Olympus-OM
[Top] [All Lists]

Re: [OM] Fwd: A question about sensors

Subject: Re: [OM] Fwd: A question about sensors
From: Moose <olymoose@xxxxxxxxx>
Date: Sat, 14 Jan 2017 22:43:51 -0800
On 1/12/2017 6:10 PM, Mike Gordon via olympus wrote:
The best FF sensors (Sonny) will outperform the best MFT sensors in 
measurements.  The  physics of 4X area with more overall signal will eventually 
trump the gains in engineering especially as one pushes up the ISO.


https://www.dxomark.com/Cameras/Compare/Side-by-side/Sony-A7R-II-versus-Olympus-OM-D-E-M5-Mark-II___1035_1006

OK, freak me out. :-) I see something like this, and feel the panic of inadequacy start to arise. But, wait, many, perhaps a majority, of my favorite images would not exist if I had this camera instead of my OM-Ds. The American Goldfinch of a few days ago? Nope. What 800 mm lens would I have, be able to carry, let alone hand hold? I've not seen anything definitive, only indications, that the Sony IBIS is far from as effective as The E-M5 II. So, 800 mm @ 1/120? Nope.

Whether that matters for moderate sized  prints or web images is another story. 
 If one likes astro-landscapes with high DR at high ISO then A7RII is way to 
go.  Sonny is still  ahead of Canyon on this.  All is not sweetness on light 
for this cam as the sensor does run hot and the A7r may be better for long 
exposures.

But what does it all mean in practical use? I'm now going to wander about, to 
end up back at this point at the end.

One can also use HDR and multishot high-resolution modes on Oly MFT to make up 
the difference for static subjects.

Yes, Ctein has commented on the virtual elimination of higher ISO noise with the HR Mode. I'd not noticed it, as I've mostly used low ISOs.


On 1/13/2017 10:59 AM, Mike Gordon via olympus wrote:
.....the frequency response of the system is a complex combination of all the 
stages

MTF of the system is the product of the components.  (one can include the eye 
in that)  Piers had a very nice link on that a few years back.

As I never have the MTFs for all the components, even leaving my eyes out of the equation, that's meaningless to me. Even comparing MTFs between brands is murky, as they all use different frequencies. Dose anyone outside the makers have the actual numbers, as opposed to little graphs?

I am compelled to bring up the ugly beast of diffraction as Dr. Diffraction is unable to do so. The smaller sensor is "diffraction limited" at larger aperture.

I miss Chuck. You may recall that, after many modest disagreements between his calculations and my experience and experiments, he finally admitted that, although the tests are 'accurate', his assumptions about the effect of the numbers overstated the visible effects.

We also have your admonitions about the loss of a stop of DR in the GM1/5 above 1/500 sec. I know I 'should' remember that, but mostly don't, when actually out shooting. So, what difference does it make?

As it happens I had just taken what I consider a high DR shot for something other than astro-landscapes and such. I'd seen a subject I liked, framed and shot - oh horror, @ 1/1000 sec.

Here's  the image. 
<http://www.moosemystic.net/Gallery/tech/Dynamics/_1000677.htm>

Ignore the B&W for the moment, I started out with a different aim than this. I did add one with the shadows raised. You can see that it's captured the large scale differences in shadow brightness. As it happens, again, I was in the same place, same time, also sunny, after a week of rain, so I took much the same shot a couple of times @ 1/500 sec. I'm pretty sure I heard the mechanical shutter operate.

I've just looked closely at the images. In addition to -0.7 EV, I shot one at 0 EV, just to put pressure on the sensor. Between the three, there isn't a spit's worth of meaningful difference. At 100%, I can see the wall texture in all three files when I pull the shadows way up. Yeah, it's a little brighter, in the 0 EV shot, but that can be matched. And the 0 EV shot doesn't need highlight recovery. Any one of them is more than sufficient for this subject. At ISO 200, with even high DR normal subjects, it's not a problem.

Then, for the first time, I actually thought Panny's logic through. With any reasonably bright subject, the shutter speed will be above 1/1000, - BUT - ISO will also be at the base of 200. When it gets dark, ISO goes up - BUT - shutter speed goes down. So, except in cooked up artificial cases, the dreaded combination of high shutter speed and high ISO together - doesn't happen.

Now, there are people who will complain about inadequate DR when that's not the problem they are experiencing at all.. Here's an almost white rose taken the same day, in direct sun. <http://www.moosemystic.net/Gallery/tech/Dynamics/_1000686.htm>

Looking at it as it comes out of the Raw converter, it could be easy to say that the highs are blocked up. OMG - inadequate DR! But a quick look at the histogram shows that it doesn't even get close to hitting the top.

The problem is an "S" tone curve that starts to level out in the highlights - combined with very subtle tonal differentiations in the subject. Change the curve, and all the highlight details appear.

So, I'm going to keep forgetting about this potential problem, and happily snap 
away.

As Moose most aptly demonstrated the MFT sensor have competitive resolution but 
my point was that the S/N and DR advantage of capturing more photons  on a 
larger sensor is perhaps even more important to consider depending on the 
application.

Now back to my question "But what does it all mean in practical use?" Most of my photography is outdoors. To get really high DRs, I have to shoot with harsh, direct sunlight, or at the sun. And that means low ISOs. When the light is lower, it simply can't generate high DRs, so when I'm shooting at high ISOs, I'm shooting subjects with lower DRs

I'm thinking that, as with DR Diff's numbers, and your stop of DR loss number, the DxO numbers you link to may mean nothing to my usual photography. Even with things like shooting a camp fire at night, the DR is surprisingly modest.

So as you say, it depends on the application, but I would word it to lean toward it being of importance in only pretty rare cases.

Diffracted to Distraction Moose
--
What if the Hokey Pokey *IS* what it's all about?
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/

<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz