8 ms·

In the Four Thirds, later Micro Four Thirds, world, Olympus and Panasonic played out some of the familiar tropes between these two corporations -- except they (

by Jedd 9y ago

In the Four Thirds, later Micro Four Thirds, world, Olympus and Panasonic played out some of the familiar tropes between these two corporations -- except they (Oly & Pana) had already agreed on the sensor size and lens mount electronics... so it was much less risky or painful to commit to one over the other. In general terms, Panasonic had the better motion video features, Olympus had the better stills, but it was / is pretty fuzzy, and a closely matched feature set. One of my top five favourite blog posts ever is an impassioned rant [1] about digital Full Frame sensors, originally published in 2008 before they became generally affordable (but while they were still very much a fantasy for many photographers who didn't really understand why they wanted one). Updated over the years, and still worth the read. As someone entering the field without a large collection of (by contemporary standards) poorly designed, overweight, ill-suited lenses -- it was a less complex decision to identify the path that most closely resembled an open standard. [1] http://www.digitalsecrets.net/secrets/FullFrameWars.html http://www.digitalsecrets.net/secrets/FullFrameWars.html

_ph_ 9y ago
The article is still quite valid. The sensor prices might have come down a bit further than expected, but a full frame camera still carries a substantial premium, especially when compared to µFT. But more important than the sensor prices are the consequences for the lenses, if you compare the sensor formats. The camera companies have solved the problems with vignetting and soft corners by designing a whole set of new lenses[1]. So a 35mm sensor performs nicely with modern lenses, but those lenses are quite expensive and most importantly, vastly bigger and heavier than their ancestors. So while a 35mm system might provide the maximum quality, for anyone with a budget, both financially and transportation-wise, the µFT system is more in the spirit of analog 35mm format, both size and performance wise, while the digital 35mm cameras drift more into the classic medium format field, in performance but importantly also price and size wise. [1] Meanwhile Leica managed to bring their digital M cameras to 35mm, getting excellent image quality even in the corners with a mix out of clever sensor design and in-camera post processing.
- achamayou 9y ago
  The M8 was an anomaly more than anything else, and probably used an APS-H sensor because Leica couldn't get a 24x36 sensor from anyone. It was a flawed camera in many ways, and they went back to 35mm as soon as they could get a sensor that size on the M9 (from Kodak iirc).
  - _ph_ 9y ago
    Actually, both the M8 and the M9 use a Kodak sensor. They are even using the same sensor design, the M9 just having proportionally more pixels according the larger sensor area. But the individual pixel design is the same in both sensors (it is also the same sensor design used in the first FT-camera, the Olympus E-1, there of course only featuring 4.9 megapixels, due to the smaller sensor size). There were considerable difficulties getting a 35mm sensor working in a Leica M camera, due to the light reaching the corners of the sensor at very oblique angles. The M9 used strong offset microlenses on the sensor to fight with that, the more recent CMOS sensors are very "shallow" and thus less sensitive to the light angles. Also, the design of the infrared filter created a lot of problems. The M8 had a to weak one - which resulted in Leica giving free infrared filters for each M8 owner. The M9 one worked, but turned out to be insufficient stable, so many M9s got a free sensor exchange (fortunately free even out of warranty). With the M(240) and especially the M10 Leica finally seem to have fully succeed in making a true digital M. The M10 is especial, as the body size is finally identical to the film based M cameras, the M8-M(240) were about 3-4 mm thicker.
- petre 9y ago
  I used to dismiss Leica M as photo gear for Magnum photographers and rich snobs, but then they began manufacturing excellent lenses for the µFT system at more acceptable prices. If an M Summicron is 3.3k, the equivalent µFT lens costs $500 - less than the Zeiss Tuit Sony E lenses, which got not so great reviews.
  - eropple 9y ago
    Leica also sells the Nocticron for a little over a grand, and that lens is a jaw-dropping MFT spin on the Noctilux. I mostly do video, but I really love MFT for photos too.

by Jedd 9y ago

_ph_ 9y ago

The article is still quite valid. The sensor prices might have come down a bit further than expected, but a full frame camera still carries a substantial premium, especially when compared to µFT. But more important than the sensor prices are the consequences for the lenses, if you compare the sensor formats. The camera companies have solved the problems with vignetting and soft corners by designing a whole set of new lenses[1]. So a 35mm sensor performs nicely with modern lenses, but those lenses are quite expensive and most importantly, vastly bigger and heavier than their ancestors. So while a 35mm system might provide the maximum quality, for anyone with a budget, both financially and transportation-wise, the µFT system is more in the spirit of analog 35mm format, both size and performance wise, while the digital 35mm cameras drift more into the classic medium format field, in performance but importantly also price and size wise. [1] Meanwhile Leica managed to bring their digital M cameras to 35mm, getting excellent image quality even in the corners with a mix out of clever sensor design and in-camera post processing.

achamayou 9y ago
The M8 was an anomaly more than anything else, and probably used an APS-H sensor because Leica couldn't get a 24x36 sensor from anyone. It was a flawed camera in many ways, and they went back to 35mm as soon as they could get a sensor that size on the M9 (from Kodak iirc).
- _ph_ 9y ago
  Actually, both the M8 and the M9 use a Kodak sensor. They are even using the same sensor design, the M9 just having proportionally more pixels according the larger sensor area. But the individual pixel design is the same in both sensors (it is also the same sensor design used in the first FT-camera, the Olympus E-1, there of course only featuring 4.9 megapixels, due to the smaller sensor size). There were considerable difficulties getting a 35mm sensor working in a Leica M camera, due to the light reaching the corners of the sensor at very oblique angles. The M9 used strong offset microlenses on the sensor to fight with that, the more recent CMOS sensors are very "shallow" and thus less sensitive to the light angles. Also, the design of the infrared filter created a lot of problems. The M8 had a to weak one - which resulted in Leica giving free infrared filters for each M8 owner. The M9 one worked, but turned out to be insufficient stable, so many M9s got a free sensor exchange (fortunately free even out of warranty). With the M(240) and especially the M10 Leica finally seem to have fully succeed in making a true digital M. The M10 is especial, as the body size is finally identical to the film based M cameras, the M8-M(240) were about 3-4 mm thicker.
petre 9y ago
I used to dismiss Leica M as photo gear for Magnum photographers and rich snobs, but then they began manufacturing excellent lenses for the µFT system at more acceptable prices. If an M Summicron is 3.3k, the equivalent µFT lens costs $500 - less than the Zeiss Tuit Sony E lenses, which got not so great reviews.
- eropple 9y ago
  Leica also sells the Nocticron for a little over a grand, and that lens is a jaw-dropping MFT spin on the Noctilux. I mostly do video, but I really love MFT for photos too.