's much easier to get a very pleasing colorgrade from the s5. The s5's internal 150mbps codec is the best internal codec in mirrorless cameras, since it preserves color details in the shadows much better then the XF avc codec on canon cinema cameras and XAVCS-I codec in sony cameras. But it's still only 150mbps, which means that it does have some blockiness if you have very fast shutterspeed and a lot of moving stuff in the frame. But that's why you get a external recorder for prores. In the end I recommend the s5 a lot more, not only does it not have the mushiness and fringing issues of the pcoket 4k, it also is a fullframe sensor, thus having amazing dynamic range and better lowlight. It also can cover the entire image circle of your FD lenses, thus I recommend it more.

Especially with how fine grained ektachrome is, honestly on hbo with it's compressed bandwith it looks just like any other 2k finished alexa project, since 35mm ektachrome don't even have that much halation, feels like a bit of a waste imo, to not even get a proper positive out of it. Love the ektachrome positive look, don't know why they cross processed it.

They serious made kodak construct an entire new assembly line just to cross process the ektachromes? The trailer looks good but honestly feels like it's entirely achievable with digital with a print film lut tbh.

They are quite lit to some degree, but yeah, most of them seem to use a hard rec709 gamma with little roll off, creating a very distinct look. where as modern grading has a lot more roll off and looks very airy.

(Music video in question) https://www.bilibili.com/video/BV1za4y1t7LX?from=search&seid=790439131910187837&spm_id_from=333.337.0.0 (please watch till the end) Hi guys, I've recently been really enjoying a band called tokyo jihen (東京事変), and I stumbled upon their music video for the song shuraba(修羅場), and found the look to be very interesting, especially since it was released in 2005. It looks very airy with a very soft highlight roll off and very beautiful skintones, any idea what they could have shot, scanned and graded this on? Since most music video of that era looks super contrasty and saturated.

Scanner and grade imo is the most important aspect, if your memories of those shows came from blu ray, then they were often scanned on spirit sdc2000 or spirit 2k, and graded on davinci 2k or comparable systems with a very high contrast curve with easily blown out highlights with little knee. If you want to know the difference that different scanner and modern grading style then look at this: https://www.blu-ray.com/movies/Y-Tu-Mama-Tambien-Blu-ray/13035/#Review (original blu ray release) https://www.blu-ray.com/movies/Y-Tu-Mama-Tambien-Blu-ray/74263/(remaster on 4k scanity and modern colorgrade). I think the camera and lenses that you shoot on is not that important imo.

Isn't the internegative 2254 quite a bit different looking then say vision 3 50d? Since it's meant to replicate a negative and not an actual real life capturing filmstock, I think the colors is going to be a bit wonky?

Is there a way to print onto negative film? Since print film's look isn't always very apparent.

As someone who is practicing cinematography, Geoff's work and videos on cookeoptics has been some of the most informative and biggest inspiration for my learning process, I am very grateful for what he have contributed to the cinematography community. I have been following his battle with cancer on his blogs so I'm glad he is able to choose his on way of going out and I wish him the best, Rest In Peace.

Hi guys, I'm trying to see how real cineon log scans from modern scanners grade, I want to see if they match the negative cineon emulation that I'm using, so if anybody can provide a completely ungraded prores video or dpx or tiff stills of the film scan off of a spirit or scanity or lasergraphics scanner it would be much appreciated, especially if there are of tests with different exposures and medium shots of faces. thank you very much.

How much stops of dr does these sensors realistically have? Especially when compared to the DFT 4k and sdc2000. I know a lot of them can do multi pass hdr but I'm curious how these sensors stack up to the dft and sdc without the helpings of hdr. And how is the resolution from these bayer sensors compared to true rgb ccd sensors of the scanity and dft 4k?

I think Rank Cintel was a very popular telecine company in the standard definition days before Spirit and it's SDC series of film scanners took over. And blackmagic bought them and developed the blackmagic cintel scanner.

Yeah Unless you really really messed up the exposure (4+ stops), it's not noticeable, since kodak vision 3 negatives aren't very dense. But the cintel scanners still doesn't have the best image quality due to the 4k cmos sensor, but it's not bad either.

Oh the FPN issue only comes up in super dense or overexposed negatives and highlights, it's not a big issue if you shoot and expose relatively correctly. But this issue is much more prevalent in reversal film, since they are extremely dense, thus the shadows can be very dark, exposing the dynamic range weakness of the cintel scanner. Silbersalz is decent for what they are offering, the cintel machine isn't that bad, as I said, still miles ahead of any scanners in any photolabs, just not as good as other higher end cinema film scanners like the scanity or the lasergraphic and Xena 6.5k machines.

Silbersalz does this as an entire service, they ship out repackaged vision 3 stocks, and you return them to their lab, so they can process it in ECN-2, and then scan it on their cintel machine, and because the film is going through their horizontally, the 4k sensor gives out a 21 megapixel scan. I really do wish if there are other labs with higher end scanners that does this though, since the cintel doesn't have a great sensor with quite a lot of FPN. So Image quality is not great, but it's still miles ahead of any standard photo film scanner like the noritsu or fuji premeire or kodak pakon. And funnily enough I was just wondering about the same thing browsing through cinelab's webiste just a few days ago. @Robert Houllahan If theoretically after the pandemic, you guys have enough work force to reopen the stills processing service, would it be charged as vista vision or 35mm in terms of costs per feet? I would looove sending some decent amount of rolls to be proccessed and scanned on the xena 6.5k machine, Can't imagine how good the result will be, especially grading them in cineon log in davinci, that would be my dream in terms of film scanning.

This is only the case for prores, but for braw there is no 4k option yet, only 5.7k and 2.8k for slow mo. and even if there is a 4k option it will be cropped 1.5x for 1:1 pixel readout on top of the 1.55x crop of the apsc sensor. Because braw can't supersample information out of a cmos bayer sensor.

Actually Mainland China back then also had quite a lot of western gear in their arsenal, arriflexes and western lenses. I'm sure that this lens was used on something, but from what I heard from the older peeps, even before the reform they were mostly using either western or a bit of russian cameras or lenses. I mean the Chinese government would rather show off the ingenuity and the manufacturing prowess of the chinese people even on limited resources, like their hongchi brand of cars and their nuclear program that were made way back in the 50-60s era. So the whole thing about purposefully hiding doesn't make sense. But this is a cool looking lens nonetheless.

Oh I see, I saw the wiki for the genesis that said it cutted the vertical resolution by half through binning ? https://en.wikipedia.org/wiki/Genesis_(camera) So I presume that this moire issue is the reason why so many manufactures decided to use a 3ccd design instead? Or is it to just maintain compatibility with the b4 lenses?

Hi, sorry for replying to such an old thread, but isn't the genesis pixel binning the 5k sensor, and the f35 is doing a full sensor readout? Thus the moire could just have been caused by the pixel binning of the genesis.