Tag Archives: picture

EX1 and EX3 Picture Profiles.

These are the picture profiles that I am currently tending to favour for the EX1, EX1R and EX3. Please remember that picture profiles are entirely subjective. These settings work for me, that doesn’t mean they are perfect or for everyone. I like the images the cameras produce when I use these profiles. Please feel free to adapt them or modify them any way you choose. They work on any of the current EX cameras.

Vivid – Designed to help match the EX to a PDW-700. Gives vivid colours with a small shift away from yellow.

Matrix – Cinema, Matrix Level +60

R-G +8,  R-B +10,  G-R 0,  G-B +15,  B-R +5,  B-G +6

Detail Level -10 Frequency +20, Crispening -40 (if using gain use crispening +14)

Gamma Cinegamma 1

Black level -3, Black Gamma -35

Low Key Saturation -10

Natural C4 – Designed to give a neutral, natural looking image.

Matrix – Cinema, Matrix Level +35

Detail level -7, Frequency +30, Crispening -40 (if using gain use crispening +20)

Black Level -3, Low key Saturation -15

AC Punch – Gives a very high contrast, bold look.

Matric – Cinema, level +40

Gamma Standard 2, Knee level 80, Slope 0

R-G 0,  R-B +1,  G-R +12,  G-B +2,  B-R +11,  B-G 0

Detail Level -10, Frequency +30, Crispening -45

Black Level -4, Black Gamma -20.

AC Good to Grade – a general purpose setup to give good grading possibilities.

Matrix – Cinema, Level +25

Gamma Cinegamma 1 (Do not use -3db gain)

Detail Level -7, Frequency +45, Crispening -45 (use +35 if using gain)

Black Level -3.

AC-SD Camera look. To mimic an older SD camcorder based on a DSR400, good for HD to SD conversion.

Matrix – Cinema, Level +15

Detail Level +20, Detail Frequency -35, White Limit +35, Black limit +45

Knee, Manual, Level 90, Slope 0.

Gamma Standard 2, Gamma Level +5

Black Gamma -10

Black Level -10

 

 

Enjoy! Any feedback or suggestions welcome. Let me know of any profiles that you come up with that may be of interest to others.

 

Advertisements

S-Log. A Further In Depth Look.

Well I posted here a few days ago about how Data was distributed across the S-Log curve. David williams (thanks David) questioned some of the things in my post raising some valid questions over it’s accuracy, so I withdrew the post in order to review it further. While the general principles within the post were correct (to the best of my knowledge and research) and I stand by them, some of the numbers given were not quite right and the data/exposure chart was not quite right.

Before going further lets consider the differences between the a video sensor works and the way our eyes work. A video sensor is a linear device while our own visual system is a logarithmic system. Imagine you are in a room with 8  light fittings, each one with the same power and light output. You start with one lamp on, then turn on another. When you turn on the second lamp the room does not appear to get twice as bright even though the amount of light in the room has actually doubled. Now with two lamps on what happens when you turn on a third? Well you wouldn’t actually notice much of a change. To see a significant change you would need to turn on 2 more lamps. Now with 4 lamps on to see a significant difference you would need to turn on a further 4 lamps. Only adding one or two would make little visual difference. This is because our visual system is essentially a logarithmic system.

Now lets think about F-Stops. An f stop (or T-stop) is a doubling or halving of exposure. So again this is a logarithmic system. If with one light bulb your scene is one stop then to increase the scene brightness by one stop you must double the amount of light, so you would add another light bulb. Now to increase the scene brightness by a further stop you would have to take your existing two light bulbs and double it again to 4 light bulbs, and so on… 2, 4, 8, 16, 32, 64….

Now going back to a video sensor, take a look at the illustrative graph below. The horizontal scale is the number of lightbulbs in our hypothetical room and the vertical scale is the video output from an imaginary video sensor in percent. Please note that I am trying to illustrate a point, the numbers etc are not accurate, I’m just trying to explain something that is perhaps miss-understood by many, simply because it is difficult to understand or poorly explained elsewhere. The important thing to note is that the plotted blue line is a straight line, not a curve because the sensor is a linear device.

s-log-sensor11 S-Log. A Further In Depth Look.
Linear Output from Camera Sensor

 

Now look at this very similar chart. The only difference now is that I have added an f-stop scale to the horizontal axis. Remember that one f-stop is a doubling of the amount of light, not simply one more lightbulb. I have also changed the vertical scale to data bits. To keep things simple I’m going to use something close 10 bit recording which actually has 956 data bits or steps (bits 64 to 1019 out of 1024 bits), but lets just round that up to 1000 data bits to keep life simple for this example.

s-log-sensor-data S-Log. A Further In Depth Look.
Sensor data and f-stops

So we can see that this imaginary  video sensor uses bits 0-50 for the first stop, 50-100 for the second stop, 100-200 for the third stop, 200-400 for the fourth and 400-800 for the fifth. So it is easy to see that huge amounts of data are required to record each stop of over exposure. The brighter the image the more data that is required. Clearly if you want to record a wide dynamic range using a linear system you need massive numbers of data bits for the highlights, while the all important mid tones and shadow areas have relatively little data allocated to them. This is obviously not a desirable situation with current data limited recording systems, you really want to have sufficient data allocated to your mid-tones so that in post production you can grade them satisfactorily.

Now look what happens if we allocate the same amount of data to each stop of exposure. The green line is what you get if, in our imaginary camera we use 200 data bits to record each of our 5 stops of dynamic range. Does the shape of this curve look familiar to anyone? The important note here is that compared to the sensors linear output (the blue line) as the image brightness increases less and less data is being used to record the highlights. This mimics the way we see the world and helps ensure that in the mid ranges where skin tones normally reside there is lots of data to play with in post. Our visual system is most acute in the mid range. that’s because some of the most important things that we see are natural tones, plants, fauna and people. We tend to pay much less attention to highlights as these are rarely of interest to us. Because of this we can afford to reduce the amount of information in video highlights without the end user really noticing. This technique is used by most video cameras when the knee kicks in and compresses highlights. It’s also used by extended gamma curves such as cinegamma’s and hypergamma’s.

s-log-sensor-log-curve S-Log. A Further In Depth Look.
Log Curve, 200 bits for each stop.

Anyone that’s seen a hypergamma curve or cinegamma curve plot will have seen a similar shape of curve. Hypergammas and Cinegammas also use less and less data to record highlights (compared to a linear response) and in many ways achieve a similar improvement in the captured dynamic range.

hypergamma-curves-jpeg S-Log. A Further In Depth Look.
Sony Hypergamma Curves

Hypergammas are not the same as S-Log however. Hypergammas are designed to be useable without grading, even if it’s not ideal. Because of this they stay close to standard gammas in the mid range and it’s only really the highlights that are compressed, this also helps with grading if recording using only an 8 bit codec as the amount of pushing and pulling required to get a natural image is less extreme. However because the Hypergammas allocate more data in the 60 to 90 percent exposure range to stay close to standard gamma the highlights have to be more highly compressed than S-Log so there is less highlight data to work with than with S-Log.  If we look at the plot below which now includes an approximate S-Log curve (pink line) you can see that log recording has a much larger difference from a standard gamma in the mid ranges, so heavy grading will be required to get a natural looking image.

s-log-hg-curve S-Log. A Further In Depth Look.
Hypergammas and S-Log curves

Because of the amount of grading that will normally be done with S-Log, recording the output using a 10 bit recorder is all but essential.

When I wrote this article I spent a lot of time studying the Sony S-Log white paper and reading up on S-Log and gamma curves all over the place. One thing that I believe leads to some confusion is the way Sony presents the S-Log data curve in the document. The exposure is plotted against the data bits using stops as opposed to image brightness. This is a little confusing if you are used to seeing traditional plots of gamma curves like the ones I have presented above that plot output against percentage light input. It’s confusing as Sony forget that using stops as the horizontal scale means that the horizontal scale is a log scale and this makes the S-Log  “curve”  appear to be a near straight line.

I have not used S-Log on an F3 yet. It will be interesting to see how it compares to Hypergamma in the real world. I’m sure it will bring some advantages as it allows for an 800% exposure range. I welcome any comments or corrections to this article.

 

PMW F3 Picture Profile Smorgasbord.

I’ve been working some more on picture profiles for the PMW-F3, mainly matrix settings. You can download the full set by clicking here: ac-profiles. Download the zip file, unzip and place the “Sony” folder in the root of an SxS card or SD card in an adapter. Place the card in the camera and go into the “picture profiles” menu and select a picture profile and then “ppdata” and “recall” to load the data into your camera. This will overwrite any PP’s you already have.

Here’s the latest settings I have:

ALL use Detail level -17, Frequency +20, Aperture +25 unless otherwise stated.

AC Warm1: Warm look, less blue/yellow

Cinegamma 1, Black Gamma -25, Black Level -2.

Matrix: Standard, level +8, R-G +14, R-B +12, G-R +4, G-B +8, B-R +4, B-G -18

AC Cool1: Stark cool look, maybe day for night.

Cinegamma 1, Black Gamma -25, Black Level -2.

Matrix: Standard, level +22, R-G -44, R-B -24, G-R -34, G-B =28, B-R -7, B-G -69

AC Elec1:  Electronic, vivid look.

Gamma STD1, Black Gamma -20, black level -3, Detail Level -10, Frequency -40

Matrix Hi-Sat,

NAT1CG-1: Neutral Look, natural colors, less yellow/green.

Cinegamma 1, Black Level -2

Matrix FL-Light, Level +3, R-G +2, R-B +2, G-R +8, G-B +8, B-R -8, B-G -6

Note that for most of these I have used a cinegamma, that is because I would assume that post work will be done on the footage. If your not planning on doing any grading or post work you should consider using a standard gamma which will give a richer looking image or cinegamma 2 which is broadcast safe.

PMW-F3 Picture Profiles. First Batch.

OK here we go. Here are some notes from testing my PMW-F3. First thing is… aliasing… a zone plate looks pretty bad with a fair amount of aliasing. I had heard rumours of this from others with pre-production units, but in the field I had not seen anything that would worry me. While the zone plate is not pretty, real world aliasing looks acceptable. I usually use brickwork and roof tiles to test for moire and these look clean on my F3. I think a fine patterned shirt could cause concern and I need to look into this further. I am surprised that there is not more about this on the web!

Excessive detail correction does increase the aliasing, however turning detail and aperture off does not reduce the aliasing significantly. Keep the detail level below -15 to avoid increasing the strength of the aliases. Above -15 the aliasing artefacts are more noticeable. Detail “Off” appears to be the same as Detail -25. Below -25 the image softens, below -45 very noticeably and there are some strange increases in aliasing below -50. For the moment I will be using detail at -17 or off.

The aperture setting can be used to add a little sharpness to the image to compensate for not using detail or a low detail setting. Aperture does not increase the appearance of the aliasing artefacts as strongly as the detail correction. I like the added crispness I can get with Aperture set to +30 combined with detail at -17. I would strongly recommend against using a raised aperture setting if you have detail higher than -15 as this will add sharpness to any detail corrected aliases and lead to twittering edges on horizontal and vertical lines.

Colours have that usual Sony look. Not bad and pretty natural looking, but for me a little on the green side. For a more natural 1:1 look I quite like these Matrix settings:
R-G +10, R-B +4, G-R 0, G-B +14, B-R +3, B-G -3, Std Matrix.

For a more Canon like look with Rec-709 Matrix I came up with these:
R-G -2, R-B +9, G-R -11, G-B +2, B-R -16, B-G -10, Std Matrix, level +14, Blk Gamma -20

For use with Cinegamma 1 I use the above with Matrix Level +25, Blk Gamma -36. Highlights are a little washy, but as with any Cinegamma the best results are obtained by grading in post production.

What is “Crispening” and how does it effect the picture?

What is “Crispening” and how does it effect the picture?

Crispening is one of the adjustments you can make in many of Sony’s video cameras that adjusts the way the image is sharpened via the detail correction circuit. On an EX1 or EX3 it is in the Picture Profiles section. If use wisely Crispening can be used to help deal with camera noise by making it less visible, thus giving a cleaner image. Crispening works across the entire luma (brightness) range. It’s really difficult to explain how the level adjustment works, it is a threshold adjustment for the detail circuit, but I’ll have a go anyway.

First off lets consider how the detail circuit works. The camera uses delay circuits to compare how the brightness (luma) levels of adjacent pixels are changing, both from left to right and line by line. If the circuit sees a rapid change from light to dark or dark to light (or light to lighter, dark to darker etc) the circuit regards this as an edge and detail correction is applied by brightening or darkening the transition, exaggerating the edge. This is seen in extreme cases as a black or white halo around edges.

On the EX cameras crispening works by adjusting the threshold at which the light to dark transition between pixels triggers the application of detail correction. So when you set a negative number, say -99 even the slightest luma difference between pixels will have detail correction applied. Set it to +99 and it takes a much greater luma change to trigger the detail circuit.

What you need to understand is that if you set crispening such that the threshold before detail is applied is 100mV (for example) then between 0v (black) and 99mV little to no detail correction will be applied, keeping blacks clean by not applying detail correction to any noise with an amplitude less than 100mV. But if there are subtle textures in the image, going say from 500mV to 599mV (mid tones) then no detail correction will be applied here either, so the image will appear a little softer, only larger luma changes of more than 100mV will have detail correction applied. These small luma changes can be anywhere within the full luma range and it is not confined just to the darker parts of the image.

Raising the crispening level setting to a positive number raises the threshold at which detail is applied to the image, so a high number prevents detail correction from being added to small luma changes. A negative number means that detail correction will be applied to smaller luma changes, this increases the appearance of noise but also makes textures appear sharper.

One thing to consider is that the noise the camera produces is not only in the blacks. If the noise amplitude (level) is for example 5mV, then if you have a subject at 500mV (mid tones) it will still have random 5mV noise added to it. It just tends to be that noise is most visible in the blacks as 5mV of noise on a 5mV (very dark) signal is modulating (varying) the signal by 100% so it’s quite obvious, however 5mV on top of 500mV is only 1% so less obvious, but still there and still visible.

You should remember that the cleaner you can make the recorded image the less stress there is on the codec. This in turn means less mosquito noise and macro blocking giving an image that looks cleaner still and grades better. I struggle to see the difference between crispening at 0 and at +20 in most normally exposed shots, but if I look closely I do see less noise in shadow and low contrast areas. Low contrast areas tend to have little detail anyway, so being able to clean these up a little helps in post production.

Sony have a PDF about it here:?http://www.sony.co.uk/res/attachment…6605183226.pdf

Brewing up a scene file: Black Gamma


In the posts above I looked at how the gamma curves effect the contrast range within the picture and highlight handling. I also noted that while I like the latitude (range) offered by using the Hypergammas that they produce a very flat looking picture. One of the adjustments that you can make to the Gamma curves is the Black Gamma. Adjusting the Black Gamma stretches or compresses the bottom part of the gamma curve, this makes the darker parts of the picture darker (negative setting) or brighter (positive setting). When setting the Black Gamma you will find 4 different ranges to choose from. Low, Low-Mid, Hi-Mid and High. These settings determine the range over which the black gamma works. Low only effects the darkest 10% of the image, L-Mid the bottom 20%(approx), H-Mid the lowest 30%(approx) and Hi the lower 35% (approx). So if you just want to make your deep shadows and blacks darker you would use Low. If you want to make the overall image more contrasty you would use H-Mid or Hi. I like to give my images a bit more impact so I often use H-Mid at -30. If the pictures are to be graded I would not use any negative black gamma.

Brewing up a Scene File: Gamma and Knee

Before anyone complains that I have missed stuff out or that some technical detail is not quite right, one of the things I’m trying to do here is simplify the hows and why’s to try and make it easier for the less technical people out there. Lets face it this is an art form, not a science (well actually a bit of both really).

So what is a gamma curve anyway? Well the good old fashioned cathode ray tube television was a very non-linear device. You put 1 unit of power in and get one unit of light out. You put 2 units in and get 1.5 units out, put 3 in and get 2 out… and so on. So in order to get a natural picture the output of the camera also has to be modified to compensate for this. This compensation is the gamma curve, an artificial modification of the output signal from the camera to make it match TV’s and monitors around the world. See Wikipedia for a fuller explaination:   http://en.wikipedia.org/wiki/Gamma_correction

So, all video cameras will have a gamma curve, whether you can adjust it or not is another matter. Certainly most pro level cameras allow you some form of gamma adjustment.

The PMW-350 has 6 standard gamma curves, these are all pretty similar, they have to be otherwise the pictures wouldn’t look right, but small changes in the curve effect the relationship between dark and bright parts of the pictures. Todays modern cameras have a far greater dynamic range (range of dark to bright) than older cameras. This means that the full dynamic range of the sensor no longer fits within the gamma curves used for TV’s and monitors. In broadcast television any signal that goes over 100% gets clipped off and is discarded, so the cameras entire brightness range has to be squeezed into 0 to 100%. The PMW-350 sensors are capable of far more than this (at least 600%) so what can you do?

The older and simpler solution is called the “Knee”. The knee works because in most cases the brightest parts of a scene contains little detail and is generally ignored by our brains. We humans tend to focus on mid-tone faces, animals and plants rather than the bright sky. Because of this you can compress the highlights (bright) parts of the picture quite heavily without it looking hugely un-natural (most of the time at least). What the knee does is takes a standard gamma curve and up near it’s top, bends it over. This has the effect of compressing the brighter parts of the image, squashing a broad range of highlights (clouds for example) into a narrow range of brightness. While this works fairly well, it does tend to look rather “electronic” as the picture is either natural (below the knee) or compressed (above the knee).

The answer to this electronic video look is to replace the hard knee with gentle bend to the gamma curve. This bend starts some way down the gamma curve, very gentle at first but getting harder and harder as you go up the gamma curve. This has the effect of compressing the image gently at first with the compression getting stronger and stronger as you go up the curve. This looks a lot more natural than a hard knee and is far closer to the way film handles highlights. The downside is that because the compression starts earlier a wider tonal range is compressed. This makes the pictures look flat and uninteresting. You have to watch exposure on faces as these can creep into the compressed part of the curve. The plus point is that it’s possible to squeeze large amounts of latitude into the 100% video range. This video can then be worked on in post production by the editor or colorist who can pull out the tonal range that best suits the production.

These compressed gamma curves are given different names on different products. Panasonic call them “Film Rec”, on the EX1 they are “Cinegammas” on the PMW-350 they are “Hypergammas”. The 350 has four Hypergammas. The first is 3250. this takes a brightness range the equivalent to 325% and compresses it down to 100%. HG 4600 takes 460% and squeezes that down to 100%. Both of these Hypergammas are “broadcast safe” and the recordings made with them can be broadcast straight from the camera without any issues. The next Hypergamma is 3259. This takes a 325% range and squeezes this down to a 109% range, likewise 4609 takes 460% down to 109%. But why 109%? well the extra 9% gives you almost 10% more data to work with in post production compared to broadcast safe 100%. It also gives you the peak white level you need for display on the internet. Of course if you are doing a broadcast show you will need to ensure that the video levels in the finished programme don’t exceed 100%.

My preferred gamma is Hypergamma 4 (4609) as this gives the maximum dynamic range and gives a natural look, however the pictures can look a little flat so if I’m going direct from the camera to finished video without grading I use either a standard gamma or use the Black Gamma function to modify the curve. I’ll explain the Black Gamma in my next post.

There are 6 standard gammas to choose from. I like to stick with gamma 5 which is the ITU-709 HD standard gamma. To increase the dynamic range I use the Knee. The default knee point setting is 90, this is a reasonable setting, but if your shooting with clipping set to 100% you are not getting all the cameras latitude (the Knee at 90 works very well with clipping at 108%). Lowering the knee down to 83 gives you almost another stop of latitude, but you have to be careful as skin tones and faces can creep up towards 83%. It’s very noticeable if skin becomes compressed so you need to watch your exposure. This is also true of the Hypergammas and with them you may need to underexpose faces very slightly. The other option is to set the knee point to 88 and then also adjust the knee slope. The slope is the compression amount. A positive value is more compressed, negative less compressed. With the knee at 88 and slope set to +20 you get good latitude, albeit with quite highly compressed highlights.

If you want to play with the gammas and knee and see how they work one method you can use is to use a paint package on your PC (such as photoshop) to create a full screen left to right graduated image going from Black to white. Then shoot this with the camera (slightly out of focus) while making adjustments to the curves or knee and record the results along with a vocal description of each setting. Import the clips into your favorite editing package and use the waveform monitor or scopes you should be able to see a reasonable representation of the shape of the gamma curve and knee.

So my Gamma Choices are:

For material that will be post produced: Hypergamma 4609 (HG4)

For material that will be used straight from the camera: Standard Gamma 5 Knee at 90 with clip at 108% for non broadcast or Knee at 88 with slope +20 with white clip at 100% for direct to broadcast.

Stetting up video cameras without charts.

There is far too much emphasis on color charts and 100% one to one – set it up with a scope settings. Very often a 100% accurate one to one response won’t look right as the video gamut is smaller and lopsided than that of the human eye so a small amount of skewing of the color gamut can often help produce a picture that visually looks more natural. One of the very best ways to set up a camera is to use a high quality color photograph of a known scene. Shoot the photograph and look at the picture on a monitor and adjust until it looks right. This will give a more natural looking image than aligning with charts and scopes and is a technique that has been used since the very beginnings of color television. I have a scene that contains vibrant colored cars, green fields and trees, buildings and blue sky. I have a dozen large copies of this picture and use it whenever I am making camera adjustments to make sure my pictures still look natural. Of course scopes should still be used if you are making any extreme settings to ensure your images are still legal, but at the end of the day what you are after is an image that looks right too you (or the producer) and whoever else will view your material, not what looks right according to a chart and a scope.

XDCAM EX Cinegammas

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%.