According to this popular blogger, M43 sensor is 60% smaller than APS-C. Is this true?

Status
Not open for further replies.

Mellow

Mu-43 All-Pro
Joined
Aug 27, 2010
Messages
1,122
Location
Florida or Idaho
Real Name
Tom
Sigh.

The sensor size is a meme over in the NEX forums too; they just can't get over it. Bottom line, when most people compare images taken with the best m43 sensors and best APS-C sensors, they can't tell the difference. Hell, there are even comparisons of the E-M1 and the full-frame A7/A7R in which the differences are shown to be very slight.

Here's the problem for those APS-C guys: they have to believe the IQ of their cameras is a lot better to compensate for the undeniable fact that their lenses are bigger. This is even more true of the FF converts now, because their lenses are going to be (notice tense LOL!) a LOT bigger. So the two advantages they will trumpet, over and over again are: better IQ (especially in low light) and shallower DOF.

As reviews have come in showing that the actual difference in IQ between the best m43 and best APS-C cameras is very slight, they now resort to something that is in fact actually measurably larger about APS-C: the sensor. They make the logically unsound argument that IQ is linearly correlated with sensor size, so their much bigger sensor must therefore correspond to much better IQ. Only two problems with this argument:

(1) there's not a linear relationship between sensor size and IQ (however you measure it);
(2) their sensor is not, in fact, all that much larger, especially when you consider the different aspect ratios. Ned (where's he been?) had a great graphic showing this; you might want to do a search to find it.
 

Reesebass

Mu-43 Regular
Joined
Nov 5, 2013
Messages
72
I agree with all that, but is that guy deliberately posting false information. My calculations show than the gap between micro four thirds and APS-C is much smaller than 65%.
Edit: im not trolling btw, just wanted to know if that blogger is a troll or serious. I mean people read that site and he gets money out of this.
 

MAubrey

Photographer
Joined
Jul 9, 2012
Messages
1,469
Location
Bellingham, WA
Real Name
Mike Aubrey
If you go to this wikipedia page, you can do the math yourself: http://en.wikipedia.org/wiki/Image_sensor_format

But in short, no. That guy has the numbers backwards. Micro Four Thirds has 61% of the area of Nikon, Sony, Pentax's APS-C and 68% the area of Canon's APS-C. The crop factor for Canon APS-C is 1.25x (i.e. for the same angle of view, you need a focal length that is 1.25x longer), So a 25mm on μ43 is equivalent in AOV to a 31mm lens on Canon APS-C. The crop factor for Nikon/Sony/Pentax APS-C is 1.33x, so for them a 33mm lens is equivalent to a μ43 25mm.

Aside from that, diglloyd's article truly a beautiful piece of trolling. He's more interested in getting a rise out of μ43 users than any thing else. Look at what he says:

The M4/3 sensor size makes possible elegantly small compact cameras with high image quality but these do not exist.

This is demonstrably false.

The M4/3 sensor size makes possible ƒ/1.0 lenses with high quality, but these do not exist except as oddball specialty lenses made mainly for video.

There's nothing "oddball" about the Voigtlander lenses

The M4/3 sensor size allows ƒ/2 lenses that are all but perfect wide open. These do not exist.

I don't know what world he's living in, but this is demonstrably false, too. In fact, we've got a bunch of f/1.8 lenses that are all but perfect wide open.

M4/3 lenses are not any less expensive than APS-C!

Other than Pentax and Fuji few system really have any decent primes explicitly designed for APS-C. Canon has none. Nikon has one. Sony has a few for NEX, but the cheap ones are cheap in more than price and the expensive ones are, well, expensive.

Anyway, I'd encourage you to ignore him.
 

thazooo

Mu-43 Regular
Joined
May 24, 2012
Messages
63
Location
Mississippi, USA
Real Name
Dana F.
I shoot both, an E=PL1 and a current APS-C. As mentioned there is a slight difference in quality. More noticeable since I'm using an older model m4/3. On the other hand the SOOC jpg blow away the jpg that come out of the APS-C. The only time I see it mattering is if you're printing a bill board :)
There are several Known name Photographers praising the current M4/3 cameras, so that speaks to the quality of M4/3.

Dana
 

b_rubenstein

Mu-43 All-Pro
Joined
Mar 20, 2012
Messages
1,446
Location
Melbourne, FL
The nominal size of a 4/3 sensor is 17.30mm x 13.00mm. An APS-C (not Canon) is 23.70mm x 15.60mm. That calculates to difference in area to about 65%. (% difference calculation gives different numbers depending on going from smaller to larger vs. larger to smaller.)

In and of itself, that a simple numeric fact and doesn't mean anything in terms of sensor performance. Now, if two sensors were fabricated with identical properties and the only difference was area, then the larger one would have 2/3 stop better noise performance. Sensors with identical pixel pitches have the same resolution capturing capability.
 

Reesebass

Mu-43 Regular
Joined
Nov 5, 2013
Messages
72
Yeah, we all know that. My problem is the guy gets paid for posting deceptive information like for example micro four thirds is cheating on the lens speed etc. He always insists that APS-C is 65% larger, but in reality it's about 30%.
 

RichDesmond

Mu-43 Top Veteran
Joined
Nov 18, 2011
Messages
573
Location
United States
Technically, he's right about the area. The area of the APS-C sensor is 369.7mm^2, m4/3 is 224.9mm^2. Formula for % change is ((x2 / x1) - 1) * 100. So ((369.7 / 224.9) - 1) * 100 = 64.4%

Of course, his larger point about IQ is completely false. Classic example of the misuse of statistics.
 

Reesebass

Mu-43 Regular
Joined
Nov 5, 2013
Messages
72
Technically, he's right about the area. The area of the APS-C sensor is 369.7mm^2, m4/3 is 224.9mm^2. Formula for % change is ((x2 / x1) - 1) * 100. So ((369.7 / 224.9) - 1) * 100 = 64.4%

Of course, his larger point about IQ is completely false. Classic example of the misuse of statistics.
Yes 65% is percentage of mft sensor area within APS-C, but not 65% bigger as he always insists,
 

dhazeghi

Mu-43 Hall of Famer
Joined
Aug 6, 2010
Messages
4,381
Location
San Jose, CA
Real Name
Dara
As reviews have come in showing that the actual difference in IQ between the best m43 and best APS-C cameras is very slight, they now resort to something that is in fact actually measurably larger about APS-C: the sensor. They make the logically unsound argument that IQ is linearly correlated with sensor size, so their much bigger sensor must therefore correspond to much better IQ. Only two problems with this argument:

(1) there's not a linear relationship between sensor size and IQ (however you measure it);
(2) their sensor is not, in fact, all that much larger, especially when you consider the different aspect ratios. Ned (where's he been?) had a great graphic showing this; you might want to do a search to find it.
Here's a visual I put together on sensor size:
Subscribe to see EXIF info for this image (if available)


I agree that by and large, people overstate the arguments for larger sensors. That said, playing devil's advocate for the moment, if you could get a 2/3 stop improvement in noise and DR, for a 20% increase in overall lens size and no additional cost, wouldn't that be a pretty significant selling point? I mean, we spend often double to get a lens that's 2/3 stop faster, so if we got that improvement for free thanks to a larger sensor, that would be pretty nice. At the moment, it's true that the gap in sensor performance is smaller than size would otherwise suggest, but I think this is simply an artifact of timing - Sony has yet to really optimize their 24MP APS-C sensors, while their 16MP m4/3 sensors have been well tuned.

Of course, this argument assumes several things, namely that the APS-C mirrorless makers can and will catch up when it comes to basic features like autofocus, and that the lens size differential can be kept within reasonable bounds. The former I expect will happen, although more slowly for some (Fuji) than others (Sony). The lens size thing is the real question for me, and the evidence so far is fairly mixed. My guess is the 20%-or-less size differential can be maintained for anything wider than 100mm, but for longer telephotos it's probably not feasible.

Yes 65% is percentage of mft sensor area within APS-C, but not 65% bigger as he always insists,
It's a matter of how you parse it, but I think most common usage would say he is correct. If you add 65% to the size of the m4/3 sensor, you get the size of the APS-C sensor. In my book, that's what 65% bigger means.
 

Reesebass

Mu-43 Regular
Joined
Nov 5, 2013
Messages
72
It's a matter of how you parse it, but I think most common usage would say he is correct. If you add 65% to the size of the m4/3 sensor, you get the size of the APS-C sensor. In my book, that's what 65% bigger means.
Damn, did my calculations, the guy is right! So is wikipedia wrong by stating that "its area, ca. 220 mm², is approximately 30% less than the APS-C sensors used in other manufacturers' DSLRs."
 

zlatko-photo

Mu-43 Veteran
Joined
Jan 8, 2014
Messages
228
It's a matter of how you parse it, but I think most common usage would say he is correct. If you add 65% to the size of the m4/3 sensor, you get the size of the APS-C sensor. In my book, that's what 65% bigger means.
In that light, it's fair to say the OP asked a fair question. I wish people would stop calling each other "trolls". That's terribly mean and dismissive. To people who use that word: would you call someone that in person? Someone should apologize to the OP.
 

Reesebass

Mu-43 Regular
Joined
Nov 5, 2013
Messages
72
In that light, it's fair to say the OP asked a fair question. I wish people would stop calling each other "trolls". That's terribly mean and dismissive. To people who use that word: would you call someone that in person? Someone should apologize to the OP.
Thanks. But this is internet after all, so someone calling me a troll won't bother me at all.
I should have used a simple percentage formula first place to get my answer, but i still think that blogger is full of ****.
Quite hard to believe that APS-C sensor is 65% bigger, doesn't seem that way if you look at visual comparison, but it's true.
 

briloop

Mu-43 Regular
Joined
May 23, 2012
Messages
171
Location
Mount Juliet, TN
...micro four thirds is doomed ...
If micro four thirds is doomed, then:

Why are Blackmagic and Kodak making bodies for the system?

Why are Schneider Kreuznach, Sigma, Voigtlander, SLR Magic, Mitakon, Samyang, and Tokina making lenses for the system?

Not to mention Olympus and Panasonic.

At least two bodies and three lenses announced in January 2014. A 4K camera is rumored to be announced next week.

If the system is doomed, then there's a quite few stupid decision-makers out there in the corporate world.
 

lightmonkey

Mu-43 Veteran
Joined
Dec 22, 2013
Messages
479
Yes mft is visibly worse than apsc at 100% for SOOC. Not enough to matter for me. At all. Even 1" which is significantly smaller than both is sufficient below ISO800.

For most people sensor size is not a bottleneck. Just fodder for argument
 

Matero

Mu-43 Top Veteran
Joined
Jun 22, 2013
Messages
528
Location
Finland
Damn, did my calculations, the guy is right! So is wikipedia wrong by stating that "its area, ca. 220 mm², is approximately 30% less than the APS-C sensors used in other manufacturers' DSLRs."
Mmmh, "How many percent smaller in area is m43 sensor than APS-C sensor?"

(370,52-224,90)/370,52 x 100% ≈ 39%

If 9%-unit difference is fair approximation, then I would say wikipedia is not wrong :smile: I would say approx. 40% smaller...
 

Reesebass

Mu-43 Regular
Joined
Nov 5, 2013
Messages
72
So how is 65% bigger at the same time 40% smaller? I really should have paid attention during math.
 
Status
Not open for further replies.
Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.
Mu-43 is a fan site and not associated with Olympus, Panasonic, or other manufacturers mentioned on this site.
Forum post reactions by Twemoji: https://github.com/twitter/twemoji
Copyright © 2009-2019 Amin Forums, LLC
Top Bottom