How much better can M43 sensors get?

F

filmguy123

Veteran
In terms of DR, highlight handling/roll-off, resolution, and lowlight?

Or are most of those properties contingent on sensor size?

What I mean is... would it ever be possible for a M43 sensor to be as good in lowlight as a 5D Mark II? to have more DR than the current BMC cameras? etc
 
If you're asking if an m43 sensor can match or exceed today's 5D Mark II -- of course it will be possible. The question is when...
 
Yes that's what I was wondering. Not so much wondering if it was around the corner, but from an engineering stand-point.

Do you mind me asking, how/why is that possible? Are these factors not contingent on sensor size? Do you mean to say that a m43 sensor size, in time, could exhibit almost identical qualities of a FF sensor in every way except FoV? If so, could you point me to some links or take just a minute to explain how/why that is possible?

Would love to understand that better! Thanks Barry.
 
As technology improves, quality improves. Right now you can get drastically better video out of a $1800 1/4" sensor camera, than we used to be able to get out of a $50,000 2/3" BetaSP camera.

The processor in your phone is probably vastly more powerful than the best desktop computer of 1998.

Obviously an m43 sensor won't be able to match the FOV of a larger sensor, but in every other category I would expect that continual improvements would mean that sooner or later, an m43-sized sensor will exhibit better sharpness and more dynamic range than today's 5D Mark II.
 
Well, I know that in a general sense! But there are also limits to physics. For example, the processor in my phone IS vastly more powerful than my desktop in 1998. However, that's because it is made on 22nm processor instead of 130nm processor. That's the equivalent of a sensor size change.

So my question is, in a specific or technical sense, how is it feasible for an m43 sensor to have the light gathering capability of FF? What technologies may enable this, and how far can they be pressed (based on our current knowledge). Of course there are always unforeseen developments, but it is a relevant question for the next 3-5 years: what specifically makes improved DR, lowlight, resolution and sharpness possible without an increase in sensor size that is on the horizon? Fancy new backlit sensor methods? Or...?

Or is that just speculation? "technology improves, so m43 will probably be like a mark II someday". Guess I was looking for a more thorough dissection, akin to something i'd find on anandtech :) If you've got anything, I would certainly appreciate it!
 
filmguy123 said:
How is it feasible for an m43 sensor to have the light gathering capability of FF?
Click to expand...

- In terms of the usable sensor area, a Full Frame sensor has a 2 stop advantage over a Micro 4/3 sensor.

- But then you have to factor in usable DOF when shooting with these two formats, where the Micro 4/3 format has a 2 stop DOF advantage over the Full Frame sensor.

When shooting with my GH2/GH3 cameras, my lens aperture is almost always in the f/2.0 - f/4.0 range, because these are the apertures that will provide me with usable DOF for my shots.

If I were to switch to a Canon 5D Mk3, my lens apertures would likely be in the f/4.0 - f/8.0 range so I would obtain the same usable DOF for my shots. The higher sensitivity of the 5D Mk3's sensor is offset by my desire for usable DOF, so generally there's not much advantage for me to shoot with a Full Frame sensor camera.

In extreme situations where I cannot light my subjects and am forced to shoot in very dim conditions, the 5D Mk3 would have a distinct advantage over the GH3 camera, but this is a very rare event for the type of work I shoot. ( corporate/medical training/interviews )

...I've noticed many very shallow DOF shots in the FX TV series "Wilfred", which is shot with Nikon D800 cameras, where I'm wondering if they are intentionally going for this very shallow DOF look, or is it simply easier to shoot shallow DOF so they don't have to boost their lighting to get deeper DOF in their shots.
 
Other than sensor size there are 2 other things things that add to the quality of a sensor. Resolution and sensitivity. If you think of the new Nokia Lumia 1020 camera phone - it has a much smaller sensor than M4/3 but great low light sensitivity and 41 megapixel resolution (okay, this is marketing nonsense but it is still high resolution).

As technology improves and becomes more affordable these kind of features will also become available in the M4/3 sensor cameras.
 
filmguy123 said:
In terms of DR, highlight handling/roll-off, resolution, and lowlight?

Or are most of those properties contingent on sensor size?

What I mean is... would it ever be possible for a M43 sensor to be as good in lowlight as a 5D Mark II? to have more DR than the current BMC cameras? etc
Click to expand...

Not only possible, but inevitable.

This is exactly the same question I've heard about physical film (remember that stuff?) for as long as I can remember. I heard it in the 1960s, 70s, 80s, 90s, and the 2000s. The people asking the questions were generally unbelieving and convinced that the current technology couldn't possibly be improved upon. The scientists and engineers working for Kodak and Fuji knew better though. As a result of their tireless efforts film continued to get better and better. Kodak's Vision 3 films are the result of 100 years of continuous R&D and manufacturing improvement. They are so much better than the film stocks from the 1970s it's hard to see how the two are related at all.

In just the last few years the question was one of "can digital possibly equal film?". Now that film has all but been beating into the ground, we get the inevitable questions about sensor sizes.

What is it about today's technology that makes you think that our scientists and engineers won't improve upon it tomorrow? What is it about history that you don't believe? And really, why is that?
 
Bruce Watson said:
What is it about today's technology that makes you think that our scientists and engineers won't improve upon it tomorrow? What is it about history that you don't believe? And really, why is that?
Click to expand...

absolutely nothing! I just want to know the specifics, not a broad "it will get better because". Wondering what is specifically done on sensors that allow a smaller sensor to become more sensitive, improve resolution and DR, etc.
 
Bruce Watson said:
Not only possible, but inevitable.

This is exactly the same question I've heard about physical film (remember that stuff?) for as long as I can remember. I heard it in the 1960s, 70s, 80s, 90s, and the 2000s. The people asking the questions were generally unbelieving and convinced that the current technology couldn't possibly be improved upon. The scientists and engineers working for Kodak and Fuji knew better though. As a result of their tireless efforts film continued to get better and better. Kodak's Vision 3 films are the result of 100 years of continuous R&D and manufacturing improvement. They are so much better than the film stocks from the 1970s it's hard to see how the two are related at all.

In just the last few years the question was one of "can digital possibly equal film?". Now that film has all but been beating into the ground, we get the inevitable questions about sensor sizes.

What is it about today's technology that makes you think that our scientists and engineers won't improve upon it tomorrow? What is it about history that you don't believe? And really, why is that?
Click to expand...

A big +1. -- (Except that film is not as near dead as some people think. For instance, most of the Oscar nominated films this past awards season were shot on film. For another thing, it's still the best archival medium available from both a cost and longevity standpoint. Both of those will change with time, of course, but in the present, rather than thinking about what's current, dead, or the trend, think about choices.)
 
filmguy123 said:
absolutely nothing! I just want to know the specifics, not a broad "it will get better because". Wondering what is specifically done on sensors that allow a smaller sensor to become more sensitive, improve resolution and DR, etc.
Click to expand...

A Google search can be your best friend.
 
filmguy123 said:
I just want to know the specifics...
Click to expand...

You and every tech. journalist and blogger to be found. Corporate R&D is private by nature. You'll know it when they announce it, not before. Unless you've got good contacts with NSA. Or you're besties with Alfred Bester (yes, yes, that reference is pretty bloody obscure, I'll grant that).

Back in the 1970s people would have been shocked at the T-grain technology Kodak released in the early 1980s. Think of yourself as that guy in the 1970s thinking "I just want to know the specifics...". If they told you, you wouldn't believe it.

Me? I try not to worry about that which I can't control. Or even influence. Unless you work in the right place... And if that's the case you don't have to ask, because you're one of the people working on the next generation(s) of sensors.
 
Last edited:
filmguy123 said:
Yes that's what I was wondering. Not so much wondering if it was around the corner, but from an engineering stand-point.

Do you mind me asking, how/why is that possible? Are these factors not contingent on sensor size? Do you mean to say that a m43 sensor size, in time, could exhibit almost identical qualities of a FF sensor in every way except FoV? If so, could you point me to some links or take just a minute to explain how/why that is possible?

Would love to understand that better! Thanks Barry.
Click to expand...

you already mentioned the Blackmagic camera, that's a camera with a smaller sensor the AF100 but has better resolution and dynamic range than the 5D even with the hack. So yeah, it's pretty east for the m43 standard to improve, look at GH1 to GH3 for example
 
From a technical standpoint just looking at future generations of computer processors can give you an idea. They're combining silicon and germanium which they've found has much better electron mobility so it makes for faster switching. Also from a photo sensitive site, germanium may be able to increase the sensitivity of current photo sensitive materials because any photon strike on a photosensitive site generates electrons which allow that strike to be detected, if current silicon resists the flow of those electrons, that reduces the gain of the photo site, so it won't be as sensitive. So yes, there are many new and exciting materials being looked at that will eventually trickle into photo sensors making them more sensitive thus making the site allowed to be smaller to give the equivalent sensitivity of current larger photo sites, and higher resolution...or keeping them the same size so as to allow lower light capabilities. But those trends are always chosen based on what the manufacturers think they can sell to the masses.
 
You must log in or register to reply here.
Back
Top