Post has attachment
Snowflake-a-Day No. 52
With blotches of colour reminiscent of a Picasso painting, this miniature snowflake feels like a work of art! It’s amazing what a few simple variables and a single ingredient can create in our atmosphere. :)

I’m forever impressed with the effects of thin film interference in snowflakes. For a primer on this, check out these pages of my book Sky Crystals: (you can buy the full book here, if you’re curious: ). One thing that I’ve continued to notice about the majority of these central bubbles however, is their shape.

Often vaguely trapezoidal, they appear in the exact same location that “dimples”, or small indentations, appear. These will fill in as the snowflake grows outward, to a standard thickness making the snowflake uniform. Here’s an example of these dimples from last year’s series. They would eventually fill it to create a smooth flat top surface:

But something else can happen. It’s my working theory based on conversations with Ken Libbrecht, physicist at CalTech, that these dimples can grow a ceiling rather than completely fill in. If the edges become very steep like a canyon wall, these edges might attract new water vapour as building blocks and cause the top edge of the dimple to start growing over the remaining surface depression. This would explain the very common shapes for these central interference colours.

This snowflake was another tough edit, with fibers from the black mitten it was photographed on appearing directly in front of the snowflake. You can shoot through them like shooting through a chain-link fence with a wide aperture, but there is still a bit of softness remaining in the lower parts of the image as a result. There was also a curious amount of debris on the snowflake, like tiny little bits of dirt. This is common when I don’t adequately clean off my tiny artist’s paintbrush before trying to manipulate a snowflake. My bad! Still, this little gem was captivating enough to become a part of this year’s series.

I’m hopeful that all of the snowflakes I share between now and the end of the series will have been shot on the pre-production Lumix S1 I was field testing – this is one of those at extreme magnification of 12:1, and then significantly cropped in to fill the frame with a crystal just under one millimeter in diameter. On such a small scale, completely imperceptible to our own eyes, there is a universe of detail that can be captured. It’s all just water and some rules for how it fits together!
Add a comment...

Post has attachment
Crystal Coalescence
Just something about a fiery orange freezing soap bubble that I have found appealing this winter, and why not combine crystal with crystal?

This is created using a large fluorite crystal specimen that I purchased for experiments like this. The freezing bubble was placed in an opening around the fluorite, and lit from behind with a filtered LED flashlight to give off an orange glow. If that was the only light I used here, we wouldn’t have anything in the foreground illuminated; there is a second LED light off to the left that is adding all of the foreground light to the image.

The lighting angles need to be carefully adjusted. The main light behind the bubble is sent through a sheet magnifier / Fresnel lens which helps keep it tight and narrow, refocused just on the bubble itself with very little spill-off to the surroundings. The white light from the left could be broader, so long as it didn’t hot the bubble significantly. It did slightly, you can see that in the upper left where the bubble and fluorite mix, but this was difficult to contain in a very dynamic environment. Remember that these bubbles freeze solid in 10 seconds or less sometimes!

This is a focus stacked image, which would largely be impossible for a freezing bubble image since the crystal structure grows so rapidly. The second frame, however, was for the more static foreground. I manually combined the two images together so that there was no confusion around the changing crystals in the bubble. It’s tough to do focus stacks with dynamic subjects, but I’ve done them on occasion – once with a bee in flight! In every case they require fully manual attention in editing.

Shot on the pre-production Lumix S1 I had for a few weeks, it was a joy to shoot with and the results all so crisp. Since there are no extension tubes for the L-mount as of yet, I adapted one of my Canon EF lenses to shoot this. A larger scale than my MP-E 65mm lens would allow for, this was done with the Meyer Optik Trioplan 100. Ironically the lens is known for its “soap bubble bokeh” when shooting wide open, but it’s a tack sharp lens when stopped down. I’ve always found this remarkable, since the original optical formula was devised back in 1916. Of course this is a newer version with the latest optical coatings (from the now bankrupt netSE that was using the Meyer Optik name), but I find it fun to think that the fundamentals of photography from a century ago still play into our images today.

On par for sharpness with my 1DX mark II, probably better if I had to pixel peep, and I haven’t even had the ability to dig through any RAW file taken with this camera yet. Even with a lot of extremes in terms of brightness, everything fell within exactly the right dynamic range here. I can talk gear all day, but I also want to remind folks that this is a very temporary sculpture, a fleeting subject that will never be the same twice in any moment in time. That is the nature of this kind of photography and the reason why I stay passionate about sharing things like this.

If you didn’t pick up my bubble recipe from previous posts, it is:
6 parts water, 2 parts dish soap, 1 part white corn syrup. All blown through a plastic drinking straw and placed where needed!
Add a comment...

Post has attachment
Snowflake-a-Day No. 51
This snowflake is simple, but special and technically complicated. It’s one of the few nearly purely “stellar” crystals I have photographed without side-branches, and there’s a good reason for that.

The idiom “six of one, half dozen of the other” is put to the test here. The reason why this snowflake has such small, almost non-existent side-branches is because it was originally growing as a twelve-sided snowflake – this is half of the dozen, but incomparable to six that would have been growing on their own. Why photograph just half of the twelve? Simple reason: attempting to move/reposition the snowflake broke the two twins apart.

Not “true” crystal twins by the molecular meanings, but smaller snowflake can sometimes glue themselves together at an almost-perfect 30-degree angle, giving each branch normally spaced out at 60 degrees the opportunity to grow, albeit in close quarters. With less available water vapour, the branches all favoured outward growth from the center, little time spent on filling in a structure that was in direct competition from a second partnering snowflake. A side-branch event might have occurred, which is usually a momentary blip in increased humidity, but those side-branches couldn’t get very far.

This was a very difficult snowflake to edit, which might be surprising based on the relatively simple structure. Here’s the problem: it was windy.

Higher wind and lower ground-level humidity can “erode” a snowflake quickly, sublimating it from a solid to a gas within minutes. Gone forever. Because there are very few side-branches, the wind could easily get all the way around every branch right down to the center. Usually a snowflake shows some level of deterioration on the outer edges between the first image taken for focus stacking and the last, but effectively every edge is an outer edge here, with so much room for air to flow around the crystal. This poses a problem for focus stacking: Even if the angle is the same, multiple passes across the crystal, or a slight time delay, will result in one image having a smaller footprint than the adjacent one.

All of these images, taken on a pre-production Lumix S1 with 9FPS – the same or more than I would have used on other cameras – had different outer edges which REALLY mucked up ever focus stacking algorithm I threw at it. In the end, I had to do some extensive re-aligning in Photoshop to get things to fit together properly. Another type of “snowflake surgery” I suppose!

All said, I don’t think this snowflake will win any awards, but it IS unique in both form and challenge to present it to you. Just imagine twice as many branches, and your imagination might make this little gem whole again. :)

Want more science and photographic help with snowflakes? The eBook of Sky Crystals will be perpetually available even though the hardcover version is sold out:
Add a comment...

Post has attachment
Snowflake-a-Day No. 50
Fifty images in this season of snowflakes so far! And with all that, we haven’t really seen a good trigonal snowflake until now. Let’s fix that!

Amidst an array of small colourful plates and broad columns, I occasionally find a snowflake where the growth on even and odd sides is synced, but not together. This three-fold symmetry is puzzling, but it does have an answer – at least in theory! Because these snowflakes are so uncommon, it’s hard to practically test the ideas that are described in this paper from Ken Libbrecht, physicist at CalTech: (you’d have to be REALLY geeky to read that)

The basic assertion is that the aerodynamic qualities of a snowflake will affect its growth. If a snowflake faces the same direction without tumbling about, certain facets will grow faster than others. If the crystal is “stable” in the air, it might turn out to be something like this. Because the air is so chaotic and a snowflake isn’t generally designed to be as aerodynamic as an air plane, these are fairly uncommon with this extreme degree of disparity between side lengths. I’ve seen much more common occurrences of snowflakes with slight differences that create three-fold symmetry instead of six-fold, but they could hardly be called “triangle flakes”.

Since many of the snowflakes from this snowfall, photographed with a pre-production Lumix S1, were able to create thin film interference colours, I should have been surprised that this triangular snowflake had the same – the icing on the cake! If you want more info on this type of strange colour, here are the pages from my book Sky Crystals that describe it:

This crystal also shows us how inward crystal growth “smooths out” the corners. There are two inset rounded triangles that you can see intersecting with the colour areas – fainter in contrast because they are on the reverse side of the snowflake. If the crystal goes through an event where the outer edge thickens, it will grow inward from this thicker edge and as it grows inward it becomes rounder. Hexagon centers will readily create circular shapes, but a trigonal snowflake would take a lot longer to reach a circle.

These orange colours are far less frequent with thin film interference, as there are fewer distances of separation that can create them. Hard to say exactly what the thickness of the bubbles were, or the ice on either side, but it’s a welcome sight. As if a Vick’s cough drop became a snowflake, this little gem is a welcome remedy for the frustrations of a cold winter day.

Also, I might not have mentioned it… but the hardcover of Sky Crystals is officially SOLD OUT! 3000 copies sold direct to customers without retail presence, I’m humbled and honoured. I may make a sequel in future years, but there is an eBook version available that has the entire photographic workflow used to create these images here:
Add a comment...

Post has attachment
Snowflake-a-Day No. 49

Hey Google+ People! Before reading this post, here are all the places you can find me online once G+ shuts down:



This snowflake might feel a little more metallic than most, due to the way shadows play across the surface. This little crystal has much higher relief than most!

Does anyone else see the Elephants, each with five trunks? Those trunks, or ridges along the main spines of a branch, often grow higher than the surrounding crystal. As you can see from the shadows they cast, I bet they’re at least twice as thick as their underlying plate! The also “cascade” their height to lower planes as they grow further towards the branch tips, with at least transitions in height noticeable on most – if you know what to look for.

If these spines (or “trunks” of these elephants) grow thick enough, something else interesting happens. More noticeable on the shorter ones furthest from the tip you’ll see them growing to be broader, as an I-beam like structure begins to form. The crystal starts to grow outward, creating a new top plane. If given enough time, this growth pattern could create an entirely new sheet of ice on top – we’d call that a “skeletal form” snowflake, quite rare. This little crystal was about to become one of those if it stayed in the clouds for longer.

The “elephant ears” appear more uniformly bright, as they have no surface details. Further up the branch towards the middle, we have a much more textured area with plenty of dimples and divots that bend the surface of the ice and create shadows. Fun fact: If I were to flip this snowflake over, those areas of high texture would be completely smooth. For reasons beyond my understanding (though I’m certain this is known to physicists or mineralogists), one side of the snowflake will always be smooth while the other has surface details, otherwise both surfaces are smooth. This can flip-flop as the crystal grows to create some very complex designs.

This is another snowflake shot on a pre-production Lumix S1 – every image I’ve edited so far has come out with greater detail than any other camera I have used to date. Panasonic doesn’t make a compatible ring flash however, so I’ve taken to using the K&F Concept KF-150 ring flash with a Bolt P12 external battery path to keep up with 9FPS shooting. Of course these flashes are intended for other TTL systems so it needs to be operated in manual mode only, but that’s not a drawback. In fact, it helps with exposure consistency which is important when focus stacking. 36 shots combined to get this one in focus from tip to tip!

More contrasty and metal-looking than most, but I think that it helps this snowflake feel more “tangible” than others. It’s almost like I could run my hand along the surface and feel all the differences in thickness from the ridges and ribs. Another unique gem for the collection!
Add a comment...

Post has attachment
Snowflake-a-Day No. 48
In all my years, in my many thousands of hours working with snowflakes, I’ve never seen anything quite like this before. That’s a solid statement – and one worth spending more words on!

The previous few snowflakes ( and ) have included elements of fusing together that creates air-gap thicknesses that can produce thin film interference effects. I’ve described this many times before, so I’ll just state that it’s the same physics as rainbows in soap bubbles… but the way it behaves here is very strange. Even more strange, is the way these snowflakes are connected.

There are really two separate crystals here forming into a single “snowflake”. I suppose the term snowflake can always mean more than one crystal, but I define it as either a single crystal, or multiple crystals that cannot be separated without destroying their collective identity. These two are intrinsically linked. There is a main “base” snowflake, the largest footprint, but then there is a smaller branched snowflake that somehow, some way, collided and “stuck” to the larger one. This part, aside from winning the cosmic lottery, I cannot explain. But this attached snowflake is very unique in its own rights!

Column-type crystals will often grow into plates on either side, running parallel and competing for available water vapour (“building blocks”) for their continued growth. When they are close, the competition is fierce and when one side gains an advantage, the other side stays at roughly the same smaller footprint it was when it’s parallel partner started growing out further. If the column is longer, however, the two plates are free to grow together without much hindrance, like we see here:

But that’s not the case here. We have very separated growth which should result in more even development of the two plates. Based on the fusing of the branches from the larger plate, it collided with the larger crystal in this form, or something very similar… and that I can’t explain either. Other than winning another cosmic lottery in the same image. Like getting attacked by a shark, and then surviving only because lightning strikes the shark. (I know, you'd probably die in that scenario too... but maybe you win the cosmic lottery and still survive!)

Then there is that same odd bubble tendril pattern that we saw on one of the earlier snowflakes linked above. I can’t identify the exact physics to explain why, but the “how” is starting to become obvious: when enclosed bubbles internally sublimate, with molecules flying from one side to another, they get more spherical. They leave behind traces though, that slowly become thin lines and eventually spherical bubbles on their own. The magic of natural physics at work that I wish I could explain!

Shot on the pre-production Lumix S1 I had on loan to me for a while, and one of my favourites from that camera not only for the mystery it provides, but for the level of clarity I am now growing very fond of.
Add a comment...

Post has attachment
Photo Geek Weekly Episode 56 is out!

On this episode of Photo Geek Weekly, fellow photo geek +Shiv Verma jumps into the conversation with a wise opinion on every topic: photo contests and image theft, Lumix Pro services now available, Kodak Alaris and their search for funds / buyers / investors, and the shuttering of one of my favourite magazines, Outdoor Photography Canada. All this and more - thanks for listening!
Add a comment...

Post has attachment
Snowflake-a-Day No. 47
This little snowflake had a lot going for it – and it also filled the frame at 12:1 magnification on a pre-production Lumix S1. Twelve times closer than the average macro lens gets! How much detail was there?

First let’s discuss how I achieve such magnifications. The lens is the Canon MP-E 65mm F/2.8 1x-5x macro lens, which be default can get me up to 5:1 magnification. If I add a set of extension tubes, that pushes a little farther to 6:1. The final ingredient is an extra piece of optics, the Canon Lifesize Converter EF which was designed to only work with the Canon 50mm F/2.5 compact macro that was, on its own, able to achieve 1:2 magnification (half life-size). Since it was effectively a teleconverter for macro magnification, and no one was stopping me from seeing if it would work to double my previous 6:1 limit. It sure did!

Here’s the setup:
Lumix S1 > Novoflex EF-to-L adapter > Lifesize Converter > Extension tubes > MP-E 65mm lens > ring flash

If you already have a teleconverter for general photography, you can often adapt it to a macro lens. They don’t normally couple due to the protruding element on a standard teleconverter, but certain types of extension tubes can fit between the TC and the lens to cover this and allow for proper coupling. You might even get more magnification that I do!... but that doesn’t mean more resolution, since you’d be working with limitations due to diffraction that soften things anyhow.

The quality in each focus slice (35 images used here) is potentially better than the Canon 1DX mark II, especially in terms of highlight details. See the colourful lines at the center of each branch tip? These very easily become over-exposed and turn just white. Even when dealing with JPG data here, I have more information than before.

As for interesting features of this snowflake, you’ll notice a smaller one fused and creating thin film interference colours in the upper left, just like the previous snowflake in this series ( ). It’s common for these types of features to be seen on snowflakes formed in the same snowfall, but these two snowflakes were photographed nine days apart. This makes them both rather unique! The flower-like center, strong symmetry, and broad branches all make for a beauty of snowflake.

The reason why the lower right branch is a little less symmetrical, I’d wager, is because another snowflake was stuck there during the final phase of growth that produced the ends of each branch. If something is missing, it’s probably because something else was in its place!
Add a comment...

Post has attachment
Snowflake-a-Day No. 46
Okay, this one requires a double-take and some curious thinking about what exactly we’re seeing. I think these two snowflakes are making their own artwork!

Here we have two plate snowflakes that have fused together. In the area that they have fused, some crazy bubble patterns and rainbow colours have emerged. It’s very important to make a distinction here: each snowflake has their own set of pseudo-symmetrical patterns, and these weird shapes and colours have formed independently of the structures in each crystal. Science! Can we figure this one out?

I’ve seen bubble patterns similar to this on other snowflakes before, including number 39 in this year’s series: . The patterns are remarkably similar, but formed in entirely different circumstances, which has me a bit puzzled. In this new image, we don’t see the funky lines at the bottom where the larger snowflake overlaps the smaller one – could there be air there? What would the air access provide to prevent this from happening?

My best guess, and it’s a total shot in the dark here: This bubble (separation between the two plates) is shrinking. As water molecules sublimate and re-attach themselves to other parts of the crystal surface, they would in turn force a certain “roundness”, just as inward crystal growth creates circular shapes, we’d have spheres. This also could attest for the smooth transition of colours in the center, something that we rarely see in optical interference from snowflakes – it’s rounding out. So this large separation is becoming smaller and smaller, but what of all the little tendrils? Why don’t they continue to close up like the rest?

Well, maybe they are. You’ll notice that a lot of the bubbles on the right side, the further away from the center the smaller they get. What if these tendrils start to close off and further compartmentalize themselves until all that is left is a series of tiny independent bubbles? We can see some of this compartmentalization on the left edge pretty clearly. This doesn’t explain why the tendrils form in the first place, but it might explain their evolution.

It might also explain the center in snowflake 39 from this series: a large internal bubble collapsing in on itself to become rounder, but mysteriously leaving neuron-like appendages around the edges. I think I might be starting to decode this one!

Also, shot on the pre-production Lumix S1 I had for a while! Absolutely love how that camera, even with JPGs (can’t edit RAW files yet), handles subjects like this. While I love micro four thirds for most macro work, the larger sensor area of a full frame camera helps defeat diffraction a little better. Having used full frame cameras since the beginning of my career, and I know them well, this is the best one I’ve used. Yes, Panasonic is a sponsor of mine, but I come by my opinions honestly – really excited to see what the full production cameras are capable of!
Add a comment...

Post has attachment
Snowflake-a-Day No. 45
This snowflake is, by percentage of surface area, the most colourful snowflake I have ever encountered. It’s almost hard to imagine that THIS is a snowflake!

It’s quite tiny, less than a millimeter across and photographed at 12:1 magnification. Even at this extreme magnification it was filling less than a third of my image sensor horizontally, we we’re pushing past the limits that diffraction imposes on resolution… it’s about as “pushing limits” as it gets!

If you’re wondering where these seemingly magical colours are coming from, check out these pages of Sky Crystals that describe it very well – thin film interference:

There are a number of things that affect resolution in photography: the optics of the lens, the size of the aperture, and the quality/number of photosites on the sensor (they aren’t “pixels” yet, which happens when the RAW data is de-mosaiced… we’ll save that rabbit hole for another time). Even with high resolution sensors and good quality optics, a very small aperture can cause a degradation in quality… light plays nice in certain scenarios!

To properly understand this, we need to figure out what the “effective aperture” is for your lens, which changes based on how much magnification you have. For example, if you’re at 1:1 magnification, which is the closest a normal macro lens can get, and you set your aperture to F/16, you’re likely shooting at F/22 as an effective aperture. It’s a simple equation for a rough estimate: for every magnification factor you have, add one stop. If I was at F/5.6 and shooting at 5:1 magnification, adding five stops brings me to an effective aperture of F/32… and this is where diffraction starts to rear its ugly head.

The smaller your aperture, the more light bends off course when it passes through it. If it bends too far, some of that light ends up hitting neighbouring photosites and effectively is colouring outside the lines. So, if I’m setting my Canon MP-E 65mm lens wide open to F/2.8, adding in extension tubes to get to 6:1 magnification and the Canon Lifesize Converter EF to double that to 12:1 magnification… we have a problem. Adding twelve stops to 2.8 brings me to F/180! Now, that’s not the true effective aperture, as there is a more complex formula that includes the pupil distance in the lens as part of the equation, but it’s a big enough number for me to realize that there is no possible way to utilize the true resolution of any modern camera. Unless you use special microscope objectives or leave light behind entirely and start using electrons to see your subject, there’s only so much you can do!

Shot on the pre-production Lumix S1 I was using for a few weeks, this camera handled as good or better than the Canon 1DX mark II I had used for years. There’s no real resolution boost from 20MP to 24MP due to diffraction (and the S1R wouldn’t have had any advantage here at 47MP, as the sensor is not the bottleneck), but diffraction aside the camera gave me the best results I could possibly hope for.

At the time of this writing, there is ONE copy of Sky Crystals left: - who’s going to take that home??
Add a comment...
Wait while more posts are being loaded