Sometimes, I am somewhat dissatisfied with the size of my digital photographs. The simplest way to improve (after the fact) is to run any Photoshop-like program, and apply a resampling (resize) preferably with the Lanczos method, in order to get 2x or 3x or 4x more pixels. It is fast and efficient, but it usually generates a slightly blurry but bigger picture. If the target is a large format print that could be watched from a short distance, or for something like printing on a very large canvas, I should have more details in the picture (more than what was initially collected by the camera). The image is still too small.

The other approach would be to use a software tool like Topaz Photo AI or the all-new Adobe Photoshop “Image Enhancer AI”. Those are Pro tools (“Pro” means “expensive”): Topaz costs 199$, its competition often relies on monthly subscriptions. For a normal guy (who only needs it a few times per year, at most), this is not applicable.

Why not look toward (free) Open Source tools? Fortunately, some are existing.

And I added another constraint: I want it to run on my own PC (no cloud resources and/or cloud subscription like what StableDiffusion requires).

The general idea

For a few months, I have been impressed by the multitude of Artificial Intelligence solutions (Didn’t you hear about something named ChatGPT?) and not only to generate human-readable text. Tools like StableDiffusion, Dall-E, or Midjourney generate spectacular pictures from a textual description (a prompt). But they often include a mechanism to extend their text-to-image activity to a much larger scale (usually they produce small pictures to sparingly use computer resources).

My solution – in details

I won’t detail the steps taken to reach that solution. But here it is.

One strong PC, with two options/directions that could be imagined.

• A powerful graphic/video card (not always, but often based on nVidia GPU).
• A super-powerful PC with (most critical) a lot of memory.

In my own case, I chose:

• i9-13900K (with a good AiO liquid cooler, because this beast can heat up quite quickly when running 100%),
• 64 gigabytes of DDR4-DRAM memory (you will hear that this is the limiting factor for the image size).

But, you could work with a smaller CPU (with a lot of patience), or a Mac computer.

Nobody would claim this is a cheap PC, but it definitely works reasonably well.

EasyDiffusion installation steps

EasyDiffusion is a free version of a tool very similar to StableDiffusion. You may install it on your PC, and it will run either on the CPU or on the video card.

• Download the installation program from the EasyDiffusion home page (according to your needs, you’ll choose the binary for Windows, Linux, or Mac).
• Follow the installation steps from the page, don’t worry about changing any option right now. You will need some patience (it goes and download additional models and other packages, it will build some of the files during the installation).
• Run the program from the Windows “Start” menu. This will open a Command window to run a few background things (no need to even look at this), then a browser window giving access to the EasyDiffusion user interface.

• For the first configuration, I would recommend clicking at the top of the page on “Settings” and selecting the following program settings:
  • Auto-save Images: YES
  • Save Location: select the appropriate directory to save your creations.
  • Open browser on startup: YES
  • GPU memory usage: High (with 64 GB of DRAM, you want to use it all or as much as possible).
  • Use CPU (not GPU): YES
  • Beta channel: NO.

Create a text file whose name should be YR_MoreResolutions.plugin.js with the following text contents:

// ==UserScript==
// @name         Ski-SDUI-MoreResolutions
// @version      0.1-YR
// @description  More Resolution Options
// @author       Super.Skirv, JeLuF, YR
// ==/UserScript==

(function() {

	options="";
	for(i=6; i<=32; i++) {
		options += '<option value="' + (64*i) + '">' + (64*i) + '</option>';
	}
    options += '<option value="' + 3072 + '">' + 3072 + '</option>';
    options += '<option value="' + 4096 + '">' + 4096 + '</option>';
    options += '<option value="' + 4640 + '">' + 4640 + '</option>';
    options += '<option value="' + 6960 + '">' + 6960 + '</option>';
    options += '<option value="' + 8192 + '">' + 8192 + '</option>';
    document.getElementById('width').innerHTML = options;
    document.getElementById('height').innerHTML = options;

})();

• Install the plugin, by copying the YR_MoreResolutions.plugin.js file in the C:\EasyDiffusion\plugins\ui directory.

Operation and typical use

When arriving at the EasyDiffusion page in your browser, you will need to feed it with a source picture to enlarge and the right upscaling options:

• Choose the source image with the “Browse” button.
• Custom VAE: None
• Image size: 3072 (width) by 2048 (height) – for 64 gigabytes of DRAM.
  • This parameter is dictated by:
    • The need to work on an image as large as possible
    • The memory size limitation. If you have only 16 gigabytes of DRAM memory, you will need to reduce these parameters each in a ratio of 2: 1536 by 1024. But you can explore the optimum for your configuration.
  • Swap the parameters if the image is in portrait (vertical) orientation.
• Inference steps: 50 (this has practically no influence, you can leave the default value)
• Prompt strength: 0 (This is the critical value: It will ensure that EasyDiffusion AI never tries generating a new artificial image from the prompt text, but concentrates on the upscaling job).
• Output format: choose the right file format for you.
• Show a live preview: NO
• Scale up by 4x with RealESRGAN_x4plus (Avoid RealESRGAN_4xplus_anime_6B which is totally not for photography)
• The prompt (at the top of the page), and the Seed value (in “image settings”) are not used, except when naming the output file.

If you don’t have enough memory, you can always reduce the “Image size” or you get a message like the following:

The message tells you what memory size it tried to allocate/reserve. This will guide you when reducing the “Image size”.

If you are correctly configured, after a few minutes, you will see the result like in:

Notice: the new resolution is now 12288 x 8192 pixels (we left from 6960 x 4640 pixels, to reach no less than 100 mega-pixels). But, above all, the Artificial Intelligence (of the ESRGAN network) determined new pixels not only interpolated but imagined as complementary to the existing ones and totally credible.

In the picture used for this example, don’t spend time checking the blurred background. But take a look at the fine details in the structure of the dragonfly wings. It may not be perfect (from the professional entomologist’s point of view, there are anatomy errors), but the ribs/veins and the light reflections are very natural-looking, even in areas where they are only half-drawn or half-blurred. The same remark applied to the eyes and antennas of the dragonfly.

On the opposite, don’t even believe the proposal from the software: It’s no use clicking the button to upscale again to 20032 x 13376; You already reached the maximum size allowed by the available memory. You can’t go further with only 64 GB.

If you want to check the results, I made the files available:

• Original
• Upscaled x4

But it does not have to be like that. There are a few important things to check and several key aspects to check during this important process. Buying a monitor is not simple, but it does not have to be a very complex protracted decision. Most of the critical factors are readily available (and often overlooked).

Screen size

The very first factor is the screen size. Choice is large from all manufacturers, but you must start by defining what is the optimal size you intend to buy. It comes from the size of your desk and the kind of confort you want to have (too large a screen means you have to be much farther from its front face).

Usually, you would not look at a size smaller that 23″ because most of the applications we use are crouding our screen surface with sub-windows, tool bars, multiple menus, etc. Those are all very useful but they eat up a lot of screen size. 23″ or 24″ screen sizes more and more appear as the minimum for confort.

But even more importantly, you should think about the resolution of your display. A large screen with too few pixels may not be worth it. So, check the maximum resolution accepted by your computer and see what maximum resolution the monitor candidates can accomodate. Today, it is fairly easy to get a 1080p/1080i monitor (displaying most commonly 1920×1080 or 1920×1200 pixels) or one of the higher resolutions like:

• 2560×1440 pixels
• 3840×2160 pixels, also know as 4k

(These figures may vary slightly with the exact size of the panel, but they are industry standards you will find easily on the market).

I would recommend that you start at 1920×1080 pixels, but also look at the higher end options too, since they bring a lot more confort (if you have enough desk space to accomodate them). 2560×1440 pixels with a 25″ to 27″ panel size is usually a very good size for our photographer’s needs. Going further would probably push you into a larger monitor. (30 inches mean a much bigger display on your desk and probably the need to watch it from a longer distance!)

Monitor technology

You will find all kinds of available technologies, with incomprehensible abbreviated names. But a few ones are important recognizing.

TN screens

Those are usually cheap… and look cheap too. Specially for us photographers. For one reason: This TN (Twisted Nematics) technology is very sensitive to the viewing angle. If you are exactly in front of the display, everything looks fine. But the more you move out of the exact axis, the more contrast drops and the more colors shift (mostly unpredictably). Things become quickly so unreliable that you should avoid them.

VA, AVA and MVA screens

Those have very good viewing angles and high contrast. But unfortunately, they are intrinsically less able to reproduce colors correctly.

IPS screens

Those are more expensive, but they have excellent contrast and are very good at displaying precise colors. Some of them are factory-calibrated to be sure that you get the maximum result and this makes quite a difference when you look at a photo.

Some technologies you should not care about

Some fine technological feats are of no use for the photographer (they are intended for other categories of monitor buyers):

• G-sync or FreeSync are strictly for video game fans
• Response times have nearly no influence on mostly static images we use
• Refresh rates should be at least 60 Hz, but which monitor is not? Higher frequencies will reduce eye strain but only marginally.

Some technologies you may want

Check for the presence of a USB hub or multiple video inputs to connect to your computer (VGA may be present but it provides a relatively limited analog input that is not up to the task of excellent colors rendering). You may appreciate the presence of loudspeakers, but this would have nothing to do with your photographer’s activity, I guess.

Color depth

The final critical feature is the size of the color space that the screen can display. SRGB has been known for years as able to produce adequate colors for a computer display. But it is also clear that if you have pictures in AdobeRGB, the color gamut (the extension of colors that are included is larger: You get deeper reds or deeper blues). Most monitors will now display 99% of sRGB.

I don’t like very much the usual graphic to display this, but the picture shown here on the right gives you a rough idea of what can or cannot be displayed.

Don’t compromise and never accept less than 100% sRGB. But pushing for near full AdobeRGB has a huge impact on cost. So, if you decide to go that way, be sure -at least- that you will have a stable light in your room (forget about a nice large window opening on the outside, lighting your desk and monitor), and that you will regularly calibrate your monitor for color stability.

Personally, I still consider that the cost of full AdobeRGB is still a bit too high (and it may stay that way for quite some years because of the inherent physical limitations of technologies currently used). But it’s really an open question if you live from your photograhic work…

After you buy

When your computer monitor arrives, the first and most important thing to do is to calibrate it. Either buy or rent a calibration sensor and you will be able to obtain the most out of your expensive purchase. Many people propose software (even free software) to do some kind of calibration. This is insufficient and very difficult to use if you are not trained for that softare.

I would consider renting a calibration sensor, contacting friends to buy one for several photographers, borrowing one from somebody else (even getting somebody to make you a gift). But do calibrate your monitor!

Until you do this calibration, you will not get the best your monitor can offer. And you will be surprised by the impact this has: It is immediately visible on nearly all the monitors (even on some of those professional monitors which come factory-calibrated).

Personally, I use a Datacolor Spyder5. Good software (easy to use), relatively fast and reliable.

• Nikon launches an 800mm f/5.6 which is becoming the biggest product of its catalog.
• Canon is still testing the all-new, all-beautiful 200-400mm f/4 (with integrated 1.4x f)
• Sony should renew its 70-400mm with an evolution whose content is still quite unclear.

These lenses will certainly be nearly impossible to purchase without winning the Lottery or holding a bank up (but from what I hear about their financial situation, even this may not be a viable option) but they may be attracting the attention of a few pro photogs and triggering some dreams from sports and wildlife photographers.

It’s now time to look back at the year behind us and try and see what is coming in 2012.

2011

Two features were really striking in 2011:

1. Natural disasters: The earthquake in Japan and the floods in Thailand have been claiming lives and hitting hard on the photography industry.
2. Sony had a glowing track of new products.

The Spring 2011 earthquake in Japan was right in front of major photo industries and we all remember that the consequences have been devastating for both the Japanese people and an industry whose main centers were very near the epicenter. Nikon was certainly among the first victims, but nearly all other companies stumbled.

As if that was not enough, Thailand experienced heavy floods that are just now beginning to really recede. Again, Nikon was a very visible victim.

The employees of all these companies fought for both their security and the well-being of their companies. Some of them even succeeded in avoiding some of the most visible impacts. For example, nearly no sales were lost during the Christmas season, but this came from hard work more than actual luck.

Despite these tough conditions, Sony has made impressive announcements both in the DSLR camera and hybrid camera markets. You may think what you want of the qualities of the new semi-transparent mirrors for the Alpha SLR series and of the APS-C hybrid NEX cameras with inter-changeable lenses, they both brought new features, new ideas and even new ways to look at a digital camera that all other photo companies now need to take into account.

The Sony Alpha 77 (and the Alpha 65) appeared as a major step forward and the NEX-7 and NEX-5n have been such a commercial success that Sony is currently reviewing their strategy to cash quickly on these.

Add to these that some actors left or nearly left the field (Kodak selling the last parts of the photography business, Olympus fighting for its life after the surprising financial decisions of its former management) and we have a pretty unusual year on our hands.

2012

So, what will a new year bring to the photographers? We all know that reading the future is an exercise that is both humbling and funny when you look back at last year’s predictions. But, there are a few elements that can already be plotted and should not be too far from the mark.

• Sony will extend quickly the lens list for the NEX family. Probably adding new cameras in the second half of 2012.
• Sony will produce two (maybe three) new Full Frame SLR cameras to replace the old Alpha 900. Resolution may be enormous since they will probably apply the technology from the existing 24-MP APS-C sensor, leading to something like a 33-MP sensor (or more if their engineers have a bit of fun with silicon wafers).
• Nikon, which has seen all its plans delayed by Mother Nature, will launch first a D4 (11 fps, 16 MP, up to ISO 102400 or ISO 204800) and a D800 (33 MP or 36 MP, up to ISO 25600, at 4 fps), then could well launch a replacement to the D7000 and D300s (either a D7100 with heavy body or D400).
• Nikon could also start at least a new entry-level camera around the end of 2012.
• Canon is going to replace the EOS 5D MkII (no surprise) but they could be preparing radically new products for entry-level photo markets.
• Pentax needs to find new directions now that it is in the hands of Ricoh. 2012 should be interesting: Either Ricoh will merely scrap the SLR business or they will launch a new family of cameras in 2012.
• Olympus cannot launch anything significant before they -really- find closure to their financial woes. It has already been so long that the chances of survival are probably under 50% now even if a buyer appears.
• Samsung is probably no longer trying to fight in the SLR market, but will keep adding to their hybrid line of digital cameras. They could become a major actor, annoying even Sony here.
• Panasonic will have a hard time finding ways renewing the 4/3 and µ-4/3 formats.
• Leica is in another world altogether.
• Sigma is already dead for the SLR, but they don’t know it yet.

All in all, I believe that the SLR market is condensing on a relatively small number of brands, while the hybrid market seems here to stay with even less players.

Since most companies want to move the awful year of 2011 back in the nightmare scene, expect some serious technology push, probably for the easy targets: more pixels, more sensitivity, more gadgets. The chances of seeing a real technology revolution seem small.

Film or cards; Kodak or Lexar; Years apart, but the same attention to production quality.

What is an EVF?

EVF = Electronic View Finder.

Be ready to see this term used more and more, even for SLR cameras. Many a compact point-and-shoot photo camera already is equipped with an EVF, but it is most common in video cameras (and it has been for many years already). In the viewfinder, instead of looking at an image coming from a more or less complex optical system, your eye is pointed at a small LCD screen (or TFT, or OLED or whatever similar technology) reproducing the picture recorded by the CMOS imaging sensor of the camera.

After all, this is only what we already know well as the LiveView mode on the back LCD of nearly all the compact photo cameras (and many SLR too). But as a small picture in the viewfinder.

An LCD display + a viewing lens = an EVF.

In some case, you could add a small mirror for space reasons, but this is a minor variation to the original tune.

EVF and Single Lens Reflex (SLR) cameras

A Single Lens Reflex (SLR) photo camera is organized around a very mature technology in which a mirror reflects the image through an eyepiece via a pentaprism (or a penta-mirror) as in the drawing on the left.

Of course, as the film (film plane and focal plane shutter) is hidden by the mirror, this one must be drawn up to take the snapshot.

All this is greatly simplified when there is an electronic viewfinder.

Advantages and drawbacks

Advantages et gains from an electronic viewfinder

The first positive side is mechanical and optical: If you remove the mirror and the pentaprism, the architecture of the camera will be greatly streamlined, lightened and its manufacturing cost will go down. For three different reasons (at least):

• reduction of the number of parts
• reduction of the assembly complexity (or of the amount of human work needed)
• increased ease of calibration (alignment) of the optical parts (an LCD and an eyepiece, rather than a whole bunch of parts)

Even better, the disappearance of a huge articulated and mobile mechanical system brings several positive effects:

• increase of the continuous shooting speed/cadence (there’s so much less hardware to move around for a photo)
• reduction of shutter noise (in a traditional SLR camera, most of the noise actually comes from the mirror up-swing and the shutter operation; It’s worth removing one of them)
• a lighter mechanical structure through reduction of the quantity of parts and since the enclosure is lighter because it has less stress to sustain in all these shocks and moves

Also, the electronic viewfinder exactly reproduces what the sensor sees. A 100% optical coverage on a viewfinder of fully optical design has always been reserved to the most expensive SLR cameras. But when you get an EVF, 100% coverage is totally natural and effortless. Luxury comes to you in the clothes (and price) of entry level.

Then, having an LCD display brings the additional possibility of adding all the data you can think of. Many brands already experimented with this and brought us some Electronic Viewfinders with features like:

• Histogram
• Artificial Horizon
• Zoom on a part of the image
• Telltales and technical displays (aperture, speed, ISO, etc.)
• Active AF zones (including face detection)

As a matter of fact, there is no other limit than what a computer screen can display: Data, image, etc. It’s only a choice from the designer.

And it’s not even the end of it. The EVF also has a major advantage when light is low: As far as the sensor can see something, it’s only a matter of amplifying light signal (like for the ISO sensitivity selection) to make the picture readable. If the photo is possible (even barely possible), showing it is possible. Think about it! A nikon D3s digital sensor could display what it sees even in the darkest dawn it can manage. Since the digital noise is only a marginal issue on an LCD display (even at 3MP resolution), you can see what the sensor will be able to grab even in a room too dark for comfortable framing with an optical viewfinder.

If you compare it with the lowest-priced optical viewfinders of entry-level SLR cameras, the EVF is an easy winner in those low-light conditions: These simple viewfinders where already unpleasant and dark, the EVF bring a new life to cameras used indoors.

Drawbacks and limitations of the electronic viewfinder

But all is not pink and shiny. EVF have the same problems as most point-and-shoot cameras find on the rear-side LCD displays used for LiveView.

To start with, even if the electronic viewfinder (EVF) is intrinsically better protected against sunlight, it is still very sensitive to bright light issues. If there was only one thing that the EVF-sensor pair does not like, this would be high lights and bright sun (sic!) highlights are easily washed white and overblown. You will find more white flat areas than nice clear zones. The worst comes when you add high contrast to the picture. The EVF has a very small dynamic range (this is not HDR!) and the designer must choose between high-lights and dark areas.

We tend to forget it, but the human eye is amazingly sensitive and adaptable. Behind a prism and a mirror, in the brightest mid-day sun light, it can marvelously adapt to over-exposure situations that totally saturate the purely digital EVF system.

Best (or worse), the eye is extraordinarily able to switch within a few hundredth of seconds from the darkest night to the most violent lighting without even thinking about the marvelous brain processes involved. On the contrary, the EVF system uses a gain correction which will change at any time the apparent brightness to track the actual scene brightness. What is clearly an advantage in a dark room, forces the system to include an auto-adaptation feature which will change the brightness of the EVF several times per second. According to what you point the camera to, the electronic viewfinder will change from lighter to darker display. Not really a problem, but this asks for some tolerance to a process which is not natural to our human eyes. You have to get used to it.

Additionally, but this is all very sensitive to the user (you!), watching a computer screen may be more or less comfortable in the long run. As a matter of fact, if you use the EVF for long hours (this is the case for the professional photographer waiting in front of the Cannes Festival stairs, for the photo safari amateur or for the intensive tourist photographer willing to bring everything on a memory card – don’t laugh, I’m sure you shoot it as much as I do), you may come back with a serious headache. This is certainly not an issue for everybody (and you’re not supposed to keep the camera to your eye if you want to immerse yourself in your subject), but this may become a real limitation to some people. At the strict minimum, you must take very seriously the setting up of the dioptric correction of the eyepiece (much more seriously than on your current camera) to limit the impact to a tolerable level. Unfortunately, the eye seems to get more prone to this issue when the photographers gets older… Our eyes are not equal in front of photo gear. Too bad, but true.

Last but not least, even if the electronic viewfinders improve in time (and Sony has shaken the market with a 2.3-million-pixel EVF in the Sony SLT-A65 and in the Sony SLT-A77 which has raised the bar to a whole new level for all the competition), the fine quality of a focusing screen in a common optical viewfinder stays incomparably better than what the best EVF can display today. Far better! This is the reason why they generally include an additional Zoom feature allowing to better judge focus through enlarging an area (On the opposite, a company like Zacuto proposes an enhanced edge sharpening to make the focus analysis easier or more natural).

If you let your camera focus for you, honestly, you shouldn’t care much about this. If you like to fine-focus your pictures (all the more if you use a USM, SSM or similar lens which allows easy manual focus correction), if you use wide-aperture lenses (which call for a very attentive selection of the focus location or AF area), you will be in a living hell playing with buttons to merely focus while you were used to just looking at the image in your old optical viewfinder. Interestingly, Sony keeping in its technological portfolio the marvelous AcuteMate focusing glass surface (bought from Minolta and still widely considered as the absolute best here), is the first to jump to the next technology.

While we talking about AF, we must stop and remind ourselves that the disappearance of the reflex mirror means that focusing is now done from the data directly available on the image sensor. Where the traditional SLR camera focuses just before the exposure, using phase difference sensors (high precision, high speed, and not very complicated), the EVF viewfinder seems to enforce the use of automatic focusing using the measurement of a contrast difference directly on the image (very consuming in terms of processor power, intrinsically less reactive). Sony, once again at the forefront, chose to insert a semi-transparent mirror to keep the fast phase detection sensors. But they are quite alone in this decision – for now.

(Temporary?) conclusion

The electronic viewfinder is no panacea. But it brings good answers to tough questions asked by the old reflex technology, starting with price reduction for the digital photo cameras which is always a major issue for the manufacturers.

The camera owners already used to the optical viewfinder of entry-level cameras will immediately recognize the many advantages and all the more if they are occasional photographers or if they come from the compact point-and-shoot camera world.

Expert users (not even pros) will certainly have more mixed feelings, even if the most advanced electronic viewfinders like the recent Sony ones progressed in great strides. As I read in one of the recent articles about these new photo cameras, Sony brought enormous (not incremental) progress but this is still an electronic viewfinder. Not perfect, but with enough advantages to balance the drawbacks.

Now, you may have to read the post again to choose which ones are the most important for you. But there is a big problem left: We usually choose our photo camera inside a relatively dark shop (perfect, ideal conditions to demonstrate an EVF) and we use it mostly during our Summer holidays (the worst conditions to use an EVF). Some people may be unpleasantly surprised if they do not think twice before choosing a camera.

If you did not have enough yet, you can always go back to the manufacturing of a Canon 500mm tele-lens (in a much older video) or you can make a virtual tour of a Canon lens manufacturing plant.

Is a UV filter harmful?

First, I would like to demonstrate once again the critical importance of having a very good quality filter rather than the usual plastic junk. Too often, we forget that the filter is degrading the image quality and that the better the lens, the worst the impact. To clearly show this impact, LensRentals tried to stack up to 50 filters on the same lens. the result is so immediately obvious that you don’t need to go pixel-peeping at 100% scale:

But even one filter will decrease image quality.

Of course, if you want to compare the impact of the impact difference between a pro filter and a cheap run-of-the-mill filter, you can stack just 5 pro UV filters and 5 cheap UV filters. Here again, you’ve got food for thought before you buy your next photo filter:

I insist! this last test does not use 50 filters, but only 5 UV filters and the result is immediately perceptible to the naked eye of the non-expert, even if you’re not looking for it. With only one filter, you can do the test by your own and decide that using filters is a matter of thoughtful choice.

How Polarizing Filters Work?

This is one of the best and simplest explanations and demonstrations of the operation of these basic filters: A Polarizing Filter on the left and the Neutral Grey Filter on the right.

Source: PetaPixel.

For a long time, we saw round and square filters, filter holders, gelatin or gel filters in colors or in grey hues, progressive graduated neutral density filters, stacks of filters, surprising little colored objects which were used to reinforce the sky, the clouds, the grass, the trees, the mountain glaciers, the lakes reflections, and which were used to smooth the sky, the clouds, the grass, the trees, the mountain glaciers, the lakes reflections. In short, apart from a few specialists as admired and as venerated as Mayans High Priests (apparently, without the need to practice human sacrifice to support their credibility, though), nobody really knew what all this was about. OK! I may be slightly off the board here (though I am sure I never heard of a photographer sacrificing live animals before shooting a landscape picture). Some people had learned a few tricks and recipes, but it was most incomprehensible to the masses like you and me.

With the advent of digital photography, everything changed: Photoshop or The Gimp or any other photographic software was able to replace all colored filters and the photographer could drop his collection of gelatins into the junk bin before going to simpler (or more complex) issues. We ran for the software manuals

But this was an error! If it stays true that the immense majority of filters are already in the first Photoshop-like software package, this is not a general rule. Colored filters are trivial to apply in Photoshop. Graduated filters are so easy that kids can use them. Color correction, exposure correction are but a click away in Photoshop and easier than a bunch of fragile gel filters attached to a complex holder. And if you do it wrong, just Undo it and try again.

But there are two optical filters that are still totally unavoidable even with the best software (and I am not speaking about the useless UV filter whose main application is protection against frontal shocks):

• Polarizing filter
• High-density neutral grey filter

Polarizing

A polarizing glass selectively filters light detecting its polarisation (a physical property quite difficult to perceive in most conditions but easily measurable on reflected light). The filter will reduce brightness of reflections without impact on the rest of the picture, an aspect that is inaccessible to Photoshop. It also plays a great role on slecting the desnity of the blue sky which varies in function of the sun light angle.

The effects of a polarizing filter on the sky in a photograph. The picture on the right uses the filter.

If you keep only one filter, this will be a polarizing filter.

Only one remark: If you find an old pola-filter from silver-analog-film times, check first if it is a linear pola. Films were not very regarding, but digital sensors can only react to circular pola filters. Note that this remark becomes less and less important since all current filters are actually circular polarizing filters.

High-density neutral grey

The other useful filter is a high-density neutral grey one. You can easily see through a low-density (pale) neutral grey filter but it is not very useful with the extreme range of sensitivity that our photo cameras exhibit today: Just rotate the sensitivity wheel a little or drop back to Photoshop. But, if you take a ND400, you’ll immediately notice that even if you can still see something through it, it nearly black now. The filter is so dense that it has the same effect as dropping the sensitivity though the floor a lot below the usual minimum of ISO 100. Consequence: Long exposures even at mid-day: Instead of 1/200s, you’ll need 5s exposures (1000 times more). Welcome to motion blur even on slow objects and under the mid-day sun.

The most common application (or the most commonly used) is landscape photography incorporating motion blur of sea water or white water river, as in the examples below.

photo credit: -Chiotas-

photo credit: ciccioetneo

photo credit: dahon©

Beware!

To answer a question from Sébastien, I will also remind you of some elementary precautions to apply when using a filter (any filter).

• Always buy a high-quality filter: You use a luxury pro lens with fluorite or ED glass optical elements; Don’t drop any plastic sheet in front of it if you want to avoid re-introducing ugly additional effects (like optical distortion and chromatic aberration). Unfortunately, the prices goes with this requirement.
• If you use a wide-angle lens (this is often the case for landscape photography, isn’t it?), select a low rim filter to avoid seeing it obstructing slightly the field of view (generating a small vignetting effect). Here, again, it comes with a price (all the more for the polarizing filters which are mechanically more complex).
• For neutral grey filters, beware of the chromatic shift introduced by the denser and cheaper filters. It is often easy to compensate in a software suite, but the filter may easily be slightly blue or orange (depending on brands).

So, what? Two filters in the bag, this is no problem.

Of course, with five lenses, you may have five different screwing diameters. But this is still much better than the whole collection of the past.

TinEye was always a little limited by the small number of searched web sites, but the results were absolutely great: They even find look-alikes, pictures that are quite similar but not identical (even deformed, re-colorized, cropped, etc.)

But here comes the web heavy-weight. Competition is going to heat up significantly: Google starts a very similar option in Google Image Search, which uses the small blue camera appearing on the right of the search bar when using the Chrome browser (and probably all browsers in the near future).

I tested this image search service to look for an image I know quite well (a giraffe photo I shot in South Africa in 2007 that I know is already quite pirated on the web – I found it in a number of different places but for those interested I have another giraffe photo which is even more pirated even if it is much less beautiful than this one, I think).

As a photographer, I’d like to find all the people who made a copy of my work and check if -a least- they took the time to attribute it to me (not enough, it’s copyrighted) or to give a link back. For YLovePhoto readers, I thought it would be interesting to do the test, review the results and share my impression about the service.

There’s no doubt in the results:

• TinEye: finds one copy on a French SkyBlog where the size was altered.
• Google: recognizes a giraffe, provides some basic data about the animal, finds some similar images (12 giraffe photos including 2 which have a very similar attitude) and finds 2 pages of blatant copies of my own photo (all sizes including a black and white modification). The two copies on my own web site are present in the list but not the one originally found by TinEye.

Temporary conclusion: Kudos to Google! But, TinEye is no junk and we should not forget it right now.

My recommendation to Google: Your results will be even better if you purchase TinEye (the founders will be happy to receive 50 millions US dollars) and keep the results presentation which are really great.

Aperture Revealed – 120 fps – HD from Camera Technica on Vimeo.

The same at a slower speed:

Aperture Revealed – 240 fps – HD from Camera Technica on Vimeo.

Thanks to Camera Technica for these videos.

Good luck.

• Nikon D5100
• Canon EOS 600D & 1100D
• Sony SLT-A33 & A55
• Lumix G3
• Samsung NX11

See also http://hubblesite.org/gallery/behind_the_pictures/ where you can see the same video in very high quality.

It appears that its an issue of trying to avoid discomfiture for the new or prospective users in front of what still is a very interesting technology. The Nikon D5100 has some night-related features, indeed. But this is not completely “night vision“. It’s more a combination of:

1. Extreme pixel amplification: Nikon did not hold any horses here. There result may be grainy as hell, but who cares if we can shoot at night?
2. Ganging up the color pixels in one monochrome photon bucket: The result is a black and white image, but it more or less quadruple the light that is collected for a single “pixel”.

The result is a grainy image. But it works up to something like ISO 102,400. It’s more a night mode than an extended sensitivity (Hi 2 is ISO 25,600).

Night vision examples from Nikon

Enables night shooting at super-high sensitivity up to ISO 102400 equivalent. Records monochrome (black and white) images.

Regular shooting:
Nikon D5100 night vision:

• All APS-C SLR cameras
• All Full Frame SLR cameras

Gee! This is going to influence the future of the Sony Alpha line.

In a nearly parallel direction, knowing that Sony decided not to announce a new Full Frame SLR camera this year (but only in 2012), it is interesting to notice the information provided by SonyAlphaRumours and telling us that Sony determined its strategy for the year 2011. For the most significant part, Sony forecasts a complete stop of manufacturing for the Sony Alpha 850 whose stocks will plummet quickly. On the opposite, the elder brother (more endowed too) Sony Alpha 900 will see its price drop progressively at the current 850 level, to boost sales, avoid leaving too much room to the competition and empty the stocks before the arrival of an Alpha 99.

Up to now, Canon and Nikon have been using mobile optical elements in the lens in order to compensate for the hand-shake of the photographer. This way, even if you’re not very steady, your lens will compensate your moves.

But the problem with mechanical image stabilization is that it’s limited by mechanical inertia. You cannot shake a lens block as fast as you’d want. This is perfectly OK for photographer shake (usually under 10 Hz) and it is adequate for vibrations on a tripod (under 40-50 Hz) if you use a really light mobile element. But it becomes tough when you want to correct easily tripod vibrations or even in-vehicle vibrations (usually 10 Hz to 500 Hz).

This is where electro-optical elements come in (and it’s not to be understood as Electro-Optical System or EOS). This is an optical element whose optical index can change under electrical voltage. In the patent drawing, you’ll notice that eo1/eo2/3d is a flat element (pressed between two electrodes). The major advantage of electro-optical elements is that the maximum frequency may be very high (probably up into the 1000’s of Hz).

With this kind of technology (already seen in other less precise patents), Canon could be preparing a third-generation IS able to adapt easily to tripods (not only accepting to leave IS on when using the lens on a tripod, but correcting in-car or in-plane shakes).

Source: Roumazeilles.net

This was really a photography marvel in a time when video was not an integrated feature of photo cameras.

Cassandra C. Jones did a relatively similar job from horse images collected on the Internet (probably with Google as her accomplice).

The future of Photoshop CS6 has been recently shown at MAX2010, the Adobe conference in Los Angeles.

I believe that the really impressive part of this technical demonstration is the set of automatisms which operate -essentially- by showing the machine an example of the result to reach rather than finding the right settings (as in this first example where the settings are automatically computed to reproduce the graphic style of the American photographer Ansel Adams). Clearly, Adobe keeps pushing Photoshop into more features built to reach a goal (sometimes a very ambitious one) rather than simply provide raw computer tools. And the magical de-blurring tool (registered trademark) is impressive too.

It’s really at the core of the idea of empowering the photographer. This is Photoshop CS6 (avant-première) before Photoshop CS6.

Via John Nack.

On the one hand, the European Space Agency (ESA) opens on its web site a presentation of its best pictures in a photo (and video) gallery where I grabbed this superb view of the East coast of Greenland.

On the other hand, here is the National Aeronautics and Space Administration (NASA), which already had a web site of the same kind, just opening a Flickr group to distribute some of its archive images coming back from the oldest sources like this solar eclipse photographed from the Apollo 12 mission.

Not willing to stay behind while they have a real treasure of exceptionally good pictures of the remotest stars, the European Southern Observatory (ESO) now has a Top 100 of pictures taken by the observatory telescopes. Like this photo of Centaurus A, a great galaxy stopped in her rotation by seevral ESO cameras in various light spectrums (visible light, X-ray, etc.)

But there was a fly in the ointment. Only a few weeks after the official launch, we learned that the digital sensor is overheating. We know that the digital photo-video sensors produce heat whenever they are used. The design of sensors for video (and thus used more or less continuously) must take that into account to avoid heat production at all costs. The design of the cameras must take into account the need to move any heat left right out of the sensor area (possibly out of the camera).

Sony definitely missed one of these, indeed. The SLT-A33 and SLT-A55 are suffering from over-heating to the point that, when reaching a certain temperature, the camera switches itself off (It shuts the sensor down). As can be seen in the table here on the side, in the worst case (when it’s hot outside, 40°C, and that image stabilization is enabled), this is going to happen fast, real fast: 3 or 4 minutes only. Probably nearly everybody will admit that this is incompatible with the normal use of video capture in a normal way.

So, no video grabbing during Summer. More or less.

If you thought that this was bad enough, there is a hint to the possible addition, unconfirmed by Sony right now, of another limitation (maybe, within a longer time frame) respective to the use of the LiveView mode (which is only a low definition video mode).

The solutions? Let’s try to think them out:

• No firmware change will reduce the thermal dissipation of the sensor.
• No firmware change will improve the heat transfer through the body of the camera.
• A firmware change could remove (more or less completely) the thermal protection. But would it be a solution to risk some other problems? (early aging of the sensor, unacceptable thermal noise in the images, etc.)
• Modifying/replacing the sensor (supposing that Sony has a better or corrected sensor)? This is going to be a MAJOR repair operation.
• If the problem really comes from a bad design of the camera body, this is worse… Nothing left in the bag of solutions.

I let you imagine Sony’s dilemma. I let you imagine why the communication will be really tough next week…

While waiting for it, for those asking if they should buy a Sony SLT-A55 and thinking that they need the video capture, I say: Wait. For those who are in a hurry, I will not less than they must forget about video on the Alpha 33 and Alpha 55 for now. Video is nearly excluded in these cameras for the time being. The small sound you can hear is those competitors (Starting with Nikon and Canon) exhaling loudly (These cameras had given them cold sweats and they now feel that this could buy them some more time to react).

Sometimes, rumours are really weird but they still have to be listened to and discussed. Recently, SonyAlphaRumors.com reported an “information” they received from an anonymous source that they indicated was new and possibly unreliable. To me, it was so weird that I thought it would not be worth mentioning (it is my role to filter out and choose the news I report here), but people keep asking questions about that.

So, let’s debunk it once and for all. The proposal is that Sony is designing the A77 with two semi-transparent mirrors instead of one. Starting from a A33/A55-type of semi-transparent SLT mirror, the anonymous source proposes that they wil add a second one essentially to mix Electronic ViewFinder and real images.

Unfortunately, it does not add up. Industrial problems crop everywhere in this design. For any reasonably seasoned engineer, it would be a nightmare to align correctly the two images, but any optician will quickly notice that super-imposing a pixelated image (70%) and a real image (30%) will probably make it only bizarre looking, not improved.

But it gets worse. The drawing provided here shows that the AF sensor would get its image from the EVF not the real word light source. Let’s say this would remove all usability of phase detection and it would be better to use the direct data from the sensor to do contrast detection as this position would only allow it. So, let’s suppose that this is slight error in the drawing and the AF sensor will receive 70% (or even 30%) of the light going up. This time, there is even less light going directly to the eye (90% from EVF and 10% from outside? What an effort for so little!)

The, the advantage of getting the real image in the viewfinder is to be able to do a fine focus on the focusing screen, but where is it? Certainly not in the usual position in the way up for light reflected from the primary mirror. This would actually reduce light going a little more and add another handicap for the AF sensor.

Let’s say that this is certainly not what Sony has in mind for the Sony A77. And let’s wait for leaked information rather than pure amateur phantasm.

photo credit: lrargerich

This is some sort of silly question, but when people look at ISO sensitivity, we tend to be impressed by the increasingly higher values (which mean that you can take pictures without much light and without adding a flash). But no manufacturer seems not willing to offer us the low ISO values of sensitivity.

Of course, it’s always possible to buy a ND filter (Neutral Density or grey filter). But then, why not integrate this directly in-camera? After all, it’s only a matter of using computer algorithms to help digital sensors not to collect all of the possible light, a thing like taking several pictures in a row but with a very high speed sequence.

The advantage would be to allow the sort of images that many a landscape photographer likes nowadays: You screw a dark ND filter, shoot at very slow speed and get the water movement as in the image here.

So, who’s going to bring that technical innovation to a DSLR camera?

photo credit: anthonycmaki

Sometimes, during Summer time, it’s fine to switch to low gear and to wander far from the normal path of things.

So, if you like lenses (specially long focal length primes), you may be interested in looking at the dinosaur age of tele-lenses: The Telescope era, which started in 1611 (nearly 400 years ago) when the Greek mathematician Giovanni Demisiani coined the word from the Greek τῆλε, tele “far” and σκοπεῖν, skopein “to look or see”; Thus, τηλεσκόπος, teleskopos “far-seeing”).

ArsTechnica has a fine 3-page feature article drawing a history of the telescope, showing how it evolved from the original 2-lens design of Galileo telescope to more modern mirror-based ones. In between the tele-lens derived from that, but it’s another story.

Fun reading. Good lesson ni optics.

As a matter of fact, an online Chinese eCommerce web site published a product page for the Nikon D800. A photo, some features somewhat far from what is expected from Nikon, but after all, why not?

First and foremost, the resolution would (fast) climb to 24 MP, or near to the current maximum for Full Frame digital sensors, indicating a possible use of the Full Frame photo sensor from Sony (already used on Sony Alpha 900 and Alpha 850, but with some video-capture ability added).

Next, we find a shutter with speeds from 30s to 1/8000s for continuous bursts of 7 frames/second, video recording in 720p and 1080p, ISO 200-6400 (ext. to ISO 100-25600) light sensitivity, exposure metering using a 1005-RGB-point sensor and system, Nikon Multi-CAM 3500FX AF 51-point AF system, a fixed LCD screen attached to the back of the camera with a 3″ (7,5 cm) LCD with 921.000 pixels.

Clearly, this would build a relatively easy to predict kind of camera (apart from the high resolution of the sensor which would indicate Nikon is leaving the lonely track of low resolution Full Frame to go and catch up with Canon and Sony where they already succeed).

But this is still surprising. If this is more than an unfounded rumour, it will allow a prediction of the direction Sony is taking in terms of FF sensors and the successor to the Sony Alpha 900 and Alpha 850 would keep their resolution at 24 MP while video capture would be added like on the Nikon D800.

Cross-section view of SLR system:
1 – Front-mount lens (4-element Tessar design)
2 – Reflex mirror at 45-degree angle
3 – Focal plane shutter
4 – Film or sensor
5 – Focusing screen or glass
6 – Condenser lens
7 – Optical glass pentaprism (or pentamirror)
8 – Eyepiece

Origin: Wikipedia

The traditional SLR camera

Let’s start with the organization of the most common Single Lens Reflex (SLR) camera, as we generally know it. On the cross-section view here, we can see the light trajectory (in yellow) when the mirror is in lower position to direct light (and the image) toward the viewfinder. At exposure time (when you press the shutter release button), the mirror moves up to let the light go straight to the sensor.

Very efficient, this configuration still has some drawbacks which have long been considered minor, but still very real.

First, during exposure, the viewfinder is totally black. It’s not very long, but the inconvenience is very observable by the user.

The mechanical design needed to move the mirror up and down is complex, fragile, but must operate very quickly to maintain a fast shooting cadence. On pro photo cameras, these mechanisms become complex and expensive to reach high frame rates. The technology progresses fast, but this is only in the most recent years that camera manufacturers have been able to provide more than 3 frames per second on standard cameras. Some pro SLRs (like the Nikon D3, for example) reach 8 fps (but the price falls in the financial investment category!)

Pellix – transparent or nearly transparent

In the 1960’s, several cameras started to use a semi-transparent mirror to avoid moving it. The idea is merely to ensure that most of the light goes straight to the film or the image sensor, but a smaller portion of it is reflected up to the viewfinder. Unsurprisingly, there are both advantages and drawbacks.

First pro: viewing is never obstructed. The photographer always keeps the camera at eye level and shoots nearly continuously (highly appreciated in sports photography, for example, to track unpredictable action).

Second pro: The removed mechanics allow to simplify the camera structure (and its cost). Light weight and low noise come to these cameras.

But the light splitting is difficult and the technology for semi-transparent mirrors is not easy to master: You want to keep transparency without the slightest color-shift (coming from the uneven transmission of all wavelengths), and without too much light loss (to have a clear and readable viewfinder, you must reflex as much light as possible, but this would be detrimental to the image exposure on the film).

The gains are perceptible but limited in front of the technical drawbacks: The market will not be convinced and, apart from random and rare occurrences like the Canon Pellix QL, this will be the end of it.

Autofocus & video

Then, come the 2000’s and point-and-shoot compact digital cameras start boasting about their video capture features. Always in search of technical innovations, camera manufacturers are happy to go from photo to video. It’s relatively easy on the a compact point-and-shoot: Either it has a separate telemetric viewfinder (no constraint), or the photo sensor is already used for LiveView (the only problem is that it engulf energy and heat the sensor up).

If you want to focus, the digital point-and-shoot cameras have a simple solution: Analyze the image on the sensor to see if it is sharp or not (this is called contrast detection). This seems simple, but it calls for a lot of computation power (this is slow) and you have to nearly randomly try various focus settings to find the best one.

On a cheap camera (or a cheaper camera that most Single Lens Reflex cameras), these drawbacks are easily forgotten and, moreover, the same approach works for video as well as for photography. So, the photo compact cameras have simple path to video capture.

An SLR with video

In 2008, Nikon decides to add video capture on a digital SLR (the Nikon D90). This should be marvelous! But the reflex architecture is totally different and there comes trouble.

First, when the mirror is up, it is impossible to use the specialized AF sensor that is present in the camera in front of the mirror, not behind (this sensor is using a method known as phase detection). The solution would appear to be easy: Go the point-and-shoot way, with contrast detection. But three problems rush in:

• SLR sensors are big and not very compatible with continuous heavy computation.
• SLR sensors are very large, favoring a small depth of field and are thus infinitely less tolerant than smaller/cheaper sensors (as a matter of fact, focusing on point-and-shoot cameras stay very imprecise, but it is efficiently hidden behind an enormous depth of field: “Everything always looks sharp”).
• The user is used to ultra-fast focusing given by phase detection, but will be hit with an enormous difference in reaction times.

The first SLR photo cameras choose to run around this problem by removing completely autofocus and most of the automated controls. Nothing elegant, nothing shiny here. But this is utterly pragmatic. And some pretty impressive cameras like the Canon EOS 5D MkII are equipped with very limited video features.

Sony Alpha SLT-A55

One exception in this landscape: Sony. Known as a historical leader of professional video, Sony cannot and will not deliver some half-baked video solution, even on a photo camera. They will only bring video capture when it is fully operational.

2010 is for Sony the year of solving this problem with a semi-transparent mirror again: Without mirror, all the troubles are gone! You can do autofocus while filming, you can keep nearly all the automated controls. The Sony Alpha 33 and Alpha 55 are entry-level photo cameras but their video capture is not relegated to the end of the feature list. Focusing is absolutely continuous during video filming (see the video demonstrations) or during continuous shooting.

Even better, on cameras priced at a few hundreds of Euros, it becomes possible to rush continuous shooting at incredible frame rates (at least out of the realm of expert and pro cameras): 6 frames per second! Or even 10 fps if you accept some constraints on available automatisms! This is comparable to what is offered by pro cameras at 10 times the price.

All this with a viewfinder permanently available. The clearer viewfinder you have, the more comfort you get.

And, if there is no longer any mirror moves, there is no longer any of the associated noises. Since the mirror mechanics generated very distinctive clangs just before and after the exposure, the cameras with a fixed mirror and remarkably silent.

Is it the end of it?

This is a good question. We understand that, now, the target is set very high. Canon, Nikon and the others competitors will be compared to a Single Lens Reflex (SLR) camera sold under 600€. Innovation is now king in the Japanese (or not) Research and Development labs.

2011 will be a very interesting year for Sony’s opponents. Will they use the same recipe or will they bring some other bright idea to bring photo and video together in D-SLR cameras?