Sony A7R on DXO - Highest full frame IQ ever

[bchernicoff]

I am not impressed by the sample images I've seen. I don't know if it's just bad in-camera processing or what. Take a look at this sample picture...click on the Original and look at the blond hairs on the horse as well as the brown hair on the forehead: http://www.dpreview.com/galleries/reviewsamples/photos/2736804/dsc00045?inalbum=sony-alpha-7r-samples-gallery

 

[LetTheRightLensIn]

There's more to a camera than its sensor, check out The Phoblographer's list of complaints about this camera:
http://www.thephoblographer.com/2013/10/30/finding-wrong-sony-a7-a7r-far/

I'd rather than 90% (or whatever percent is appropriate) the IQ 99% of the time vs. technically great stills that are out of focus, missed, underexposed, etc. most of the time.

Yes and that of course makes it more restrictive than if Canon could just put such a sensor in a 5D4 or 1DX2 or whatnot, but it least we now have an option for all the landscape type work where you can just focus by liveview zoomed mode and none of those complaints matter a whit. For a 5D3/1DX this won't replace those cameras but just be in supplement to get the MP and vastly better low ISO DR.

 

[LetTheRightLensIn]

There's more to a camera than its sensor, check out The Phoblographer's list of complaints about this camera:
http://www.thephoblographer.com/2013/10/30/finding-wrong-sony-a7-a7r-far/

I'd rather than 90% (or whatever percent is appropriate) the IQ 99% of the time vs. technically great stills that are out of focus, missed, underexposed, etc. most of the time.

Yes, these are points that are weak in the A7r. But the sensor is Canon's weakness and they should try to improve it. It seems really hard from Canon lovers to recognize problems without trying to find faults in competitors. Yes, Canon's great, but they can be greater.

I hear you, but I think the sensor plays a smaller role in the final image than most people think (where post-processing plays such a huge role). If it didn't why would any (un-sponsored) pro shoot with Canon? If dynamic range and color depth were so critical wouldn't it be obvious which camera was used to shoot a landscape or portrait? When I flip through National Geographic, which is probably the purest in terms of post-processing, I can't tell you if it was a Nikon, Canon, or Hasselblad, and neither can anyone else. At least not by looking at the photos alone.

I think DxOMark is a good resource, particularly for lenses, but just about any current DSLR can give you amazing results suitable for publication or exhibition.

That doesn't say anything at all. Of course you can't tell because the images that don't work out they don't publish!

And since there always other shots one can take that need less DR of course you can use a lesser sensor and still produce enough good shots but that doesn't mean that those shooting the lower DR sensors are not having to give up potentially really cool shots here and there all the time.

Anyway, at least for stuff where you can do liveview manual focusing we in the Canon world can now escape Canon's tiresome milking of decade old sensor tech and just nab a Sony and an adapter.

 

[LetTheRightLensIn]

DxOmark just published the full Sony A7r sensor test. And according to their measuresand ranking system the camera scores the same points as the Nikon D800-D800E (and that for $500 less). And it literally kicks the Canon5DIII in the butt.

I have thousands of images from the D800 and from the 5D3, and there is no butt-kicking happening in either direction. While both do a fine job, for me and my work the disadvantages of the D800 outweigh the advantages.

DxO charts are interesting if you are just buying a sensor. But photographers evaluate photos and usability for their tasks more than sensor charts. The sensor is the least interesting thing about the A7/A7r, which looks to be a very interesting camera otherwise.

The (mostly) anonymous online complaints about Canon's sensors don't stop anyone from making absolutely awesome photos with Canon's sensors in a great variety of circumstances. This is proven day after day by thousands of photographers, including some of the best in the business. When I hear about or meet a photographer whose work I've admired for decades and their main camera today is the 5D3, that tells me a lot more than anonymous complaints on the Internet.

Indeed, I don't give a shite about absolute perfection in sensors or DR. I run a portrait business... and often speak with other colleagues doing the same. Not once has anyone cared about whether you shoot Canon, Nikon, or whatever. It's always more about how the gear performs in the field and if it's more of a hindrance or asset. Todays offerings from the camera world will more than cover the needs of most in the IQ department... missed focus and poor metering though will be a hindrance any day of the week.

Yeah but did ever think that not everyone is in the studio portrait business where you can control the lighting exactly as you wish???

 

[LetTheRightLensIn]

Indeed, I don't give a shite about absolute perfection in sensors or DR. I run a portrait business... and often speak with other colleagues doing the same. Not once has anyone cared about whether you shoot Canon, Nikon, or whatever. It's always more about how the gear performs in the field and if it's more of a hindrance or asset. Todays offerings from the camera world will more than cover the needs of most in the IQ department... missed focus and poor metering though will be a hindrance any day of the week.

Well said. IQ is generally very high today and some measured difference at certain ISOs is not that special. Too many other factors come into play when making photographs.

+1
Nobody cares what the dynamic range and IQ is of an out of focus picture....

-1
what about all the landscape stuff where you can use a tripod and liveview AF? if you are using a T&S lens for them you'd already be doing that anyway

but i guess it is a bad and useless thing to be able to get around the sensor limitations for some scenarios and still be able to use canon glass : if canon isn't printed on it somewhere it's useless trash right? :

 

[Drizzt321]

Ok, I'll be the first one to jump in. DxOMark. Ugh.

While some of their measurements are meaningful to me, the whole 'rescale to 8MP' for most of the measurements of the sensor is just...WTF?! Anyway...good for the bits that are useful.

The rescale is not WTF in the slightest, it doesn't even matter what they rescale it to so long as to the same count for all bodies. It's just normalization. It's the only sensible and fair way to compare across bodies.... but if you think some old 4MP body is cleaner than an 18MP 1DX then go ahead and be my guest and don't bother with any tests that do normalization .

Well, I understand the normalization is important for some kinds of comparisons, but from my basic understanding (could be wrong) of how DxO does some of their scoring, things like Noise & Dynamic Range tend to prefer higher MPx sensors because when you down-scale those, with a good algorithm, move in a positive direction. But that isn't necessarily the true measure of the sensor, and while lots of us do shoot and down-scale for the web, many don't and for big prints we use the full resolution image. Now, I'll grant you, it'd be hard to design a good, scientific & repeatable test for many of the things they use the re-scaled images for (hmmm...do they keep using the exact same rescaling code every time? or do they keep using newer algorithms/code in, say, PS?), but surely the really smart people at DxO could come up with some other means that also makes sense.

 

[JohnDizzo15]

I'm interested to find out how many people in this forum and elsewhere are going to be getting this body specifically to shoot landscape with their Canon glass?

My second question is, how many of you in here will be taking photos on a regular basis where the extra 2.7 stops of DR will make or break your capture?

Just curious.

 

[Skulker]

Anyway, at least for stuff where you can do liveview manual focusing we in the Canon world can now escape Canon's tiresome milking of decade old sensor tech and just nab a Sony and an adapter.

If you really feel like that, and I'm sure you do, why don't you do just that? What is it that makes you feel the need to make such comments?

Your comments are so obviously heart felt and it seems to matter so much to you. You seem to believe that Canon are such a poor company to buy from I just can't get my head around why it matters so much to you.

 

[zlatko]

Indeed, I don't give a shite about absolute perfection in sensors or DR. I run a portrait business... and often speak with other colleagues doing the same. Not once has anyone cared about whether you shoot Canon, Nikon, or whatever. It's always more about how the gear performs in the field and if it's more of a hindrance or asset. Todays offerings from the camera world will more than cover the needs of most in the IQ department... missed focus and poor metering though will be a hindrance any day of the week.

Well said. IQ is generally very high today and some measured difference at certain ISOs is not that special. Too many other factors come into play when making photographs.

+1
Nobody cares what the dynamic range and IQ is of an out of focus picture....

-1
what about all the landscape stuff where you can use a tripod and liveview AF? if you are using a T&S lens for them you'd already be doing that anyway

but i guess it is a bad and useless thing to be able to get around the sensor limitations for some scenarios and still be able to use canon glass : if canon isn't printed on it somewhere it's useless trash right? :

Every camera in the world has sensor limitations. Knowing and working with sensor limitations is a part photography, no matter what camera you are using. Ansel Adams frequently wrote about sensor limitations.

 

[Pi]

Ok, I'll be the first one to jump in. DxOMark. Ugh.

While some of their measurements are meaningful to me, the whole 'rescale to 8MP' for most of the measurements of the sensor is just...WTF?! Anyway...good for the bits that are useful.

This does not mean that they actually resize to 8mp, they just use as a base to report the results. Like choosing mm or feet, etc.

 

[mackguyver]

LetTheRightLensIn, you're awfully grumpy today. I understand that you want better DR and such, and yes, Canon is behind Sony (and by virtue, Nikon) in sensor tech, but my point is that DxO's measurements would make you believe that the 5DIII and 1DX are absolute garbage when in fact a huge number of commercial shots are produced with them. Even Nikon's main man, Thom Hogan is making a similar point in his latest blog post:
http://www.dslrbodies.com/newsviews/the-full-frame-debate.html

 

[Apop]

The sensor is great all right. We know that since Nikon has (almost) the same sensor in the D800.

The question here is, great IQ with what lenses? Only two (rather slow) native lenses to speak of and don't get me started on adapters. Even DxO with its myopic "scores" came out with an ultimate assessment that the IQ even with a great sensor will only be as good as the lens you put in front of the camera.

Will this go the way of the D800? With a number of compromises made just for a smaller body, I guess it will sell worse. (People who have wanted the extra MP / DR have already switched to Nikon while those on the DR / MP hobby horse can be found trolling in the CR space)

BTW, I've read that there is no confirmation Sony has microlenses to adjust for the problems with the short flange distance with wide lenses that lead to side to side color problems? Leica does this in its cameras but what about Sony?

But with a metabones smart adapter you can best of both worlds , a7r sensor+canon glass (even though AF is gonna be slow).
An a7r ,35 2.8 , metabones adapter, and grip sounds like a pretty nice addition for someone with canon bodies/glass wanting that sensor performance in such a compact housing.

Actually, most of those a7r sales are probably gonna be canon users wanting 36mp in a compact body.
I am wondering who sony is targeting with the a7r, an a7r with 2 lenses seems pretty pricy for a family, especially with all those phones around.


If I was new on the market , I would not invest in sony ( lets say an a7 and couple of those new lenses)
The lenses are smaller, hence slower, but surprisingly I find them rather expensive.
The same goes if you intend to use sony's adapter to use a mount lenses, you do get the a77 AF capabilites (which is build into the adapter), but again quite expensive ( especially 300 2.8 and 500 f4 compared to others).

 

[MichaelHodges]

Canon is getting trounced in IQ, period. I can say that without my upper lip quivering or my eyes moistening, because I did not exchange vows with the gear when I ordered it.

There's no question it's getting harder and harder to sit and watch these stunning sensor made available as Canon spins tires.

 

[MichaelHodges]

just about any current DSLR can give you amazing results suitable for publication or exhibition.

One can apply this logical fallacy to almost any technical discussion. But what we are talking about here is the specific aspects of the new Sony sensors that apparently makes them superior to the rest.

 

[MichaelHodges]

The (mostly) anonymous online complaints about Canon's sensors don't stop anyone from making absolutely awesome photos with Canon's sensors in a great variety of circumstances. This is proven day after day by thousands of photographers, including some of the best in the business. When I hear about or meet a photographer whose work I've admired for decades and their main camera today is the 5D3, that tells me a lot more than anonymous complaints on the Internet.

No offense, but this paragraph doesn't really say anything at all.

 

[Lichtgestalt]

YAWN... it´s basically the same sensor so who had thought otherwise?
i mean it´s not as if this is a suprise.

david noton is shooting with canon... after the the D800 was released i asked him if he is switching back to nikon. i mean as a professionell landscape shooter you must have the D800 or your nothing.. right?

his answer was short: no.

 

[Lichtgestalt]

The (mostly) anonymous online complaints about Canon's sensors don't stop anyone from making absolutely awesome photos with Canon's sensors in a great variety of circumstances. This is proven day after day by thousands of photographers, including some of the best in the business. When I hear about or meet a photographer whose work I've admired for decades and their main camera today is the 5D3, that tells me a lot more than anonymous complaints on the Internet.

No offense, but this paragraph doesn't really say anything at all.

then it is in line with most stuff writen here about DR and sensor performance....

 

[EchoLocation]

people can make as many excuses or give as many reasons as they want why these results don't matter.
the fact is, if this score was had by a Canon camera, people here would be doing backflips and high fives for weeks.
end of story.

 

[Woody]

I am wondering if we'll ever see DR and other sensor improvements in Canon DSLR sensors. In Canon's latest financial report (Oct 2013), they indicate they are continuing their crusade for cost cutting. As far as I know, the only way to improve their sensor performance is to shift to new silicon processes that require new machines. If they want to cut down on operational costs, they will probably be very reluctant to make this change.

Sigh.

 

[Mt Spokane Photography]

http://photography-on-the.net/forum/showthread.php?t=1339351

DxOmark just published the full Sony A7r sensor test. And according to their measuresand ranking system the camera scores the same points as the Nikon D800-D800E (and that for $500 less). And it literally kicks the Canon5DIII in the butt.



Source: http://photography-on-the.net/forum/showthread.php?t=1339351

Well played Sony! And thanks to the wonderful metabones adapter, I guess we could say that this really is the "Canon body" we have been waiting for.

You seem to be confused. DXO does not rate the IQ of the camera, they do not even take a photo or mount a lens for this test. All they did was test the sensor without a lens.

At least DPR tests the camera, not just a sensor. DXO tests the sensor but gives the camera a rating when the images might be horrible, or it might have some really nasty flaws.