I've had this question I wanted to clear out before I shot my upcomming project next week.
Usually when I shoot any type of video I do my best at having information on all the zones of luminance of my frame, that usually results on low contrast images etc...which I find better to work with on post.
Later down the process I contrast the image and color correct it to my desired look. Now, regarding the blacks, which is the best method of getting as much of information as possible, setting the camera up for a 7.5% pedestal (and taking the master ped up to a +1 or +2) or leaving it at 0 and going to the positive scale on the master ped, or a mix of both?
Also, most of the time I use the Cine like V gamma setting with the HVX 200.
I'm not sure that if raising the camera setup to 7.5 and leaving the master ped at 0 or +1 will rescue more information from the toe rather than leaving the setup at 0 and rasing the ped to the +4 or +5. WHICH IS BETTER FOR THE GRAIN? Shouldn't I be messing with the 7.5 setup?
Thanks guys, this is just one of those things that i've been doing since the pd-150 era and didn't really think it up for the recent times.
p.s. I will not be able to conduct any testing with a HD monitor with the color corrected images prior to production in order to evaluate the noise with different camera settings... So i'm hoping to hear your opinions and stories.
I too would be interested in knowing others setup, as this is certainly a subjective topic. Personally, I haven't spent much time tweaking and testing the various settings, so I know I'm not an authority in this area, but that said I keep all things set to zero and I have a gamma setting that I like to throw on in FCP. (Its usually in the 1.1-1.4 range)
I've heard people saying that on the 1080 setting, the blacks (with emphasis on nite ext) are very noisy. Thats why I only shoot 720pN. But under this setting (720pN) I really have little info on the pedestal and camera setting (7.5 or 0) for acceptable noise on the toe of the curve.
Another thing i've noticed on the camera's menu is the zebra detect: it goes up to 105 ire. Does this mean the camera handles hilights up to 105, or is the 105 mark is just there for a reference of how high is your hilight's going?
Hey guys,- I found this thread after making a similar post...sorry for redundancy.
Please let me know what you found out. I have similar questions...
I have been using the HVX with great results in 720 24PN in all kinds of exterior daylight. My next shoot will be at night, on the streets of a big city with all the light sources you can imagine, from sodium vapor street lights, to tungsten and neon.
I have been shooting with a consistent under exposure to preserve detail in the highlights, but now i am worried about losing detail in the shadows. Is there a particular gamma curve (cine Like V or D... not sure what they mean...) or 24PN compatible setting that you can advise me to try? Is there a setting for black levels that will allow me to keep my beautiful film like look and recover shadow detail?
thanks for helping me learn!
shoot starts Monday 8 /6 I appreciate your prompt response.
The setup (0 or 7.5) won't do anything in P2 mode, its intended for tape mode.
As far as pedestal settings, higher settings will give your image less contrast, but noise "grain" more visible. I would shoot some tests and see how they look in your editor, also looking at the waveforms would help to see if any parts of the image are getting heavily clipped. Having an external monitor like the Marshall helps as you can quickly adjust the brightness/contrast and see what is not visible on the HVX lcd.
For night time shooting, you'll probably have the lens wide open and the hvx will still want more light. Avoid turning up the gain, but go through the presets and find the one that gives you the most data in the darks and bright areas in the least contrasty gamma. You want to get the most image data you can, go through the presets and do some tests if you still have time. Then view each in your editor and see how well it color corrects.
Im still trying to find the best setting for this camera (hvx200) with exactly the same frame-of-mind, which is to get the maximum latitude and keep the color and contrast for post.
on the DVX100 i used to use the CINELIKE D , which seems to me gives the most latitude. on the hvx200 that doesnt seems so, the HD NORMAL setting seem to do a better job. in any case the CINELIKE V compress the whites and black and therefor not a good choice if you want more latitude, you better be using the CINELIKE D in my opinion.
anyway, to me it looks like the HD NORMAL is good and also the CINELIKE D, both of them can be manipulated nicely in post. i shot my first short film with CINELIKE D and it looks great. now im using the HD NORMAL for a feature documentry and it also looks great. it looks realy close and more natural even..
it just so hard to find a real HD monitor and HD system for a real test
Check out this comparison of the HVX's different gamma presets. What Adam doesn't mention in the article is that the different gamma presets are designed to be used in conjunction with exposure to render proper midtones. So a gamma like "HIGH" that looks washed-out in the test will actually look quite good with good shadow detail, once you iris down for proper midtones. But his examples as shown are necessary so you can see how the different curves are processing the signal.
For my money CINELIKE D gives the most information for post correction, since it's closest to a "film look" from the presets in this camera. I haven't tested the HD NORM curve for the extremes of highlight and shadow detail, but all of the preset curves are just pushing around the luminances that the camera is already capturing. Except for the application of knee compression, you're really not getting any more dynamic range with one curve over another -- it's all the same info, just pushed up or down across the luminance range. In my mind, the less you have push those luminances in post, the fewer artifacts you will have. Also keep in mind that the CINELIKE color matrix is designed to compensate for the softer-looking color you get as you flatten out the gamma.
Noise is inherent to the signal, and becomes more visible as you boost it -- doesn't matter if it's boosted in camera or in post. Recording format compression can actually make it appear worse, so you don't really gain anything (no pun intended) by shooting with a gamma preset that keeps the blacks down and lifting them in post, vs. shooting with a gamma preset that gives you brighter, noisier shadows in camera.
Like many digital cameras, the HVX captures (and records) luminances above 100%. If you want your signal to be broadcast safe (and also have a more "organic" film-to-video look), you'll want to somehow clip or compress those highlights to not exceed 100% on output. I've had good luck simply lowering the RGB levels down to 100% in post, sometimes boosting the midtones up a little to compensate for the slightly lowered luminance. It rounds off the highlights very beautifully, and gets rid of that harsh "digital" look when going out to DVD.