ISO Invariance

Posted on February 22, 2019 by Andrew Astbury

Is ISO invariance really important, should I worry about it or should I buy a new camera that has it because my current one isn’t ISO invariant?

I get asked these questions A LOT because there’s a lot of talk about new camera sensors and ISO invariance on the bigger YouTube photography channels.

If your camera is NOT ISO invariant is it therefore an inferior camera – OF COURSE IT ISN”T!

ISO invariance is becoming something of a ‘buzz-term’ touted by the bigger photography influencers as if it’s the Holy Grail of camera sensor performance, and this is something I find somewhat annoying.

Why?

Because there are plenty of other sensor attributes that are FAR MORE important, such as dynamic range for instance!

The only real advantage of a true ISO invariant sensor is that it would give you a ‘get out of jail free’ card if you screw up your shot by massively under exposing it.

But the necessity for it is easily avoided if only folk would learn how to use their equipment properly in the first place.

Correct exposure is everything in photography, and ISO invariance – for the most part – is just a way of circumventing it!

🤘🤔Nikon D800E, D810 and D850 Usable Dynamic Range Test and Budget Buying Advice👌🤘

Posted on February 7, 2019 by Andrew Astbury

Like it or not this video compares the real usable dynamic range of Nikons’ three most used cameras for landscape photography. Everyone bangs on endlessly about dynamic range when in fact most of them have no clue what they’re talking about. If you want to see the truth about dynamic range improvements since 2012 then the results of this video may well come as a shock!

If you want to see the tonal response curves of the three Nikon models AND the Canon 5DMk3 then click the image below to view at full size:

Camera Imaging Area

Posted on November 6, 2018 by Andrew Astbury

Camera Imaging Area

For God’s Sake! Another bloody idiot YouTuber uploaded a video the other day where they were trying out the Fuji GFX 50, and just moments into said video he came out with the same old pile of junk which amounts to “it’s a bigger sensor so it soaks up more light”.

So now his 76,000+ subscribers are misled into believing something that is plain WRONG.

For anyone who didn’t manage to grasp what I was saying in my previous post HERE let’s try attacking this crackpot concept from a different angle.

Devotees of this farcical belief quite often liken larger sensors to bigger windows in a room.

“A bigger window lets in more light” they say.

Erm…no it doesn’t. A bigger window has an increased SURFACE AREA that just lets in a larger area of THE SAME LIGHT VALUE.

A 6 foot square pane of glass has a transmission value that is EXACTLY the same as as a 3 foot square pane of the same glass, therefore a ‘BIGGER WINDOW’ onto the outside world does NOT let in more light.

Imagine we have a room that has a 6×6 foot and 3×3 foot window in one wall. Now go press your nose up to both windows – does the world outside look any different?

No, of course it doesn’t.

The only property that ‘window size’ has any bearing on is the area of ‘illumination foot print’.

So basically the window analogy has ZERO bearing on the matter!

What lets light into the camera is the LENS, not the damn sensor!

The ‘illuminant value’ – or Ev – of the light leaving the back of the lens and entering the imaging plane DOES NOT CHANGE if we swap out our FX body for a crop body – DOES IT!!??!!

So why do these bloody idiots seem to think physics changes when we put a bigger sensor behind the lens? It’s pure abject stupidity.

The imaging area of a sensor has ZERO effect on the intensity of light striking it – that is something that is only influenced by aperture (intensity) and shutter speed (time).

With digital photography, exposure is ‘per unit area’ NOT total area. A dslr sensor is NOT a unit but an amalgamation of individual units called PHOTOSITES or pixels. Hence it is the photosite area that governs exposure NOT sensor total area.

There is a sensor on the market that blows all this ‘sucks in more light’ crap clean out of the water, and that sensor is the Sony IMX224MQV. This is a 1/3 class sensor with a diagonal dimension of just 6.09mm and 1.27Mp. By definition this one hell of a small ‘window’ yet it can ‘see’ light down to 0.005lux with good enough SNR to allow the image processor to capture 120 10bit images per second.

A cameras ‘window onto the world’ is the lens – end of!

Imagine going back to film for a moment – correct exposure value was the same for say Ilford FP4 irrespective of whether you were using 35mm, 120/220 roll film, 5×4 or 10×8 sheet film.

The size of the recording media within the imaging plane was and still is completely irrelevant to exposure.

Bigger recording media never have and never will ‘suck in’ more light, because they can’t suck in more light than the lens is transmitting!

The only properties of the sensor within the imaging area that WILL change how the it reacts to the light transmitted by the lens are:

Photosite surface area – number of megapixels
Sensor construction – FSI vs BSI
Micro lens design
CFA array absorption characteristics

After my previous post some stroppy idiot emailed me saying Ken Wheeler AKA The Angry Photographer says the Nikon D850 is a full frame Nikon D500, and that because the D850 is an FX camera and has better dynamic range then this proves I’m talking bollocks.

Well, Ken never said this in terms of anything other than approximate pixel density – he’s not that stupid and dick-heads should listen more carefully!

The D500 is an FSI sensor while the D850 is a BSI sensor and has a totally different micro lens design, AD Converter and IP.

Out of the 4 characteristics listed above 3 of them are DRASTICALLY different between the two cameras and the other is different enough to have definite implications – so you cannot compare them ‘like for like’.

But using the same lens, shutter speed, ISO and aperture while imaging a flat white or grey scene the sensor in a D850 will ‘see’ no higher light value than the sensor in a D500.

Why? Because the light emanating from the scene doesn’t change and neither does the light transmitted by the lens.

I own what was the best light meter on the planet – the Sekonic 758. No where does it have a sensor size function/conversion button on it, and neither does its superseding brother the 858!

There are numerous advantages and disadvantages between bigger and smaller sensors but bigger ‘gathering’ or ‘soaking up’ more light isn’t one of them!

So the next time you hear someone say that increased size of the imaging area – bigger sensor size – soaks up more photons you need to stop listening to them because they do not know what the hell they’re talking about.

But if you chose to believe what they say then so be it – in the immortal words of Forrest Gump ” Momma says stoopid is as stoopid does………”

Post Script:

Above you can see the imaging area for various digital sensor formats. You can click the image to view it bigger.

Each imaging area is accurately proportional to the others.

Compare FX to the PhaseOne IQ4. Never, repeat never think that any FX format sensor will ever deliver the same image fidelity as a 645 sensor – it won’t.

Why?

Because look at how much the fine detail in a scene has to be crushed down by the lens to make it ‘fit’ into the sensor imaging area on FX compared to 645.

Andy your talking crap! Am I ? Why do you think the worlds top product and landscape photographers shoot medium format digital?

Here’s the skinny – it’s not because they can afford to, but rather they can’t afford NOT TO.

As for the GFX50 – its imaging area is around 66% that of true MF and it’s smaller than a lot of its ‘wannabe’ owners imagine.

Sensor Size Myth – Again!

Posted on October 23, 2018 by Andrew Astbury

Sensor Size Myth – “A bigger sensor gathers more light.”

If I hear this crap one more time either my head’s going to explode or I’m going to do some really nasty things to someone!

A larger sensor size does NOT necessarily gather any more light than a smaller sensor – END OF!

What DOES gather more light is BIGGER PHOTOSITES – those individual light receptors that cumulatively ‘make up’ the photosensitive surface plane of our camera sensor.

Above we have two fictional sensors, one with smaller physical dimensions and one with larger dimensions – the bottom one is a ‘larger sensor size’ than the top one, and the bottom one has TWICE as many photosites as the top one (analogous to more megapixels).

But the individual photosites in BOTH sensors are THE SAME SIZE.

Ignoring the factors of:

Micro Lens design
Variations in photosite design such as resistivity
Wiring Substrate
SNR & ADC

the photosites in both sensors will have exactly the same pixel pitch, reactivity to light, saturation capacity and base noise level.

However, if we now try to cram the number of photosites (megapixels) into the area of the SMALLER sensor – to increase the resolution:

we end up with SMALLER photosites.

We have a HIGHER pixel resolution but this comes with a multi-faceted major penalty:

Decreased Dynamic Range
Increased susceptibility to specular highlight clipping
Lower photosite SNR (signal to noise ratio)
Increased susceptibility to diffraction – f-stop limiting

And of course EXACTLY the same penalties are incurred when we increase the megapixel count of LARGER sensors too – the mega-pixel race – fueled by FOOLS and NO-NOTHING IDIOTS and accommodated by camera manufacturers trying to make a profit.

But this perennial argument that a sensor behaves like a window is stupid – it doesn’t matter if I look outside through a small window or a big one, the light value of the scene outside is the same.

Just because I make the window bigger the intensity of the light coming through it does NOT INCREASE.

And the ultimate proof of the stupidity and futility of the ‘big window vs small window’ argument lies with the ‘proper photographers’ like Ben Horne, Nick Carver and Steve O’nions to name but three – those who shoot FILM!

A 10″x8″ sheet of Provia 100 has exactly the same exposure characteristics as a roll of 35mm or 120/220 Provia 100, and yet the 10″x 8″ window is 59.73x the size of the 35mm window.

And don’t even get me started on the other argument the ‘bigger = more light’ idiots use – that of the solar panel!

“A bigger solar panel pumps out more volts so because it gathers more light, so a bigger sensor gathers more light so must pump out better images………”

What a load of shite…………

Firstly, SPs are cumulative and they increase their ‘megapixel count’ by growing in physical dimensions, not by making their ‘photosites’ smaller.

But if you cover half of one with a thick tarpaulin then the cumulative output of the panel drops dramatically!

Also, we want SPs to hit their clip point for maximum voltage generation (the clip point would be that where more light does NOT produce more volts!).

Our camera sensor CANNOT be thought of in the same way:

We are not interested in a cumulative output, and we don’t want all the photosites on our sensors to ‘max out’ otherwise we’ll have no tonal variation in our image will we…..!

The shot above is from a D800E fitted with a 21mm prime, ISO 100 and 2secs @f13.

If I’d have shot this with the same lens on the D500 and framed the same composition I’d have had to use a SHORTER exposure to prevent the highlights from clipping.

But if bigger sensors gather more light (FX gathers more than DX) I’d have theoretically have had expose LONGER……….and that would have been a disaster.

Seriously folks, when it comes to sensor size bigger ones (FX) do not gather more light than smaller (DX) sensors.

It’s not the sensor total area that does the light gathering, but the photosites contained therein – bigger photosites gather more light, have better SNR, are less prone to diffraction and result in a higher cumulative dynamic range for the sensor as a whole.

Do NOT believe anyone anywhere on any website, forum or YouTube channel who tells you any different because they a plain WRONG!

Where does this shite originate from you may ask?

Well, some while back FX dslr cameras where not made and everything from Canon and Nikon was APSC 1.5x or 1.6x, or APSH 1.3x. Canon was first with an FX digital then Nikon joined the fray with the D3.

Prior to the D3 we Nikon folk had the D300 DX which was 12.3Mp with a photosite area 30.36 microns²

The D3 FX came along with 12.1Mp but with a photosite area of 70.9 microns²

Better in low light than its DX counterpart due to these MASSIVE photosites it gave the dick heads, fools and no-nothing idiots the crackpot idea that a bigger sensor size gathers more light – and you know what……it stuck; and for some there’s no shifting it!

Hope this all makes sense folks.

Don’t forget, any questions or queries then just ask!

If you feel I deserve some support for putting this article together then please consider joining my membership site over on Patreon by using the link below.

Alternatively you could donate via PayPal to tuition@wildlifeinpixels.net

You can also find this article on the free-to-view section of my Patreon channel by clicking this link https://www.patreon.com/posts/sensor-size-myth-22242406

If you are not yet a member of my Patreon site then please consider it as members get benefits, with more membership perks planned over the next 3 months. Your support would be very much appreciated and rewarded.

Before I go, there’s a new video up on my YouTube Channel showing the sort of processing video I do for my Patreon Members.

You can see it here (it’s 23 minutes long so be warned!):

Please leave a comment on the video if you find it useful, and if you fancy joining my other members over on Patreon then I could be doing these for you too!

All the best

Andy

Two Blend Modes in Photoshop EVERY Photographer Should Know!

Posted on October 14, 2018 by Andrew Astbury

Two Blend Modes in Photoshop EVERY Photographer Should Know!

The other day one of my members over on my Patreon suggested I do a video on Blending Modes in Photoshop.

Well, that would take a whole heap of time as it’s quite a big subject because Blending Modes don’t just apply to layers. Brushes of all descriptions have their own unique blend modes, and so do layer groups.

There is no need to go into a great deal of detail over blend modes in order for you to start reaping their benefits.

There are TWO blend modes – Multiply and Screen – which you can start using straight away to vary the apparent exposure of your images.

And seeing as my last few videos have been concerned with exposing for highlights and ETTR in general, the use of the Multiply Layer Blending Mode will be clear to see once you’ve watched the video.

Hope the video gives you some more insight folks!

My Members over on Patreon get the benefit of being able to download the raw files used in this video.

All the best.

Spot Metering – In Camera vs Separate Meter

Posted on October 1, 2018 by Andrew Astbury

Landscape Photography Exposure, ETTR and Highlight Spot Metering Accuracy

CLICK ME to watch the Video!

In this short(ish) video I want to show you why your camera spot meter can be something of a ‘let down’ in exposure terms when you are trying to obtain an accurate highlight reading for your scene.

Most ‘in camera’ spot meters are a lot more imprecise than the user imagines.

Nikon spot meter ‘spots’ are generally 4mm wide. That means 4mm ON THE SENSOR!

On an FX camera the sensor is roughly 36mm wide, so the ‘spot’ actually has a ‘window’ or ‘measuring footprint’ that is 1/9th of the viewfinders horizontal field of view.

And don’t think that because you use a Canon you’re any better off – in fact you’re worse off because Canon spots are a tiny bit BIGGER!

In this example I use a shot taken with a Zeiss 21mm – this lens has a horizontal angle of view of 81 degrees.

So the 4mm Nikon spot has an angle of view equivalent to 1/9th the frame and hence 1/9th the horizontal AoV of the lens, in other words 9 degrees.

Aimed at the brightest highlight in the sky its footprint takes in sky tones that are dramatically less than highlights. So the reading it will give me is ‘darker’ than it should be.

My D800E has it’s highlight clipping/blow point 3.6 stops above its mid tone.

If I then apply ETTR to this reading by exposing at +3 to +3.3 stops it will result in blown highlights.

But if I use a 1 degree spot meter aimed at exactly the same place its much narrower angle sees ONLY THE BRIGHT AREA I’m aiming at. This gives me a much BRIGHTER reading, allowing me to push the exposure by +3.3 stops without blowing any of my highlights.

Hope this all makes sense folks.

Don’t forget, any questions or queries then just ask!

If you feel I deserve some support for putting this video and article together then please consider joining my membership site over on Patreon by using the link below.

Alternatively you could donate via PayPal to tuition@wildlifeinpixels.net

All the best

Andy

Dynamic Range, Mid Tones, Metering and ETTR

Posted on September 28, 2018 by Andrew Astbury

Dynamic Range, Mid Tones, Metering and ETTR

I recently uploaded a video to my YouTube channel showing you an easy way to find the ‘usable dynamic range’ of you dSLR:

The other day I was out with Paul Atkins for a landscape session in the awesome Dinorwic Quarry in Llanberis, Snowdonia. Highly dynamic clouds and moody light made the place look more like Mordor!

Looking towards the top of the Llanberis Pass from the middle level of Dinorwic Quarry and Electric Mountain.

Here are the 6 unedited shots that make this finished panoramic view:

As you can see, the images are are shot in a vertical aspect ratio. Shooting at 200mm on the D800E this yields an assembled pano that is 16,000 x 7000 pixels; the advantages for both digital sales and print should be obvious to you!

As you can see, the bright parts of the sky are a lot brighter in the captures than they are in the finished image, but they are not ‘blown’. Also the shadows in the foreground are not choked or blocked.

In other words the captures are shot ETTR.

Meter – in camera or external.

Any light meter basically looks at a scene (or part thereof) and AVERAGES the tones that it sees. This average value value is then classed by the meter is MID GREY and the exposure is calculated in terms of the 3 variables you set – Time, Intensity and Applied Gain, or shutter, aperture and ISO.

But this leads to all sorts of problems.

All meters are calibrated to an ANSI Standard of 12% grey (though this gets a bit ambiguous between manufactures and testers). But you can get a good idea of what ‘light meter mid grey/mid tone” looks like by mentally assigning an RGB value of 118,118,118 to it.

However, we – humans – find 18% grey a more acceptable ‘mid tone grey’ both in print and on our modern monitors.

NOTE: 18% grey refers to the level of REFLECTANCE – it reflects 18% of the light falling on it. It can also be reproduced in Photoshop using a grey with 128,128.128 RGB values.

So problem number 1 is that of mid tone perception and the difference between what you ‘see’ and what the camera sees and then does in terms of exposure (if you let the camera make a decision for you).

128RGB grey versus 118RGB meter mid grey

Click on the pano image from Dinorwic to view it bigger, then try to FIND a mid grey that you could point your camera meter at – you can’t.

Remember, the grey you try to measure MUST be exactly mid-grey – try it, it’ll drive you nuts trying to find it!

This leads us to problem number 2.

Take your camera outside, find a white wall. Fill your frame with it and take a shot using ZERO exposure compensation – the wall will look GREY in the resulting shot not WHITE.

Next, find something matte black or near to it. Fill your frame with it and take another shot – the black will look grey in the shot not black(ish).

Problem number 3 is this – and it’s a bit of a two-headed serpent. An exposure meter of any kind is COLOUR BLIND but YOU can SEE colours but are tonally blinded to them to some degree or other:

Simple primary red, green and blue translate to vastly different grey tones which comes as a big surprise to a lot of folk, especially how tonally light green is.

Scene or Subject Brightness Range

Any scene in front of you and your camera has a range of tones from brightest to darkest, and this tonal range is the subject brightness range or SBR for short. Some folk even refer to it as the scene dynamic range.

If you put your camera meter into spot mode you can meter around your chosen scene and make note of the different exposure values for the brightest and darkest areas of your potential shot.

You camera spot meter isn’t the most accurate of spot meters because its ‘spot’ is just too big, typically between 4mm and 5mm, but it will serve to give you a pretty good idea of your potential SBR.

A 1 degree spot meter will, with correct usage, yield a somewhat more accurate picture (pun intended) of the precise SBR of the scene in front of you.

Right about now some of you will be thinking I’m hair-splitting and talking about unnecessary things in todays modern world of post-processing shadow and highlight recovery.

Photography today is full of folk who are prepared to forego the CRAFT of the expert photographer in favour of getting it half-right in camera and then using the crutch of software recovery to correct their mistakes.

Here’s the news – recovery of popped highlights is IMPOSSIBLE and recovery of shadows to anymore than a small degree results in pixel artifacting. Get this, two WRONGS do NOT make a RIGHT!

If the Mercedes F1 team went racing with the same attitude as the majority of camera users take pictures with, then F1 would be banned because drivers would die at an alarming rate and no car would ever make the finish line!

So, one way or another we can quantify our potential scene SBR.

“But Andy I don’t need to do that because my camera meter does that for me…….”

Oh no it does NOT, it just averages it to what IT THINKS is a correct mid tone grey – which it invariably isn’t!

This whole mid tone/mid grey ‘thing’ is a complete waste of time because:

It’s near impossible to find a true mid tone in your scene to take a reading off.
What you want as a mid tone will be at odds with your camera meter by at least 1/2stop.
If you are shooting wildlife or landscapes you can’t introduce a ‘grey card’.
Because of the above, your shot WILL BE UNDER EXPOSED.

“Yeah, but I can always bracket my shots and do an exposure blend Andy so you’re still talking crap….”

Two answers to that one:

You can’t bracket shots and blend if your MAIN subject is moving – de-ghosting is only effective on small parts of a scene with minimal movement between frames.
The popular “shoot and bracket two each end” makes you look like total dickhead and illustrates that you know less than zero about exposure. Try doing that on a paying job in front of the client and see how long you last in a commercial environment.

By far the BEST way of calculating exposure is the ETTR method.

ETTR, Expose to the Right.

If you meter for a highlight, your camera will treat that as a mid tone because your camera ASSUMES it’s a mid tone.

Your camera meter is a robot programmed to react to anything it sees in EXACTLY the same way. It doesn’t matter if your subject is a black cat in the coal house or a snow man in a snow storm, the result will be the same 118,118,118 grey sludge.

Mid tones are as we’ve already ascertained, difficult to pin down and full of ambiguity but highlights are not. So let’s meter the brightest area of the image and expose it hard over to the right of the histogram.

The simplest way to achieve this is to use your live view histogram with the camera in full manual mode.

Unlike the post-shot review histogram, the live-view histogram is not subject to jpeg compression, and can be thought of as something of a direct readout of scene tonality/brightness.

Using your exposure controls (usually shutter speed for landscape photography) you can increase your exposure to push the highlight peak of the histogram to the right as far as you can go before ‘hitting the wall’ on the right hand side of the histogram axis – in other words the camera sensor highlight clipping point.

Of course, this has the added benefit of shifting ALL the other tones ( mids and shadows) to the right as well,resulting in far less clipping potential in your shadow areas.

So back to Dinorwic again and here’s a shot that has been exposed ETTR on the live view histogram using spot metering over what I deemed to be the brightest area of the sky:

The red square indicates the approximate size of the spot meter area.

I was a naughty boy not recording this on video for you but I forgot to pack the HDMI lead for the video recorder – I’ll do one shortly!

The problem with using the Live View Histogram is that it can be a bit of a struggle to see it. your live view screen itself can be hard to see in certain light conditions outside, and the live view histogram itself is usually a bit on the small side – no where near as big as the image review histogram you can see here.

But looking at the review histogram above you can see that there’s a ‘little bit more juice’ to be had in terms of exposure of the highlights because of that tiny gap between the right end of the histogram and the ‘wall’ at the end of the axis.

Going back to the video the maximum ETTR ‘tipping point’ was centered around these three shots:

Clipped

Not Clipped (the one we allocated the star rating to). Exposure is -1/3rd stop below clipped.

Safe, but -2/3rd stop below Clipped.

The review histogram puts the Dinorwic shot highlights firmly in the same exposure bracket as ‘Safe, but -2/3rd stop below Clipped, and tells us there is another 1/3rd stop ‘more juice’ to be had!

So lengthening the exposure by 1/3rd stop and changing from 160th sec to 1/50th sec gives us this:

The red square indicates the approximate size of the spot meter area.

Live View Histogram ETTR

Live View Histogram plus 1/3 stop more juice! Highlights STILL below Clipping Point and shadows get 1/3rd stop more exposure.

That’s what it’s all about baby – MORE JUICE!

And you will not be in a position to confidently acquire more juice unless you find the USABLE DYNAMIC RANGE of your camera sensor.

The whole purpose of finding that usable DR is to discover where your highlight and shadow clipping points are – and they are very different between camera models.

For instance, the highlight clipping point value of the Nikon D850 is different from that of the Nikon D800E, but the shadow clipping point is pretty similar.

There is an awful lot more use to discovering your cameras usable dynamic range than a lot of folk imagine.

And if you do it the precise way then you can acquire a separate meter that will accept camera profiling:

You can create a dynamic range profile for your camera (and lens combo*) and then load it into the meter:

and then have your cameras usable dynamic range as part of the metering scale – so then you have NO EXCUSE for producing a less than optimum exposure.

(*)Note: yes, the lens does have an effect on dynamic range due to micro-contrast and light transmission variables – if you want to be super-picky!

AND THEY SAY HANDHELD METERS ARE DEAD, OLD TECH and of NO USE!!!

Anyone who says or even thinks that is a total KNOB.

Your camera dynamic range, the truthful one – FIND IT, KNOW IT, USE IT.

And don’t listen to the idiots and know-nothings, just listen and heed the advice of those of us who actually know what we’re doing.

NOTE: The value of grey (gray) cards and how to use them for accurate measurement is a subject in its own right and provides the curious with some really interesting reading. Believe me it’s far more expansive than the info I’ve given here. But adopting an ETTR approach when exposing to sensor that you KNOW the physical behavior of (dynamic response to light/dynamic range) can alleviate you of all critical mid-tone concerns.

This article has taken me over 8 hours to produce in total, and is yours to view for FREE. If you feel I deserve some support for doing this then please consider joining my membership site over on Patreon by using the link below.

Alternatively you could donate via PayPal to tuition@wildlifeinpixels.net

ETTR High Contrast Scene Processing.

Posted on August 19, 2018 by Andrew Astbury

ETTR High Contrast Scene Processing.

When faced with a high contrast scene like this most photographers would automatically resort to bracketing shots.

Sometimes you will be in a situation where shooting a bracketed sequence is difficult or impossible.

But a single image exposed to the right of the histogram – ETTR – where highlights are recorded at their maximum level of exposure can allow the camera sensor to capture far more detail in the darker areas than Lightroom will allow you to see at first glance.

Exposing to the right (of the in-camera histogram) correctly means that you expose the brightest scene highlights AS HIGHLIGHTS.

But it’s a balancing act between exposing them fully, and ‘blowing’ them.

Getting the ETTR exposure correct invariably means that the sensor receives MORE exposure across all tonal ranges, so you end up with more usefully recoverable shadow detail too.

In this video I show you a full Lightroom and Photoshop workflow to produce a noise-free image from a raw file exposed in just such a way.

Members of my Patreon site can download the all the workflow steps together with the raw file so that they can follow my processing, and perhaps come up with their own versions too!

Lumenzia Plugin for Photoshop: https://getdpd.com/cart/hoplink/21529?referrer=c0vpzfhvq7ks8cw8c

Lumenzia + Comprehensive Training: https://getdpd.com/cart/hoplink/21529?referrer=c0vpzfhvq7ks8cw8c&p=165704

Just to keep you up to speed on my video channel, here’s my previous video from last week which illustrates how I do my dust-spot and blemish removal in Photoshop:

Exposure Value – What does it mean?

Posted on August 3, 2018 by Andrew Astbury

Exposure Value (Ev) – what does Ev mean?

I get asked this question every now and again because I frequently use it in the description annotations of image shot data here on the blog.

And I have to say from the outset the Exposure Value comes in two flavours – relative and absolute – and here I’m only talking mainly about the former.

So, let’s start with basic exposure.

Exposure can be thought of as Intensity x Time.

Intensity is controlled by our aperture, and time is controlled by our shutter speed.

This image was shot at 0.5sec (time), f11 (intensity) and ISO 100.

exposure value

We can think of the f11 intensity of light striking the sensor for 0.5sec as a ‘DOSAGE’ – and if that dosage results in the desired scene exposure then that dosage can be classed as the exposure value.

Let’s consider two exposure settings – 0.5sec at f11 ISO100 and 1sec at f16 ISO 100.

Technically speaking they are two different exposures, but BOTH result in the same light dosage at the sensor. The second exposure is TWICE the length of time but HALF the intensity.

So both exposures have the same Exposure Value or Ev.

The following exposure of the same scene is 1sec at f11 ISO 100:

exposure value

The image was shot at the same intensity (f11) but the shutter speed (time) was twice as long, and so the dosage was doubled. Double the dose = +1Ev!

And in this version the exposure was 0.25sec at f11 ISO 100:

exposure value

Here the light dosage at the sensor is HALF that of the correct/desired exposure because the time factor was halved while using the same intensity.

So half the dose = -1Ev!

Now some of you will be thinking that -1Ev is 1 stop under exposure – and you’d be right!

But Ev, or exposure value, is just a cleaner way of thinking about exposure because it doesn’t tie you to any specific camera setting – and it’s more easily transferable between cameras.

What Do I Mean by that?

Example – If I use say a 50mm prime lens on my Nikon D800E with the metering in matrix mode, ISO 100 and f14 I might get a metered exposure shutter speed of 1/10th of a second.

But if I replace the D800E with a D4 set at 100 ISO, matrix and f14 I’ll guarantee the metered shutter speed requirement will be either 1/13 or 1/15th of a second.

The D4 meters between -1/3Ev and -2/3Ev (in other words 1/2 stop) faster/brighter than the D800E when fitted with the same lens and set to the same aperture and ISO, and shooting exactly the same framing/composition.

Yet the ‘as metered’ shots from both cameras look pretty much the same with respect to light dosage – exposure value.

Exposure Settings Don’t Transfer between camera models very well, because the meter in a camera is calibrated to the response curve of the sensor.

A Canon 1DX Mk2 will usually generate a evaluative metered shutter speed 1/3rd of a stop faster than a matrix metered Nikon D4S for the same given focal length, aperture and ISO setting.

Both setups ‘as metered’ shots will look pretty much the same, but transposing the Canon settings to the Nikon will result in -1/3 stop under exposure – which on a digital camera is definitely NOT the way to go!

‘As Metered’ can be regarded as +/-0Ev for any camera (Note: this does NOT mean Ev=0!)

Any exposure compensation you use in order to achieve the ‘desired’ exposure on the other hand can be thought of as ‘metered + or – xEv’.

exposure compensation

Shot with the D4 plus 70-200 f2.8@70mm in manual exposure mode, 1/2000th sec, f8 and ISO 400 using +2/3Ev compensation.

The matrix metered exposure indicated by the camera before the exposure value compensation was 1/3200th – this would have made the Parasitic Jaeger (posh name for an Arctic Skua!) too dark.

A 1DXMk2 using the corresponding lens and focal length, f8, ISO 400 and evaluative metering would have wanted to generate a shutter speed of at least 1/4000th sec without any exposure compensation, and 1/2500th with +2/3Ev exposure compensation.

And if shot at those settings the Canon image would look pretty much like the above.

But if the Nikon D4 settings had been fully replicated on the Canon then the shot would be between 1/3 and 1/2 stop over exposed, risking ‘blowing’ of some of the under-wing and tail highlights.

So the simple lesson here is don’t use other photographers settings – they never work unless you’re on identical gear!

But if you are out with me and I tell you “matrix/evaluative plus 1Ev” then your exposure will have pretty much the same ‘light dosage’ as mine irrespective of you using the right shutter speed, aperture or ISO for the job or not!

I was brought up to think in terms of exposure value and Ev units, and to use light meters that had Ev scales on them – hell, the good ones still have ’em!

If you look up the ‘tech-specs’ for your camera you’ll find that metering sensitivity is normally quoted as an Ev range. And that’s not all – your auto focus may well have a low light Ev limited quoted too!

To all intents and purposes Ev units and your more familiar ‘f-stops’ amount to one and the same thing.

As we’ve seen before, different exposures in terms of intensity and time can have the same exposure value, and all Ev is concerned with is the cumulative outcome of our shutter speed, aperture and ISO choices.

Most of you will take exposures at ‘what the camera meter says’ settings, or you will use the meter indicated exposure as a baseline and modify the exposure settings with either positive or negative ‘weighting’ via your exposure compensation dial.

That’s Ev compensation relative to your meters baseline.

But have you ever asked yourself just how accurate your camera meter is?

So I’ve just stepped outside my front door and taken these two frames:

EV=15/Sunny 16 Rule 1/100th sec, f16, 100 ISO – click to view large.

Matrix Metering, no exposure compensation 1/200th sec, f16, ISO 100 – click to view large

These two raw files have been brought into Lightroom and THE ONLY adjustment has been to change the profile from Adobe Color to Camera Neutral.

Members of my subscription site can download the raw files and see for themselves.

Look at the histogram in both images!

The exposure for xxx164.NEF (the top image) is perfection personified while xxx162.NEF is under exposed by ONE WHOLE STOP – why?

Because the bottom image has been shot at the camera-specified matrix metered exposure, while the top image has been shot using the good old ‘Sunny 16 Rule’ that’s been around since God knows when!

“Yeah, but I could just use the shadow recovery slider on the bottom shot Andy….” Yes, you could, if you wanted to be an idle tit, and even then the top image would still be better because there’s no ‘recovery’ being used on it in the first place. Remember, more work at the camera means less work in processing!

Recovery of either shadows or highlights is ‘poor form’ and no substitute for correct exposure in the first place. Digital photography is just like shooting colour transparency film – you need to ‘peg the highlights’ as highlights BUT without over exposing them and causing them to ‘blow’.

In other words – ETTR, expose to the right!

And seeing as your camera meter wants to turn everything into midtone grey shite it’s the very last thing you should ever allow to dictate your final exposure settings – as the two images above prove beyond argument.

And herein lies the problem.

Even if you use the spot metering function the meter will read the brightness of what is covered by the ‘spot’ and then calculate the exposure required to expose that tonal brightness AS A MID TONE GREY.

That’s all fine ‘n dandy – if the metered area is actually an exact mid tone. But what if you were metering a highlight?

Then the metered exposure would want to expose said highlight as a midtone and the overall highlight exposure would be far too dark. And you can guess what would happen if you trusted your meter to spot-read a shadow.

A proper hand-held spot meter has an angle of view or AoV of 1 degree.

Your camera spot meter angle of view is dictated by the focal length of the lens you have fitted.

On my D800E for example, I need to have a lens AoV of around 130mm focal length equivalent for my spot to cover 1 degree, because the ‘spot’ is 4mm in diameter – total stupidity.

But it does function fairly well with wider angle lenses and exposure calculations when used in conjunction with the live view histogram. And that will be subject of my next blog post – or perhaps I’ll do a video for YouTube!

So I doubt this blog post about relative exposure compensation is going to light your world on fire – it began as an explanation to a recurring question about my exif annotation habits and snowballed somewhat from there!

But I’ll leave you with this little guide to the aforementioned Sunny 16 Rule, which has been around since Noah took up boat-building:

To use this table just set your ISO to 100.

Your shutter speed needs to be the reciprocal of your ISO – in other words 1/100 sec for use with the stated aperture values:

Aperture	Lighting conditions	Shadow PROPERTIES
f/22*	Snow/sand	Dark with sharp edges
f/16	Sunny	Distinct
f/11	Slight overcast	Soft around edges
f/8	Overcast	Barely visible
f/5.6**	Heavy overcast	No shadows
f/4	Open shade/sunset	No shadows

* – I would not shoot at f22 because of diffraction – try 1/200th f16

** – let’s try some cumulative Ev thinking here and go for more depth of field using f11 and sticking with 100 ISO. -2Ev intensity (f5.6 to f11) requires +2Ev on time, so 1/100th sec becomes 1/25th sec.

Over the years I’ve taken many people out on photo training days, and a lot of them seem to think I’m some sort of magician when I turn their camera on, switch it manual, dial in a couple of settings and produce a half decent image without ever looking at the meter on their camera.

It ain’t magic – I just had this table burnt into the back of my eyeballs years ago.

Works a charm – if you can do the mental calculations in your head, and that’s easy with practice. The skill is in evaluating your shooting conditions and relating them to the lighting and shadow descriptions.

And here’s a question for you; we know our camera meter wants to ‘peg’ what it’s measuring as a midtone irrespective of whether it’s measuring a midtone or not. But what do you think the Sunny 16 Rule is ‘pegging’ and where is it pegging it on the exposure curve?

If you can answer that question correctly then the other flavour of exposure value – absolute – might well be of distinct interest to you!

Give it a try, and if you use it correctly you’ll never be more than 1/3rd of a stop out, if that. Then you can go and unsubscribe from all those twats on YouTube who told you it was out-dated and defunct or never told you about it in the first place!

I hope you’ve found the information in this post useful.

I don’t monetize my YouTube videos or fill my blog posts with masses of affiliate links, and I rely solely on my patrons to help cover my time and server costs. If you would like to help me to produce more content please visit my Patreon page on the button above.

Many thanks and best light to you all.