This blog is a user's perspective on the Micro Four Thirds camera system. Read more ...

Lens Buyer's Guide. Panasonic GH4 review.

My lens reviews: Olympus 9mm f/8 fisheye, Lumix G 12-32mm f/3.5-5.6, Leica 25mm f/1.4, Lumix X 12-35mm f/2.8, Lumix X 35-100mm f/2.8, Sigma 30mm f/2.8, Sigma 19mm f/2.8, Lumix X PZ 14-42mm f/3.5-5.6, Lumix X PZ 45-175mm f/4-5.6, Olympus M.Zuiko 45mm f/1.8, Panasonic Lumix G 100-300mm f/4-5.6, Panasonic Leica Lumix DG Macro-Elmarit 45mm f/2.8 1:1 Macro, Panasonic Lumix G 45-200mm f/4-5.6, Panasonic Lumix G 20mm f/1.7 pancake, Panasonic Lumix G 14mm f/2.5 pancake, Panasonic Lumix G HD 14-140mm f/4-5.8, Panasonic Lumix G HD 14-140mm f/3.5-5.6, Panasonic Lumix G 8mm f/3.5 fisheye, Lumix G 7-14mm f/4, Samyang 7.5mm f/3.5 fisheye, Tokina 300mm f/6.3 mirror reflex tele, Lensbaby 5.8mm f/3.5 circular fisheye lens
The blog contains affiliate links. As an Amazon Associate I earn from qualifying purchases.

Sunday 27 December 2015

Using V-Log L with GH4

Some months ago, Panasonic released a V-Log L video option for the Lumix GH4. Here, I intend to give a short introduction to what this is, and how you get started with this video profile.

This is, of course, not a complete course. That would require a lot more than a short read. But this is intended to show you what V-Log L is, and how you get started, if you are interested.

For most people, myself included, V-Log L usage is not a big deal. Most of the time, I am happy with the normal video output from the camera, and I don't spend the extra time and effort with the V-Log L profile. So keep in mind that this is optional: You can still enjoy the camera without caring about V-Log L.

Some would compare the V-Log L profile to using the RAW image format, rather than JPEG, when taking still images. It is not exactly that: The video is still compressed (unlike RAW), and it is processed (not straight from the sensor, like RAW normally is).

However, this comparison still makes some sense, as you would use the V-Log L output much in the same way as you would use RAW: To be able to make adjustments to the colour and tone curves of the image. For video, this process is called "colour grading".

When enabling the V-Log L profile, the output is flat. What I mean with flat, is that it looks grey, with little colour saturation, and little difference between the light and dark areas. If you uploaded the video straight to Youtube, people would think it looks quite boring.

That is why you go through the grading process first. Here, you apply a Look-Up Table (LUT) to convert the flat profile back to something that looks normal, and you can apply colour and tone corrections on top of that.

Recording in V-Log L

First of all, of course, activate the V-Log L profile, and you should see the corresponding symbol in the top left corner:

As you see here, the image is "flat" in the sense that it is grey and has low contrast.

You can use the normal exposure modes also for V-Log L. However, keep in mind that the camera will often underexpose in V-Log L, and mostly you'll want to correct for that. I usually leave the exposure correction at +1 stop when using V-Log L, as you see above.

For "run and gun" use, where you have varying lightning, this works well. I guess some seasoned videographers would frown at using automatic exposure, as opposed to full manual, but I don't think you should worry about that. Using auto exposure is a good way to get started.

If you use manual exposure, you'll need some tool to make sure that you have the right exposure. As the profile is so flat, it is hard to notice from the screen if the exposure is too low or too high. The "zebra" stripes is a good tool. Set to "Zebra1":

And adjust that to 80%:

That way, anything which is overexposed (burnt out) will get the zebra stripes in the display. Adjust the exposure so that you are just not getting the zebra stripes, for the optimal exposure. Some zebras are ok, but keep in mind that those areas don't get any details in the final output, due to overexposure.

In V-Log L, you can only use ISO between 400 and 6400. You can still use Auto-ISO.

Editing the video

As the video output is so flat, you cannot just upload it to Youtube, you must first grade it. I have been using Adobe Premiere Pro, but most serious video recording software can handle this.

You bring the footage into the editor as usual, and note that it still has the flat look, as it is not yet graded.

To get the right Look-Up Table (LUT), you can get Panasonic's official one from here, scroll to the bottom and look for "LUT (Look-Up Table)." This is a zip file, and inside you will find a file called "VLog_to_V709_forV35_ver100.cube". This is the one you input into the video editor software.

Add the Video Effect "Lumetri Color" to your video stream. Inside this effect, choose "Custom" from the "Input LUT" option. Here, you select the file you just downloaded from the Panasonic site.

After having done this, you can select the output style you want in the "Look" section. I have just selected "Neutral". You can look at the other ones as well, and see if you would like to test some retro film looks. And remember to select "Active" for it all to work.

This makes the video look normal again:

On top of this, I thought the shadows had some green tint, so I moved the centre of the "Colour Wheels" a bit for the Shadows and Midtones. Also, I took down the Saturation a bit, as I thought the colours were a bit strong:

These colour wheels are central to tweaking the video. A common cliche the last years, is to push the shadows towards the blue, and the midtones towards orange. This creates a contrasty effect, and makes the skin tones stand out a lot. You'll see this effect used a lot in, e.g., "Transformers" and "Iron Man 2".

Using an effect like this is going to make your video look modern today, but look dated in some years time. I prefer to keep them more neutral.

After having done all this, I can render the video as usual. Here is the output in Youtube:

Sadly, the focus was a bit off during a lot of this movie. I used the Lumix 25mm f/1.4 lens at f/1.6 and didn't foresee the focus issues it would have. The problem here is that the microphone in the front attracts the focus some of the time, and when the musician moves, the background looks more stable and easier to focus on for the camera. I could have prefocused on the musician's face, and turned off autofocus, that would have solved the problem. This is a lesson learnt for me.

The exposure was: f/1.6, 1/60s, ISO 640.

This video could have been graded better still. As you see, the face of the musician is a bit too red. However, I had a hard time finding the right white balance. I guess an experienced grader would have made this far better.


The V-Log L video option opens up for more possibilities to use custom video profiles in post processing. However, you need to spend more time on editing the video files, and you must invest in an editing software, like Adobe Premiere Pro.

Finally, you can get by pretty well without worrying about the V-Log L at all. Here is an example video I recorded using the normal video profile (not V-Log L), and uploaded directly to Youtube without any post processing:

Most of the time, this looks totally ok. Personally, I am not using the V-Log L a lot. For normal hobbyist use, the standard profile is just fine.

For  a professional, having V-Log L is crucial, since it allows for colour grading clips to match, even if they are recording under different lightning condition, or even by different cameras. But for amateurs, it adds an additional step of post processing, which may not add much extra value.

Sunday 20 December 2015

Waiting time

We are in the "Advent" time in the Western hemisphere, which means expectant waiting. What we are waiting for is probably quite personal and individual.

One thing I may have in common with the readers, though, is waiting for camera news. So what cameras are we waiting for?


The oldest Olympus Micro Four Thirds camera in the active lineup now, is the Olympus OM-D E-M1. Announced in September 2013, it cleverly replaced the E-5 Four Thirds DSLR, and became the top Micro Four Thirds camera at the same time.

It achieved this by employing on-sensor PDAF sensors, still the only M4/3 camera with this technology, to be able to focus the legacy Four Thirds lenses. It is the only M4/3 camera which can autofocus all the legacy Four Thirds lenses.

This camera is, in my opinion, the Olympus camera with the best ergonomics. Technology wise, though, it is lagging behind. I would guess that a replacement should be due soon, although I think no concrete rumor exists that it will be replaced soon.

Anyone with a cache of high quality legacy Four Thirds lenses, like the 150mm f/2, or 50-200mm f/2.8-3.5, are sure to be eagerly waiting for an updated E-M1, or any other camera with PDAF technology.

We have had specific rumors that Olympus is launching a new line of cameras, though, which may be called "PEN F". These are probably going to be similar to the Olympus E-PL7, but with a built in viewfinder, probably in the rangefinder style. A camera with the newer 20MP sensor, currently used in the Lumix GX8 will probably be announced. I would expect to see an announcement at the CES 2016 show, January 6-9.

In terms of lenses, after a long wait, we are expecting to see the Olympus 300mm f/4 lens at CES:

The lens will have optical image stabilization (OIS), making it useful also for Panasonic Lumix camera users. Further, it will be weather protected, like the other Olympus PRO lenses. The expected price is around US$2500.


The current top dog, the Lumix GH4 is starting to show its age now, as it was announced almost two years ago. However, it still isn't really lacking anything, so why update it? The GH4 was also given a shot in the arm recently when the V-Log L was added with a firmware update. I would guess that we don't see the GH5 announced at CES, but it will probably come in the first half of 2016.

A new GH5 will most likely feature the newer 20MP sensor (used in the Lumix GX8). Also, the GH4 has a drawback when it comes to 4K videos: They are recorded with a crop factor of 1.3x, read more about it here. This means that a 14mm lens behaves like a 18mm lens in 4K video mode. The GH5 will surely use the whole sensor for 4K videos, hence there will be less need for wide angle lenses.

Perhaps the GH5 will also allow for 4K video recording at a higher framerate. Today, the limit is 30FPS with all the 4K capable M4/3 cameras.

Other Pansonic cameras due for an upgrade (for the 4K treatment), include the Lumix GM5 and Lumix GF7. A version of the GM5 with 4K video recording is expected fairly soon.

Just like Olympus, Panasonic is expecting to launch a long tele lens, the Lumix/Leica 100-400mm f/4-6.3 OIS:

And just like the long lens from Olympus, this one is also expected to debut at the CES in January, and sell at approximately the same price.


The big announcement due now, is for the Nikon D5. It is expected to come at CES (early January). Replacing the Nikon D4S, this camera will mostly be "more of the same". It will have a 20MP fullframe sensor, and most likely, will feature 4K video recording. This is a very expensive, large, and heavy, DSLR. It is for professional users, who want the very best in terms of ergonomics, ruggedness, reliability, and speed.

You could be wondering why the upcoming Nikon D5, the most expensive camera, only has a resolution of 20MP. Especially when you can get cameras at half the price with twice the MP count, e.g., Nikon D810 (36MP) or the Sony A7R II at 42MP.

The answer is that for most uses, 20MP is sufficient. And that is what the professional users of this series knows: With 20MP, you can cover almost any client need, including photos for glossy magazines. Some fashion photographers like to use medium format cameras with 40-50MP, but it is hardly needed.

The professional line of cameras has had a moderate MP count for a long time: Nikon D3 with 12MP in 2007, Nikon D4S with 16MP in 2012, and now the Nikon D5 with 20MP in 2016.

In the more distant future, I would expect an upgrade to their high resolution fullframe DSLR, the Nikon D810. Probably, a new camera with the 42MP sensor from the Sony A7R II, will be announced in 2016.

In the crop sensor DX line, all the cameras have been rehashed fairly recently. What enthusiasts are waiting for, though, is Nikons answer to the Canon 7D II: A rugged crop camera, especially for bird/wildlife enthusiasts. Such a camera will probably be called Nikon D500, and has been expected for years. Will it come in 2016? I think that is anyone's guess, right now.

Nikon has their Nikon 1 line of mirrorless cameras with a one inch sensor. This line recently saw the release of the Nikon J5, with a newer 21MP sensor from Sony, that gives much better image quality than earlier model, a much awaited development.

Now, the Nikon 1 fans are waiting for a replacement of the Nikon V3 enthusiast camera model. The V4 should have the same sensor as the J5, as well as the deeper buffer and quicker handling associated with the V line of cameras. And, not least, an eye level viewfinder (EVF).

There is a lot of uncertainty as to the future of the Nikon 1 line currently. I think at least one more generation of the V line will be produced. It may not see an announcement at CES, but it could come during Q1 next year.


With Samsung, people are not waiting for one specific camera model, but they are waiting for news of the future of the NX system: Will there be any more cameras? As the system appears to be discontinued in parts of Europe, that does bring some doom and gloom feelings.

The Samsung NX1 is their top camera, styled like a pro DSLR, and it has a class leading 28MP APS-C sensor, capable of 4K video recording. The same sensor, and much of the same features, sit in the entry level Samsung NX500, very competitively priced.

Still, it is hard for them to gain a market foothold, especially since photography interested people tend to be conservative, and only reluctantly look outside of the "big" camera brands. On the negative side, though, the Samsung NX lens lineup is a bit uninspiring, even if it should be more than sufficient for most.


Recently, Sony have focused almost all their effort on fullframe cameras and lenses. And for a good reason: This is where the bigger margins are, and this is where they can differentiate themselves from the competition: No other manufacturer has a fullframe mirrorless system. I am disregarding the Leica SL system here, as it is hideously large and expensive.

Still, we know that Sony has not dismissed the APS-C sensor mirrorless system completely: A camera replacing the Sony A6000 has been expected for some months now, perhaps it will debut at the CES.


From Pentax, the expectation is that they will soon release their fullframe DSLR. How can they compete against Canon, Nikon and Sony here, though?

Their camera is going to be fairly compact, and weatherprotected, at a reasonable price. That will be their niche.

Also, Pentax still has quite some loyal fans, with collections of older fullframe Pentax lenses. They will probably contribute to keeping the sales up. And, they are already quite good at waiting. Historically, Pentax has not been very good at delivering on time.

Saturday 19 December 2015

Blog economy

Some of you have probably noticed that there are links to webshops on this blog. And here is a disclosure: If you click on the link, and buy something, I'm getting a commission from the purchase.

So does this mean that writing this blog is my day job? And should you also start a blog, and get rich quick?

Blog traffic

Anyone can write a blog, but to get people to actually read it takes a lot of time and effort. Here you can see the statistics for my blog:

I haven't done much in particular to attract the readers. I could have been more active on social media and so on to draw more readers, but I prefer not to be pushy.

Rather, most of the traffic comes from people using search engines. And it takes time before search engines consistently direct traffic to your blog. You should expect to write quality articles for a long time, probably at least a year, before you can expect to gain any significant traffic from search engines.

From the statistics, you can see that the traffic peaked around late 2013, and has since been in decline. This is partially due to that I don't write quite as frequently, but mostly, I think, due to more competition. Competition from fellow bloggers, but more importantly, competition from serious review sites like DPReview.

Back when I started, there was not much resources about Micro Four Thirds on the internet, and more people, relatively, came to my blog. Nowadays, more alternatives exist, and, to be frank, at a higher quality, too. But I think some readers like to see independent articles as well, and some of those find their way to my blog.

I think there are some handful of individuals who started blogging very early, and are now getting a lot of search engine hits. They probably get a lot of traffic. But for a new blogger to reach this level takes a long time. Expect to work hard on the writing for years to get many readers.


I'm using the American Amazon store for the webshop links. If anyone buys something I have linked to, I get a commission: 4% of the sales amount. For non-electronics items, the commission can be higher, around 6-7%.

For each click on a webshop link, around 2.5% actually buy something. This is called the conversion rate, the rate of clicks converted into a sale. Perhaps you think 2.5% sounds low, but this is a quite common level.

Actually, I think the level normally is a bit higher, around 3-4%. But non-American readers of my blog are less likely to buy anything from the American Amazon shop, and that brings the rate down a bit for my blog.

I could have implemented a system to direct readers to a webshop in their own country. However, to make those changes now would be extremely time consuming for me, and the reward would be fairly marginal. So I retain only the American links for now.

When someone buys something, most of the time it is something inexpensive. Like lens caps or adapter, which give me only cents in commission. Once in a while, someone will buy a lens, which earns me some tens of dollars. But it is rare.

On a monthly basis, this commission amounts to around US$70. Given that I usually spend around 20-30 hours per month on the blog, this is of course far from sufficient to pay for my time.

And here I come to my point: Bloggers should not expect to get any significant monetary reward. Only some very few bloggers reach a number of readers that makes them a normal wage. Most end up with peanuts.

If you work hard and write good articles frequently, you will end up with a fair number of readers. But breaking even, in terms of getting well paid for the work, does not come easy.


You've probably seen the ads on youtube videos. Surely, those who make the videos get rich?

The most viewed video I have made, is this one, showing the very impressive IBIS inside the Olympus OM-D E-M5 II:

This video was featured on the technology portal Gizmodo about a year ago, and has had a total of 115.000 views at this time of writing. Now, it has around 500 new views per month.

The total ad income on this video is US$50. Given that I spent around five hours creating the video, at least this gained me an ok hourly wage, didn't it?

Sort of. But this simple calculation fails to take into consideration the 100 other videos I have uploaded, which have not gained any significant income. Considering also the other videos, the hourly payment again goes down to a negligible level.


You should not start blogging, expecting to get paid well. Only some very few, very active bloggers can expect that. Most get paid next to nothing.

If you do start blogging, it should be out of a genuine interest in the writing. As a hobby. That is what I do.

Sunday 13 December 2015

Google Photos review

Some time ago, Google added the Photos service to their portfolio. So, what is it, and should you use it?

The Google Photos service aims to help photo interested people by providing a free cloud service, where it should be easier to search for your images later. There are essentially two versions of the service: The one most people probably use is the free - in the gratis sense - version, where your photos and videos are stored compressed. You can still upload an unlimited number of items.

There is a second version where Google stores your material uncompressed, but then you need to pay for the storage through their Drive service.


To upload your images, it is easiest to use a batch uploader, available for Mac and Windows. This is a simple program running in the background, which scans for new images and videos, and uploads them for you. I've used this program for Windows, and found that it is a big unstable.

Sometimes, it crashes, it does not always find new photos automatically, hence, I must restart it to have it upload new images. Also, it seems to skip the largest videos, in size above around 1GB. I've found that I often need to upload these manually.

Which brings me to the manual upload possibility. From the main page, you can click on the cloud symbol (far right below) to upload single files, or multiple files in a batch.

The advantage of this latter method is that you can upload photos from everywhere. If you are on the move, for example, you can upload your photos over a wifi network in a cafe, not needing to wait until you are home.

The downside is that Google will never store the catalogue location of your photos, even if you use the batch upload program: The directory path where your images resided, e.g., C:\photos\holiday\2015\London, is lost, and you cannot search for these keywords later. So how do you navigate your photos, you may ask yourself?

Browsing photos

Google Photos offers a handful of ways to browse your photos. First of all, you can scroll them all in chronological order. Which is all well if you have some tens of photos, but if you have thousands, that becomes increasingly difficult.

Next, you can navigate by places and things:

Places categorizes your pictures into where you took them. If your camera or smartphone has a GPS device, this is the source for the geographical location of each photo. However, the photos and videos can also be categorized by some pattern recognition algorithm.

This means that when someone uploads photos that are GPS tagged, the information about what objects reside where is used to map other people's photos which are not GPS tagged. This seems to work quite well.

When a picture is GPS tagged, you can see a map showing where it is taken. If it was not GPS tagged, but located with the pattern recognition algorithm, you can only see which city, typically, it was taken in, not the exact location.

Things aims to group your pictures according to what you have photographed. The categorization is solely automatic, you cannot yourself select which category each picture should go into. This works quite well, even if it is not very accurate. In the examples above, you see scooters categorized as "bikes", and to the right, you have a hawfinch in a pine tree categorized as a "flower".

You can also search for any word, and the results are sometimes helpful, sometimes not. Here, I have searched for "sparrow". This yields a lot of birds, but none of them are actually sparrows:

EXIF information

If you select one photo, you can see some key EXIF informaion: Time and date, original filename and resolution, aperture, shutter speed, ISO and focal length. It also shows what type of camera was used. However, you cannot search from this information.

So you cannot search for photos taken with the Lumix GH4 camera, for example, or images taken at 14mm focal length. Which is a strange limitation. If you key in "Nikon" in the search field, you will get pictures of Nikon cameras, not pictures taken by Nikon cameras.


Image resolution is reduced to a maximum of 16MP, which is mostly quite sufficient. In addition, they are also compressed further in a new JPEG file. The original JPEG file you uploaded is not saved, unless you are using the pay-for-storage solution.

You could upload RAW files as well, but then Google will convert them to JPEG and discard the original file.

Video files are compressed to a maximum of 1080p.

Here is an example picture. You can download the original photo, and the compressed photo created by Google Photos:

Original, click to download4608x3456 pixels, 8.54MB
Compressed, click to download4608x3456 pixels, 2.14MB

To help you in comparing them, here are 100% crops from both images:

It is hard to find any real evidence that the compressed image is worse.


If you upload certain types of photos, Google will create "Creations" for you. For example, if you pan while taking several photos, you will get a "Creation" which is the stitched panorama. If you photograph a similar scene several times, you will get an animation.

And if you use the bracketing feature on your camera, Google will make an HDR for you. Here is an example of the latter: I used the five picture bracket feature on the Lumix GH4 with the Samyang 7.5mm f/3.5 fisheye lens, and set the difference to one stop. This gives me five pictures:


Google combines these for you automatically, and gives you this HDR picture. It looks somewhat cheesey, but is sure to give you many likes on social media:


Google Photos can be a good search and archive tool. But it is not perfect: It doesn't have the functions you would typically want, like searching for images taken with specific cameras, or taken at specific focal lengths or apertures.

People who are concerned, or, if you want, paranoid, may say that Google will use the information from the images you upload to profile you, and generate more specific ads tailored for you. That is probably true. So the service is not truly free, you are probably paying with your own personal information.

Sunday 6 December 2015

Ultrawide lens compared with fisheye

One of the early Micro Four Thirds lenses was the Lumix G 7-14mm f/4 ultra wide angle zoom lens (my review). In 35mm film camera equivalent terms, it goes as wide as 14mm (7mm times the crop factor of 2).

This is a very wide rectilinear lens, but not quite as wide as you can get with other formats. For Canon Fullframe cameras, you can get the amazing EF 11-24mm f/4, and for crop DSLR cameras, you can get the Sigma 8-16mm f/4.5-5.6, with a 12mm equivalent wide end.

The Lumix G 7-14mm used to be the most expensive M4/3 lens. It is no longer, due to other more expensive lenses, but it is still the widest rectilinear zoom lens. But how does it compare with a fisheye lens, in terms of field of view? Here are some example images for comparison (click for larger images):

The fisheye image (right above) can be defished (click to see how it can be done using the free software Hugin). Defishing yields this result:

Samyang 7.5mm defished
Samyang 7.5mm desfished and cropped

When defishing using a rectilinear projection, like I did above, the image gets very stretched in the corners. To avoid this effect, some use the Panini projection in Hugin.

Comparing the two fields of view shows that you get a lot wider images using the fisheye lens:

As you can see, the fisheye lens gives you much wider images, but keep in mind that the corners are quite stretched, and you may not be able to use the whole picture when defishing. Still, even if you crop it a bit, it is way wider than what you get from the Lumix G 7-14mm f/4 ultra wide angle zoom lens.

Other lenses

Another very wide rectilinear zoom lens is the Olympus 7-14mm f/2.8 PRO. It is similar to the Lumix G 7-14mm f/4 in terms of field of view, but adds another stop of aperture, at the expense of a larger, heavier lens, and a steeper price tag. You shouldn't expect better images from this lens, but you get the possibility to use f/2.8. Also, the lens is weather protected, meaning that it is more likely to survive some water splashes.

The Samyang 7.5mm f/3.5 that I used above is a manual focus lens. There are also two autofocus capable fisheye lenses:

The Lumix G 8mm f/3.5 is one of the smallest autofocus capable fisheye lenses, but not any better optically than the Samyang lens, see a comparison here. The lens is still very good, but a tad expensive.

Olympus also has a fisheye lens, with a record large f/1.8 maximum aperture, the Olympus 8mm f/1.8 PRO. This large aperture is rarely needed for real life use, in my opinion. However, people who are interested in astrophotography will find it interesting. Also, it is good for use underwater, inside a waterproof housing.

LensLumix G 7-14mm f/4Samyang 7.5mm f/3.5Olympus 7-14mm f/2.8Lumix G 8mm f/3.5Olympus 8mm f/1.8
Weight 300g179g534g165g314g
Length 83mm48mm106mm52mm80mm
Diameter 70mm60mm79mm61mm62mm
Lens elements/groups16/129/714/1110/917/15
ProsSmall, light, not too expensive these daysVery good image quality, fantastic value for moneyFast aperture, weather protectedCompact, good image qualityVery fast for a fisheye, weather protected
ConsSharpness could be better at 7mm f/4, but improves when stopping down. Some purple flareNo autofocusLarge, expensiveSomewhat expensiveDo you really need a so fast fisheye?

Sunday 29 November 2015

Myths in photography

Photography and camera equipment is a conservative business, and there are a lot of myths out there. Myths that perhaps were true back when SLR cameras were popularized (1960s-70s), or were never true. Here are some examples that I often come across.

Sunday 15 November 2015

Fisheye lenses compared

I like fisheye lenses: They can cram an impressively wide field of view into the image frame, and create perspectives that you would not be able to see with the human eye. Another aspect of fisheye lenses is that they create a lot of barrel distortion (rounded images), which you can remove through a defish process, or retain in the final image.

Here is a collection of fisheye lenses for Micro Four Thirds and other systems:

From the left: Samyang 7.5mm f/3.5 (my review), Lumix G 8mm f/3.5 (my review), Yasuhara Madoka 180 7.3mm f/4 (my review), Olympus 9mm f/8 (white) (my review), Lensbaby 5.8mm f/3.5 (the biggest) (my review)

There are basically two types of fisheye lenses: A circular fisheye lens renders a circle in the centre of the image frame, which usually extends to 180° all around. A full frame/diagonal fisheye, on the other hand, renders the full imaging sensors, and usually extends to 180° from corner to corner.

Sunday 8 November 2015

Trend: Large sensors in compact cameras

Most photo enthusiasts now understand the value of a large imaging sensor: It enables better resolution, better dynamic range, and better low light sensitivity. However, large sensors have typically come in larger camera systems.

And a large sensor DSLR system, with a long register distance, also needs large lenses, especially if the aperture or the zoom ratio is large.

Hence, among enthusiasts, there has been a desire for more compact large sensor cameras, to bring along more easily in a jacket pocket, for example. We have waited a long time for these types of cameras to arrive, and in the last years, we have gotten more to choose from.

To illustrate this trend, I have compiled this diagram of compact, premium cameras from the last decade. The measure I have used along the left axis is the imaging sensor size (in mm²), divided by the maximum aperture. The bigger this figure, the better.

In the case of a zoom lens, I have divided by the average over the long and short end maximum aperture, which is a bit unfair towards those with a longer zoom range, but life is unfair anyway.

Sensor area in mm² divided by maximum aperture size

Saturday 31 October 2015

Lumix tele lenses compared

Panasonic makes an impressive number of tele zoom lenses. This is not a stupid choice, after all, the interchangeable system camera market relies on the consumer actually buying the extra lenses. So there needs to be some lenses to choose from. And a tele zoom lens is a popular aftermarket choice, probably the most popular together with a fast prime lens.

In this article, I will be comparing six of them:

Wednesday 14 October 2015

Product news

We have seen some interesting new photo products recently:


Samyang is a South Korean third party lens manufacturer. They specialize in classic manual focus lenses, and try to fill holes in system camera makers portfolios by mostly making very wide and very fast lenses at a reasonable cost. They are often marketed under other names like Rokinon or Walimex.

Only fairly recently have they been making lenses especially for mirrorless cameras. With the shorter register distance of mirrorless cameras, wide lenses can be made more compact.

Their two most recent lenses are made especially for mirrorless cameras, and hence, are quite compact and light:

Samyang 21mm f/1.4Samyang 50mm f/1.2

These are good for photographers who enjoy classic manual focus lenses with a metal construction. Their focus and aperture rings are usually well dampened and a joy to use.

Considering the speed of these lenses, f/1.4 for the 21mm lens, and f/1.2 for the 50mm lens, they are very compact, and also quite competitively priced. A 50mm f/1.2 lens is something you either have to buy second hand, or at an extremely high price new. Until this lens launch.

Already, they have a very good fisheye lens for Micro Four Thirds, the Samyang 7.5mm f/3.5 (my review), which is one of the true gems for M4/3: Very good optical quality at a competitive price.

Also look out for the Samyang 12mm f/2 (my review), which can be had at around US$300 these days, and is a well performing wide angle manual lens.


A surprise announcement from Sony today, the RX1 II fixes the two major flaws of the original Sony RX1 fullframe compact camera: The lack of an electronic viewfinder (EVF), and the slow focus speed.

The new camera will retail at around US$3300 in November, and features the same 42MP sensor as found in the Sony a7R II. The sensor has on-chip Phase-Detection Autofocus sensors (PDAF), which are designed for faster and more accurate autofocus, especially for moving objects.

Unlike the a7R II, though, the RX1 Mark II does not have in-body image stabilization (IBIS). Given that the camera has a fixed 35mm f/2 lens, IBIS could have been very useful.


Having seen that the premium large sensor compact market is where the margins are, Canon are launching more of them. Both the Canon G5 X and Canon G9 X contain the Sony one inch 20MP sensor, a sensor size also used in the Nikon 1 mirrorless line, and compact cameras from Sony and Panasonic.

Canon G5 XCanon G9 X

The Canon G5 X is obviously targeted towards serious enthusiasts, with the built in EVF, and fully articulated LCD screen. It has a 24-120mm equivalent lens with a very fast f/1.8-f/2.8 aperture range.

It is natural to compare it with the Sony RX100 IV, which uses the same sensor. The Sony lens costs much more, though, at US$950, while the Canon G5 X costs US$800. I guess it comes down to compactness: If you want the most compact camera, get the Sony RX100 IV at a higher cost. However, if you want the best ergonomics, in a somewhat larger package, get the Canon G5 X.

The Canon G9 X is relatively cheaper, with a more modest 28-84mm equivalent lens, with an aperture range of "only" f/2-f/4.9. But then again, it is truly pocketable. It is a good choice for those who want a pocketable camera at a reasonable cost.

Canon also appears to want to restart their EOS M mirrorless line, by launching a completely new camera kit, the Canon EOS M10 with 15-45mm f/3.5-6.3. Both the camera and kit zoom lens appears to be intended to be as small as possible.

The camera does, however, contain the relative large APS-C sensor with hybrid CDAF and PDAF focus system, promising better focus performance than the original EOS M camera. I think the lens is sensible. Starting at 15mm, you get a true wide angle at 24mm equivalent. I think more basic kit zoom lenses should start at 24mm.

Canon still needs to launch more lenses for their EOS M line to make sense. So far, they have been taking their time, with only five lenses in total, whereof the 11-22mm wide angle zoom lens is only available in Japan.


Pentax is gearing up towards their upcoming fullframe DSLR, which will probably become the smallest weatherprotected FF DSLR on the market.

To complement their upcoming premium DSLR, they have launched the lenses to go with it, the D FA* 24-70mm f/2.8 and the existing D FA* 70-200mm f/2.8, making up the typical professional standard zoom lenses.

Sunday 20 September 2015

Sigma 60mm long portrait lens

Both lenses used here are designed for a larger sensor than in the cameras pictured:

The Sigma 60mm f/2.8 DN used on the Olympus E-M5 II (left) is designed for mirrorless cameras with an APS-C sized sensor, e.g., Sony E, where it becomes equivalent to 90mm, a classic portrait lens. However, it also comes with a Micro Four Thirds mount, where it becomes a 120mm equivalent lens, i.e., a long portrait lens.

Tuesday 15 September 2015

Get GH4 V-Log profile for free

Some fans have been angered by the news that Panasonic would charge people to use the new V-Log profile on the Lumix GH4. This is an video profile that allows for higher dynamic range than usual.

However, with a simple trick, you can get this profile for free. Here is how:

  1. Make sure you have the Firmware 2.3 installed.Get it here.
    It seems Panasonic have closed the download, but you can try this alternative source.
  2. Turn on the camera, and enable the Wifi mode. This turns on the blue light on the top of the camera.
  3. Connect the camera to your smartphone with the Panasonic app.
  4. Set the camera in a video mode, e.g., by setting the mode dial to the creative video mode:
  5. Open the Panasonic Image App on your smartphone. Go to the "Remote operation" function:
  6. Inside the Remote operation, tap the "Q.MENU" item:
  7. Tap the first option, "Photo Style":
  8. Select V-Log L:
  9. Look at your camera: It now has the V-Log L mode:
  10. Save this mode for later use. Push the "Menu" button on the camera, and go to "Cust.Set Mem." to store it as one of the custom functions (C1, C2, C3-1, C3-2 or C3-3):

And that is it! Now you can test out the V-Log L mode for free, and see if you think it will improve your videos.

If you want to use the V-Log V profile for several different video modes, e.g., 4k and 1080p, then repeat this process for each mode. For example, start off with 4K before starting the Panasonic App to retrieve the V-Log L profile, and you will get a 4K V-Log L profile that you can save as one of the custom functions.

Note that successfully using the V-Log L profile requires skillful postprocessing in a program like Adobe Premiere Pro. Without this video grading, the V-Log L video will just look flat with boring colours.

Anyone who is not familiar with video grading are better off using the normal video modes. So V-Log L is not a miracle cure that makes your videos look fantastic. Rather, it enables competent users to get the most out of the video stream later.

The GoPro cameras have a similar feature called "ProTune". This creates a video file which is more flat, with less sharpness and less colour saturation. Here is what they write about this mode:

What this means for the every-day user is that Protune footage requires more editing to get that traditional GoPro look. That’s great if you want to spend more time editing your footage to put your own unique spin on it, but if you want to do simple editing and already like the GoPro look and feel that you’re used to, we'd recommend leaving Protune turned off.

Essentially: If you don't know what you are doing, don't use ProTune. Panasonic could say the same about V-Log, and in a way, they do, by charging an extra cost for it.

A new 2.4 firmware is coming on September 17th, which will disable using the V-Log L profile without purchasing it. Also, newer upgrades of the Panasonic Image App will also disable this.

Sunday 13 September 2015

Olympus E-M5 Mark II sensor shift demo

All Olympus Micro Four Thirds cameras have had in-body image stabilization (IBIS), however, it was not until the Olympus E-M5 Mark II that this feature became truly useful also for stabilizing video recording. To illustrate how it works, I have mounted the Lumix GH4 facing into the E-M5 II using the Leica 45mm f/2.8 macro lens, both mounted to a dual camera bracket:

Before actually starting the recording, I put some transparent plastic around the lens, between the cameras, to make the lightning more even. Here is the outcome, in an animated GIF:

Sunday 23 August 2015

Impact of sensor size

Looking back, we have had "the megapixel race", in which camera makers aimed put more and more resolution into their cameras. And for compact cameras, we have had the "superzoom race", so far culminating in the mindboggling 83x built-in optical zoom in the Nikon Coolpix P900.

In the enthusiast segment, though, there is a very clear trend at the moment: The importance of the sensor size. We have the very successful Sony RX100 large sensor compact camera sporting a "one inch" sensor, while previously cameras from the same segment typically had 1/1.7 inch sensors.

So why, exactly, is the sensor size important? It does lead to larger and more expensive cameras, and larger lenses, so there must be positive aspects as well to balance this out.

One such positive is bokeh: The larger the sensor, the thinner the depth of focus is. Meaning that the foreground and background will be more out of focus, everything else equal. Read more about it here.

Also, the larger the size of each individual photosite, the better the quality. At least in theory. Hence, one would normally expect less noise and better dynamic range from a larger sensor than from a smaller sensor. To illustrate this, I have taken the same pictures using three different sensor sizes:

From left to right: Nikon 1 V3 (one inch sensor size), Olympus OM-D E-M5 Mark II (Four Thirds sensor size), and Nikon D3300 (APS-C sensor size).

Here is a relative comparison of the sensor sizes: