All posts by Biolapse

Botanical Time-lapse tips. Part 1) Dealing with Flicker

Hello folks!

I hope you have all been well! It has been a little while since I put up a blog post. Lack of time would be the reason. Summer is coming, nurseries are selling more plants which are getting ready to bloom, and there is just a lot of shooting to do, and I am not the only one doing this stuff either. It seems to have been a… blossoming (hahaha) of other time lapse photographers that are giving a hand at botanical timelapse photography! I am very excited about this too! I love seeing how other people go about things and I love watching their results.

I have a lot of information buried within this blog, and figured maybe it would be nice to put out a Tips & Tricks blog for the aspiring botanical time lapse photographers. I have been focusing on plants now for 4 years, I have learned a lot, made many many many mistakes, some of them pretty costly, and if I can help others out to save them time, money, and energy, well, that’s what I love to do.

 

Avoiding Flicker

The most common issue I see with botanical time lapse is flicker. Many of the photographers experimenting with this seem to have flicker to some degree, and there is just no need for it. None of mine have any flicker whatsoever since 2014, I never use de-flickering software, and I am happy to share my secret.

It took me a while to figure it out, but the flicker can be a combination of several things. Bulbs not warming up, exposure inconsistencies, lack of control of light in a room. This is the most common thing I have seen on many of the new botanical time lapses I have seen.  So lets just go ahead and crush this one now and you will NEVER have annoying flicker in your time-lapses again.

Lighting

I only use LED lighting. I have tried many forms of light.  Traditional bulbs. Florescent. Strobes. Speed-lights. None of these in my view are very good.

  • Bulbs work “OK” However as they warm up I sometimes run into color shifting which can cause some white balance flicker. if the bulb is on NON-STOP it works pretty well, but if you have it on steady through the day then have it turning on and off during the night cycle, you will probably notice a slight hue shift during the evening. The on-off switching can take its toll as well, and there may be some overall dimming. With bulbs its best just to turn it on and keep it on the entire time, this is good with shorter timelapse sequences, but shooting much past a few days the plants will become distressed.
  • Florescent. Sort of the same as bulbs, but worse if you run them constant for a day cycle then turn them on and off for night. There will be some very heavy issues with exposure as it takes several minutes for them to get to max brightness.  Works OK if you are just doing a few days of shooting.
  • Studio Strobes. When I first started I was using some Elinchrom studio strobes. At first I actually had pretty good results, but after a while the bulbs in the strobes started to be less and less consistent as they aged.  I am not sure how may shots you should be able to get with a tube, but a week of shooting will rack up hundreds of discharges. If I remember correctly, after a few thousand discharges I started getting quite a bit of flicker.
  • Speed-lights, same issues as with the Studio Strobes but you also have to keep replacing the batteries. No good for this work.
  • LED. Yes. This is it. This is what you want. LED Lights. Nothing beats them for this type of work. There is no warm up time, they produce very constant and even lighting. you CAN still get flicker if you are shooting with a fast shutter speed and using PWM. But this is easy to overcome. You can turn LED on and off as many times as you want, the output is always the same.  I only use LED period. If you look at my work over the last few years, there is ZERO flicker, and much of this has to do with the LED lighting.  And the best part is the LED lighting does NOT need to be expensive, and high power LED is useless for the time lapse portion (but it is beneficial for macro video work). So there is really no need to go run out and spend thousands of dollars on LED panels. I use 2 cheap ones off amazon and cost me under $100 each, and even then I spend more than I probably needed.. Look for something dim-able (pwm is fine unless you are doing video).

Shutter Speeds. 

Woah there buddy. Slow down. The plants are barely moving. Fast shutter speeds are your ENEMY.  Let me explain why. Shutter speed is usually pretty constant, however there can be very minor inconsistencies on the speed of the shutter.  I am going to throw some numbers out just to demonstrate why, but these numbers should be taken with a grain of salt as they are probably inaccurate in reality, but accurate for the example.

1/1000th of a second is a single Millisecond.  Lets say your camera shutter has a variance of 1 millisecond.  When shooting with speeds of 1/10th of a second, that means your shutter speed would be anywhere between 99 to 101 miliseconds. So you have a variation of approximately 2%.

If you take an image, and make a 2% exposure adjustment in photoshop, you can see the difference between the two images. This is what causes shutter speed Flicker.

Now lets move that shutter speed to 1 second.  That means with the 1 millisecond variance your shutter speed is 999 to 1001 miliseconds, meaning you have a variation of approximately 0.2% instead of 2%.  That 0.2% exposure adjustment is far more difficult to see than the 2%.

Of course, camera shutter speed variances are far less than 1 millisecond, as most cameras can shoot 1/4000th of a second and faster with pretty consistent looking results.

PWM Flicker

Lets say you are using LED lighting. It may be a voltage regulator adjusting the brightness, but many use PWM to adjust brightness, which essentially is just making the light work like a strobe so fast that it seems solid. This shows up on video quite often as rolling lines of lighter and darker areas due to the way the image sensor pulls its data.

Lets say you have your LED’s nice and dim, 10% and 20% power.  What you see below, the x shows when the LED is off, the 1 shows when it is on.  10% has half the number of ON as the 20% making it 1/2 the power.  100% means it is steady on.

10% = xxxxxxxxx1xxxxxxxx1xxxxxxxx1xxxxxxx1
20% = xxxx1xxxx1xxxx1xxxx1xxxx1xxxx1xxxx1xxxx1
100% = 11111111111111111111111111111111111

If the LED cycles 100 times, at 10% during a 1/100th shutter speed, you will get variance as sometimes you will get the LED on 99, or 100 times during that exposure, it depends where in cycle when the exposure starts. This gives a 1% variance that can potentially show up as flicker depending on how many times the LED burst while the shutter was open.

if you have a 1 second exposure, the LED would pulse on about 9,999 to 10,000 times during that 1 second, meaning the variance is substantially lower, meaning a far smoother output with virtually no visible flicker.

(once again, numbers for illustrative purposes only)

Aperture. 

I shoot non native lenses. So the aperture is locked in place the whole time,  but just light shutter speed on native lenses there WILL be some aperture slop that will cause flicker. This is well known within the landscape timelapse community, and many people use Nikon Lenses adapted to whatever body you are using.

Light pollution Flicker

This one you have to get a handle on, even if just shooting for an afternoon. My recommendation is a windowless room, or blackout curtains. Any light from the outside can mess things up due to clouds, reflections, sunny morning and cloudy day, etc.

if you are unable to get some space with complete light control you might want to look into shooting within a grow tent, they provide excellent light control to prevent light contamination from the outside world.

If that is not an option, you might just have to look for a day where the forecast shows full sun and keep the shoot to mid-day. there are quite a few flowers you can find that will bloom in less than an hour such as daffodils.

Easy breezy

In the end, the recipe to 100% completely eliminate flicker is simple. LED lighting. Slow shutter speed. Locked Aperture.

When shooting plants often you are using close focus or macro lenses, so it is far more beneficial to shoot at f/16 or f/22 for the larger depth of field (area in focus), which also makes it easier to use longer shutter speeds.

When I am shooting my LED panels are normally at about 10-20% brightness. My shutter speeds are always longer than 1 second. My aperture is almost NEVER wider than f/11.

Oh and one more benefit of a slow shutter speed, if there are any tiny bugs flying around they wont show up in the image.

 

Hopefully if you are into botanical time lapse and have been running into issues with flicker this may help you out.
I will be doing a few more of these blog posts in the next few weeks. If anybody has any topics they would like me to cover let me know. 🙂

I already did that.

So sometimes I start a project and lose track of  SQUIRREL!!!!!!!!!!!!!!!!!!!!!!!!!!

HAH.  I mean sometimes I get distracted and lose track of shit. For every 10 ideas i have of something I want to do, only 1-2 of them come to be realized.  A lot of this just has to do with a finite amount of time. Being a single parent, having a full time career, working on Biolapse, and attempting to have some sort of a social life, I just have too many things I want to do.

Sometimes though I get pretty far on something and completely forget that I had even done it. Case in point, the BCM2. I would like to build a better version of the BCM.  There are two main concerns I have to work with. The hardware, and software.

In order to keep things tidy,  I usually try to design the PBC in Designspark then fire it off to DFRobot since they have a super affordable prototyping service where small runs of circuit boards can be manufactured for very low cost.

I was spending some time today trying to arrange and organize a PCB that would house eight a4988 stepper drivers for a Arduino Mega shield.  This is part of a project where I plan to gut the eMotimo and put my own electronics in it, converting it to a Dragonframe controlled stepper driver hub.

Before anyone gets too excited, this will be an AC powered studio rig, not really something portable to the field without some generator of some sort.

I had created a template for the stepper drivers, and I was trying to locate it when I ran across a BCM PCB design file.

I popped it open, and holy crap. I literally have no memory of doing this.

8 relay outputs? CHECK!
Multiple Optically isolated outputs? CHECK!
Joystick, Encoder, and Buttons for inputs? CHECK
Real Time Clock Module? CHECK
DHT 22 Temp/Humidity interface? CHECK
I2C connection for display? CHECK!

NICE! I had completely forgotten about this. The file is from 2015.  All i need to do is review it and make sure it is setup correctly, fire it off to DFROBOT and then start ordering the parts for the BCM2.

Now if i just had a DFMOCO arduino shield already designed……

 

 

Shooting Wildflowers

Hey folks.

I am currently out in Sedona Az on vacation. I always love this time of the year as the wildflowers are starting to bloom.  This week I have been using the Laowa 15mm f/4 1:1 macro lens on a Sony A7 and I absolutely love it. I figured this may make a fun blog post to do sort of a review on this lens/camera combination for this purpose.  I have always been into macro photography but 60 and 100mm are always so tight. You can get the subject but the background is usually pretty non existent, at least, it does not really become as big of a part of the photograph as the main subject.

Up to this point my favorite wildflowe lens was the Nikon 10.5mm fisheye, while it is not a “macro” lens it had such a huge DOF that you could pretty much focus on something that was touching the lens. However the fisheye aspect made it a bit tricky to use.  I had that lens for about a year and ended up selling off all my Nikon gear in a move to mirrorless. Personally I really prefer mirrorless over DSLR,  the battery life is not awesome, but the cameras are small and light, they take very little room in a backpack, and are not as conspicuous as a large DSLR.

They almost always include an articulating screen which I find remarkably handy with timelapse work, and there is no mirror assembly to get jammed up. Not that there is a huge problem with that, but the less moving parts the less potential mechanical failures.

I loved my setup from Fuji, I had an Xe1, Xm1 and an XT-1, all which produce fantastic results, however I never had a very good close focus wide angle lens that would get the results I wanted with wildflowers.

Until now.

The moment I saw the Venus Optics Laowa 15mm F/4 macro lens I knew I wanted one. It is a very niche lense to be sure. 15mm is extremely wide on a full frame camera, and it has fairly low distortion which would make it a pretty solid landscape lens. However the macro aspect is what really fascinated me.

I finally picked one up in late july 2016, well after the wildflower season. I played around with it a bit but missed the prime time to use it for what it would excel with.

The lens itself is very well built. it appears to be all metal and glass. It has a very solid feel to it that many lenses are missing these days. Physically it has a few quirks. First off, the aperture ring and the focus ring are reversed. This takes some time to get used to. even after using this lens non stop for a week, I still want to reach to the end of the lens to adjust the focus.

The focusing is very precise but a bit odd. Being a wide angle lens it naturally has a deep DOF, and the first 75% or so of the rotation shifts the focus only a few cm. From 1 m to infinity is almost instant. They are marked in meters which yankees like me sometimes struggle with. I know what 4 inches, 1 foot, 1.5 feet look like, but i get a bit hazy with 17cm. So it takes a few shots before I usually get it right.

Shooting Bees

Both the focus ring and aperture ring feel fantastic. There is a very smooth yet solid feel to them. No slop, no play.

The image quality is outstanding. I have never looked for MTF charts, but I would expect the sharpness scores very high. Off to the corners it does soften a little bit, but not enough that I have noticed it to be a problem. There is an ever so slight vignetting that occurs, but I personally find some slight vignetting to be aesthetically pleasing and it brings attention to the center of the frame(as long as it is not drastic)

The bokeh is pretty good too. While this lens is f/4 and wide angle it is not really great at blowing out the background. I normally find myself shooting between f/11 to f/32 in full daylight. It still manages to render the background content out of focus enough to isolate the subject.

This lens is also sort of a tilt shift lens too. The lever for it totally sucks, and it is one of those scenarios where you push the level and start trying to shift the lens, and it finally gives free and moves to the extreme. Same when returning it back, so you have to put effort in and pace your motion in order to move it partially. When using the tilt shift the vignetting gets very severe at the opposing side. It does do a good job of correcting distortion though. I spend some time playing with that today at Courthouse Rock.

Lens flare can be a bit of an issue. I have shot with some lenses that can give very pleasing lens flare, this is not really one of them. It manifests as a very sharp tight chain of shapes, and I tend to try to avoid lens flare with this one.

My main gripe with it is the barrel flange is stupid huge. it extends well out further than it really needs to. It has a 77mm filter thread and could easily have been just a 70. The flange flare at the end is just obnoxious and creates shadows where you dont want it. The close focus on this lens is literally just a few mm away from the glass, so with the size of the flange you take up nearly half the light hitting the subject. Good luck getting 1:1 with bugs unless they are walking on the lens……. which gives me a neat idea….

I have found to get real close you have to really pick and chose your shots. I picked up a cheap flash from Newwer. I generally dont like using flash, and have had some pretty hit and miss experiences with it. I much prefer natural light, but you have to work with what you got.  I picked this one up because it provides TTL metering and only cost about $60. I used to have several Nikon flashes, two SB-600’s and an SB-900, and after experimenting around with them for several months I ended up hardly ever using them. So cheap was a factor. I liked the design of this flash as it moves the strobe head forward closer to the front of the lens. Then I picked up this cheap flash modifier from amazon. The idea was to curl the front of it down to roll the light over the flange and illuminate the subject.

After spending a few days I found that the flash works fairly well when aimed about 45 degrees to the side and hand holding the bounce to reflect the light back in, but my best results are without the flash, and just holding the bounce and using it as a reflector.

All the nitpicks aside, this has become my favorite lens hands down. The more I use it the better the results. Shooting Bee’s has been a LOT of fun. I set it to f/16 , iso 800, and 1/-500-1/800 shutter speed and set the focus about 1.5cm away from the front element, then just start shoving the camera into the faces of the bees and snapping away. I get quite a bit of bad ones, but some really good ones too.

Check out this bee’s butt.

As for the A7, well, its a decent body. I really wish Sony would pick up on fujis control scheme. I love the control of the XT-1, and pretty much always shoot manual because of that camera. The sony allows for manual control but I dont like the front and rear dials.  If i could stuff the Sony’s guts into an XT-1 it would be a perfect camera for me. I really do dig the sensor of the A7 though. Beautiful colors.

I am eyeing the Sony A7r II very closely right now. Getting such high ISO performance AND IBIS would be huge with this lens.

BCM2

So my Biolapse Control Module has a relay that is starting to flake out. This is the 2nd time this has happened, but to be fair that relay has probably turned off 10-12k times in the past 2 years.

It is going to be a PITA to swap out. For the most part the BCM has run flawless, but there are some nitpicks I have about it. Heck I basically hacked a chronoshield and cut some traces and added new ones to get a lot of the circuit work done. Most of it works, but this Biolapse project is here to stay for the foreseeable future, and I would like to build something a bit more flexible.

Last night i started sketching out the concept behind the BCM2. How could I build upon the original, how can I improve. I had thought about doing this a little while back but I got distracted with Otto.

My requirements,

Better control interface, larger screen for more data
Dual power rails, one protect, one non.
8 assignable outlets, 3 protect, 5 non
Day/night cycles
Grow light control
Fill light control
Temp control (high/low mark)
Humidity control (high/low mark)
Pump control for watering cycles
Adjustable pre/post shot buffers for grow lights and fill lights
Easy to replace relays, plug n play as much as possible.
Input trigger for external timer
2 -3 triggers for camera or external moco
Better chassis, better build.

I will build myself 2 for sure. I like having a backup. If there is interest I might actually make a small limited run of these, 7-8 total and put some of them up for sale. Considering how much work is involved in building/coding these things, they would not be cheap, but at least obtainable. Depending on the difficulty, maybe a grand. May sound like a lot for a glorified set of relays, but it will be a very small run of hand made systems. Very time consuming for a single father with a full time career and a could side businesses.

Something like this is hardly needed for flower blooms as those are usually done in a few days and you can just keep a light on the whole time, but any sort of seed-adult type work, or anything that may take a few weeks the plants start to suffer as fill lights dont hit the spectrums they need very well, and the lack of day/night cycles can stunt growth or even kill the plants. So this really is a specialty system. But i might as well make a couple extra.

Ill be sure to blog about the status and how things are going.

 

 

 

Planet Earth II

Planet Earth II

I know a lot of people on the other side of the world have already seen all of these, but here in the US we are making our way through this series for the first time.  This series is spectacular. I watch it and see shot after shot wondering how the hell did they manage to film that?????

I am just amazed when I see their botanical work.  Timelapse itself is something that takes quite a bit of time, research, trail, and effort to learn.  Botanical timelapse is an entirely different beast than landscape work. You go from being at the mercy of the environment, to being required to fully control the environment.  I have been filming plants for about 3 years now.  During this time I have had a lot of failures, a lot of wasted footage. Quite a few successes, but the stuff that they have on Planet Earth just blows my work out of the water. No comparison.

To be fair, I am completely out of my element when it comes to botanical timelapse, which may sound strange coming from somebody who spends so much time trying to do it.

I am not a photographer. I picked up my first SLR about 7 years ago. I have never taken a class, never had a photography business, never studied under anybody, never had an apprenticeship.  I have done maybe a dozen shoots for friends and such, but never been hired for a shoot. I am just a self taught amateur at best.

I don’t know shit about plants.  I really don’t. I am excellent at killing houseplants. I am somewhat decent at keeping carnivorous plants alive, but that is about it. I am no good at looking at a plant and understanding what its problem is.  I have been learning, but I dont really have any access to botanists to know much about them. Is this something I cut? should that be at an angle? is this too much light? too little light? is this soil ok? how often should I give plants food? Most of those questions are a mystery to me.

I have always had an artistic streak in me, but I wouldn’t consider myself an artist at all. More like a xerox machine.  I have spent time painting, airbrushing, carving, sculpting, sketching, etc. Enough that I can build some rudimentary sets but they all look pretty fake to me. 🙂

I have a very strong technical background, but I am not an engineer. If you want somebody to discuss the finer points of VoIP with SIP protocol, I’m your guy. If you want an understanding on how telecommunications work from user to carrier to user, I know how it all works. 8xx, CNAM, CID, SS7, 5ess, DMS, EWSD, CAS, ground start, loopstart, q931, all of it.  However the mechanical engineering stuff, building robots, programming micro-controllers, etc. That is all pretty far outside my wheelhouse professionally, and there is really no carryover from that technical world to robotics. It may as well be comparing a  catfish to a doorknob.

I suspect that is not the case with the people working for the BBC gathering that sort of footage. I would imagine they have real photographers, set designers, mechanical engineers, motion control experts, time lapse experts, and the sort working together as a team. I could be wrong, if they have a one person Op doing that work I would love to be able to talk them up.

 

I suppose the one thing I really have going for me is I am too dumb to know when I am out of my league. When I find something of interest I tend to achieve a laser focus and manage to push through until I accomplish what I am looking to accomplish. On the same token, that can also be a hindrance, as sometimes I don’t know when it is time to pull in additional resources. Maybe I need some help with this project.

I have several Biolapse projects I am working on. The thing is they are very slow going because I am trying to build living sets now, and it can take quite a bit of time for those sets to establish (if they don’t just outright die).  I feel like I am lacking a clear direction though. Building the robots, setting up the studio, all those had clear end goals.

I suppose this is sounding like I am being pretty down on myself, that is not the case though. I have only been filming plants for 3 years, and I have had to undertake every challenge on my own, and i have accomplished quite a bit.  It may not be the quality you see on the BBC.

Yet….