Wednesday, April 25, 2012

Two video versions of the same trip

In December I made this video from my summer trip to Lapland. I was not very satisfied with it, though the main problem was that I messed up some clips by filming them with too fast a shutter speed (newbie mistake). Also loosing my tripod head in the beginning of the trip led limited the possible filming angles.

When I got the Sony Vegas Movie Studio 11, I though about giving another shot at the fairly limited raw material. The result is seen here.

I'm not fully satisfied with this one either. The uneven motion caused by using too fast a shutterspeed is very visible in some places and even seems to have become worse in the Vimeo version. The few places where I'm holding the camera while riding look rather unsharp after having been processed with the Vegas stabilizer tool.

Comments? Which one is better and why? What could be improved?

Tuesday, April 10, 2012

Camera support for video

For photography I have three different tripods for different purposes. The smallest one is a GorillaPod that holds a compact system camera with a big zoom, but not much else. It can be wrapped around branches and narrow tree trunks, which makes it quite versatile. The next one is a small Slik Sprint Mini II GM, which is stable when extended up to 1 meter, but not much more. I've used it quite a lot, and this is the one I bring when going out for a bike ride. Trying to attach some sort of pan head would be too much, though. My big tripod is an ancient Manfrotto 055C with a Slik Pro Ballhead 800. This combination is much too big and heavy to be transported any longer distances with a bike or by foot.

For videography I'm considering a panning head, though I don't know which one yet. It would also be used with the bird spotting scope, which I have used with the Ballhead 800 for now. The panning head would have to be light (and not too expensive), since I don't plan to use any heavy equipment with it. A long term purchase would also be a new big tripod, which in practice would have to be of carbon fibre to be light enough for me. I doubt that I can afford it soon, though.

For video shots not taken on a tripod I bought the CamCaddie, but it turned out to be a small disappointment for me. It didn't provide any additional stability over my monopod, a Slik Monopod 350 EX, while being a lot less versatile. For the most part the video below is filmed with the camera on a monopod (no image stabilization) with only a few static parts on a tripod:

I think this monopod technique is very usable and with some practice I should also get better at it.

Thursday, March 29, 2012

Some progress

I finished the video from the snow crust overnighter in the beginning of March some time ago. When I started editing it, I almost immediately noticed that I lacked some shootage, but I guess this is normal and is part of the learning process. Maybe a good idea would be to actually do some concrete planning. While I did have some ideas when setting out, I had no real plan. The result is this (to be viewed fullscreen with a higher video resolution):

I also lacked the suitable step-up rings for the ND filters, so the parts where there is lots of light might not have as good quality as possible. I think with an f-value of around 18, the diffraction might actually affect the sharpness even in video resolution. Since this video, made when I had owned the camera for just a few days and did not really know how to use it, I've changed my settings for the camera video mode. This should give better head room for applying small corrections in the video editor.

One of the things I'm still learning is what shutter speed to use in different situations. The general rule is to use a shutter speed twice that of the frame rate, i.e. for 25 fps the shutter speed should be 1/50, which does limit you quite a bit and necessitates the use of ND filters for many situations. I think one can get away with higher shutter speeds when an object is mainly view from or towards the camera, but for panning or objects moving across the picture a fairly slow shutter speed is necessary, especially when using a frame rate of 25 fps (and I have my reasons for that). The clip below demonstrates what happens when the shutter speed is too high. It was shot during my Tour de Pöyrisjärvi last summer, but did not make it into the video, since the movement is much too stuttery.

For video editing I've settled for the Sony Vegas Movie Studio HD Platinum 11, which I bought in a suite that gave some sound editing tools and other stuff as well. It gives a lot of features for a modest price, but takes a while to get used too. I think I'm getting the hang of it, but there are still lots of details I'm not sure about. E.g. the book DSLR Cinema mentions that transcoding (converting) the AVCHD clips made by the camera to a more editor friendly format using e.g. MPEG Streamclip is necessary to get the best color correction possibilites, something which sounds completely wrong in my opinion. This just shows that I have a lot to learn.

Finally, it is perfectly clear that YouTube pretty much destroys the technical quality of the videos. I wonder if Vimeo would give better results?

Tuesday, March 13, 2012


I did an overnighter in nice conditions one week ago and brought no less than three cameras: The Panasonic GH2, the Olympus E-PL1 and the Canon S90. Fortunately these cameras and associated lenses are small enough to fit into the LowePro Slingshot 102 AW bag without problems. This is not the normal situation, though, but I wanted to test the new cameras. The S90 was used for a little time lapse sequence, which it is well suited to with the CHDK. The E-PL1 works as expected. It has less controls than the E-P2, but since is still very usable. The image quality good, as expected, and mostly depends on the lens.

I did some video filming with the GH2, but I haven't yet assembled the material. It is perfectly clear that it will take some time to learn to use the it efficiently and to find the correct parameters. The camera itself contains a lot of functionality and then there is also a jungle of unofficial firmware enhancements (hacks) that change the data video encoding parameters for even better video quality at the cost of  more storage space. The improvement in video quality should be quite big with some of the patches, even though the standard version in my opinion seemed to have noticeably better resolution than that of the EOS 60D, which I've had before. I installed one of the more popular ones, the FlowMotion patch. To evaluate these for yourself would be very time consuming, so I just have to rely on what other have written. Still, with or without any of these patches there are a lot of parameters to experiment with. I filmed last week's fatbike overnighter using the 1080/25p HBR mode with Cinema film mode and the other parameters at default. I later read that it would be better to use the Smooth film mode with the contrast and sharpness values turned down a little, since this would give better possibilites to do color corrections afterwards. It is fairly simple to turn up the contrast and sharpness in the video editor, but impossible to correct afterwards if the video has been recorded too much contrast that have exceeded the dynamic range of the sensor. A simple test (be sure to click it fullscreen and a higher resolution) can be seen below. The first section looks rather dull and flat with the Smooth film mode, but the second section shows the same material with some corrections applied.

Well, there is a lot to learn about the camera, video making and video editing. I've now reached the stage where I have a lot more questions than before, but that is just a natural progression.

Saturday, March 3, 2012

New gear

A desire to start doing some more serious video work has slowly grown stronger. Finally I took the plunge and bought one of the best video capable interchangeable lens cameras, the Panasonic GH2. This camera uses the same m43 system as my Olympus E-P2 camera, meaning that all the lenses can be used on both cameras. I ordered the kit with the 14-42 lens, which actually was only marginally more expensive than just the camera body, to a local shop, Kamera-Aitta. It was a little more expensive than it would have been from a German internet store, but I am willing to pay a little more to get it locally.

Parallell to this I streamlined my existing system a little, due to some ridiculous rebates at another local camera store, Rajala Pro Shop. I sold some existing lenses and the Olympus E-P2 body, since I can't afford and don't want to have stuff I don't use.

The net result of this is seen in the following picture:

The E-PL1 is a cheaper version of the E-P2, which I sold. It has fewer controls, a slightly less advanced image stabilizer and a fastest shutter speed of 1/2000s. The image quality should theoretically be slightly better: It has the same sensor, but a weaker AA-filter that should help resolving a little more detail. I doubt I will be able to notice any difference, though. Somewhat surprisingly, the E-PL1 does have a small internal flash that can be used to trigger external Olympus flashes, giving full wireless TTL control over several external flashes. I tried it and it really does work. This camera should be a good second camera body to have.

I had contemplated the Panasonic GH2 for some time, but wanted to see how the new Olympus m43 flagship, the OM-D E-M5, would turn out. The E-M5 is a very interesting camera, with most features better or at least equally good as those of the GH2. According to the first reports, the E-M5 has an exceptionally good image stabilizer that works wonders for videos as well. Unfortunately, the video capabilities of th E-M5 are otherwise no match for the GH2, so I ultimately decided to get the GH2. It has the best sensor of the current m43 cameras (the performance of the E-M5 sensor is still undetermined) and is especially good for videos. The GH2 offers a lot of functionality and it will take some time to learn it to get the most out of it. There is even a bunch of hacks for the GH2, which improve the video quality quite a bit. I tested out a simple and well supported version to get a higher bitrate, which translates to more details when a lot happens in the picture, and high ISO in manual video mode. ISO 8000 did seem quite usable for video, at least when using the maximum  noise reduction setting. A lot to learn and test out.

These two cameras and the lenses all follow the m43 standard, and hence all the lenses are compatible with both cameras. I will probably not bring the entire system at the same time, unless explicitly doing a photography excursion.

In addition to the cameras above, I still have the Canon S90 pocket camera. It gives good image quality in a truly pocketable format and has a bright lens with a maximum aperture of 2.0, so it is quite usable when the sun isn't shining as well. I also have the CHDK software for the S90 and intend to use it for some time-lapse photography.

Sunday, January 22, 2012


Time for a pure photography excursion. BioFoto Finland held the year meeting at Utö. This was the first visit to Utö for me, though I technically had set my foot on the island once before, when the ferry spent the night there in shelter for a hard January storm before finally going to Jurmo. Utö is a place visited mostly by photography and bird enthusiasts, at least outside the summer season. The place became more accessible some five years ago when the military left the island.

You get to Utö with a ferry from Pärnäs/Pärnäinen and the journey takes a little under five hours. Our group, consisting of almost twenty persons, met at the ferry on Friday evening. We arrived at 22:40 in the evening at Utö Havshotell, an old military barrack converted to a hotel of a good standard. A totally different accomodation than on my usual outings.

After a good nights sleep, the morning looked quite promising, despite the rather depressing weather forecast. A quick glimpse outside the hotel before the breakfast buffet showed the following view.

The morning colors were perfectly acceptable.

The Utö lighthouse...

... and a previous version?

While the military no longer have an active presence on the island, it is an important base for the coastal pilots and the coast guard, hence the radar surveillance.

I was not the only one out photographing.

Some morning pictures.

After some time the sun disappeared behind the clouds, changing the light.

A poor gull had lost its feathers (and life).

The small chapel on the island.

The lighthouse again.

We even got to see the inside of the lighthouse. It has a small ballroom and even a small chapel, available for weddings.

Occasional ice patches.

After lunch the weather changed and it started to snow a little. When we arrived there was no snow on the island, but now the ground got a thin cover of snow.

The small village. There is probably a between ten and twenty permanent residents, with a lot more people living there in the summer.

Port lateral mark.

The waters around Utö have always been dangerous for mariners, and there are lots of ship wrecks there. This is a memorial for SS Park Victory, which sank during a storm in 1947.

Changing light.

Another view.

Nature first.

Another house.

The harbor.

This is how the coastal pilots get out to the ships they are to assist. In all possible weather conditions.

The sun set behind the clouds at 16:20. It had been an interesting day with quite varying weather. The morning temperature was a little below freezing and there was no wind. Later the temperature rose a little and there was some wind. It was also snowing for a short while.

The following morning was cloudy and windy, with a less inspiring light, but there was still something to photograph.

Exposed and windy.

A Mute Swan enjoying the wind. Or not.

Our transport home, M/S Eivor, formerly Baldur from Iceland. There was plenty of time to go through the pictures and write this on the way home.

Since this was my first visit to Utö, there are a bunch of ordinary tourist pictures, but hopefully a few pictures still rise above that category.

I used my newish lightweight photo gear and did not feel restricted by it, but a longer tele would possibly have enabled some bird shots. The others all had very serious camera gear, but I have made my choice and strive to make the most of it. This time I did have my old heavy Manfrotto 055C tripod with me, since I thought a lighter tripod would not have been able to handle the wind. On the Sunday morning in the hard wind I also think a heavier camera would been more stable in the wind. The Olympus E-P2 with the 40-150 lens felt like the wind caused some high frequency vibration on it when mounted on the tripod.

Sunday, December 25, 2011


I'm an available light photographer foremost. The newer digital cameras allow you to take photographs in many situations, also indoors, without a flash if you have fast (=big aperture) optics. Still, I bought a small flash for my Olympus E-P2 as an early Christmas gift for myself. The Olympus FL-300R is a small flash made for the new PEN cameras. The built-in wireless functionality does not work with my E-P2 camera, though, but I ordered a cheap long TTL cable to test some fill flash for my bike photography. The TTL cable, which haven't arrived yet, is needed to get it away from the optical axis, in order to avoid the hard and ugly flash light which plagues most of the flash pictures you see.

The first test for indoor photography went fine. The flash is (just) strong enough to be bounced via the ceiling to give a more natural looking light. Combining it with e.g. the Panasonic 20/1.7 lens and a suitable long exposure times enables getting a good balance between the indirect flash light and the ambient lights.

I also brought the flash with me on an overnighter with the bike to test out some slow-sync flash pictures. In these situations the flash should be set to fire when the second curtain of the shutter closes, to get the light trails to end with the flash instead of to begin with it.

One test showed a funny side effect of the helmet light on half power. Since this particular Led light is dimmed by PWM (the light is pulsed) the light trail looks funny. It shows that the light is not on all the time.

Another funny effect was when the fairly strong lights burned out the foreground before the bike arrive at the place where the flash fired, causing the wheels to partially disappear from the picture.

Finally, the following picture shows what happens when the flash synhronizes on the first curtain instead of the second. The effect is not particularly visible on this picture, but one can still see the the bike and ride have moved slightly after the flash fired. In this case the effect actually brought a little more movement to the picture, but usually second curtain synchronization works better for this purpose.

The last picture is closest to what I envisioned using a fill flash for. There are lots of situations in the forest where it is too dark to get a fast enough shutter time to freeze the action. By using a shutter time e.g. between 1/15 s and 1/50 s the surroundings would be correctly exposed and a flash fired towards the bike rider would render a sharp image of the rider as well. The possible unsharpness of the rider that comes from the long exposure time is much weaker and only contributes to the feeling of movement. More experiments ahead...