Leaving Micro 4/3... But what for?

This is true for photos taken in the ideal circumstances. When I use controlled studio lighting, my Panasonic G2 almost always produces wonderful results, as does my Samsung EX1. Even my cellphone takes wonderful pictures in the studio :D

However, in real-world situations there are many factors that make Micro 4/3 perform significantly worse. The problem No. 1 for me is G2's relatively narrow dynamic range, especially when coupled with its live view histogram which is far from truth. And don't even get me started on Micro 4/3's perennial difficulties with brightly-colored objects (yellow, orange, red, violet etc.) which inevitably lead to single-channel clipping and all kinds of weird color shifts. There was a thread on Mu-43.com dedicated to exactly this issue and it was pretty clear from the image samples in this thread that even GH2 in that respect is closer to small-sensor compact cameras than to APS-C DSLRs and CSCs. For example, I incidentally discovered that there are whole families of flowers whose colors can not be reproduced with reasonable accuracy by any 12 MP M4/3 camera (for example, most species in the genus viola). I found it out when a friend of mine asked me to shoot her prized flowers — I was forced to rent a Canon DSLR to finish this relatively simple task.

Very interesting. As is often the case, it can come back to what an individual shoots, as well as how. There is no doubt in my mind that some of the strengths of other sensors can become apparent in some shots, and of course if there are many small differences, they can cumulatively make a real difference in some circumstances. For example, there are some photographs made in a session where I have used several cameras that I can identify clearly as coming from my full-fram Canon. Many (most) times though, I cannot.

Do you know anyone who has one of the contenders for your new camera? If so, can you borrow it for a day or two? I suggest that you do a real-world shoot-out, using RAW images from the borrowed camera and the G2. Shoot the things you normally do, and of course some of the things where you have experienced challenges. Of course you will pixel-peep the images - but then process them to your preferred output. Look for real differences that are meaningful to you in the finished product.
 
Stratokaster,

As Don mentioned I had a 60D, and it was a great camera. I actually sold it and moved to a T3i/600D when it came out and have been very pleased with that camera. The move was mostly subjective and mostly because I wanted a little smaller form factor as I travel with the camera often. My decision to stay with a DSLR, at least for now, is mostly due to my need for a sports shooting camera, as I have a very active teenage son who plays three sports. That is an area where the m/43s format is still lacking, although it is getting better.

Anyway, give the 600D a good look vs. the 60D and if you don't need the few addtional features of the 60D, the 600D is a great performer. You will find a number of comparison articles floating around the web on that subject. PM me if you have any specific 600D questions. I will take a hard look at the Nex 7, but it will take a lot to get me to switch right now based on my 600D experiences. Good luck.
 
However, in real-world situations there are many factors that make Micro 4/3 perform significantly worse. The problem No. 1 for me is G2's relatively narrow dynamic range, especially when coupled with its live view histogram which is far from truth. And don't even get me started on Micro 4/3's perennial difficulties with brightly-colored objects (yellow, orange, red, violet etc.) which inevitably lead to single-channel clipping and all kinds of weird color shifts. There was a thread on Mu-43.com dedicated to exactly this issue and it was pretty clear from the image samples in this thread that even GH2 in that respect is closer to small-sensor compact cameras than to APS-C DSLRs and CSCs. For example, I incidentally discovered that there are whole families of flowers whose colors can not be reproduced with reasonable accuracy by any 12 MP M4/3 camera (for example, most species in the genus viola). I found it out when a friend of mine asked me to shoot her prized flowers — I was forced to rent a Canon DSLR to finish this relatively simple task.

I don't accept that this is a Micro 4/3 issue. Yes, Micro 4/3 sensors have lower DR than the best current APS-C sensors. However, every camera I've ever owned (including at least a dozen DSLRs from Canon, Nikon, etc) has clipped the colors on bright yellow/orange/red flowers if left to usual autoexposure. Look at any Flickr group for flowers, and you'll see tons of shots like this one, taken with a Sony A700 DSLR: La vall de Boí - Flores en el jardín | Flickr - Photo Sharing!

Canon's sensors aren't great for dynamic range either. Right now, Sony stands alone in that area.
 
I don't accept that this is a Micro 4/3 issue. Yes, Micro 4/3 sensors have lower DR than the best current APS-C sensors. However, every camera I've ever owned (including at least a dozen DSLRs from Canon, Nikon, etc) has clipped the colors on bright yellow/orange/red flowers if left to usual autoexposure. Look at any Flickr group for flowers, and you'll see tons of shots like this one, taken with a Sony A700 DSLR: La vall de Boí - Flores en el jardín | Flickr - Photo Sharing!

Canon's sensors aren't great for dynamic range either. Right now, Sony stands alone in that area.

Right. Which is partly why I suggest a real-world comparison. As I noted in an earlier post, there are circumstances in which various cumulative factors together may make a particular sensor stand out for certain of my photographs. Generally though, for web display and prints to, say, 11 x 17 inches, I think there is relatively little difference. Of course then other factors come into play, such as portability, weatherproofing etc: these may or may not be factors for an individual. There is a the little thing of lenses too...

As far as DxO Mark is concerned, it may be useful in some respects, but the chuckle example there is the M9 - it really doesn't do very well.
 
As far as DxO Mark is concerned, it may be useful in some respects, but the chuckle example there is the M9 - it really doesn't do very well.

That's because it really does not very well in low light. This alone results in a rather low overall score. But CCD sensors generally don't do as well in low light as modern CMOS sensors. So there is no surprise here. In my opinion the overall score of DxOMark is not really a value one should rely on.
 
That's because it really does not very well in low light. This alone results in a rather low overall score. But CCD sensors generally don't do as well in low light as modern CMOS sensors. So there is no surprise here. In my opinion the overall score of DxOMark is not really a value one should rely on.

Yes, but also note that the M9 is pretty much on a par with the GH2 when it comes to dynamic range. Better not get an M9. Dynamic range is like an M4/3. You won't get your shot... ;) :D
 
Yes, but also note that the M9 is pretty much on a par with the GH2 when it comes to dynamic range. Better not get an M9. Dynamic range is like an M4/3. You won't get your shot... ;) :D

Dynamic range is not my main concern with µ4/3.
 
Dynamic range is not my main concern with µ4/3.

Oh, sure, I noted that. My point was part of the larger discussion, as raised by the OP - dynamic range is certainly part of the stated concern. My observation was not aimed at you or your concerns.

My point, really, is that for most output and most photography, the dynamic range differences between M4/3 and it's nearest competitors is slim technically, and probably not noticeable to the eye in final output.

There are many great reasons not to use M4/3, but for most photography DR is not one of them.
 
To be honest, I could live with all technical imperfections of µ4/3. I would really like better ISO 800 and ISO 1600 than I have now, but I suppose that this may be solved by the next sensor generation. The gear is so light, that I would take these problems as part of the compromise I chose. My main problem with µ4/3 is my frustration that it takes weeks and months (sic!) until I can use the lenses I buy for that system, because the service is so slow (missing spare parts, ...). Maybe I just had bad luck, but these problems and the technical imperfections together are too much for investing further into that system.
 
To be honest, I could live with all technical imperfections of µ4/3. I would really like better ISO 800 and ISO 1600 than I have now, but I suppose that this may be solved by the next sensor generation. The gear is so light, that I would take these problems as part of the compromise I chose. My main problem with µ4/3 is my frustration that it takes weeks and months (sic!) until I can use the lenses I buy for that system, because the service is so slow (missing spare parts, ...). Maybe I just had bad luck, but these problems and the technical imperfections together are too much for investing further into that system.

Ouch. I wasn't aware you had those problems. Service issues always leave a deep and lasting impression with me too. I get it that things break or have faults, but if a company wants to play in the global game, then service is every bit as important as the physical product. Sorry to hear of your frustration.
 
Well, maybe my biggest problem with Micro 4/3 is the feeling that I'm not getting the best possible image quality. :D Also I still believe that issues with individual channel clipping are more of a problem for Micro 4/3 than for some other cameras. Maybe I'm wrong, but this is my lasting impression.

Anyway, my friends at Panasonic kindly offered to loan me a Panasonic G3 which has better dynamic range and noise performance than my G2. I will use it for a month and see if I can live with its rather limited movie mode (after all, I can always use my Zoom H1 to record audio and sync later).

I will also try the all-new Olympus 45/1.8 to see if it satisfies my needs for a portrait lens. If both answers are positive, I will not switch systems (mainly to keep the size & weight advantage I currently enjoy).
 
I've shot Canon APS-C (XTi, XSi) and FF (5Dc), Sony NEX3 and m43 (EP1 and EP3) within the last 12 months.

The only real, noticeable difference IMO is FF vs. everything else. APS-C vs. m43 in my experience is very marginally different. Yes, the DR is better in APS-C generally, and yes the noise can be lower, but I still find images from both systems a bit flat feeling. Even putting range finder lenses on the NEX I find to be sub-par. You end up cropping most of the character of the lens right out of the image, and legacy lenses in general on the digital backs lack contrast.

I have bought back a Canon XSi (after selling my 5D to get an EP3) and also an EF 35mm 2.0, because I like the combination of APS-C and that specific lens. The color, contrast, vignetting and DOF of the 35/2 on APS-C is a very nice combo, is relatively compact, and only cost me about $500 combined. But I also bought a couple of old film rangefinders because I don't want to carry a 5D anymore, and I can't afford an M9, so the only way to get FF is film.

My kit(s) for now will be: Pen with some native lenses for walk-around easy shooting (and family snaps), XSi + 35/2 for digital quality shooting, and FF film for the FF experience.

I really hope beyond all else that SOMEONE will release a new FF mirrorless that is less than $2,000 for the body. I wanted Ricoh to have done that with their M module. My next hope is maybe Fuji, but the easy money I think is on Sony. I think I'll pass on the NEX 5n or 7. I just don't see what they really, substantively bring to the table. It's splitting hairs, IMO, to move from m43 to NEX.
 
My Panasonic G3 impressions

I will make it clear: I don't like Panasonic G3.

Its image quality is very good. It produces much better results than my G2. ISO 1600 and even ISO 2500 in incandescent light are perfectly usable. In low daylight, even ISO 6400 can produce acceptable results after some careful processing. AF is much faster when using the kit lens, but with Panny 20mm the difference between G2 and G3 is not really noticeable in real-world shooting.

Dumbed-down control scheme is also fine by me. Having 2 customizable buttons means that I can adjust the camera to my tastes.

This may sound strange, but I'm pleased by the removal of the eye sensor.

Video mode is better than G2. The camera now features a stereo microphone. It even has a 3-level sound adjustment, which is nice. However, it doesn't accept external mics... Oh well, I could live with that.

However, I absolutely hate three things about the camera.

1. Build quality is much worse than that of the G2. For example, battery cover wiggles slightly every time I take the camera in my hands. The display hinge is also noticeably less secure.

2. The click-wheel is now much smaller and made of cheap hard plastic. It's also much harder to turn. The overall experience is similar to hard plastic gears found in some cheap Chinese toys. And its position, because of the smaller size of the camera, is awkward. Every time I use the click-wheel on the G3, I'm afraid of dropping the camera.

3. Battery life. The first time I used the camera, I spent significant time customizing it to my preferences and managed to get only about 200 shots before the camera turned off. In regular use, I found I can make about 320 shots per charge. This is, again, significantly worse than Panasonic G2 which regularly manages 500 shots per charge.

I guess, overall G3 is a nice camera. It just doesn't excite me.

I definitely sense a NEX-7 in my future...
 
Back
Top