Low light capability of "foreign" bodies

[xps]

I read an review on Dpreview of the Samsung NX1.

On the page http://www.dpreview.com/reviews/samsung-nx1/10 you can compare some Cams.

I am surprised, that the Eos 6D and the 70D are much nore noisy - compared to other manufacturers.
(Nikon 750, Sony A7...) (Iso 100/ +...EV)

What is your opinion? Are these Cams so much better?

 

[ajfotofilmagem]

In my opinion, is it totally useless and misleading, someone test cameras at ISO100, pushed 3 stops in post production.
It just simulates a user who does not know how to use a camera properly, and not how a serious photographer will do in the real world.

 

[Loswr]

To serious photographers, the number of stops you can push a shot with the lens cap on is of paramount importance.

 

[ajfotofilmagem]

...What is your opinion? Are these Cams so much better?

These cameras are best for people unable to properly exposed photos.

In ISO800 or above, Canon cameras usually have better images than its competitors. Only you need to know to properly photometry.

 

[tpatana]

To serious photographers, the number of stops you can push a shot with the lens cap on is of paramount importance.

Really serious photographers also shine flashlight through the top LCD panel while taking picture with the lens cap on.

 

[xps]

But why do they do this?

Why not rising the ISOs to compare them with?
Whats the sense in this kind of comparison?

 

[Loswr]

To serious photographers, the number of stops you can push a shot with the lens cap on is of paramount importance.

Really serious photographers also shine flashlight through the top LCD panel while taking picture with the lens cap on.

Really, really serious photographers also shine the flashlight on the lens release button or the lower left part of the lens mount while taking picture with the lens cap on.

 

[Dantana]

To serious photographers, the number of stops you can push a shot with the lens cap on is of paramount importance.

Really serious photographers also shine flashlight through the top LCD panel while taking picture with the lens cap on.

This is why I have started hoarding high end flashlights, so I can be as serious as possible.
Really, really serious photographers also shine the flashlight on the lens release button or the lower left part of the lens mount while taking picture with the lens cap on.

 

[Deleted member 91053]

To serious photographers, the number of stops you can push a shot with the lens cap on is of paramount importance.

Really serious photographers also shine flashlight through the top LCD panel while taking picture with the lens cap on.

Really, really serious photographers also shine the flashlight on the lens release button or the lower left part of the lens mount while taking picture with the lens cap on.

Don't the really, really, really serious photographers also have a friend to shine a flashlight into the viewfinder as well?
After all a really, really,really, really,really, really serious photographer would never take pictures without someone there to see how really serious they are would they?
Just kidding!

 

[Aglet]

What is your opinion? Are these Cams so much better?

While some of the CR members here will disparage the poster and the question, the simple fact is
that the other cam's have sensor systems that ARE better than Canon's (astro use might be an exeption) and that allows users a LOT more leeway when editing raw files with much less noise control work. It also tends to provide better color accuracy (less color noise) than Canon's and that can be noticeable.

Canon's system for getting data off the sensor and into the processor was a comparatively good implementation when they came up with it over a decade ago.

Since then they've stuck with the same method. It's evolved; gotten worse (50D), then better (70D).
Meanwhile, all the other mfrs are using a better circuit-system design that's leap-frogged Canon's and is also continuing to evolve and improve to the point that micro-four-thirds sensors, about half the surface area of Canon's APS-C sensors, outperform the Canon stuff by nearly a stop. So that's nearly 2 stops worth better performance per surface area that everyone else can do using different sensor readout and data-transfer technology (Sony's).

As for Canon still being better above 800 iso - not any more. Everybody else has closed the gap and-or beat Canon's higher ISO performance too. That region is now bumping into the limitations of physics so improvements there are more in how the software interprets and manages (smooths) the data.

FWIW, with the level of software now available for managing noise and recovering detail, as long as you choose one of the newer Canon bodies that no longer produce a picket-fence of noise stripes in the shadow areas, you can get good image results. Canon's 70D and 7D2 seem to be pretty good now, if not quite as noise-free and color-pure as the competition. Hopefully the new 5DS series are similarly free of that nasty pattern noise, for the sake of Canon fans and users.
I switched to Anything-But-Canon a few years ago and could not be more pleased with the improvements in raw file quality. That saves me a lot of trouble prepping large format prints.

.. wanna make a Canon comparison look really bad?...plug a Fuji XT1 into the comparison tool ;}
Or a bottom-end Fuji XA1, even more fun. Fuji knows how to do some sneaky in-camera processing that rivals what can be done with a desktop.

 

[daniela]

Hey, XPS!

I checked your cited site and I can say the writer of the articel seems to be right.
I rented an Alpha 7R with an Zeiss 1.8 55mm and the shot are great. Not good, but great.

My tip: You come from south Germany, right? I´ll post you an shopadress, where you can rent an D810, an Alpha 7R and an 5DIII for one week each. and the compare your shots.

I hope Canon will close the gap in the near future.
I know, Canon lenses are mostly seen the best on the market - but the bodies will have to move upwards ind IQ and all the other frequent discussed items.

Dani

 

[raptor3x]

.. wanna make a Canon comparison look really bad?...plug a Fuji XT1 into the comparison tool ;}
Or a bottom-end Fuji XA1, even more fun. Fuji knows how to do some sneaky in-camera processing that rivals what can be done with a desktop.

Assuming you're talking about the high ISO performance, part of the apparent difference comes from the higher degree of implicit chroma noise reduction in the X-Trans demosaicing algorithm (see the difference between the X-A1/X-M1), the other part comes from Fuji straight up lying about their ISO values. In reality the Fuji bodies are no better (but also no worse) than the rest of the APS-C field.

 

[Mitch.Conner]

To serious photographers, the number of stops you can push a shot with the lens cap on is of paramount importance.

You keep saying stuff like this, and calling people iirc "lens cap shooters". Usually in DR arguments, but now with ISO.

What is the origin of this? Who takes photos with the lens cap on? Is that some sort of testing metric I've never heard of?

EDIT: I finally found some articles talking about it. I'd never heard of it before. Weird.

 

[ajfotofilmagem]

To serious photographers, the number of stops you can push a shot with the lens cap on is of paramount importance.

You keep saying stuff like this, and calling people iirc "lens cap shooters". Usually in DR arguments, but now with ISO.
What is the origin of this? Who takes photos with the lens cap on? Is that some sort of testing metric I've never heard of?
EDIT: I finally found some articles talking about it. I'd never heard of it before. Weird.

Every day I find strange things that humans do.
Psychology explains these behaviors as:

Need to prove you're right and others are wrong.
Need to display using very expensive cameras.
Need to prove that you have a stick larger than the other. ???
Need to prove that you have more DR than the others. :-X

The world seems strange every day.

 

[Loswr]

To serious photographers, the number of stops you can push a shot with the lens cap on is of paramount importance.

You keep saying stuff like this, and calling people iirc "lens cap shooters". Usually in DR arguments, but now with ISO.

What is the origin of this? Who takes photos with the lens cap on? Is that some sort of testing metric I've never heard of?

EDIT: I finally found some articles talking about it. I'd never heard of it before. Weird.

1. Capture RAW image at ISO 100 with lens cap on.
2. Push exposure 5 stops in post.
3. Examine noise.
4. Make broad, sweeping generalizations about camera performance.
5. Congratulations, you've joined the lens cap shooters club.

 

[Aglet]

Assuming you're talking about the high ISO performance, part of the apparent difference comes from the higher degree of implicit chroma noise reduction in the X-Trans demosaicing algorithm (see the difference between the X-A1/X-M1), the other part comes from Fuji straight up lying about their ISO values. In reality the Fuji bodies are no better (but also no worse) than the rest of the APS-C field.

Assuming you're talking about the high ISO performance, part of the apparent difference comes from the higher degree of implicit chroma noise reduction in the X-Trans demosaicing algorithm (see the difference between the X-A1/X-M1), the other part comes from Fuji straight up lying about their ISO values. In reality the Fuji bodies are no better (but also no worse) than the rest of the APS-C field.

hi iso, of course, we all know it mops the floor with Canon at base.
yes, Fuji's ISO's inflated, kinda like they're on permanent HTP mode
boost a stop and they're still outperforming most for luma noise
boost 2 stops of ISO and the Xtrans still looks better than Canon for luma noise and by then the Fuji is likely using higher real ISO sensitivity.

Even the bayer'd Fuji X-A1, a mere $400 camera, has sensor performance that Canon should aspire to with their latest crop bodies which demand much higher prices.

I'd really like to see DxOmark get on with measuring X-trans, maybe then we could get some comparable number on their sensor metrics. Otherwise it's more anecdotal and rather dependent on the raw converter.

Hmmm... That reminds me... If I ever have some time, I could compare an X-A1, X-M1 and a 16MP & 24MP nikon body for accutance, possible using the same lens.

 

[ishdakuteb]

"...
I'd really like to see DxOmark get on with measuring X-trans..."

but i would like to see right images captured by you... of course new ones since none of them in http://www.a2bart.com, which is owned by you, looks right to me... lol...

i'll show you mine capturing today, for fun but does look better than yours...

 

[Hillsilly]

What is your opinion? Are these Cams so much better?

This particular test on exposure latitude shows that a Samsung sensor outperforms a Canon sensor. But so what? Given that it is marketed as a performance orientated camera, I'd rate autofocus speed and reliability as more important than exposure latitude. And then you have build quality, lens choices,flash options and other accessories etc. I can't see too many areas where the Samsung camera beats a Canon camera.

Besides, if exposure latitude was your main criteria for purchasing a particular camera, wouldn't you just shoot colour negative film?

 

[raptor3x]

hi iso, of course, we all know it mops the floor with Canon at base.
yes, Fuji's ISO's inflated, kinda like they're on permanent HTP mode
boost a stop and they're still outperforming most for luma noise
boost 2 stops of ISO and the Xtrans still looks better than Canon for luma noise and by then the Fuji is likely using higher real ISO sensitivity.

I really have to disagree with about the luma noise. I own an X-T1 and it's really no better than any of the other APS-C bodies out there. Below is a comparison of the X-T1 and the 7D2 taken from DPReview with both cameras set to ISO 6400. I adjusted the exposure of the X-T1 shot such that the histogram was as close as possible to the 7D2, added a tiny amount of chroma NR to both (zero luma NR), and then exported them both to 16MP jpegs. Can you tell which is which? The only differences that I see are a slightly different character to the noise and one maintains detail better than the other.

 

[Mt Spokane Photography]

To serious photographers, the number of stops you can push a shot with the lens cap on is of paramount importance.

You keep saying stuff like this, and calling people iirc "lens cap shooters". Usually in DR arguments, but now with ISO.

What is the origin of this? Who takes photos with the lens cap on? Is that some sort of testing metric I've never heard of?

EDIT: I finally found some articles talking about it. I'd never heard of it before. Weird.

Basically, all modern digital cameras are so good that you cannot reasonably detect the difference in a properly exposed image. This has led to some underexposing images by 4 or 5 stops and then raising the exposure in post production to show that a given sensor works better that way. It only happens at low ISO values, and the reverse may be true at high ISO settings, so the camera sensors are only rated at ISO 100. There is some value in that, since auto exposure of various camera scenes is imperfect such that a stop or even two variance occasionally happens. It could potentially let you recover a poorly exposed image.

Some actually compare images taken with a lens cap on, exposure boosted a huge amount, so that the quality of the noise can be discussed.

However, its very important to look at a whole camera system rather than just one aspect. Everything is a compromise, so pick the set of compromises that works best for you.

Its common to see posts wishing that a camera had features that its missing. However, if all the cameras were identical, then there would be only one brand, and no reason to improve it. Competition is a driver of improvements, because the best camera can generate more sales. The downside is that higher prices due to exotic options deter sales. Its a compromise of features versus cost. And, there are always some who want the best of the best with gold plating, no matter what the cost. Selling just a few gold plated cameras can generate big profits even if they are otherwise identical.