Cine gamma settings in the EX1
 

I’m trying to find some info on the Cine gamma settings in the EX1 and how or what each setting is doing.
I currently use Cine 4 for all types of shooting indoor and outdoor but often find that the top end whites seem to be a bit milky or washed at times with Cine 4.
I’m experimenting now with Cine 1 but would like a bit more knowledge if the top end whites are being compressed or a Knee is applied.
Cine 1 seems good for outdoor shooting in bright natural light but is a bit dark for indoor shooting under lights I find.
I’m thinking of Cine 4 for indoor and Cine 1 for outdoor.
Can anyone shed their experience using different Cine’s for indoor or outdoor shooting.

Thanks

The entire picture profile topic- including the Cine Gammas- is rather complicated. You should definitely browse thru the Picture Profile Settings thread at the top of the forum, in the "Sticky" section.
However, as a basic concept, I think of Cine 1 as having more dynamic range. It is good for bright, colorful, high contrast outdoor shots. If you use auto iris, the footage will tend to be underexposed. I shoot manual iris and push the histogram towards the right, often allowing a little 100% zebra on the hot spots for Cine 1.
Cine 4 is good for lower light, lower contrast scenes and auto iris is usually pretty close to right on. Actually, Cine 4 is a good, reliable, all around gamma setting, particularly for run n' gun shots.
There is a huge amount of info & opinions on this stuff- probably more than you want to know, but in Adam Wilt's original review of the EX1:
ProVideo Coalition.com: Camera Log by Adam Wilt | Founder | Pro Cameras, HDV Camera, HD Camera, Sony, Panasonic, JVC, RED, Video Camera Reviews
he provides very good summary information regarding most of these settings. When I read carefully & gave it some thought, I was able to come away with some very useful concepts, a sort of basic framework for understanding what its all about.

Thanks Robert,
I shoot manual everything and have, are re-reading all the posts I can find.
Thanks for the link I forgot that one.

Cheers
Edit: I just found on the Sony brochure the cuve diagram which shows the Cine curve for each cine.
Also on the Sony UK site. XDCAM EX Shooting Tips: Issue 7 shows some diagrams

I have been doing a lot of research into the best gammas to use on the EX's for different lighting situations. The cinegammas are designed for shooting footage that will be graded, the images they produce are not entirely natural looking, however they do maximise dynamic range by compressing highlights and at the same time allocating a large part of the recorded signal range to mid tones and shadow detail. This is why shadows can look washed out or milky. However this also gives you more to play with in the grade.

Cinegamma 1 is tailored for shooting bright scenes or scenes where there will be large areas of highlights. CG1 is tailored for maximum highlight handling with lower shadow dynamic range compared to CG3 and CG4.
Cinegamma 2 is essentially the same as CG1, except the overall level is reduced making it broadcast safe at 0db. Cinegammas 1,3 and 4 all record up to 109% at 0db and 104% at -3db.
Cinegamma 3 has strong highlight compression but the compression starts later than CG1 so it's not as compressed as CG1. Midtones and shadows are stretched more than CG1. This gives more dynamic range to mid tones and shadows compared to CG1 at the expense of some highlight handling.
Cinegamma 4 is similar to CG3 but with the mid tones lifted still further so that it gives a brighter looking picture overall.

My preference is to use CG1 for outdoor, brightly lit scenes or scenes where highlight handling is critical. Then I use CG3 for indoor and scenes on dull days where extreme highlight handling is less critical, but shadow detail becomes more important. What I have also found is that when shooting interviews the cinegammas work best when they are slightly under exposed compared to standard gammas and then graded in post. If using cinegammas I tend to expose skin tones at around 60%.

Cinegamma 1 on the EX is the same as Hypergamma HG4 on the PDW-700, F900R etc and cineegamma 2 is the same as Hypergamma HG2. With CG1/HG4 : 460% D-range is compressed to 109%.

Cine4 provides the greatest dynamic range but that doesn't make it best for all situations. It rolls over highlights more than the other gammas (see attached) so if you want contrast in those regions (e.g. sunsets) you're better off with cine1. Cine4 gives you a lot of latitude for good middle exposure without clipping highlights, but you need your important areas of the image in the more linear portion of the gamma curve. Exposing to put the histogram to the far right (as the guiding principle) is likely to over expose important parts of the subject, rendering skin (for example) in unfortunately orange tones. Having said that, Cine4 is my default setting.

EDIT: that was written before seeing Alister's note. Good point about grading in post --- I tend to assume that and forget to mention it.

Alister,
I watched your presentation on the Sony site with footage taken from the EX1 and this got me thinking that Cine 4 was blowing the highlights out on certain situations.
Cine 4 seems to have the best overall brightness for a particular indoor setup but I guess as will most things it’s one’s own personal taste.
I have copied this fron the Sony site so I hope this is within the rules?
• CINE1 preset smoothes the contrast in darker areas and accentuates gradation changes in brighter areas.
• The CINE2 preset gives almost the same results as CINE1, and limits video signals under 100% video level.
• Select CINE2 when you wish to limit video signals under 100% video level in case you have constraint in white level head room at 100% in post production environment.
• The CINE3 preset emphasizes the contrasts in bright and dark areas more than CINE1 and CINE2, and accentuates gradation changes on the black side.
• The CINE4 preset emphasizes the contrast in dark areas more than CINE3. (Contrast in dark areas is weaker than STD3, and contrast in brighter areas is stronger.)

Edit:
Serena,
Not sure I understand that graph with Cine 2 and 3. Are they being clipped at 65 and 80.

Simon, not being clipped. I was only interested in comparing 1 and 4 to STD 1, so didn't do the extra work. The curves go to 109, but I stopped at 100.

Those Sony descriptions of the gammas I always found rather uninformative (other than cine2), even after I quantified the actual curves.

Yeah Cine 2 is clipped or rolled off to be legal it looks like.
I'm with you on Sony's gamma descriptions.

Serena, I think leaving out the top 10% of the curves is a mistake as much of the important highlight compression takes place in the top end of the curves. The top 15% of the cinegammas can contain as much as 40% of the input signal and this has a great bearing on how bright scenes will look.

I would agree. It seems from Serena's curve extrapolations, that Cine3 might actually yield higher latitude than C4 or even C1, depending on where things get clipped at 109%. The problem with extrapolations are that they are based on some mathematical model/curve fit routine. It is extremely difficult to decide what algorithm gives the best fit, so everything is a rather of a guess.

I agree with the comments. The trend indicates how each curve will treat highlights, confirmed by practical assessments. The intention was to quantify the differences between the curves as a basis for photographic testing. I could repeat for greater precision but it would be more to satisfy my curiosity than for practical needs. I should say that the curves were not merely extrapolations, as I think Bill implied. I did try to fit a mathematical curve to them, but all were a poor fit.

I read that the BBC recommend Cine 3 for best over all shooting. I found this on their web page for the EX1,3 settings.

The BBC recommendation is interesting as it implies several things. One is that all footage will be graded as Cine 3 will produce illegal levels and somewhat washed out blacks (in a gradable way). The issue that the BBC have when making such recommendations is that they are looking for a general purpose setting that will work for most applications. This doesn't mean it's always the best setting. I would have recommended a standard gamma ( 2 or 3) with a fixed knee at 85 as this will work for most shoots without producing illegal levels.

The problem with just using the cinegammas without grading is that while it might look nice in your edit, if your programme then goes through any kind of limiter or legaliser the highlights get chopped off. So anything broadcast stands to look a lot worse than when it was shot if the material is not carefully graded or legalised.

There is no right or wrong gamma to use, it will depend on your own specific requirements, the look you are trying to achieve and your workflow. The only gamma I will never use is standard gamma 1 as it is very noisy and never use cinegamma 2 at -3db as it gets clipped at 94% restricting lattitude.

Depends which document this was from. If it was from Alan Roberts documentation then generally he was after the best result for grading and 'film look'. For general use, as you say, STD gamma 3 with the set knee is the best option, and the reason why this is the default setting, although STD gamma 4 is also a good option if you wish to stay in that realm but want more scope for a grade afterwards.

I'm tending to use Cine 1 a lot for my outdoor shots now. It gives great results when combined with Tiffen DFX.

I think it's important to mention that when delivery is only planned to the computer screen or BD, those values recorded above 100% which would be illegal in broadcast or SD DVD are completely OK, and can contain information that can be displayed without clipping.

Therefore, I'm basically using 4 PP's, based on the following gamma curves:

- Cine1 for bright outdoor scenes without backlight (clear sky)
- Cine4 for dull outdoor scenes with strong backligh (like from white cloudy sky)
- Std1 for low contrast indoor scenes, with low light but without backlight
- Std4 for low light indoor scenes with backlight

I'm using Std gammas for lowlight indoors mainly because the overall brightess they yield is substantially higher than that of Cine gammas.

I'm largely with Piotr with the exception of using Standard gamma 1 in low light. Std 1 is very noisy and in low key shots the last thing you want in my opinion is a noisy picture.

I would use std 4 with some negative black gamma to achieve a similar look to std1 but without the noise.

Interesting observation, Alister - I haven't noticed that; will have an eye on the Std1 noise. Of course you're right that (especially with Std curves), you can basically use any single one, and tweak its black gamma / knee to mimic the other ones...

BTW. I use Std1 very sporadically indeed - only when I'm after the very punchy look with crushed shadows; it's usually accompanied by lowered detail settings and detail Frequency and Crispening way up... Perhaps this is the reason I haven't noticed any excessive noise.

I dont leave the Cinegammas when i shoot on my EX1.

Im still using Bills True Color Profiles with a few minor tweaking.

Cine1 for Exterior bright days.
Cine3 for General all round use - it also provides a nice normal saturated look.
Cine4 for lowlight night footage or where ever i really need to stretch the blacks.

All my profiles run -3 gamma and matrix on Hisat.

My matrix settings are all the same settings as Bills TC Profiles.


If im multicam shooting with EX3s i leave everything stock and run with Cine Gammas, Matrix on HiSat and -3 Gamma.

Currently grading a project that has footage shot both with Cine Gamma and Standard Gamma settings from multiple operators.. And judging from how much work i have to do to the Standard Gamma footage im never going to even think twice about only using Cine Gamma again. Standard gamma is just horrible.

Matter of taste and depending on the scene being shot, Joe. Technically, there is nothing "horrible" with STD gammas (except for the "abrupt highlights clipping" phenomenon, as discussed here:

http://www.dvinfo.net/conf/sony-xdca...-clipping.html

- but this can be easily avoided by setting Knee Auto=OFF).

Serena..

There's a few mathematical models that fit quite well, not the least of which is a cubic equation, but also, a sigmoidal model, an exponential growth model, a hyperbolic model, and a power model. I'm sure there are many others. At least as far as the one curve I studied, C4.

My point is that no matter what kind of mathematical model you choose, you're predicting, from some sort of linear regression, points that are more than 100% away from the data. In most circles, this is known as extrapolation. I can be fairly confident that most scientists would not put much credence in an extrapolation of more than 10%. Given the number of possible mathematical models available, the uncertainty is high enough to question the validity in the region of interest, near the 90-100% subject brightness intercept.

Well put Bill,

mind you I didn't understand a word of what you are talking about.

The key to good photography and video, is to get out and shoot some material, take a look at it and if it looks good then it probably is.

Bill is 100% right in doubting the value of far extrapolation (as opposed to interpolation, which usually gives reliable results).

Indeed, no extrapolation method (however sophisticated) can guarantee the shape of the unmeasured gamma curve fragment - Sony may as well have implemented some sort of a hard knee there...

OK, I will take your word for this.

I wonder if Degas talked with Monet about how many hairs they had on their brushes, and if the end result was better with 335 hairs, compared to a brush with 257 hairs.

All very interesting stuff.

I can remember trimming some of my prints for an exhibition, and carefully measuring each print to a fraction of an inch. My father (who was an artist) took the blade and did the whole lot by eye.

The lesson here was to do your creative work by eye, rely on your own judgement. Sure a Cinegamma 1 will give a different look to Cinegamma 4 or 2 or 3, but experience and practice will tell you which one to use for a particular scenario.

Ha ha, you're right, Vincent.

Both me and Bill are engineers, but as far as I'm concerned, I always judge my video picture (both quality and aesthetic - wise) by eye.

Obviously I didn't try hard enough, limiting myself to the polynomials available in Excel; I was only interesting in creating a nicely drawn line over my measured points and no curve I tried was satisfactory over the whole range. On extrapolation I agree totally with your comments. However I point out, again, that I didn't extrapolate the curves by any means other than measurement.

Have to agree with Piotr,
Joe may prefer Cine gammas for many reasons , perhaps all are legitimate but to declare that std gammas are horrible is off the charts to me. i use them all the time with excellent results.

The gamma curve options, like all the other camera options, are tools. Understanding the available tools and how and where to use them is essential for good results over the wide range of situations we may encounter. While I use Cine4 as default, there are many situations where the std gammas give far superior rendering. We need the whole tool-box.

Piotr, true. That's why it is important to know what is there. Be nice if Sony published the real curves, rather than just the crude representations in the brochure.

I am left wondering why Sony is so secretive about their gamma curve data.

Serena...in the FWIW category, there is a nice plugin for EXCEL that does a VERY nice job of curve-fitting data. It has a library of algorithms one can apply. It is rather pricey at $795, but, a good tool for engineers and scientists...
http://www.excelcurvefitting.com/XLfit5_new.html
...and they offer a free 30 day trial.

I am finding that I don't like the Cinegammas at all because of the nasty looking skin tones they tend to impart. There is nothing wrong with the camera because even the demo DVD Sony distributed for the EX1 that I saw before I bought the camera had those nasty orange / red skin tones. There are some Hollywood movies that have those skin tones as well. To each his own I guess, but for my work the standard gammas look much better to my eyes.

John

If you can to post some Settings for Profiles using the Standard Gammas I'd definitely be keen to put them into my camera and give them a try!

The few times i have shot with standard gamma and have had poor results and as soon as i found Bills TC profiles i was hooked on the Cines.

I have also seen a lot of bad EX1 footage around the place, and i can tell by either cutting the footage, or just watching it in shows on TV that there are many people here in NZ shooting with the standard gammas and the camera still setup out of the box. So i guess this has just put me off doing more experimentation with them.

Every client i shoot for, whether it be interlaced or progressive has liked the Cine Gammas.

But like i said, if you want to put up some profile settings or link me to a post that has some good ones im keen to have a play.

If you plan to use the Std Gammas instead of the Cine Gammas then make sure to turn off Auot Knee (which is on by default) and then set your knee point and slope. There is nothing wrong with Std Gammas but the Cine Gammas will give you more dynamic range in general. As has been said before Cine 1 & 2 will compress highlights so that you will get more range in bright scenes. Cine 3 & 4 bring up the low end of the gamma curve so they work better in low light scenes. I would never use Cine 3 or 4 outside on a bright day, unless I was shooting into the shadows.

Gamma isn't changing your skin tones, it's not a color change at all. It's more like a contrast change in the mid tones.

John, I'm not sure what you mean by nasty skin tones. If its a color issue that should have nothing to do with cine gammas. However I suspect you are seeing compression in the brighter parts of the flesh tones that looks nasty with the Cine Gammas. That's why I don't use them based on my first tests with the camera.

I suspect that can be overcome by exposing a little lower than you normally do to keep flesh tones lower out of that compressed area. I plan to do some testing on this one of these days to see if I can make the cine gammas work for me. But I thought it was dangerous wherever there were skin highlights. Especially on backlight subjects.

How do those of you using cine gammas deal with this?

The cinegammas are designed to be graded. They work best when your skintones are right in the middle of the curve so somewhere between 55% and 65%. This is a little lower than you would expose using the standard gammas. The idea being that skin tones are in the most linear, uncompressed or stretched part of the curve. That gives you the maximum leeway in the grade. If you over expose skin tones even a little using the CG's will will start running into compression issues which will skew the colours.

If you are finding that your overall pictures are red or have a red shift try using the Cine Matrix with the level set to +35. I find the cine matrix has a more natural response than the standard matrix.

You can get just as much lattitude using the standard gammas as the cinegammas, but you will have to set the knee point and slope to the appropriate levels for the scenes you are shooting. Careful use of Knee and black gamma can give very good results without affecting the way you expose skin tones and requires less grading for a pleasing looking picture. I use standard gamma 2 or 3 with the knee set to 85 and slope at +10 as a general setup where I don't want or need to grade.

Alister,

I also go with Standard gammas ( STD3 ) and agree about no Auto Knee, but my tests found setting the point at 85 to be dangerous ( unwanted compression) though I probably tested with a slope of 0.

I've set mine at 93 with 0 slope as a general setting but I'll try yours out.

85 is generally considered low for standard gammas. The danger is that if you expose skin tones in STD gamma at around 75% or even 80% there is a risk of compression in harsh lighting when you need to compromise with the exposure.

95% I think it more sensible for general use in many cases, although 85% can be used with care. I wouldn't touch the knee slope unless you can measure things properly. Set up correctly, as Alister states, you can actually get just as much range out of the STD gammas as the Cines. But you really need to know how to set up the black gamma and knee slope.

A good starting point is STD gamma 4 ("BBC gamma"). The trouble is that setting up the standard gammas in this way leads to less pleasing roll off in highlights because unlike the cinegammas there is no smooth transition from the shadows to the midtones to the highlights AFAIK.

The Cinegammas were meant to be graded, however I would really apply this more to 1 and 2. I regard Cine 4 to be more like Panasonics Video Rec cine curve and good for general use even without grading.

Regarding exposure with the Cine curves, you have to be careful with skin. As Alister rightly points out you should expose at a lower level to what you might be used to on STD gammas. One issue here is that you also have to be careful of under exposing. There is a sweet spot that is just right, but miss it and things can look a bit odd, especially in the shadows too.

Anyone know roughly how many stops of latitude the EX1 has? Say, for example, using Cine Gamma 1?

I have done a few test with the Cinegammas.
This is nothing new for anyone but I want to explore what is happening.
I set up a shot with my Wife with natural light coming into the room and exposed to her face @ -3db,f 4.8. Zebras set to 70% and then 80% moving to 100%, lighting stayed the same exposure for this quick test.
This is by no means a proved scientific method but I use what I have.
Cine 4, exposed Zebras all over face @ 70%, less @80% reducing to none @ 100%
Cine 3, exposed Zebras 1/3rd of face @ 70%, less @80% reducing to none @ 100%
Cine 2, exposed Zebras none over face @ 70%, none @80% none @ 100%
Cine 1, exposed Zebras just a strip @ 70%, none @80% none @ 100%

So my conclusion is I found Cine 3 to produce the best. A pleasing tone in the blacks and also a nice soft tone on the top end.
Cine 4 is great for situations where overall brightness need to come up in the scene such as a wedding reception room or something similar.
Cine 2 is great for bright harsh outdoor situations where top end needs to be clipped.
Cine 1 is also great for outdoor situations but leaves the top end a bit more intact.

Conclusion for me to use on my next shoot tomorrow. Live gig to Imag
Cine 3 best overall for indoors.
Cine 1 solid for out door situations.

Cheers

The EX has around 11 stops of latitude when using CG1, CG3, CG4 or with Std gamma plus correctly set knee.

I find having the knee on the EX at 85 with the slope at 15 works well for me, but then most of what I do doesn't involve interviews, instead having to deal with high contrast exteriors.

OK, I've redone the gamma curve plots. Just for Cine 4, Cine 1 and STD 1 (knee off). The wiggles at the bottom of the curves due to data uncertainty.