Sensor size, FF vs APS-C, Micro Four Thirds vs 1” sensors and any other combination one can think of

This is yet another very common discussion which we aren’t really supposed to be having. “Arguments” about how much of a difference makes a difference, assumptions about what is good enough for everyone and statements about how good photos can be taken with any camera. All of this intermixed with real world comparisons of sensors from different generations and from different brands with their respective pricing. Endless.

However, in reality, it is all tradeoffs and subjective decisions, rather than broad generalizations about what is good enough. It is funny how we tend to think that our point of sufficiency is valid for everyone else.

“A good photo can be taken with any camera”
Absolutely true. It can be small, slightly out of focus, with visible flare etc. and still be inspiring or delivering its message. But why wouldn’t people buy the best image quality and flexibility they are willing to spend money on?

That statement is a great argument to not obsess too much about the gear and that understanding composition, color, light etc. is the priority. However, it doesn’t mean that one shouldn’t also invest in good gear, it means only that one shouldn’t think of gear as a replacement for the aforementioned artistic skills.

Photography, Sony and Minolta's legacy

For the last… for as long as I can remember, there have been these comments about Sony - that they are not a photography company, but an appliance one and that their cameras are not designed by photographers, but by Playstation engineers. With some variations, of course… and not always limited to Sony.

As someone who has used (or tried) cameras from pretty much every brand, I can say that I absolutely do not understand what these people are talking about. I doubt that they themselves do. Almost always, when pushed to explain what they actually mean, it tends to end up with something about dedicated shutter dials, overcrowded menus, feature overload and brand legacy, i.e. complete nonsense that has almost nothing to see with photography. It also shows how many have no idea that Sony has a history of making cameras way before they acquired a part of Minolta in 2006.

Part I - a rant

Camera interfaces have changed significantly over time and will continue to change. Some people, however, seem to be stuck in a specific period of camera development and are unable to get over it. In itself, this is not much of an issue (I also have my own specific preferences), but to claim that their chosen period is more photography-oriented than another, seems a bit self-serving and is certainly a very limited view. 

• One controversial topic is the viewfinder, where the optical one is shown as some ultimate achievement that allows photographers to have a “better connection with the scene”. I don’t necessarily disagree with that, but what does that mean for photography? Let’s look at how DSLRs superseded rangefinders (for a good reason; unless one thinks that it wasn’t for a good reason, in which case I would be very interested to hear their argument).

SLR OVFs are more versatile; they can be used for macro, longer lenses or whatever; they show you what you will achieve; are less complicated and prone to misalignments etc. All of these advantages were accepted even though the SLR (and now DSLR) OVFs had the disadvantage of being smaller and dimmer, and not providing that expansive view of the surroundings that rangefinders can (not to mention all the other benefits of RFs).

Well, EVFs are superseding DSLR OVFs in pretty much the exact same way and they require much less compromise, if at all. They are more versatile, allowing whatever information you need to be displayed; zoom around for precise focus; shoot video etc. They also show what you will capture in an even better way than DSLR OVFs and completely avoid any misalignment issues. On top of that, being independent of any light-related mechanics, not only are they cheaper, but they can be made as big as one wants. In discussions about OVFs vs EVFs I would often hear “my D700 this, my a900 that”, i.e. people forgetting that there are many more APS-C than FF digital cameras out there. How does that VF in a 7D compare to the one in theX-T1? (I won’t even address entry-level DSLRs) Sure, there are still some problems (I am sure somebody will cry foul and explain how panning is horrible), but, overall, OVFs have been surpassed. And, note, I am not arguing for the disappearance of OVFs (I certainly would like to have that option in the future). So, I find it really strange when some Canon representative says something like this: “DSLRs can capture the moment better than mirrorless, because you’re viewing directly, not through an LCD.” An utterly ridiculous comment that actually means nothing; such empty statements wouldn’t make sense even in the context of a personal opinion.

• Another topic would be that shutter speed dial that some are so enamored of. This is often encountered when talking about Fuji (or Leica). Fuji is apparently making cameras for photographers, unlike all the other clueless manufacturers. That is actually a very interesting point of view, as there is really nothing that different about Fuji cameras, with the exception of the retro analog interface. So, back to that shutter dial - it is a very inefficient implementation, especially for those it is supposed to address the most - manual mode shooters. It is simply an inconveniently placed stepped dial for an often used feature. A regular thumb/index step-less dial makes much more sense here (hence why it is on every camera except Fuji’s and Leica’s), especially when considering the wealth of information available on the LCD/EVF; it is quicker and more comfortable. We are not limited by mechanics anymore. Again, note that I completely understand the appeal for some people and that it is great that we have it. But to pretend that it is somehow more photography-oriented?

• Yet another issue raised by purists is the sheer amount of features in modern cameras (be it gimmicks or useful things). We take too many photos, we take them too fast, don’t compose as carefully etc. And, yet again, I understand that feeling. I do consider myself to be one of these purists and I also have issues with my shooting discipline, which are difficult to address on digital. However, that doesn’t make some camera more photography-oriented, just because it lacks video or AF. It is simply a personal issue requiring a personal solution and not some tool to fix it. 

• Lastly, the LCDs and, more specifically, the tilting/swivel ones. I understand that these aren’t useful to everyone, but it boggles my mind that some can’t realize the convenience these can be for others. The usual nonsense here is how they introduce complexity, can potentially break and add to the cost of the camera. After all those years (fifteen, is it?), I have yet to see or hear about a single hinge failing, but it seems theoretical worries override empirical evidence. As to the cost, this really seems like a strange thing to worry about when one pays hundreds (and often thousands) of dollars for a camera. Most importantly, I simply cannot understand how it can be called a gimmick, when on my diminutive digital camera I have the modern (and improved) version of a waist-level viewfinder (in addition to an eye-level one, mind you).

Progress happens by making things cheaper and easier, i.e. more convenient and accessible, not by sticking to some random subjective standard, just because it happens to be yours and you are comfortable with it.

Even Art Filters, which I consider a gimmick simply because in 99% of the cases one will pass the files through a computer, are still light years ahead in convenience over what was available in the film era. For people who pretend to be all about photography, I find it bewildering to not be amazed by the amazing tools we have nowadays.

Part II - Some Sony history

Getting back to Sony’s history. Some say that Sony has clearly followed in the footsteps of Minolta (some say otherwise), but, as I see it, they are simply following in their own footsteps, while taking strong cues from the Minolta legacy.

Here is a list of notable (in my opinion) Sony cameras. It is extremely shortened though and doesn’t discuss all they have brought to the camera industry - links to more information can be found at the end of this post or by googling.

Before Minolta :

Sony Mavica (1981) - this is where the story begins with a prototype for an electronic SLR still video camera - it was the first of its kind and the direct predecessor of digital cameras; the first cameras were released in the late 80s, although, apparently, the prototype has been used during the Olympics in 1984;

Sony Mavica MVC-C1 (1988) - Sony’s first consumer electronic camera;

Sony DSC-F1 (1996) - the first Cybershot;

Sony DSC-D700 (1998) - Sony’s first foray into all-in-one SLRs;

Sony F828 (2003) - the first consumer CCD digital camera;

Sony DSC-T1 (2003) - a real compact at 17mm thickness;

With Minolta : 

Sony a100 (2006) - albeit a fine camera, this one was really a regular DSLR, but it shows how Sony followed firmly in Minolta’s footsteps just after the acquisition; 

Sony a900 (2008) - best viewfinder, highest resolution sensor and first FF camera with in-body image stabilization;

Sony a300 (2008) - dual-sensor design for improved Live View and camera performance (Olympus had this approach first, but with a lesser implementation);

Sony a850 (2009) - cheapest FF camera on the market;

Sony a55 (2010) - first SLT camera and the switch to EVFs for Sony; SLT was the natural evolution of the dual sensor setup;

Sony NEX-3/NEX-5 (2010) - Sony’s first mirrorless cameras;

Sony NEX-7 (2011) - Sony’s first advanced mirrorless cameras and, as far as I am aware, the first mirrorless with an integrated EVF;

Sony RX1 (2012) - first digital FF compact;

Sony RX100 (2012) - yet another proof of Sony’s miniaturization prowess;

Sony a7 (2013) - first FF mirrorless, if we are not taking into account Leica rangefinders;

Sony a7 markII (2014) - introducing in-body image stabilization to the a7x line, still unique to Sony for FF cameras;

As one can see, Sony have been at it for a long time and this without even taking into consideration the video side, which I find to be very relevant. In terms of innovation (relevant to photographers), the only other company that has done as much during the last ten, or so, years, would be Olympus, in my opinion. In the consumer and prosumer markets, Sony have constantly delivered some of the most amazing packages and have a long history of sophisticated all-in-one designs and miniaturization. Actually, it looks like the period just after Minolta’s acquisition was the most problematic one, maybe because they tried to follow too closely in Minolta’s footsteps. I am confused as to the main reason of their purchase, but my uneducated guess would be that they simply saw a cheap way to gain market share and take on the big two; the know-how that also came with the package was just a bonus. I would dare to say that they didn’t do a great job in the beginning, simply relying on existing Minolta technologies, and that cost them. However, currently, they seem to be back on track and concentrating on their strengths.

From Sony:

Camera history:

Articles from long ago:

Reviews of some Sony cameras - more links after the links:

Photo of the Sony Pro Mavica MVC-7000

Zeiss announces the Loxia 21mm 2.8!

As expected, the lens is 40% heavier than its ZM equivalent, same as for the other two Loxias. The price however is lower than I expected, although still very pricey and would make me hesitate - the Voigtlander Ultron 21mm 1.8 can be had for significantly less money and even with the close focus adapter, it is still cheaper. Still, the new Loxia seems to be very compact and the 52mm thread is very welcome.

Zeiss Loxia 21mm site.

Early review at PassportsAndLenses.

Voigtlander lens tests from around the internet

Given my interest in Voigtlander lenses, I decided to make a list with the best reviews I find and it only makes sense to share it. I will be concentrating mainly on tests done with Sony cameras and will be updating this post over time.

15mm f4.5 Heliar III (M-mount)
Review from Ian Norman at Photoncollective - with great landscape samples

Review from Ian Norman at Lonelyspeck - astrophotography

Review by Alik Griffin


21mm f1.8 Ultron (M-mount)
Review on 3DKraft - a comparison of various wide angles


35mm f1.7 Ultron (M-mount)
Review by Phillip Reeve - plenty of samples

Voigtlander e-mount lenses

Voigtlander announced 3 wide-angle lenses for e-mount today! I find them particularly interesting, as equivalents simply do not exist or are troublesome - modern wide lenses strive for speed, older wide lenses have issues on digital sensors. Here are the basic specifications:

• 10 mm F 5,6 Hyper-Wide-Heliar

• 12 mm F 5,6 Ultra-Wide-Heliar

• 15 mm F 4,5 Super-Wide-Heliar

So, the existing 12mm is apparently being redesigned, same as the 15mm was recently - "Voigtländer continues its VM-Mount digital optimization with the new 12 mm F 5,6 Ultra-Wide-Heliar III". However, the really interesting part is the 10mm Heliar, as I am unaware of anything else that wide (I wonder if it is going to be also available in M-mount).
I am confident that these will be great lenses performance-wise, but we will have to wait until early 2016 for the pricing. Meanwhile, here's a link to an excellent review of the redesigned 15mm in M-mount on a Sony a7r. And another one, but with fewer samples.

Now, onto some speculation. Judging by the picture posted on Voigltander's website (if it is a picture of the final designs and if they are to scale relative to each other), the 12mm looks entirely different and will possibly sport a filter thread similar to the redesigned 15mm, i.e. 58mm (or slightly bigger). Same for the 10mm. As for pricing, the redesigned 15mm had a significant price increase over the previous version and the conversion to e-mount will probably add even more. I wouldn't expect any of these lenses to go above 1000US, but they will probably be close, especially the 10mm. If they do go above, then the M-mount versions will make much more sense ( I expect Voigtlander to provide both the 10mm and the 12mm also in M-mount at the time of release).

Voigtlander rumors

Apparently, Voigtlander is going to announce new lenses for Sony's e-mount. If that isn't good enough, there's even a rumour about them releasing digital cameras. That would be really excellent, considering the amazing value that most of Voigtlander's products are. 

 Montreal, Voigtlander 20mm SLII.

Montreal, Voigtlander 20mm SLII.

However, if these are simply existing Voigltander lenses converted to e-mount to provide EXIF data and a few other modern niceties, for a higher price, I'd rather stick to the adapter route, especially for M-mount, where the Voigtlander close-focus adapter provides some interesting flexibility.

As to the digital camera part, it would be nice to have again an M-mount camera that isn't a Leica. And if Zeiss aren't going to create a digital Ikon, than the next best thing would be a digital Bessa. Preferably, with an EVF.

 Montreal, Voigtlander 20mm SLII, I think.

Montreal, Voigtlander 20mm SLII, I think.

The upcoming Zeiss 21mm Loxia - a hypothesis

Being a huge fan of wide angle lenses, I am waiting impatiently to see what Zeiss will pull off with its upcoming 21mm Loxia

The Loxia line seems to have only minor changes compared to the ZM one (this redesign, by the way, in my opinion makes it clear that there won’t be a digital Ikon camera, which is a pity). However, the lenses are more expensive (10 to 20%) and heavier (by roughly 40%); although they do feature a de-clickable aperture, which certainly accounts for some of these increases. By following this tiny statistic, the upcoming 21mm 2.8 lens will be 420g and will cost between 1600US and 1700US. However, the current Biogon 21mm has issues on Sony cameras, which means that some kind of redesign will be necessary, which may also affect the numbers.

Lenses and diminishing returns

Here’s a short test I did a year ago. I rarely make such comparisons, so don’t take the images below too seriously.

Voigtlander 58mm 1.4 (V), Sony 50mm 1.8 (S) and Sony Zeiss 55mm 1.8 (Z), tested on a NEX-7. 

Looking at the Zeiss and its glowing reviews, something is wrong. Maybe I have a bad copy, maybe the NEX-7 sensor is being picky, maybe I messed up the focus (probably the latter, as I am actually very happy with the lens). Still, is the FE55 worth twice as much based on what you see here? Even if there’s an issue with the test, how much sharper would one need their lens to be when looking at the Voigtlander and the SEL50F18 (and how much sharper the FE55 can be)? 

I omitted F5.6 to save some space, as there isn't much difference between F4.0 and F8.0.

It is a good idea to put things into perspective occasionally. As it is the norm today, there are plenty of measurements, 100% crops and overblown statements. Lenses are primarily reviewed and compared based on their sharpness and sharpness has undeservedly risen to an unprecedented level of relevance (mostly thanks to modern high-resolution sensors). It doesn’t take long to find a thread explaining how a lens is “crap”, because it is less sharp than another one worth three times more and weighing twice as much. 

Lens “performance” is actually a very stretchy concept, depending on lens, camera, output, personal preferences and budget. What good is sharpness if there’s a huge colorful blob in the middle of your image? Or, if the lens is so big or expensive that you hesitate to take it with you? Suddenly, it all becomes very relative. 

Now, let's look at a couple of more realistic scenarios.

 ISO 100, 15 sec, can't remember the aperture setting, NEX-7 on a tripod

ISO 100, 15 sec, can't remember the aperture setting, NEX-7 on a tripod

 100% crops, default sharpening

100% crops, default sharpening

The above was shot with a Nikkor 135mm F3.5. For an almost 40-year-old lens, it works quite well (well, I don't know of a really bad 135mm lens, but still). This is my weakest lens in terms of sharpness, but I wouldn't have any issue printing quite large with it (after some proper sharpening) and I often crop heavily. It was cheap even at its release and I paid around 2000 yen for mine. So, if we look exclusively at sharpness, it would make most modern lenses look quite bad in terms of value. While something like the new Zeiss 135mm Apo will be better in every aspect of image quality (not to mention being faster), it certainly isn't 100 times better. Hence, the diminishing returns.

Sensor size and its importance

There’s a reason why people say that a good photographer can make a good photo with any camera or that pretty much any camera nowadays will deliver great image quality. Half of it is because sensor size matters less than one might think.

We have all read the endless debates on various forums and comment sections about sensor size, with the predominant topics being Full Frame vs APS-C, where the sensors really do get close in dimensions. The best argument for a Full Frame is that it is the biggest affordable sensor on the market, both in terms of cameras and lenses. The best argument for APS-C is that it is good enough. But, whether the difference in image quality is marginal or negligible, this discussion is mostly of academic interest, because one is buying a camera, not just a sensor. Everything else is not equal and that does matter.

To be able to squeeze out that better image quality from the bigger sensor, one needs to have the skills, both photographical and in image treatment. To be able to take the photos one wants, it would be necessary to have the right lenses and accessories in their system of choice. To be efficient and comfortable while shooting, one needs to have the right features in their system of choice. However, systems are not equal, they have different lenses, different features, different ergonomics. Add to that subject matter - not every subject will benefit equally from a bigger sensor; or add the final output required; or the audience; or the photographer’s idea of what is acceptable to transmit his message or idea or concept. In this big scheme of things, the size of the sensor suddenly becomes just one small part of the equation. 

In short, where one draws the line in terms of size/price/IQ is subject to so many other variables and the final result is so much dependent on the photographer's skills, that simply saying one sensor size is better than another does not mean much, however true. It amounts to nothing more than claiming “my wide angle is better than your nifty-fifty!”.