Black
Photographer.
"Go on sir .... it is only waffer thin"*
*Possibly obscure Monty Python reference?
Michael Markey
Veteran
Michael Markey
Veteran
user237428934
User deletion pending
"lose", "win", "beat the pants off".... this sounds so boring and tech oriented. Who on earth thinks in terms of "virtual quality points per $" just like a mileage? Strange.
Corran
Well-known
It's just facts. It's just as valid as someone saying they like the ergonomics of whatever camera more than another, but that isn't quantifiable. For instance, I know a lot of folks who love Canon's ergonomics. I hate them, can't use their cameras at all without cramping my hand.
airfrogusmc
Veteran
The MM is absolutely no joke. And as far as price I remember when Leica M was twice the price of the top-o-the line big two now they are even so Leica M is now the bargain wouldn't you say? I had a lot more issues with my Canons by the time the were the same age and shutter count as my MM. I had a complete shutter fail (Canon)in the middle of shooting for NATO when they were here in Chicago May 2012.
I had some hands on this past Saturday and I didn't expect to like it as much as I did.
The Leica digital hate gets old as does the film hate on the Nikon and Canon forums. They are just tools, find ones that work for you.
GO BLACKHAWKS!!!!!
phatnev
Well-known
Doesn't the MM have the same defective sensor as the M9....?
airfrogusmc
Veteran
I haven't had any issues with mine. Over 30K and 2 1/2 years.
Dante_Stella
Rex canum cattorumque
Having now used the electronic shutter on the X100T for a couple of months, this rolling shutter issue strikes me as (1) related to video only or (2) an imaginary problem for most people who are not shooting F1 races from 10 feet away.
If you didn't have rolling shutter with 35mm cameras, which at 1/1000 scan slits much more slowly than an electronic shutter (35mm cameras scan at 1/30-1/250 when at anything above their flash sync speed), then there is nothing to worry about but your wallet.
D
willie_901
Veteran
The lens aperture doesn't matter. The data stream signal-to-nose ratio does.
Bill Claff's data shows the M240 outperforms the OM-D E-M5 up to about ISO 1600. At that point they become equal.
n.b. Claff publishes dynamic range vs ISO data based on statistical analysis of un-rendered raw data. It turns out dynamic range is directly related to the datas' signal-to-noise ratio.
Corran
Well-known
Of course the lens aperture matters. If you have a faster lens, you can use a lower ISO. Especially with a smaller sensor and correspondingly wider focal length and more DOF for a given aperture/field of view.
And there's more to the results than the DR. How about noise performance and resolution?
And there's more to the results than the DR. How about noise performance and resolution?
willie_901
Veteran
DR depends directly in SNR. The two are physically co-dependent. One can't obtain excellent DR without excellent SNR and vice-versa.
The lens aperture is irrelevant because one can maximize exposure (= maximize SNR) at any aperture. Lens aperture plays no fundamental role in SNR or DR.
However a wide aperture means the one can maximize exposure at shorter shutter times and this is often convenient. Convenience does not appear in the parameters that model the inherent SNR (which determines the DR) for a raw file.
Of course the inherent lens/sensor resolution is invariant. Yet resolution is obscured by noise. As the SNR decreases, the uncertainty in the data increases. So a one-stop SNR advantage minimizes the loss of resolution caused by uncertainty. Ad-hoc inspection of shadow regions (where the read noise surpasses the photon noise) is where the effect on resolution is most apparent. Aesthetically, I often prefer letting shadow regions remain dark (which renders resolution differences moot). Still, at high ISOs (say 1600) where the meter recommends the sensor be under exposed by 3 to 4 stops, most of the image is a composed of shadow regions. In this case the rendered resolution increases as the camera's data stream SNR increases.
Additionally, noise filtering (more commonly referred to as noise reduction) also compromises resolution. Noise filtering cleverly averages pixels with less certainty with pixels with higher uncertainly. Because an image's total information content must remain constant, the rendered resolution must decrease. It is possible to filter the noise such the balance between resolution loss and perceived gains in contrast and tonality is beneficial. A one or two stop increase in rendered SNR makes this easier.
The lens aperture is irrelevant because one can maximize exposure (= maximize SNR) at any aperture. Lens aperture plays no fundamental role in SNR or DR.
However a wide aperture means the one can maximize exposure at shorter shutter times and this is often convenient. Convenience does not appear in the parameters that model the inherent SNR (which determines the DR) for a raw file.
Of course the inherent lens/sensor resolution is invariant. Yet resolution is obscured by noise. As the SNR decreases, the uncertainty in the data increases. So a one-stop SNR advantage minimizes the loss of resolution caused by uncertainty. Ad-hoc inspection of shadow regions (where the read noise surpasses the photon noise) is where the effect on resolution is most apparent. Aesthetically, I often prefer letting shadow regions remain dark (which renders resolution differences moot). Still, at high ISOs (say 1600) where the meter recommends the sensor be under exposed by 3 to 4 stops, most of the image is a composed of shadow regions. In this case the rendered resolution increases as the camera's data stream SNR increases.
Additionally, noise filtering (more commonly referred to as noise reduction) also compromises resolution. Noise filtering cleverly averages pixels with less certainty with pixels with higher uncertainly. Because an image's total information content must remain constant, the rendered resolution must decrease. It is possible to filter the noise such the balance between resolution loss and perceived gains in contrast and tonality is beneficial. A one or two stop increase in rendered SNR makes this easier.
Corran
Well-known
I'm not saying aperture effects DR. But you're missing the forest for the trees.
If, for a given exposure, you are shooting a FF camera with a 50mm at f/5.6, 1/250, ISO 400, you would get the identical field of view and depth of field with a 2x crop camera using a 25mm at f/2.8, 1/250, ISO 100. You could of course instead shoot at 1/1000 and ISO 400 instead, but that's irrelevant in the point I'm making.
In a low-light scenario, you can use a wider aperture on a smaller sensor camera and get the equivalent photo with a lower ISO setting. In the specific example, using the data you linked, the EM5 @ base ISO 200 has a better/higher DR than the Leica M240 at ISO 800.
If you must use the concurrent higher shutter speed instead, for action, that means the M240 wouldn't even have enough light to freeze the motion adequately anyway, without boosting the ISO, so it's a non-issue.
In good light where you can use the lowest ISO settings, of course the FF wins. This is just obvious.
If, for a given exposure, you are shooting a FF camera with a 50mm at f/5.6, 1/250, ISO 400, you would get the identical field of view and depth of field with a 2x crop camera using a 25mm at f/2.8, 1/250, ISO 100. You could of course instead shoot at 1/1000 and ISO 400 instead, but that's irrelevant in the point I'm making.
In a low-light scenario, you can use a wider aperture on a smaller sensor camera and get the equivalent photo with a lower ISO setting. In the specific example, using the data you linked, the EM5 @ base ISO 200 has a better/higher DR than the Leica M240 at ISO 800.
If you must use the concurrent higher shutter speed instead, for action, that means the M240 wouldn't even have enough light to freeze the motion adequately anyway, without boosting the ISO, so it's a non-issue.
In good light where you can use the lowest ISO settings, of course the FF wins. This is just obvious.
jloden
Established
That's demonstrably not the case. I shoot with an electronic shutter every week (sermon photos on a GH3), and I lose shots to rolling shutter artifacts & distortion all the time.
Any moving subjects can be affected, though it doesn't mean it'll ruin every shot or that it's not a useful feature. It just has limitations that are worth being aware of, that's all.
The plus side is that electronic shutters are getting better at reducing rolling shutter effect by improving read-out speed. I suspect the Q will do comparatively well with such a fast processor and presumably fast sensor/data pipeline.
GaryLH
Veteran
If I understand correctly, how noticeable the effect of rolling shutter is dependent on different factors
- sensor tech, ie,
-- how fast can u read each row of the sensor out
-- is tech such that u are reading all of the sensor at once or multiple rows at a time.
- how much movement is present on subjects going across the screen as well as speed of that object
- how recognizable distortion is to the said subject
Compared to the earlier cameras, these newer ones seem to be less prone to it.. There is mention in the q advertisement about improvements to sensor readout design.
Gary
- sensor tech, ie,
-- how fast can u read each row of the sensor out
-- is tech such that u are reading all of the sensor at once or multiple rows at a time.
- how much movement is present on subjects going across the screen as well as speed of that object
- how recognizable distortion is to the said subject
Compared to the earlier cameras, these newer ones seem to be less prone to it.. There is mention in the q advertisement about improvements to sensor readout design.
Gary
willie_901
Veteran
This is evidence Fujifilm's ES implementation is flawed. I have seen numerous and diverse examples of Fujifilm ES distortion when even camera/subject motion is limited. It seems the effects of limited motion increase as the lens focal length decreases.
I expect Leica will use a superior system such as the one on your GH3.
willie_901
Veteran
We are just thinking about this differently. I am concerned with the recording maximum possible SNR/DR possible in the raw data.
You are addressing maximizing exposure in the most convenient and practical way possible. For M240 , below about ISO 1600, increasing ISO when shutter time has little or no benefit for the rendered image. The recorded signal level is reduced by under exposure whether ISO is 200 or 1600. But increasing ISO is convenient when one wants to review the in-camer JPEG.
All things being equal, sensor area increases signal amplitude. Should the the cameras' read noise levels be equal, increased sensor area means more SNR which also increases DR. However, I had a 24 X 26 mm sensor camera with less SNR and DR than an APS-C camera. The signal and noise levels should not be considered independently
Corran
Well-known
I agree, I'm thinking about it in actual practical photographic application.
raid
Dad Photographer
This a great point&shoot Leica travel camera!
Used, and at $2700, it is a good option for a travel camera.
Maybe a chat with the Leica CEO will clarify what Leica was thinking when ....
Used, and at $2700, it is a good option for a travel camera.
Maybe a chat with the Leica CEO will clarify what Leica was thinking when ....
I dont have a camera that has an electronic shutter, but I've seen several examples of X100T images posted that show rolling shutter artifacts. And no, they were not F1 cars.