[JJ_SO]

Are you trying to confuse me?   

 

If the aperture is f/2.8 and the shutter speed of two systems is the same, ISO will also be the same - the effect will be different because of the multiplication to "the (bloody) same size of picture" - this idea of "same size picture" is wrong!

 

Why fooling around with multiplication factors and pretend the only thing what counts is a final image of THE SAME size for all equivalenced sensor? One doesn't need to be a scientist to remember the effect of different film format sizes but that shitty waste of time called "equivalence" NEVER mattered in times of analogue film - why this ongoing useless debate today while none of the equivalence preachers can truly deliver the exact equivalence - which also goes for the amplification of different sensors. Who ever said a small smartphone sensor has to deliver the same poster size picture like a medium format?

 

Working with false assumptions is not helpful

[wim]

I'm sorry, but I can't leave this uncommented:



Right, and one other is ISO. Equivalency is about comparing the end result, the final image, and you can't ignore ISO, or to be more precise the level of noise in it.


So, in practical terms, with a 400/5.6, you'd just raise ISO by two stops and end up with the same DOF, same FOV, same shutter speed and same noise level as with the 200/2.8 on m43.

 

The question really comes down to what is acceptable to a user. Personally, I still shoot 5D II, and yes, you lose some detail from 6400 iso and up, although 6400 iso is almost acceptable. With the Pen-F and E-M1 II it gets to that level slightly above 3200 iso,. IOW, same ISO for both I will use when pushed, namely 3200 iso.

 

Personally, I do not care, I do not need crazy ISOs. I used to shoot at concerts just fine with Tri-X pushed to 1600 ISOo, and I can assure you it was a lot grainier and had much less IQ, let alone DR, than anything I have ever done with digital.

 

I also appreciate the fact that even the GF2, the MFT camera I started with 6, 7 years ago, was way, way, way better than the 350D I started with as a dslr 11 or so years ago.

 

In the end it is horses for courses, as it is all about what the prerequisites are for specific shooting conditions. In my case it often is about space available, IOW, size of package and weight. MFT wins hands down in that regard over FF. 12 lenses and 2 bodies for MFT take up about the same space as 1 FF body with 4 lenses, at least the stuff I tend to carry around. Do note: IQ is absolutely excellent with MFT, for real life purposes (i.e., say, 40 cm x 60 cm or 45 cm x 60 cm prints).

 

From my POV equivalency is only useful in the end in order to establish what aperture you need to get equivalent DoF, or to establish whether you are happy with the embedded noise. Personally, I took that decision, for myself anyway, years ago, because, again, it is about needs and conditions.

 

In short, equivalency is only part of all the practical terms, and only a minor part for that matter, and in the end is only a matter of tech geekery rather than actual photography. Photography has always been, for me anyway, to get the best about what you use. Photography to me is about art, realizing your vision, not to get the ultimate representation of reality (which the human eye cannot perceive anyway).

 

This doesn't mean we should not strive for (technical) perfection, because it always helps to have good tools, and they will allow for leeway when conditions aren't perfect, but the best tools also depend on the circumstances for which they are used. It always is a balance. For me that means I know when I should use FF, and to know when I am better off using a different format, like MFT. Funnily enough, considering the iQ the modern tools provide, most often MFT will do much better than just nicely.

 

Kind regards, Wim

[thxbb12]

Are you trying to confuse me?   

 

If the aperture is f/2.8 and the shutter speed of two systems is the same, ISO will also be the same - the effect will be different because of the multiplication to "the (bloody) same size of picture" - this idea of "same size picture" is wrong!

 

Why fooling around with multiplication factors and pretend the only thing what counts is a final image of THE SAME size for all equivalenced sensor? One doesn't need to be a scientist to remember the effect of different film format sizes but that shitty waste of time called "equivalence" NEVER mattered in times of analogue film - why this ongoing useless debate today while none of the equivalence preachers can truly deliver the exact equivalence - which also goes for the amplification of different sensors. Who ever said a small smartphone sensor has to deliver the same poster size picture like a medium format?

 

Working with false assumptions is not helpful

 

I've never shot film and I can tell you equivalence is extremely helpful. Especially today with the numerous camera systems using various sensor sizes.

 

BC's explanation above is crystal clear and simple. It's got nothing to do with assumption, it's merely a way of comparing systems in a meaningful way.

 

It's arguably the best way to compare systems in terms of outputs.

[mst]

Are you trying to confuse me?   

Not intentionally

If the aperture is f/2.8 and the shutter speed of two systems is the same, ISO will also be the same

Yes, but that's not what we're talking about. It's about creating the same result with two different systems, and that means comparing images created with equivalent, not same settings.

Your point was: you can shoot with f/2.8 on m43 and thus have faster shutter speed compared to shooting with a f/5.6 lens with equivalent focal length on a full frame camera.

My point was: you're ignoring ISO, because on the full frame camera (assuming similar resolution and sensor on roughly the same engineering level), you just bump up ISO by two stops, because that will lead to the same image quality in terms of noise and result in the same shutter speed, too.

In other words: there is no speed advantage of a 200/2.8 on m43 over a 400/5.6 on full frame.

[Brightcolours]

Equivalence ALWAYS mattered "in times <sic> of analogue film". Photographers who used to shoot with 6x9 folder cameras and wanted to start using a small 135 format Leica rangefinder instead, indeed looked at what would be an equivalent lens for what they used on their old 6x9 MF folder.

Photographers who used large format technical cameras and a Nikon F depending on the job, of course knew what was equivalent from one format to the other.

And those photographers NEVER argued that one should shoot each format at ISO 100 and f2.8 exclusively, and that somehow f2.8 meant "exposure".

[MatjazO]

Silly me, and I was thinking photography is about subject, light and right moment.

[Brightcolours]

You seriously mean that you never consider FOV and DOF? So basically you can be happy with a smartphone with a single FOV lens?  :o

[mst]

Silly me, and I was thinking photography is about subject, light and right moment.

Of course. But as BC already pointed out: at least for me, the focal length and the aperture (or the corresponding DOF) are my main settings I adjust on the camera to create the effect I want. Sometimes shutter speed, too, if it's about freezing motion (or the contrary, when panning for example).

[JJ_SO]

Be it FF or µ 4/3, at base ISO I want to see the guy who is able to detect a difference for quite a range of output formats. Therefore I see the calculations with ISO extremely useless. Sensors are an ongoing development, today we see sensors we never thought about going eater that high in ISO or that low in noise or that high in DR - and each system has rather particular advantages.

 

And being able to shoot at f/2.8 instead of f/5.6 IS a speed advantage.

 

Trying to equivalence systems which are in no way scaled in the same proportions, on every single aspect which matters, will lead to meaningless results, simply because there are no mathematically perfectly scaled two systems. That's what I'm talking about when I say false assumptions. There are cameras around with small sensors which make bigger ones look pretty pale. There is post production of RAW. There are back and front lit sensors. Lenses cannot be scaled equally otherwise I already would have seen the same results. Production tolerances - you guys pretend these are also scaled, but the machines for the medium format lenses are the same like the ones for µ 4/3. There are so many factors in these theoretical assumptions that I can't believe you're not able to really that reality doesn't follow mathematical models, especially these kind of simple linear ones.

 

Try to scale down a Porsche to a 1/30 scale - the theoretical speed should be around 300/30 = 10 km/h (166.66 m/min) then? Now you also scale down the engine, the screws and the manufacturing tolerances. A nozzle in the engine of ø 1 mm with tolerance of ± 0.03 mm will become ø 0.0333 mm and a tolerance of 0.0001 mm - good luck in manufacturing that... And how do you scale down the petrol's energy to about 1/30 of it's power? How do you weld the sheet metal from 1.5 mm to 0.03 mm? And get the stability of it? Think your "scientific" equation until it's end and with all components, not just the selected ones you like to talk about.

 

Equivalence is not helping me in any way, so why bother? Like Wim says, most things can be done with µ 4/3, others better with FF, fine, what more do I need to know? Telling others "oh this 200/2.8 is pretty lame, on FF it's only 400/5.6" is soooo useless, that I just start with counting:

1. how many fast lenses for µ 4/3 are around to choose from? µ 4/3 owners will be pretty happy about, at least the ones with big enough pockets
   
2. how many really competing FF lenses 400/5.6 are around? Some old bottle bottoms and the rest is already 400/4 and not so easy to finance. That makes this comparison pointless.
   
3. I daresay a FF lens in this quality at 400/5.6 won't be much cheaper.
   
4. But might come with a more nervous bokeh.
   
5. but will have a much longer minimal distance
   
6. and need to bump up ISO by two stops, that can make the difference between a nice memory and a shot to sell
   
7. thxbb12, have you used a 400 mm on your FF camera, before you went mirrorless? It's a question to find out if equivalence in this case was helping you much
   

Very often when I read this kind of equivalence statements, it's simply not helping the people for who the lens is addressed to, often the lens choices are limited in mirrorless land. People who already started with µ 4/3 don't benefit - they read "400mm" think, oh that must be big - and when you mount a 400 on a FF it's pretty disappointing, because even with 600 mm the birds stays a small animal.


Be it FF or µ 4/3, at base ISO I want to see the guy who is able to detect a difference for quite a range of output formats. Therefore I see the calculations with ISO extremely useless. Sensors are an ongoing development, today we see sensors we never thought about going eater that high in ISO or that low in noise or that high in DR - and each system has rather particular advantages.

 

Trying to equivalence systems which are in no way scaled in the same proportions, on every single aspect which matters, will lead to meaningless results, simply because there are no mathematically perfectly scaled two systems. That's what I'm talking about when I say false assumptions. There are cameras around with small sensors which make bigger ones look pretty pale. There is post production of RAW. There are back and front sensors. Lenses cannot be scaled equally otherwise I already would have seen the same results. Production tolerances - you guys pretend these are also scaled, but the machines for the medium format lenses are the same like the ones for µ 4/3. There are so many factors in these theoretical assumptions that I can't believe you're not able to really that reality doesn't follow mathematical models, especially these kind of simple linear ones.

 

Try to scale down a Porsche to a 1/30 scale - the theoretical speed should be around 300/30 = 10 km/h (166.66 m/min) then? Now you also scale down the engine, the screws and the manufacturing tolerances. A nozzle in the engine of ø 1 mm with tolerance of ± 0.03 mm will become ø 0.0333 mm and a tolerance of 0.0001 mm - good luck in manufacturing that... And how do you scale down the petrol's energy to about 1/30 of it's power? How do you weld the sheet metal from 1.5 mm to 0.03 mm? And get the stability of it? Think your "scientific" equation until it's end and with all components, not just the selected ones you like to talk about.

 

Equivalence is not helping me in any way, so why bother? Like Wim says, most things can be done with µ 4/3, others better with FF, fine, what more do I need to know? Telling others "oh this 200/2.8 is pretty lame, on FF it's only 400/5.6" is soooo useless, that I just start with counting:

1. how many fast lenses for µ 4/3 are around to choose from? µ 4/3 owners will be pretty happy about, at least the ones with big enough pockets
   
2. how many really competing FF lenses 400/5.6 are around? Some old bottle bottoms and the rest is already 400/4 and not so easy to finance. That makes this comparison pointless.
   
3. I daresay a FF lens in this quality at 400/5.6 won't be much cheaper.
   
4. But might come with a more nervous bokeh.
   
5. but will have a much longer minimal distance
   
6. and need to bump up ISO by two stops, that can make the difference between a nice memory and a shot to sell
   
7. thxbb12, have you used a 400 mm on your FF camera, before you went mirrorless? It's a question to find out if equivalence in this case was helping you much
   

Very often when I read this kind of equivalence statements, it's simply not helping the people for who the lens is addressed to, often the lens choices are limited in mirrorless land. People who already started with µ 4/3 don't benefit - they read "400mm" think, oh that must be big - and when you mount a 400 on a FF it's pretty disappointing, because even with 600 mm the birds stays a small animal.

[Brightcolours]

Be it FF or µ 4/3, at base ISO I want to see the guy who is able to detect a difference for quite a range of output formats. Therefore I see the calculations with ISO extremely useless. 

Be it FF at ISO 400 or MFT at ISO 100 I want to see the guy who wants to argue that they are able to detect a difference. Therefore I see this constant hammering of "metering has to be done at base ISO/ISO100" as particularly silly.