FS7: What is the actual base sensitivity of the FS7

[Adam_Levins]

I have been shooting at Ei2000 but a colorist on set yesterday said he did extensive testing and found the least noise in the blacks at Ei1280 and he considered that to be the native sensitivity of the camera.

What do you fine shooters think? Cheers, Adam

 

[Liam Hall]

ISO2000.

There will be less noise in the blacks at ISO1250, but that's not the same thing as native sensitivity. He's talking bollocks.

 

[morgan_moore]

Base sensitivity is entirely arbitary based upon your own desire for stops of headroom and your own definition of black clip (ie where you decide the sn is unnaceptable)

For a non linear camera baking a harsh knee this number is much less flexible.. but for Slog3 the doors are open to your own interpretation (else they wouldnt make EI_800)

Sure sony may publish a number.. usually a high one as 'low light monsters' sell (to idiots)

Personally Id rate the camera at.. about 1200

I always consider the lowest plus a stop or a half to be the base.. eg red was about 400-500.

S

 

[Grug]

I haven't used the FS7, but if indeed it is the same sensor as the F5 (and assuming I don't have a broken/wonky sensor in my camera), then the native sensititivity is 1000 ISO.

I've tested this with a waveform monitor, and to bring a grey card up to Sony's stated value of 32 IRE for SLOG2 - the sensor has to be rated at 1000 ISO. At 2000 ISO it's clearly underexposed.

 

[Liam Hall]

I haven't used the FS7, but if indeed it is the same sensor as the F5 (and assuming I don't have a broken/wonky sensor in my camera), then the native sensititivity is 1000 ISO.

I've tested this with a waveform monitor, and to bring a grey card up to Sony's stated value of 32 IRE for SLOG2 - the sensor has to be rated at 1000 ISO. At 2000 ISO it's clearly underexposed.

Sorry, you're wrong. The FS7 is ISO2000. A number of good souls have measured it using hardware waveform - they're not just making a guess based on another camera system.

If people want to push or pull the exposure to allow for more shadow detail or better highlight retention (as I often do), that is a creative decision, it doesn't mean the camera is not ISO2000.

 

[morgan_moore]

Is ISO not mesured with a 360 shutter? (that most of us dont use)

-

Coming back to actually using the camera there are considerations..

how does the image hit your NLE after lut application (lots of ETTR can baffle a colourist)

how is the image looking in terms of highlight retantion and shadow noise.

Even if Liam is technically correct I still feel it is your job to push or pull (expose) in a manner that works for your and your post team. Most find that 1200 or suchlike achieves that.

S

 

[Liam Hall]

I rarely use a light meter these days and don't use LUTs to judge exposure, so ISO is pretty irrelevant for me. What I like to know is where my mids are hitting, where my blacks are falling and where my highlights are at. I can do all that with a quick glance at the waveform. If I need to lift the exposure a stop I can see that in the waveform, if I need to close down, I can see that too.

 

[Viddovation]

This topic gets a bit complicated to discuss, and as you can see, there is some passion involved.

The OP asked for the "actual" base sensitivity, so, in one respect, giving him the pure math answers is appropriate. But what is worth understanding is that the light sensitivity of the image sensor photosites is one thing, and the apparent brightness at the base ISO (GAIN ZERO), is another. The manufacturer has some discretion and secret sauce in how they process the image from the sensor, through the analog to digital converter (A/D) and through the digital signal processor (DSP although, there may be more than one).

Now, before we go further, let's review how light works in relation to our eyes and to digital sensors.
Because photosites capturing light as voltage intrinsically work in a linear fashion, where twice as much light means twice as much voltage, this means dark areas of the sensor receiving little light are far more susceptible to fluctuation in voltage being read as higher or lower light values. Bright areas don't do this, because they receive exponentially more voltage, and minor fluxuations become so subtle that they typically go unnoticed, and unrecorded.

So, where do this relate to the original topic? Because the manufacturer chooses how much to brighten up those lower values from the sensor. With the FS7, they made separate choices for each mode of operation, and further complicating is that in custom mode you can push and pull the black levels yourself. But all of this goes to say, it's one thing for the manufacturer to rate their camera as they please, and to define the operation at the (GAIN ZERO) of the camera, but it is entirely different, and completely valid, for DPs and colorists to review the manufacturer decisions, and decide whether they like how noisy the floor is. In respect to that, many choose to rate the camera closer to 1250, which trades a bit of dynamic range for darks with less noise in them. It's a valid choice.

 

[allenfacemire]

This topic gets a bit complicated to discuss, and as you can see, there is some passion involved.
In respect to that, many choose to rate the camera closer to 1250, which trades a bit of dynamic range for darks with less noise in them. It's a valid choice.

Trying to keep up here. I understand the nuences of ISO (remember when it was ASA?) but the question was "What is the native ISO of the FS7?"
When I restore to factory settings, it seems the camera reverts to ISO 800. Now the question that begs to be asked is should I go into my settings and make my base setting (Gain switch set to L), ISO 1250 and then doubling from there as I go from L db to M and H?
Or am I getting Gain and ISO rating horribly confused?

 

[Viddovation]

Trying to keep up here. I understand the nuences of ISO (remember when it was ASA?) but the question was "What is the native ISO of the FS7?"
When I restore to factory settings, it seems the camera reverts to ISO 800. Now the question that begs to be asked is should I go into my settings and make my base setting (Gain switch set to L), ISO 1250 and then doubling from there as I go from L db to M and H?
Or am I getting Gain and ISO rating horribly confused?

The ISO that the FS7 considers to be "zero gain" or "base" varies depending on the gamma mode you're using. For instance, set to Custom/SLog2, it offers ISO2000 as base. Set to Custom/Standard, it offers ISO1000 as base. Set to CineEI/Slog2, uses 800EI as base.

I would suspect your camera is in CineEI after the factory reset?

 

[nsoltz]

The sensor always captures at ISO 2000. What you are seeing in CineEI mode is the Exposure Index for gain set in the low position. You can change the pre-set L-M-H EI values in the menu structure. The values may be expressed either in db or EI depending upon how you've set your camera. To take advantage of EI, you turn on a LUT, set EI and expose to that LUT. Thus, for example, if you are shooting EI100, you are -1 stop; EI 800 -1 1/3 stop; EI 500 -2 stops. The Sony Catalyst apps read the metadata and display accordingly.

But the base settings have noting to do with the base sensitivity of the FS7 sensor. Sony says it is 2000. Rate it according to shooting scenario, your presonal preferences, or essentially where you want to place middle gray.

I think this is hat Allen is asking.

Ned Soltz

 

[Doug Jensen]

When I restore to factory settings, it seems the camera reverts to ISO 800. Now the question that begs to be asked is should I go into my settings and make my base setting (Gain switch set to L), ISO 1250 and then doubling from there as I go from L db to M and H?

The default 800 EI setting that happens when you reset the camera is a firmware mistake by Sony. It should be 2000 EI because that is the base sensitivity of the camera in the Cine EI mode --- not 800. So after you reset the camera, change the EI to 2000. It won't make any difference at all to what is captured to the memory cards, but if you use a MLUT while setting exposure or if want your metadata to be correct, you need to change that number.

 

[Ebrahim Saadawi]

What I like to do, test the camera extensively from the moment I receive it, finding the best ISO value for ME. You first test how DR changes across the ISO range by keeping the exposure constant. Then you test noise levels across the ISO range also by keeping exposure constant, then I test skin tones across the ISO range keeping exposure constant. Then you'll be able to make an informed decision on what exact ISO value/s to use. After choosing the best ISO value, I then test what's the best exposure value the camera gives best colours, lowest noise and highest DR, testing again how each of these three elements are affected by over/under exposing in 1/3 stop increments. These few tests tell me for once and for all how to expose and set ISO for the sensor.

Don't just take a manufacturers word and use that, you can miss a lot of your camera's performance for your needs doing so, or not, that's why, test, as scientifically and logically as possible, before shooting a real job.

 

[Doug Jensen]

Ebrahim, I'd be very interested to hear what you discovered in your detailed testing of the FS7. Do you disagree with Sony's numbers for Cine EI, Hypergamma, or Standard?

 

[Viddovation]

To quote from - http://www.xdcam-user.com/2013/12/c...buffer&utm_medium=twitter&utm_campaign=Buffer

Sony’s native ISO ratings for the cameras, 1250ISO for the F55 and 2000ISO for the F5 are values chosen by Sony that give a good trade off between sensitivity, noise and over/under exposure latitude. In general these native ISO’s will give excellent results. But there may be situations where you want or need different performance. For example you might prefer to trade off a little bit of over exposure headroom for a better signal to noise ratio, giving a cleaner picture. Or you might need a very large amount of over exposure headroom to deal with a a scene with lots of bright highlights.
The Cine EI mode allows you to change the effective ISO rating of the camera.
With film stocks the manufacturer will determine the sensitivity of the film and give it an Exposure Index which is normally the equivalent of the films measured ASA/ISO. It is possible for a skilled cinematographer to rate the film stock with a higher or lower ISO than the manufacturers rating to vary the look or compensate for filters and other factors. You then adjust the film developing and processing to give a correctly exposed looking image. This is a common tool used by cinematographers to modify the look of the film, but the film stock itself does not actually change it’s base sensitivity, it’s still the same film stock with the same base ASA/ISO.

So again, there's the manufacturer rating, and then there is the users taste, and the tools to either expose hotter and noisier, or darker and cleaner. Certainly, the base sensitivity of the imager to light does not change when modes or gammas are changed, but gammas do determine how the sensor data is read, and not all the gammas are rated at 2000 ISO - for instance, on my camera the Standard Gamma (R709) reads out as 1000ISO at zero gain.

Doug - if you use the 800EI setting, the monitoring shows proper exposure for 800EI, with the MLUT on or off. So if the user is visually exposing properly at 800EI, then there's not a problem - the metadata will record 800EI, and the software will playback at 800EI, and the proper exposure seen on set will remain. There's only a problem if you're not getting a proper exposure at 800EI, or otherwise intentionally wish to expose for a different EI. Certainly, if 800EI is too dark, switching to 2000EI to expose properly on set is something you should do (or add light!).

Ned - I don't quite agree with your phrase "the sensor always captures at ISO 2000." I would agree with "The sensor always reads photosites with the same base sensitivity, which the manufacturer rates at 2000 ISO." I think, some people would argue that the noise at 2000 ISO is not what they would consider base (ie, too noisy), so said people don't consider the camera to actually be 2000 ISO, regardless of manufacturer specification.

 

[alohype]

Data coming in Cine EI is coming almost straight of the sensor, we see noise at 2000 but it's not meant to be used like that (what people don't understand is that NR in most cameras even if turned off is still there. CineEI is completely different - here it's truly turned off for your flexibility). You have to use noise reduction in post. Using lower EI is a way to get rid of noise, but once NR will be available in CineEI I think 2000 will be usable as well straight of the camera. But base iso 2000 will always be 2000 no matter how you rate it - it's where widest dynamic range is.

 

[Viddovation]

What I like to do, test the camera extensively from the moment I receive it, finding the best ISO value for ME. You first test how DR changes across the ISO range by keeping the exposure constant. Then you test noise levels across the ISO range also by keeping exposure constant, then I test skin tones across the ISO range keeping exposure constant. Then you'll be able to make an informed decision on what exact ISO value/s to use. After choosing the best ISO value, I then test what's the best exposure value the camera gives best colours, lowest noise and highest DR, testing again how each of these three elements are affected by over/under exposing in 1/3 stop increments. These few tests tell me for once and for all how to expose and set ISO for the sensor.

Don't just take a manufacturers word and use that, you can miss a lot of your camera's performance for your needs doing so, or not, that's why, test, as scientifically and logically as possible, before shooting a real job.

Agreed. And think about this for a second - the ideal base for one project may not be the ideal base for another. Let's say you have four projects, each consisting of one scene: a low budget daytime desert piece, a well budgeted studio set piece, a low budgeted, minimal lighting night piece, and a very well budgeted glossy night scene piece.

For the desert piece, you'll never need high sensitivity for lack of light, but you don't have huge HMIs and grip gear to always fill in the suns shadows. Using 2000 ISO as your project base might make sense to maximize dynamic range.

For the well budgeted studio piece, you'll never lack lights, or control there of. You don't need low light sensitivity, and since you have complete control, you don't need exceptional dynamic range. Using 800 ISO as your base would mean cleaner darks, and a potentially glossier final product.

For the under budgeted night time piece you would quickly choose 2000ISO as a base, and push farther on shots that needed it. (Or heck, get an A7S!)

For a glossy, handsomely budgeted night time piece, with large generators and huge HMIs, you might choose 800ISO as a base, since you'll never need tremendous dynamic range (you have control of all the bright lights!), and avoiding noisy darks makes the piece more polished.

 

[Doug Jensen]

This post is not aimed at anyone in particular, but I am skeptical that some of the things being said are on this thread are true. Or if they technically true, that they actually make any difference in the real world. A very slight difference that is almost imperceptible to the naked eye is the same as no difference at all, in my opinion. Unless someone is willing to put the time and effort to backup their claims by posting some actual side-by-side testing to prove their point (all they way through post), I will continue to believe what I have already determined for myself through actual testing to be true about the best settings to use on the camera. So I'm not saying anyone is wrong, but let's see some examples, because so far this thread is all talk and theory that may only make a difference on paper. I would like to see proof of the statements people are making.

 

[Viddovation]

This post is not aimed at anyone in particular, but I am skeptical that some of the things being said are on this thread are true. Or if they technically true, that they actually make any difference in the real world. A very slight difference that is almost imperceptible to the naked eye is the same as no difference at all, in my opinion. Unless someone is willing to put the time and effort to backup their claims by posting some actual side-by-side testing to prove their point (all they way through post), I will continue to believe what I have already determined for myself through actual testing to be true about the best settings to use on the camera. So I'm not saying anyone is wrong, but let's see some examples, because so far this thread is all talk and theory that may only make a difference on paper. I would like to see proof of the statements people are making.

Doug, I understand not trying to call anyone out, but if you aren't clear about what claims you find suspect, I'm not sure how the readers should know what they need to respond to.

As for the comments I've been making - the OP presented a question where he ran into a colorist who stated they preferred to rate the camera at 1280 ISO, based on what the colorist felt was an acceptable noise level, and the OP wanted clarification on what the ISO of the camera should be considered as. He was given some responses that affirmed the camera is indeed 2000 ISO, mathematically, and so forth. My comments have not been intended to contradict those statements on the math - but to address why the colorist's opinion has merit, subjective as it is. The nuance here is, what do you consider an acceptable noise floor. If you're shooting a shadowy scene at 2000ISO, and you can't see noise in those shadows, then your eyes aren't as sharp as mine - I definitely can. Whether that amount of noise is acceptable is, at least in my opinion, conditional on the shooting situation. Personally, when I'm not pressed for light quantity or dynamic range (when I have sufficient lighting control and budget) the noise at 2000ISO on the the FS7 is not acceptable to me. I like richer blacks with less noise whenever possible. In some circumstances where high dynamic range is required or adding light is not practical, I certainly take advantage of the 2000ISO setting and accept the higher degree of noise.

The above is all fairly subjective, so I don't know what sort of documentation I could provide.

Please clarify what you disagree with so I can address it more directly.

 

[morgan_moore]

I would like to see proof of the statements people are making.

I performed my tests months ago. Here is one - different EI levels, noise/DR trade off..

The methodology is to shoot at three different stops and then attemp to match the broad exposure in post and then examine the highlight and shadow rendering..

https://dl.dropboxusercontent.com/u/33167160/ei.jpg

Revisiting Id say that 500 has the cleanest shadows but harsh highlight 4000 is noise central with colour noise clearly appearing in the shadows and even creeping onto the tire .. and 1200 seems the best trade off.

S