Tricks and tips for a first session with models

A couple of days ago I was browsing a subreddit where someone asked for advice when photographing a model for the first time. One day afterwards, another person asked mostly the same, tricks to make the model feel comfortable. This has made me think that perhaps there are many people around here who would like to start taking pictures about models and do portraiture, but they don’t feel able to manage a session and the responsibilities that it implies.

In my opinion, most of the insecurity a photographer suffers before doing his/er first session is just a consequence of not knowing what to expect, and disappears soon after the beginning of the session. Besides, I thought that the advice I provided could be useful for many people beginning to take portraiture pictures, so I decided to share them here too.


Know your ene… your friend!

One of the key advice I can give for this kind of session is: know the person you are going to photograph. It can be very uncomfortable and cold to take pictures of someone while both stay silent not knowing what to say. Confidence is important, as it gets reflected in the expression of the model.

I only photograph people I know, friends, people I’m confident with. This way there is no problem in finding a topic to talk about during the session and conversation will flow naturally. But, of course, you also want to photograph people you have never seen before. What I do is to convert them in “friends” before the shooting. A couple of days before the session I like to meet them to take a coffee (or tea if they are English gentlemen) and explain some details of the session. In this meeting (around 30 minutes, but it can be longer if the feeling is good) I explain the idea I have about the session so they can think about it. If they know what to expect and have a couple of days to think about it, they will feel more confident as all the process will sound familiar to them. I explain what pictures I plan to take, some details about the material, why I chose the place or any tips about posing that may be useful.

But I also use this meeting for another thing: knowing the model. The model is the center of the shooting, everything moves and evolves around him/er. This means that knowing how s/he thinks, what does s/he expects from the session and other personality features will help to design and adapt better the session. The meaning of a portraiture session is to show the person as it really is, capturing the essence. Also, knowing a little bit about the model allows to find topics in common to talk about during the session. This way is harder to get to a dead silent point where no one knows what to talk about.

A famous photographer once said that you cannot take a good portrait of someone without falling in love with them a little bit. Once you get to that point, you start to see the real subtle aspects of the person that make him/er unique, and you can use photography to share them with the rest of the world in a way that when people see the picture will exclaim: “wow, this picture is really him/er”.

You know... Just hiding

Even if your model is shy, you can use it as an advantage to take some pictures that show the personality. It’s important to adapt to the model, not making the model adapt your ideas of the session.


Have I ever told you about the weather in Madagascar?

This is another of the golden topics, so important that has already appeared in the previous one. You should never stop talking for more than a minute. It might seem hard at first, but it is really important.

In order to achieve a connection and complicity a good flow of ideas needs to be present. It doesn’t matter about what you talk as long as you keep a natural an fluid conversation. Use what you learnt about the model to bring those topics to the conversation. In the worst case make the model talk about himself. People like to talk about their lives and will bring up some conversation for a while.

Also, you can use some time to explain what you are doing, why you do it, and teach a little of basic photography. Knowing a little bit about the process will make the model feel more involved in the shooting, and by extension more confident about posing.

If the model gives you some ideas, never discard them. Even if it sounds dumb, you never know how well an idea will perform until you try it, and you might find it provides a good picture at the end that you would have never got if you hadn’t considered it. This will also make the model feel a part of the process, and feel his/er ideas valued and respected. Even if the idea doesn’t fit with the theme you had in mind for the session, taking a few more pictures is not expensive in time and money nowadays, and the model will be happy to have them. If you don’t like them, you are not forced to publish them in your social networks or blog.


Use props to distract the attention

Holy Week V
The prop the girl is using provides more interest and context to the picture than a picture just of her face. Her gaze from behind the prop is what gives strength to the composition

Another trick that I read some time ago and worked very well for me is bringing a small prop to the session. Sometimes the object is related to the theme of the shoot or something related to the model. Other times it is just something innocent that doesn’t hurt the image, like a lollipop, a small ball, a ring, or a bell. The point is that when the model has something to fiddle with, gets distracted with it and feels less nervous. It works as an axiety release valve and helps the model to feet attached to something. If the prop doesn’t belong to the session and is cheap, you can give it to her as a present, so she has a nice memento of the shooting.

If the object is related to the shooting theme, you can also use it artistically to show a new perspective of the model’s personality. For example, if the model plays a musical instrument, you can make some pictures using it. If s/he likes an sport, you can use clothes and a ball related to that sport or the favourite team. This way the model has a familiar object to distract with, and it can add a new dimension to the portrait and what you want to show.


Respect the space of the model

This is a so important advice that takes a paragraph only for it. Give the model space to move and act without interfering with him/er. The minimum distance should be the one that, even if you extend your arm completely to the front, you won’t be close to touch him/er. And also never touch the model to correct the posing. If you need to give indications, do it by voice, or do it yourself and let the model mimic you. Touching a model is usually one of the fastest ways to incommode him/er and make the session unpleasant.


Technical advice, because I’m never too much technical

Also some technical advice can help to obtain better pictures and feel better about the result of the session:

  • Use a short tele for portraiture. Between 70-200mm is recommended. Take into account the crop factor of your camera. In APS-C format, for example, a 50mm works like an 80mm, which is a good focal for this kind of photography. A zoom adds versatility, a prime lens allows you to take pictures with lower illumination level and take more advantage of the ambient light. Wide angle lenses might give a good portrait if you know how to use them artistically, but they tend to distort the face proportions, so use them with care. More than 200mm should not be used, as it compresses so much, yielding flat faces with low attraction.
  • Some people claim that for wide faces a narrow lighting should be used, and for a narrow face a broad lighting is the correct one. I believe it’s not a strict rule, and you can use both creatively, but taking this advice in mind gives you a good starting point for the session. This rules are made by experience and usually work well, but can also be broken if the creativity allows a better picture. Never avoid trying new approaches and experimenting a little bit, as you will be expending just a little bit of time in exchange for the possibility of achieving a great picture; but keep the rules at hand, as an average they will work very well.
  • If you do a close-up, you can cut slightly the forehead, but never cut over the neck. In case of doubt it’s better if you take the picture a little bit wider capturing everything and you recompose afterwards on the computer cutting on the right place.
  • If you feel that the model is appearing in the pictures with the face wider that what it should, make him/er to move the forehead slightly to the front. This will change the perspective and will help to dissimulate some defects. This Youtube video explains it very graphically.


The Pearl of Japan (Reprise)

This picture, taken using a focal length of 67 mm (full frame) shows the face keeping the proportions natural, providing a natural appeal.


With all this advice in mind, I think any first session with models will go slightly better. Do you have any other tip or trick you use in your sessions? Share it in the comments.


One daisy glowing in the night

One daisy glowing in the night

This is the second picture selected from the set I took at the fireworks in my city a few days ago. I believe that the first one, Dandelion forged in strontium, was a little bit more impressive (without being marvelous) but this one is also pretty in a minimalistic way. The composition in this one is shifted to the left. Instead of showing the part of the funfair on the horizon, this time I chose to show the buildings in the city. They are dark and blurry (thanks, wind!) but at least they are visible enough to provide a context: A synthetic flower over a forest of concrete buildings.

As I promised a couple of days ago, I’m going to talk about firework’s colors.

Matter is composed of atoms, that are linked together to form the compounds that conform everything around us. There are very few kinds of atoms, 118 discovered until now, but only 78 are abundant enough to be used in everyday processes (I don’t include uranium, plutonium and other radioactive elements which are unstable and created just for their consumption in very specialized industries). What makes matter different is not only the kind of atoms they are made of, but also the number of each it has and how they are linked together, in the same sense as Lego pieces form bigger figures. Those atoms are composed by a core made of protons and neutrons, and a “shield” made of electrons. And those electrons are the magic of the color in the fireworks (in fact, they are the magic of almost everything, but now I’m talking about fireworks).

The electrons of an atom are characterized by their energy. When they are left alone, they have the minimum possible energy to be linked to whatever they have around. But it might happen that, when you provide them with more energy, they will accept it. This increase of energy will be retained by them for a short period of time (usually the time is less than 1 nanosecond) and afterwards they will release it again. The most typical way of providing them with energy is by heating (for example, with a flame), and releasing the energy as heat is also very common. But also some of the energy is released as light, and depending on the amount of energy the color of the light will be different. This amount on energy that an electron can gain and lose is characteristic of every type of atom, and it is useful for identifying them (for example, in a mixture).

A firework is composed mostly of gunpowder, which works as a propellant so it can fly, and a shell, that also includes gunpowder to make it explode. And the explosion provides a lot of energy that any present atom can use to transform it into colored light. Many of the atoms provide light in energies that cannot be seen by the eye, but a few of them provide colored light, and those are the most useful when making fireworks:

  • Barium: Barium emits mostly green light when excited, so any firework that contain barium will be green. Usually it is not a very catchy green, so it is usually used combined with other salts to increase its appeal and tune the color (make it a yellowish green or a bluish green). Barium is also what they make you eat when they want to take X Ray from your bowel. Bowel tissue is transparent to X rays, so they won’t see anything (like taking a picture to the glass of a window), but barium is opaque. Filling the bowel with barium is like staining the glass with dark powder so you can see any fissures it might have, and reveals the shape of the tissue.
  • Sodium: Sodium salts, like sodium chloride (table salt) provide a yellow-gold hue. Sometimes it is used combined with other salts to modify their color a little bit. Sodium can be found in our bodies, and is one of the elements that sustain our electrical system (nerves, but also charge potential useful in our cells).
  • Calcium: Calcium salts provide a very deep and intense orange, like in orange juice.  Calcium salts are famous for being good for your bones but they are also used in construction, toothpastes and optical glasses. Calcium is a very versatile element!
  • Lithium: Lithium provides a very intense oragish-red color, very attractive to the eye. It is used to provide red color in the main parts of the firework. Lithium is also used in some medicines for mental diseases, as it connects well with the brain chemistry.
  • Strontium: Similar to lithium, it provides a red color. But in this case this red is less brilliant and deeper in hue. It works well to provide a red color to accessory parts of the firework. It was extensively used a few years ago to build the screens of color television’s cathodic tubes and other similar screens.
  • Copper: Copper provides blue light when is excited. It can be used on its own, to provide a cyan color, but also combined with strontium to create purple hues. Copper is a very versatile metal apart from fireworks, and has many uses pure (wires, pipes…) and as an alloy (brass, bronze).
  • Cesium: Cesium creates indigo light when excited. It is not very used as it is relatively expensive and you can achieve the color with some mixtures of other salts, but if you require the exact hue, then it is used. The most known use of cesium apart of fireworks is to make atomic clocks measure time.
  • Magnesium: Magnesium provide white light, but extremely intense. It can be used on its own to create white light but also combined to any of the other elements to slightly increase their intensity.  Magnesium is used as a metal to create alloys that are highly resistant without increasing their weight (aeronautics), but also as a salt in many ways: optic glasses, paper industry, fireproof woods…

Also, some other metals are combined to those above to provide special effects. Those are some of them:

  • Aluminum: Aluminum can be used to create bright sparkles that burst out of the explosion. The brightness is a property of the metal, while the way the sparkles happen can be controlled with the shape of the aluminum particles.
  • Antimony: Antimony creates a glitter effect, as it reflects the light in a very characteristic pattern. Apart of alloys, it is mainly used as a fire retardant.
  • Iron: Iron particles can change their color when temperature varies. They are used to create long lasting particles that change from bright yellow to dark red with time.
  • Zinc: Zinc is not colored by itself, but it is used to create smoke effects. The small particles of smoke can reflect the light of other compounds in a diffuse way.
  • Phosphorus:  It works in a similar way to “glow in the dark” stickers or figures. The difference with the previous elements is that it can retain the energy for longer periods of time (sometimes over 1 second, which means 1000 million times longer). The color is usually yellowish-green, but it is used for effect that last for many seconds.

If you think this a matter of magic and not an everyday issue, think again! When you light a propane or butane cooker, the flame is blue. Those gases are mainly composed of carbon, and carbon emits blue light when is excited with heat (and a butane flame is very hot, more than 400 ºC). Have you ever tried to boil something and the water spilled over the flame? If you remember, when this happens, the flame turns yellow. This is due to the amounts of sodium chloride the water have (especially if you added salt to it).

Do you want to try it yourself? Use a little flame, like the one in your cooker or a candle (warning, this is hot and you know it, be careful if you try) to burn small amounts of different salts solutions. You can take a rigid wire and make a tiny “O” on one extreme (like the device you use to make soap bubbles, but in a smaller scale), then take a little bit of liquid with it and put it over the flame. The flame will change the color for a moment depending on the salt used. If you use table salt, you are using sodium. If you use low sodium salt, you will be using potassium (reddish color). Do you want to use copper? Add a small amount of vinegar to the water and leave a piece of copper for a few minutes, until the solutions starts to turn green/blue. You can try that solution. And remember, you don’t need a high amount of the salts for this to work.

So, after this long explanation, next time you see a good fireworks show take a second to think about all the elements, compounds and knowledge that is needed to make them.

Can you identify the elements in the last two published pictures?

Dandelion forged in strontium

Dandelion forged in strontium

A couple of weeks ago I stated how much I hate glasses on portrait photography. Yesterday, I discovered something I hate even more: Wind at night. With glasses at least you have some tactics to try to avoid the distortion they generate on the face, but with wind… I found nothing that could really help.

Today’s picture depicts the fireworks that traditionally mark the end of the funfair in my city. Some years ago I managed to capture a few nice pictures from a close distance. This year I wanted to try something different. Instead of staying close and using wide angle lenses, I took a friend and we went to a near hill, 3 km away from the launching spot. Although I though the distance would be too much (it is the nearest elevated point to that place), a focal length of 200mm solved the problem much better than expected.

Of course, when we had everything under control, the problem begun.

We haven’t counted with the wind. In my area, wind is usually not a big problem, as hardly ever blows faster than a breeze. But, from time to time, it has a harder day. In the middle of the city, surrounded by buildings, you don’t notice it. But after the sunset is when it blows harder, and in the top of a hill it’s difficult to cover from it. The result: a lot of movement on the camera and a great loss of sharpness. Do you notice those waving lines in the trails? That’s a product of the wind. Of course, in that part it’s even aesthetic, but if you look closely to the city, the blurred and soft aspect it offers it is also due to the wind.

In many places they will offer you tips and tricks to deal with this event:

Use a sturdy tripod (check!), and don’t use the final segment of it, as it offers less stability than mounting the camera directly on the three legs (check!).

Although it is recommended to turn the stabilizer off when shooting on tripod, we left it on to compensate the wind movement. It worked, but only a little bit. Better than off, but not for much.

Put your body between the wind and your camera, to shield it. If it worked for us, we didn’t notice it. Perhaps in the fastest gusts it helped a little bit, but on the long run it seems like it did nothing. Still we followed this trick, as it was easy to do and provided the moral calm of knowing that you were trying all you could.

Tie your camera bag to the bottom of the tripod. It will add weight and help to stabilize the camera. In my case, this didn’t help. The tripod is heavy enough and it wasn’t moving. The problem was in the camera and the long lens it had attached, which was acting like a lever, multiplying the force of the wind. The resistance to movement overall is only as good as the weakest part. Important: Only do this if your bag is heavy enough to stand the wind. If it doesn’t, it might swing and crash the tripod.

Use a fast shutter speed. Impossible in this case, as we were interested in the trails the fireworks leave, not in freezing the hot spots. The same happens with increasing ISO and aperture.

– Of course we didn’t try some absurd tips for the situation, like moving to another place to shield from the wind (we wouldn’t be able to take the pictures we wanted) or waiting for the wind to stop (fireworks are not going to wait for us). But they might work for you in another situation.

At the end, nothing really worked well, but at least it helped to turn a bad picture into an acceptable one. Knowledge is not usually a magic bullet to achieve things, but when the situation gets hard, at least it can help to turn the odds into a more favorable result.

What do you think? Do you like it? In a couple of days I will add one or two more of this session, and explain a little bit about the colors of the trails.

Warm light filter

If you look through the internet you will see that many popular pictures are characterized for having a very warm and catchy light, which makes the picture very appealing. If you take the time to analyze the pictures you see you will realize that many times the illumination they are depicting is not possible in the real life. It is easy to find thousands of very red dawns and sunsets on the internet but… When was the last time you saw a sunset so intense with your bare eyes? Probably never.

Many of those pictures have the light enhanced by filters or other photo-artistic means. Sometimes is on purpose, by adjusting parameters as the contrast, vibrancy or hue. Others, is just by chance: many mobile devices tend to modify colors in automatic mode to make the pictures more appealing to the viewer. Some of those modifications may be drastic and easy to detect, but in many cases the modifications are subtle and faint, and only the trained eye can detect them. In all those cases, the modifications were made not to get a faithful picture, but an artistic one, a picture able to impress us with its beauty and to transport us to a magical place.

This process I’m explaining here is not in the subtle category, but it can be made as subtle as needed and combined in different ways to get amazing atmospheres. The following procedure creates a global warm light atmosphere for the entire picture, but with a little of patience it can be adjusted for only some areas of the picture, and applied selectively to the appropriate sections. Anyways, even in its global form, it can be useful to enhance a sunset landscape, or, like in this example, to include some artistic warmness into a cold atmosphere picture.

Base: For this example, I’m using a picture of two friends of mine. The base picture is already processed, the color is reliably adjusted and all sharpness modifications have already been done. It’s what it can be considered a faithful finished picture. As the copy is stored in a lossless format in 16 bits, working with it doesn’t mean losing any quality.

00 Base
This is the starting picture. The atmosphere is extremely gelid for such warm gazes.

Step 1: In order to adjust the atmosphere, I’m going to separate it into two parts: highlights and shadows. This separation will allow me to adjust hue and relative brightness considering those two important parts of the image. To adjust the tone of the highlights we will create an adjustment layer using the mode Gradient Map. We have to create a gradient using white for the light part of the gradient, and in this picture I’m using a pale orange / sepia color for the dark part. To get the magic done, set the blending mode of the layer to Multiply. This mode will add the hue of the dark parts of the gradient to the  dark parts of the image but won’t affect (almost) the bright parts. This way we will get a dim painting of the light parts, that will be more evident as the color of the picture gets further from pure white. Of course, we can adjust afterwards the gradient parameters to obtain a better result for our picture. Just play with the controls until you are satisfied.

01 lights parameters
Conditional blending values for highlights.

The problem in this step is that the layer is also modifying our shadows. To avoid it we need to modify the blending options (double click on the layer, to open the blending options box). In the section of “Blend if” we will use the Grey option, and pressing the Alt key we will move the dark triangle of the “Underlying layer” to a middle point, around 128. This way what we are doing is deactivating the light modification gradually as the image gets darker, but without using any drastic cutout that can leave artifacts in the picture.


01 lights
The cold light got a little bit warmer. Still, the dark parts of the picture didn’t realized that the temperature is raising.


02 shadows parameters
Conditional blending values for shadows.

Step 2: We repeat the same procedure as in step 1, but this time we will use a color for the gradient a little bit reddish and darker. To affect the darker areas, we will use the “Linear Burn” blending mode and, because the effect is very intense for this mode, we will reduce opacity under the 50%, to a point where we are satisfied with the result. Again, this method affects also the highlights part of the picture, so we have to modify the “Blend If” controls. This time we will move (pressing the Alt key) the white triangle of the underlying layer to a satisfying middle point.


Of course, this is a coarse adjustment. We can add any other adjustment layer over each one of those, like levels or curves, and using the “Clip to layer” option it will modify only the gradient layer effect, not the entire picture.

02 shadows
The overall tone now is more adequate, according to the situation. Also, the light now blends well with the girl’s hair, which is an incentive for using this processing on this picture.

Step 3: To refine the process, we will add a global warming effect (in this case it is not environmentally dangerous, as it doesn’t require CO2). This will allow fine adjusting the global tone of the image. We have to create a Solid Color adjustment layer, and choosing a color. As we are moving in the warm area of the spectrum, I used a golden color. For a subtle effect, use Linear burn blending mode with an opacity of 20%. For a more drastic effect, use Soft Light with a 5-10% opacity. The first one will tune the color focusing on the dark or more saturated parts while the second one will focus on clarifying the picture and increasing contrast. The decision is up to you. In this case, I used the first one.

In the picture you can see the final result. What do you think? Do you know any other way to add a magic light to a picture? If so, please comment and share your “secret”.

03 Highlights final
A subtle golden glow tone finishes the atmosphere. No more cold white light!



Today is time for a little self-portrait. A couple of days ago a received my new Canon 70-200 f/4 L IS, and I wanted to test it. I don’t have much free time now to make a session, so I decided to try it on me. Perhaps I’m not the best subject to photograph, but at least I would try it.

I’m not very proud of my self-image, and I don’t like being photographed. If I could, I would substitute my image for a grayed-shilouette, like unlocked characters in videogames. But even if you don’t like it, you are projecting your image to the world every day, all days. People will see you the way you are, so avoiding taking pictures, even if you don’t like them, is not a solution, just a way to delay the problem until the next time you get out of your house.

But I have to admit that any photographer should do self-portraying from time to time. It makes you think on what you are doing from other perspective, and you have to develop new strategies to take a good picture. Autofocus is nice until the moment when you can’t be there to press the button or, if you are there, there is nobody in the place to focus on. Some people change the tripod position to focus and afterwards change to manual. It works fine when the depth of field is large enough, but when it reduces to a few cm the method gets too imprecise for a correct measure. Eventually you manage to find a way, getting a dummy on the place and focusing on it, or just estimating the distance by trial and error. This time, I have to admit, focusing wasn’t an issue and I got the correct distance easily.

There are a few things I learnt while doing this picture:

– Working on 4.5 square meters makes you feel like in the old times when people sent ships to the moon using technology that we can now find on our watches, or when people built operative systems in 32 kb of memory. It requires a lot of optimization, especially when your main umbrella is 109 cm in diameter. Everything has to be calculated to the millimeter. You learn how to compensate lights using 2 A4 paper sheets as diffusers, and eventually to accept that those pesky shadows won’t go away no matter what you do.

– The 70-200 f/4 L IS is an exceptional lens for portraiture. Not only it doesn’t distort the image, it is really sharp and keeps the face proportions in an attractive way. The color is wonderful and focusing is precise as can be. Even for my small improvised studio the focal length wasn’t too long. I started working at 70 mm and I managed to go up to 93 mm without any problem. Of course, using a full frame sensor helps, with an APS-C sensor I would have been forced to change to another location, or change the lens.

– I hate glasses. Perhaps small low graduation glasses might be fine, but big and high graduated glasses are pure hell in photography. I don’t know how to manage them, and I would really appreciate if any professional photographer would tell me. The little bit I moved my face to any side, the distortion the glass created was so big that almost shrunk half of my face to half of its size. Too many halves for my taste. I had to keep my face perfectly aligned to the lens axis to avoid (or at least dissimulate) the distortion, what makes a boring portrait. No broad or short lightnings, no creative profiles, just boring frontal planes. Of course, the first picture is nice, but when all the good pictures you manage to get are those that seem like taken for an ID card you start thinking on how much you miss landscape photography. And of course, taking the glasses away wasn’t possible. Not only because I need them to see the result while I’m taking the pictures, but also because they are a part of me, and I wanted a portrait of my whole me (avoiding a problem doesn’t solve it). Too many restrictions make difficult pictures.

Anyway, this is the result. I won’t probably take any other picture of myself in the next months (perhaps years). But I guess it’s something you have to go through from time to time, at least as a didactic exercise, to know how it feels to be on the other side, while still keeping control on the picture. Like playing chess against yourself.

The Photographer’s Ephemeris

tpe_expanded_500x500The Photographer’s Ephemeris (from now on TPE) is one of those apps that will help us to plan our sessions better. It is rather similar to Exsate Golden Hour, but it includes some differences that help to make TPE complementary, instead of an alternative. This program shows a map of the desired place, and overlaps a circular graphic showing the position of the moon and sun for an specific period of time. This way you can use it to plan a session knowing where will the sun be at the golden hour, or at what time the moon will be in an specific position to photograph it.

Compared to Exsate Golden Hour (EGH) it has advantages and disadvantages.


-As with EGH, there is an app for Android and iOS, so you can look at it anywhere. It requires a data connection in order to access the map. But it also has a desktop app, that you can access from your computer for free. For decisions made on the spot it is usually enough the mobile app, but for a good planning the big screen of a PC and the precise use of a mouse is an advantage.

-It takes into consideration the shape and altitude of the horizon (the terrain, in general), which is the best and most useful advantage of TPE over EGH. One of the problems that normal calculations have is that they consider the horizon to be flat. Many programs can tell you that an object will be visible just because it’s 10 degrees over the horizon. But if it happens that you have a mountain in front of you, that covers 15 degrees high in altitude, it will cover the object and the program won’t know about it. Also, sunset is achieved before the expected time when there are mountains around, and happens later if you are in the top of an elevation and the horizon is flat. TPE has all of this into account. Apart of the main point, that you set in your camera position, you have an auxiliary marker that you can set in different points of the map, and it will calculate the relative elevation of the point respect the main marker. The difference of altitudes, given in meters and degrees, allows simple calculations of the exact time for an event.

-Based on the altitude of the main marker, it can estimate the position of the real horizon. This way you can estimate which part of the scenario will be visible in your landscape, or in which range can be annoying a present object that might interfere with the expected picture. Everything outside of the horizon mark can be ignored most of the times.

TPE example
This example is settled in Byparken, a park in the center of Bergen. The city is surrounded by mountains. In this example, the annoying one is the one on the right: Fløyen. The predicted point of sunrise in the example is marked by the yellow line on the right, but this assumes a flat horizon. As the grey marker of the right states, the mountain at that point is 417 meters higher than the main marker (the red pin in the center). As the information bar states, that supposes an elevation of 8.37 degrees. The thin orange line on the right shows the point when the sun will be over the real horizon. The official sunrise happens at 05:24, but the real one happens at 06:51, almost one and a half hour afterwards. The exact point where the sun will rise was calculated by trial and error, but it took a couple of minutes: Set the marker on the highest point and calculate the time the sun will be there. If the sun is higher than the point, move the sun until its altitude equals the grey marker. Now move the marker to that position, if the sun is now lower, repeat the procedure until there is no variation.


– Although both programs include information about sunrise, sunset and twilight, EGH adds more information, like the exact range of time that the blue hour lasts, or the time of the golden hour.

– TPE lacks of predictive behavior. You cannot set some rules and predict when will be a good moment to take the picture. In TPE you have to look at the data manually and predict the desired conditions, time and place moving the dials.

-TPE doesn’t include meteorological forecast. The predictions in EGH adapt to weather conditions like clouds, rain or clear skies. In TPE you have to take care of this parameter on your own.

So, after seeing all of this, the question is which one is the better to use. My decision is clear: both of them. They are free and they act as a complement of each other. I usually use TPE when I have to plan for a new place. It allows me to know the terrain, and to take care of what I really want to do. Once I have chosen the place and the time, I use EGH to refine the moment, to set a condition for the weather or repeating times when the event will happen. In a sentence: I use TPE to discover new places and find where and when I want to place my camera, and I use EGH to remind me of every time my desired conditions are met, and to refine them to include additional parameters.

Do you know any other programs or apps that are similar? Just tell us in the comments section!

Ease of use: 4.5 / 5 (The higher the better)

Specificity: 4 / 5 (Higher doesn’t mean better)

Applicability: 4.0 / 5 (Higher tends to be better)

Name: The Photographer’s Ephemeris

Producer: Crookneck Consulting LLC

Platform: Android, iOS, Desktop.

Price: Free

Download: The official page (Links to every platform) , Direct access to the desktop app

Photography basics: Subject to background compression

It is usually believed by novel photographers that focal length (also called zoom at those stages) is just a measure of how close you are to the subject. If you take a picture at 50 mm it is expected to appear the same as if it is made with a 100vmm, as long as you get close enough to make the subject appear the same size in the frame. And those novel photographers usually will pass the following years believing this, as they take thousands of pictures, wondering why some lenses seem to take more appealing pictures of some subjects, why other lenses give best results for another kind of pictures.

The use of a short length on the picture exaggerates the size of the shoe, compared with the rest of the body. This provides strength and dynamism to the picture. Picture by Keia Eskuetan.

But it turns out that when you modify the focal length of the lens, you are not only modifying the apparent size of the subject, but also the relationship between the distances in the picture, and this makes a great impact in the mood of the scene. When taking pictures with a wide angle lens, usually all distances tend to appear exaggerated, especially on the borders.  Even the tiniest distance in the frontal axis tends to get amplified. This can be a problem or a feature, depending on how you work it: for a portrait, it tends to exaggerate the facial features, something that can be seen unpleasing when looking for beauty and perfection. But also, this enhancement of the perspective can provide of original compositions when used creatively, or when a more comical approach is desired. This is due to the increase of the field of vision the lens has, capturing more surrounding area and “bending” the light so everything can fit on the frame.

On the contrary, when using a telephoto lens, all distances in the lens axis get compressed. On the real world, a distance of 50 cm and 100 cm are seen as the second one being the double of the first one. But on the final picture this relation doesn’t hold, and the second one only seems slightly larger than the first one. When using a telephoto in portraiture, it usually yields flat faces, with little detail. It is useful, for example, to hide a prominent jaw or nose.

Como buitres al acecho
On this example, the “doves” are separated a distance similar to the one between the shoe and the knee in the former picture. In this case both seems almost equal, while in the previous picture the distance seemed much bigger.

This effect is what it is known as the subject to background compression of focal length. This is the reason that a wide angle lens is usually preferred for landscape photography. It is not only that it allows more field of view, which means capturing more scene. Also it increases the apparent distances, providing the landscape with volume. This is one of the main differences between a good picture and the typical one you get with your phone on a rush… the second one seems flat, boring and artificial. Also, in portraiture, a moderate telephoto is preferred for this reason (usually between 70 and 125 mm). It requires getting further from the model to get the picture, what can be seen as “cold” sometimes, but it helps to avoid giving too much importance to facial defects. Using a shorter length usually produces long noses, big ears or an exaggeration of the distances of the face, making them seem long. Using longer lengths usually provides flat and boring portraits, without personality. Anyway, depending on the face, sometimes getting out of the standard focal lengths can be a good option.

Also, an Instagram video has become popular in the last days, which shows why the camera seems to make us fatter. This is also related with this topic. As it can be seen in the video, as we increase the focal length, the face of the boy seems to get bigger. The fact is that there is no real difference in the size of the face by itself. The posterior parts of the face, which are clearly distinguished with a short focal length, tend to become “close” to the frontal of the face as we increase the zoom, which make us more difficult to differentiate, for example, where the face ends and the neck begins. The effect is also more notable because all the pictures are compared as a succession, giving the impression that the low length picture is more “thinner” and the long length picture is “fatter” than they really are if you look only at one, isolated. The more appealing picture, in fact, would be one of the intermediate ones.


How the face proportions are modified as the focal length increases. Gif by Dan Vojtech.

For more information about this topic, and a graphical example of this effect, I recommend reading the article Exploring How Focal Length Affects Images, by Andrew Childress , where a wider and more complete explanation, with example pictures, can be seen. So, the next time you want to take a picture, stop for a second and think: how do I want the background relate to the foreground? How do I want the distances in the lens axis relate with the distance? And once you answer to those questions, choose the right focal length, even if it means having to move a little bit. Remember: you should always move to adapt the picture to the chosen focal length, and never adapt the focal length to the position you have chosen for taking the picture.