Saturday, February 8, 2014

Watches in Water - Title of Irony

What time is it? Time to drop my watch in a measuring cup of course!

I did this kind of photo last summer with dropping ice cubes into water with food coloring. The results turned out really cool. I don't think tonight turned out quite as good, but then again I had to end early to take care of the liter of water on the floor...

I am happy with tonights photo, namely how the timing turned. All the photos I did this summer didn't have to be to precise as far as timing goes. I had maybe 1/3 of a second to decide when to trigger the shutter. That doesn't sound like much time, but you'd be amazed at how fast you're brain can process information. The photos were more to capture the water splash after the impact, so I had a much, much bigger window.

Tonight was much different. My vision was to get the exact moment of the watch hitting the water. I estimated I would have a window of less than 1/200 of a second to trigger the shutter. So basically it'd be all luck. Even when shooting video at 60 frames per second, the gap between frames is large enough to miss the moment of impact.

WARNING: Tech talk right here...

So, let's talk about the timing a bit more. The moment I wanted to capture was a single moment, I didn't want motion blur. So I had to use strobes to freeze the action as the ambient light was too dim (plus, ambient light usually sucks in dorm rooms). To freeze the watch, I needed a "shutter speed" ( t.5 time...or is it t.3 time? always forget. discussion for another day) of at least 1/300 of a second. Does this mean I had 1/300 of a second window to trigger the shutter? No, not at all. I did some tests with motion blur, and discover that I had a window of about 1/200 of a second to capture the image I wanted. That's .005% of a second. Even shooting a video at 200 hundred frames per second, you could theoretically miss the decisive moment. At 200 FPS, each frame has 5 milliseconds to expose, and you have to at least have a shutter speed of 1/200. As I just said, 1/200 isn't a short enough time to freeze the falling watch, you need at least 1/300. There for, to put it into perspective, you would have to shoot at least 300 frames per second to have any chance at capturing the right moment. That gives you a window of .003% of a second, or 3 milliseconds.

But that's talking about shooting video right? How does that apply to still images. Well, we've established that the decisive moment occurs in less than 1/300 of a second. Thus, we have to have a exposure time of less than that to freeze the moment we are looking for. This is really subjective based on the speed of your subject and what the length of time of your decisive moment is.

The decisive moment for the watch hitting the water occupies about 5 millimeters of linear space. That translates to 1/300 of exposure time, as I just said. So, if we want to capture every moment in the decisive moment we have to expose for 1/300 of a second or less.

And yes, I think the deceive moment is an infinite number of moments within a set interval of time, not a single point in time.

But we don't want to capture every moment in the deceive moment, at least in this case. The reason being is that there would be too much motion blur. Again, this is subjective based on the interval of the decisive moment, speed of the subject, amount of blur desired, relative size of subject to linear distance of the decisive moment, lens chose, distance from lens to subject, etc... It's a case by case basis...all determined by your brain in a split second. It's seriously amazing how well your brain knows physics without even trying. Everything I've written so far is calculated in the brain in a percentage of a millisecond.

But back to the problem at hand. So how much motion blur did I want? Well, not enough for the viewer to notice. Based on experience, that would require a exposure time of about 1/350 or 1/400 of a second. I could achieve this with my strobes by using a really low power setting. A low power setting means a lower t.1 time.

Quick explanation. The t.1 time of a strobe is the time it takes for the strobe to drop below 90% of it's peak power. Basically, that's how long it take the strobe to "flash." So if my t.1 time was around 1/400 of a second, I'd be golden. Even if my t.5 was around there (time to drop below 50%), I'd be OK.

I figured out that 1/32 power gives me the t.1 times I need. It's not perfect, but it's really close. There's not enough blur to detect except upon close inspection.

To recap, I had about 1/300 of a second to trip the shutter. Even I, with years of practice at this, wasn't that good. Capturing the moment I did was a combination of luck and gut instinct. I had to account for the time it takes for my finger to move, and for my brain to process and tell my finger to move. I estimate it takes me about 150ms to make the decision, and another 50ms for my finger to move. Therefor, I had to take the photo 200ms earlier than I though I needed too. The rest was totally luck and gut.

Turned out though.

My word I haven't even talked about set up yet...I've been at this for 3 hours now...

Okay so set up. I had my power setting right, now to balance that with my ambient. I did want some ghosting going on, as well as enough ambient to get a nice glow in. A shutter of 1/15 of a second did the trick with that. An aperture of f5.6 balanced depth of field with the output of the strobes and the ambient. I had to use a higher power setting on the strobe, and thus longer t.1 time, but the sacrifice was necessary to get the motion blur and depth of field I wanted. And I used ISO200 btw because ISO100 wasn't nearly bright enough.

Here's the set up....


Pretty self explanatory. Two flashes to freeze the action and add some color. You can see they are both gelled. I also had two pieces of white computer paper over them to diffuse the light.

I'm guessing after all that you'd probably like to see the photo...



You can see the ghosting caused by the lower shutter speed, but then you can easily see the frozen action from the strobes. I think it worked out pretty perfect. I quite like it. There's some really cool color in the glass too, that green gel did wonders.

In Lr, I first raised the exposure and clarity to make it brighter as it was a little underexposed. Shadows came up a bit too. I added a vignette a bit too, just for drama. The colors weren't quite right though, so I moved to Ps.

First thing I did was add a lens flare. Hehehe I know it's evil but who cares? That added a cool effect, but the image was too...yellow. A couple cooling filters later, duplicated over the sprays of water, and it was much better. Back into Lr.

The greens still weren't popping enough. So I changed the aquas, yellows, and greens to be more...well..green. Did I mention my favorite color is green?

Did I mention it's now hour four of writing? Please excuse the sudden drop of quality...

I think that's it for tonight. I've run out of things to talk about...except for this.

To gamers complaining about frame rate...read this.

Isn't it ironic that the photo is a watch and I talked about time basically the whole time? Haha pardon my pun.

See you tomorrow.

No comments:

Post a Comment