Fred Miranda comparison confirms DXO results. 5D III blown away.

Started Apr 26, 2012 | Discussions
5tve
Contributing MemberPosts: 652Gear list
Like?
Re: Exactly what's going over Your head!
In reply to qianp2k, Apr 27, 2012

qianp2k wrote:

I have read several threads in this form regarding 5D3 vs D800 high ISO. After downsamping D800 to the same 22mp, I don't see D800 holds more details in high ISO.

Sure it's just a JPEG comparison at the moment and they are going to do in RAW comparison. But CameraLabs correctly downsampling D800 to the same 22mp rather upsampling 5D3 to 36mp. It shows 5D3 has better high ISO noise and holds a bit of more details actually in high ISO.

Amateur Photographer magazine measured the resolution of the camera's at ISO 25600
D800= 30 & 5DMKIII= 24

Sure the Canon has the cleaner Jpegs at high ISO but if you shoot raw & use dark frame subtraction followed by careful noise reduction before downsizing to 22MP the D800 will have more detail at the highest ISO's

 5tve's gear list:5tve's gear list
Nikon D7000 Sony a6000
Reply   Reply with quote   Complain
qianp2k
Senior MemberPosts: 9,988Gear list
Like?
Re: Exactly what's going over Your head!
In reply to 5tve, Apr 27, 2012

5tve wrote:

Sure the Canon has the cleaner Jpegs at high ISO but if you shoot raw & use dark frame subtraction followed by careful noise reduction before downsizing to 22MP the D800 will have more detail at the highest ISO's

http://forums.dpreview.com/forums/read.asp?forum=1032&message=40909924

from this RAW comparison when compared at the same size, I don't see either one has obvious detail advantage over others (D4 is the best while 5D3 and D800 basically tired). I agree D800 is an amazing camera and seems holds high ISO well till ISO 6400 but 5D3 seems has better high ISO above ISO 6400 from what I have seen in a few comparison. Nevertheless I have to say 5D3 doesn't have much better high ISO than D800, and D800 doesn't have much more details than 5D3 either. Their difference is rather small. After of all 36mp is not really way more resolution than 22mp, not in the same degree from 12mp to 21mp.

http://www.dpreview.com/previews/canoneos5dmarkiii/12

Just look this DPR lab test and move around different areas. D800 certainly has more details but not really in big margin. 5D3 has better high ISO over 6400 but also not in large degree.

-- hide signature --
 qianp2k's gear list:qianp2k's gear list
Sony RX100 Canon EOS-1D Mark III Canon EOS 5D Mark II Canon EOS 5D Mark III Sony Alpha 7R +20 more
Reply   Reply with quote   Complain
SubPrime
Senior MemberPosts: 1,233
Like?
Re: Exactly what's going over Your head!
In reply to qianp2k, Apr 27, 2012

qianp2k wrote:

from this RAW comparison when compared at the same size, I don't see either one has obvious detail advantage over others (D4 is the best while 5D3 and D800 basically tired).

You obviously missed the part that said "These normalised to the D4 size". It's no secret that donwnsampling high res image will sacrifice detail. In fact, the D800 maintains more detail than the 5D3 right p to 12,800.

I agree D800 is an amazing camera and seems holds high ISO well till ISO 6400 but 5D3 seems has better high ISO above ISO 6400 from what I have seen in a few comparison.

Really? Can you pick which is which at 12,800 ISO

Or 25,600 ISO?

Nevertheless I have to say 5D3 doesn't have much better high ISO than D800, and D800 doesn't have much more details than 5D3 either. Their difference is rather small.

Ya think?

http://www.bezergheanu.com/TestNikon/Test-Nikon-D800/22087378_KqWcB7# !i=1763885715&k=BN6QTnD&lb=1&s=O

Reply   Reply with quote   Complain
qianp2k
Senior MemberPosts: 9,988Gear list
Like?
Re: Exactly what's going over Your head!
In reply to SubPrime, Apr 27, 2012

SubPrime wrote:

qianp2k wrote:

from this RAW comparison when compared at the same size, I don't see either one has obvious detail advantage over others (D4 is the best while 5D3 and D800 basically tired).

You obviously missed the part that said "These normalised to the D4 size". It's no secret that donwnsampling high res image will sacrifice detail. In fact, the D800 maintains more detail than the 5D3 right p to 12,800.

No I am not, I said compared at the same size. But why I upsampling if 21 or 22mp is already enough for me? I don't print more than 30x20" then I just don't see D800 has noticeable better print than 5D3 upto 30x20" as someone already did that and said no difference.

I agree D800 is an amazing camera and seems holds high ISO well till ISO 6400 but 5D3 seems has better high ISO above ISO 6400 from what I have seen in a few comparison.

Really? Can you pick which is which at 12,800 ISO
Or 25,600 ISO?

I don't trust your hands as you're obviously biased. Let's wait DPR official review. Nevertheless maybe we should compare at their original size or per-pixel level.

Here are DPR lab samples and I can see 5D3 is noticeable better in these two high ISOs. Sure D800 ones look bigger but 5D3 hold better finer details and resolve the textual details better.

Nevertheless I have to say 5D3 doesn't have much better high ISO than D800, and D800 doesn't have much more details than 5D3 either. Their difference is rather small.

Ya think?

http://www.bezergheanu.com/TestNikon/Test-Nikon-D800/22087378_KqWcB7# !i=1763885715&k=BN6QTnD&lb=1&s=O

What special really that something 5D3 even 5D2 cannot do in controlled studio?

-- hide signature --
 qianp2k's gear list:qianp2k's gear list
Sony RX100 Canon EOS-1D Mark III Canon EOS 5D Mark II Canon EOS 5D Mark III Sony Alpha 7R +20 more
Reply   Reply with quote   Complain
Rick Knepper
Veteran MemberPosts: 9,906Gear list
Like?
Re: Photo stitiching
In reply to qianp2k, Apr 27, 2012

I know exactly what I am talking about. Re-read your little article again. Then read what I wrote. You are completely turned around.

More MPs at capture given the same sensor area produces more detail no matter what slice of the image you use or how many you stitch together. On crop cameras, it's why you got the reach effect given the same MPs but a smaller area.

You can stitch 32 21 MP and 24 36 MP images together (I know my math is off), crop and resize them any way you want and you will see that the final image made with the 36 MP sensor produces more detail. This is not up for debate. It is a fact.

And it is why you or at least people who actually know what they are talking about use a 21 or 36 MP sensor now to stitch their images rather than stitching 144 6 MP images together - because there is more detail in a single 21 MP image than in a single 6 MP image.

qianp2k wrote:
You don't know what you're talking about.

http://www.cambridgeincolour.com/tutorials/digital-panoramas.htm

Digital photo stitching for mosaics and panoramas enable the photographer to create photos with higher resolution and/or a wider angle of view than their digital camera or lenses would ordinarily allow—creating more detailed final prints and potentially more dramatic, all-encompassing panoramic perspectives.

I used 70-200L/4.0 IS in two pano photos I demo in my earlier post. Each one combined around 32 21mp photos together. The file size is around 800mb. You got kidding me that their resolution is way more than a single photo that D800 can generate. Do you ever stitch photos? By the way 36mp is not much higher resolution than 21/22mp.

Rick Knepper wrote:

qianp2k wrote:

Oh, if you want more resolution, photo stitching is a great choice.

IMO, stitching with 21-22 MPs does not capture the same fine detail because you are stitching panels shot with a 21-22 MP FF (area) sensor vs a 36MP FF (area) sensor. All you are doing is blowing up an image that didn't capture the same fine detail as a 36 MP sensor. Shoot the same scene with a 36 MP sensor, crop and resize and you will get more, finer detail that can actually be seen at small or large sizes.

To me, stitching is mostly about increasing your field of view which, by the way, pushes detail away from you.

-- hide signature --

Rick Knepper, photographer, photography never for sale, check my profile for gear list and philosophy.

-- hide signature --
-- hide signature --

Rick Knepper, photographer, photography never for sale, check my profile for gear list and philosophy.

 Rick Knepper's gear list:Rick Knepper's gear list
Nikon D3X Nikon D800E Canon EOS 5D Mark III Canon EOS 6D Canon TS-E 17mm f/4L +17 more
Reply   Reply with quote   Complain
qianp2k
Senior MemberPosts: 9,988Gear list
Like?
Re: Photo stitiching
In reply to Rick Knepper, Apr 27, 2012

You still have no clues. Just think about what FOV means.

Give an example let's shoot a building. In single photo you need to use wider lens (said 24mm) to fit the building into the frame. Now by stitching, I use much longer focus lens said 70mm or even 200mm or even 500mm (actually whatever lens) but only shoot small portion of building (much narrower FOV). Do you suggest 36mp at 24mm (FOV) can resolve more details than 22mp at 70, 200 or even 500mm FOV? You got kidding me. I'd assemble 100 individual photos from 500mm lens with 22mp into a gigantic photo with much higher resolution (you can imagine it's way bigger X * Y pixels and huge file size), way bigger than your single 36mp file! All birding photog know that a crop 16mp APS-C camera with super-tele lens 500mm resolves lots lots more details than a FF camera (even at 36mp) with 200mm lens. Otherwise by your theory all birding photog now should all buy D800 with 70-200 lens and no longer need to carry bulky and super expensive super tele-lens, wow!

Rick Knepper wrote:

I know exactly what I am talking about. Re-read your little article again. Then read what I wrote. You are completely turned around.

More MPs at capture given the same sensor area produces more detail no matter what slice of the image you use or how many you stitch together. On crop cameras, it's why you got the reach effect given the same MPs but a smaller area.

You can stitch 32 21 MP and 24 36 MP images together (I know my math is off), crop and resize them any way you want and you will see that the final image made with the 36 MP sensor produces more detail. This is not up for debate. It is a fact.

And it is why you or at least people who actually know what they are talking about use a 21 or 36 MP sensor now rather than stitching 144 6 MP images together - because there is more detail in a single 21 MP image than in a single 6 MP image.

qianp2k wrote:
You don't know what you're talking about.

http://www.cambridgeincolour.com/tutorials/digital-panoramas.htm

Digital photo stitching for mosaics and panoramas enable the photographer to create photos with higher resolution and/or a wider angle of view than their digital camera or lenses would ordinarily allow—creating more detailed final prints and potentially more dramatic, all-encompassing panoramic perspectives.

I used 70-200L/4.0 IS in two pano photos I demo in my earlier post. Each one combined around 32 21mp photos together. The file size is around 800mb. You got kidding me that their resolution is way more than a single photo that D800 can generate. Do you ever stitch photos? By the way 36mp is not much higher resolution than 21/22mp.

Rick Knepper wrote:

qianp2k wrote:

Oh, if you want more resolution, photo stitching is a great choice.

IMO, stitching with 21-22 MPs does not capture the same fine detail because you are stitching panels shot with a 21-22 MP FF (area) sensor vs a 36MP FF (area) sensor. All you are doing is blowing up an image that didn't capture the same fine detail as a 36 MP sensor. Shoot the same scene with a 36 MP sensor, crop and resize and you will get more, finer detail that can actually be seen at small or large sizes.

To me, stitching is mostly about increasing your field of view which, by the way, pushes detail away from you.

-- hide signature --

Rick Knepper, photographer, photography never for sale, check my profile for gear list and philosophy.

-- hide signature --
-- hide signature --

Rick Knepper, photographer, photography never for sale, check my profile for gear list and philosophy.

-- hide signature --
 qianp2k's gear list:qianp2k's gear list
Sony RX100 Canon EOS-1D Mark III Canon EOS 5D Mark II Canon EOS 5D Mark III Sony Alpha 7R +20 more
Reply   Reply with quote   Complain
orlandoom
Senior MemberPosts: 1,028
Like?
Re: Exactly what's going over Your head!
In reply to qianp2k, Apr 27, 2012

carlk: You have talked many weeks. Why you don't get D800 or D800E and move to Nikon? You don't have much lenses built-up and will not lose lot if you sell them. You keep whining but without an action is beyond me. I'd love to see your D800 photos in Nikon forum one day. However if you don't think you can get good photos from 5D3 or even 5D2, I doubt you will suddenly get from D800/D800E. Glass is not always greener at the other side of the fence.

+1
Good post.
Don

Reply   Reply with quote   Complain
SubPrime
Senior MemberPosts: 1,233
Like?
Re: In the real world, higher MPs capture more detail
In reply to qianp2k, Apr 27, 2012

qianp2k wrote:

SubPrime wrote:

qianp2k wrote:

Extreme pulling usually will only generate surreal HDR-alike photos which is not a taste to my people. I don't deny Sony sensor advantage in DR and no intention to downplay D800 which is a great camera but just warn not to over exaggerate.

It's not a matter of exaggeration. DR is much more than just pushing shadows, it means better gradation from highlights to shadow, more shadown in shadows and mod tones, and more latitude.

Sony Exmor sensor's 14 stop DR doesn't jump out by itself in photos. You need to pull shadow in extreme to leverage it. By doing that it usually generate surreal HDR-look alike photos that I have seen many. I am just not convinced that extra 2-stop advantage in shadow noise at base ISO is the main factor to decide IQ or such big deal. I just don't feel that way in real world photos. I think lens choice (Canon gives me much better choice on my needs), color tonality (I prefer Canon colors) are (far) more important. As I said I just don't see Nikon/Sony cameras have obvious advantages in IQ than Canon ones in real world photos.

I have both Canon and Nikon. I had the D700 then picked up a 5D2. While I love the 5D2, the first thing I noticed in PP was the limited latitude the Canon files produced. I wasn;t able to do nearly as much with them.

I don't have that feeling from Canon files. I own 60D, 5D2/5D1, 1D3. I virtually find no needs to pull 4-stop from shadow but only need to pull moderately, just my taste. I have more concerns in highlights that fortunately Canon matches to Nikon basally (although latter is slightly better).

You don't see much difference in real world photos from Nikon and Canon cameras in reality.

The D800 has only been around a few weeks, so give it time. The studio shots I am seeing from the D800 are simply amazing, and I suspect we'll see a lot more that will begin to really differentiate the D800 from all ofhter DSLRs.

I have also seen many stunning studio photos from 5D2, and 5D3 is only better. I only can see D800 generates bigger studio photos but not really (much) better.

Oh, if you want more resolution, photo stitching is a great choice.

Not really. Photo stitching is an option but not a great choice. It requires mutiple exposures, which means that you're screwed if your subject matter has moving objects.

For landscapes, I find photo stitching is better than UWA, much less distortion, much wider and certainly much higher resolution. I don't see moving subject is an issue in landscape stitching in most times especially in open space landscape (not in busy city streets).

Here are two samples I took in recent trip, all hand-held.

Great images.

Thx

-- hide signature --
Reply   Reply with quote   Complain
Rick Knepper
Veteran MemberPosts: 9,906Gear list
Like?
Re: Photo stitiching
In reply to qianp2k, Apr 27, 2012

qianp2k wrote:

You still have no clues. Just think about what FOV means.

Give an example let's shoot a building. In single photo you need to use wider lens (said 24mm) to fit the building into the frame. Now by stitching, I use much longer focus lens said 70mm or even 200mm or even 500mm (actually whatever lens) but only shoot small portion of building (much narrower FOV). Do you suggest 36mp at 24mm (FOV) can resolve more details than 22mp at 70, 200 or even 500mm FOV?

No I am not suggesting anything along these lines. Listen, I had to check your nationality first before pointing out that you have one of the worst reading comprehension and composition problems I've ever seen on this forum and thought maybe it was a language translation problem. Good luck to whatever it is you do.

You got kidding me. I'd assemble 100 individual photos from 500mm lens with 22mp into a gigantic photo with much higher resolution (you can imagine it's way bigger X * Y pixels and huge file size), way bigger than your single 36mp file! All birding photog know that a crop 16mp APS-C camera with super-tele lens 500mm resolves lots lots more details than a FF camera (even at 36mp) with 200mm lens. Otherwise by your theory all birding photog now should all buy D800 with 70-200 lens and no longer need to carry bulky and super expensive super tele-lens, wow!

Rick Knepper wrote:

I know exactly what I am talking about. Re-read your little article again. Then read what I wrote. You are completely turned around.

More MPs at capture given the same sensor area produces more detail no matter what slice of the image you use or how many you stitch together. On crop cameras, it's why you got the reach effect given the same MPs but a smaller area.

You can stitch 32 21 MP and 24 36 MP images together (I know my math is off), crop and resize them any way you want and you will see that the final image made with the 36 MP sensor produces more detail. This is not up for debate. It is a fact.

And it is why you or at least people who actually know what they are talking about use a 21 or 36 MP sensor now rather than stitching 144 6 MP images together - because there is more detail in a single 21 MP image than in a single 6 MP image.

qianp2k wrote:
You don't know what you're talking about.

http://www.cambridgeincolour.com/tutorials/digital-panoramas.htm

Digital photo stitching for mosaics and panoramas enable the photographer to create photos with higher resolution and/or a wider angle of view than their digital camera or lenses would ordinarily allow—creating more detailed final prints and potentially more dramatic, all-encompassing panoramic perspectives.

I used 70-200L/4.0 IS in two pano photos I demo in my earlier post. Each one combined around 32 21mp photos together. The file size is around 800mb. You got kidding me that their resolution is way more than a single photo that D800 can generate. Do you ever stitch photos? By the way 36mp is not much higher resolution than 21/22mp.

Rick Knepper wrote:

qianp2k wrote:

Oh, if you want more resolution, photo stitching is a great choice.

IMO, stitching with 21-22 MPs does not capture the same fine detail because you are stitching panels shot with a 21-22 MP FF (area) sensor vs a 36MP FF (area) sensor. All you are doing is blowing up an image that didn't capture the same fine detail as a 36 MP sensor. Shoot the same scene with a 36 MP sensor, crop and resize and you will get more, finer detail that can actually be seen at small or large sizes.

To me, stitching is mostly about increasing your field of view which, by the way, pushes detail away from you.

-- hide signature --

Rick Knepper, photographer, photography never for sale, check my profile for gear list and philosophy.

-- hide signature --
-- hide signature --

Rick Knepper, photographer, photography never for sale, check my profile for gear list and philosophy.

-- hide signature --
-- hide signature --

Rick Knepper, photographer, photography never for sale, check my profile for gear list and philosophy.

 Rick Knepper's gear list:Rick Knepper's gear list
Nikon D3X Nikon D800E Canon EOS 5D Mark III Canon EOS 6D Canon TS-E 17mm f/4L +17 more
Reply   Reply with quote   Complain
qianp2k
Senior MemberPosts: 9,988Gear list
Like?
Re: Photo stitiching
In reply to Rick Knepper, Apr 27, 2012

Rick Knepper wrote:

qianp2k wrote:

You still have no clues. Just think about what FOV means.

Give an example let's shoot a building. In single photo you need to use wider lens (said 24mm) to fit the building into the frame. Now by stitching, I use much longer focus lens said 70mm or even 200mm or even 500mm (actually whatever lens) but only shoot small portion of building (much narrower FOV). Do you suggest 36mp at 24mm (FOV) can resolve more details than 22mp at 70, 200 or even 500mm FOV?

No I am not suggesting anything along these lines. Listen, I had to check your nationality first before pointing out that you have one of the worst reading comprehension and composition problems I've ever seen on this forum and thought maybe it was a language translation problem. Good luck to whatever it is you do.

But this is exactly you said 36mp D800 resolves more details than my 32-photos stitching pano, didn't you? I am waiting to see your real photos to show us your theory. What lens you can fit the scenes from my two sample pano into a single frame, fisheye seems is the only option, then how D800 will have more resolution is beyond me. Now you desperately resort to the low-end tactic of questioning my language skill. I have no problem to read your posts and you have no problem to understand what I said that I am very sure about. Now you question my nationality. I am a proud American and a citizen of the United States of America, all right?

You got kidding me. I'd assemble 100 individual photos from 500mm lens with 22mp into a gigantic photo with much higher resolution (you can imagine it's way bigger X * Y pixels and huge file size), way bigger than your single 36mp file! All birding photog know that a crop 16mp APS-C camera with super-tele lens 500mm resolves lots lots more details than a FF camera (even at 36mp) with 200mm lens. Otherwise by your theory all birding photog now should all buy D800 with 70-200 lens and no longer need to carry bulky and super expensive super tele-lens, wow!

Rick Knepper wrote:

I know exactly what I am talking about. Re-read your little article again. Then read what I wrote. You are completely turned around.

More MPs at capture given the same sensor area produces more detail no matter what slice of the image you use or how many you stitch together. On crop cameras, it's why you got the reach effect given the same MPs but a smaller area.

You can stitch 32 21 MP and 24 36 MP images together (I know my math is off), crop and resize them any way you want and you will see that the final image made with the 36 MP sensor produces more detail. This is not up for debate. It is a fact.

And it is why you or at least people who actually know what they are talking about use a 21 or 36 MP sensor now rather than stitching 144 6 MP images together - because there is more detail in a single 21 MP image than in a single 6 MP image.

qianp2k wrote:
You don't know what you're talking about.

http://www.cambridgeincolour.com/tutorials/digital-panoramas.htm

Digital photo stitching for mosaics and panoramas enable the photographer to create photos with higher resolution and/or a wider angle of view than their digital camera or lenses would ordinarily allow—creating more detailed final prints and potentially more dramatic, all-encompassing panoramic perspectives.

I used 70-200L/4.0 IS in two pano photos I demo in my earlier post. Each one combined around 32 21mp photos together. The file size is around 800mb. You got kidding me that their resolution is way more than a single photo that D800 can generate. Do you ever stitch photos? By the way 36mp is not much higher resolution than 21/22mp.

Rick Knepper wrote:

qianp2k wrote:

Oh, if you want more resolution, photo stitching is a great choice.

IMO, stitching with 21-22 MPs does not capture the same fine detail because you are stitching panels shot with a 21-22 MP FF (area) sensor vs a 36MP FF (area) sensor. All you are doing is blowing up an image that didn't capture the same fine detail as a 36 MP sensor. Shoot the same scene with a 36 MP sensor, crop and resize and you will get more, finer detail that can actually be seen at small or large sizes.

To me, stitching is mostly about increasing your field of view which, by the way, pushes detail away from you.

-- hide signature --

Rick Knepper, photographer, photography never for sale, check my profile for gear list and philosophy.

-- hide signature --
-- hide signature --

Rick Knepper, photographer, photography never for sale, check my profile for gear list and philosophy.

-- hide signature --
-- hide signature --

Rick Knepper, photographer, photography never for sale, check my profile for gear list and philosophy.

-- hide signature --
 qianp2k's gear list:qianp2k's gear list
Sony RX100 Canon EOS-1D Mark III Canon EOS 5D Mark II Canon EOS 5D Mark III Sony Alpha 7R +20 more
Reply   Reply with quote   Complain
SubPrime
Senior MemberPosts: 1,233
Like?
Re: In the real world, higher MPs capture more detail
In reply to qianp2k, Apr 27, 2012

qianp2k wrote:

The above samples are typical Nikon skin tone and color tonality that I don't envy.

No, that is typical of the Sigma 150 f2.8 macro, which is notorious for producing a yellow cast.

Have you read many complaints in Nikon forums for orange and yellow color casts?

Easily solved with an Xrite passport profile or by adjusting white balance. I do find it odd that you describe your needs as being lascape relatedm, but flip to examples fo protraiture to buttress your argument.

Reply   Reply with quote   Complain
SubPrime
Senior MemberPosts: 1,233
Like?
Re: In the real world, higher MPs capture more detail
In reply to qianp2k, Apr 27, 2012

qianp2k wrote:

I have seen those real world photos from D800/D800E in Nikon forum and I quoted a few, nothing 5D2/5D3 cannot do really.

We hear Canon shooters making thsi claim repeately, but never able to produce an example.

DR is not about pulling in extremes, it is able tonal gradations and color depth. There is a reason why the Nikon D800 raw files are doubt the size of the 5D3 - there is simply more color information there.

Reply   Reply with quote   Complain
thanhuy
Regular MemberPosts: 288
Like?
Re: Old news
In reply to David Hull, Apr 27, 2012

I think you probably the only one tend to use high ISO for nature photography. I don't think many people use higher ISO for that type of photography unless in some circumstance... I rarely use ISO above 400 for nature photography. I always keep it at base ISO as much as possible. For nature, you don't handhold your camera during sun rise or sun set. Use your tripod.

David Hull wrote:

carlk wrote:

stephenmelvin wrote:

This was reported a few days ago, and it's not a surprise, really.

Thing is, the DR advantage of the D800 goes away rapidly as the ISO goes up. It's about 2 1/2 stops at ISO 100, but by ISO 160, it's under two stops (9.12 vs 10.84). At ISO 320, the advantage is about a stop (8.89 vs 9.95) and by ISO 1280, they're practically the same.

So while the low ISO dynamic range of the D800 is extremely good, depending on how a photographer uses his camera, it may matter greatly, or it may not matter at all.

As for me, maybe one photo in ten is taken at ISO 100. I'd like to have the extra DR, sure, but it's not where I make my living.

Op said he's a nature shooter. I'm sure he's not too interested in high ISO shooting. Matter of fact no body should be shooting high ISO if can be avoided because it always compromises the IQ no matter what camera you're using.

The other way to put it is the best IQ D800 can produce (@ ISO100) can be better than the best IQ 5DIII can. That point should not be missed by people who are looking for ways to get the best IQ they could.

This, of course, depends on the nature of his "nature" photography. Not all nature photography is done at ISO 100. I would be willing to bet that not a lot of it is done there. The best shots are often golden hour shots in the early morning or early evening. There are also situations where the subject is in the shade (animals up early, out late etc.).

As the guy above correctly points out, the two stop advantage of the Sony design only exists at base ISO and is down to one stop by 400 and essentially gone at 800. For all practical purposes -- with the exception of purely contrived examples such as the one being discussed here – the output of these two cameras will be indistinguishable.

Better advice for the OP would be to take inventory of his shooting scenarios, determine what ISO ranges he needs and whether he will gain any advantage from the DR. The resolution is another story, depending (again) on his subject matter, the extra cropping might come in handy.

Reply   Reply with quote   Complain
qianp2k
Senior MemberPosts: 9,988Gear list
Like?
Re: In the real world, higher MPs capture more detail
In reply to SubPrime, Apr 27, 2012

SubPrime wrote:

qianp2k wrote:

The above samples are typical Nikon skin tone and color tonality that I don't envy.

No, that is typical of the Sigma 150 f2.8 macro, which is notorious for producing a yellow cast.

I have seen the same Sigma lens on 5D2 without this yellow cast. How about the other two links that they don't use Sigma lens. Nikon yellow/orange cast in well known in all Nikon models as said by many.

Have you read many complaints in Nikon forums for orange and yellow color casts?

Easily solved with an Xrite passport profile or by adjusting white balance. I do find it odd that you describe your needs as being lascape relatedm, but flip to examples fo protraiture to buttress your argument.

If you read you will know where this came from. I answered Carl's post. But I keep my opinion that Canon is better in color tonality and skin tone and I am not alone.

-- hide signature --
 qianp2k's gear list:qianp2k's gear list
Sony RX100 Canon EOS-1D Mark III Canon EOS 5D Mark II Canon EOS 5D Mark III Sony Alpha 7R +20 more
Reply   Reply with quote   Complain
SubPrime
Senior MemberPosts: 1,233
Like?
Re: In the real world, higher MPs capture more detail
In reply to qianp2k, Apr 27, 2012

qianp2k wrote:

I have seen the same Sigma lens on 5D2 without this yellow cast.

That's because it was removed in PP.

How about the other two links that they don't use Sigma lens. Nikon yellow/orange cast in well known in all Nikon models as said by many.

The D800 has actually been criticized for having a magenta cast, so that's old news.

If you read you will know where this came from. I answered Carl's post. But I keep my opinion that Canon is better in color tonality and skin tone and I am not alone.

Historically yes, but it comes at the expense of being unable to handle reds, which the 5D is notorious for failing at.

Reply   Reply with quote   Complain
SubPrime
Senior MemberPosts: 1,233
Like?
Re: Exactly what's going over Your head!
In reply to qianp2k, Apr 27, 2012

qianp2k wrote:

No I am not, I said compared at the same size. But why I upsampling if 21 or 22mp is already enough for me? I don't print more than 30x20" then I just don't see D800 has noticeable better print than 5D3 upto 30x20" as someone already did that and said no difference.

No, they said the D800 had more detail, as one would expect.

I don't trust your hands as you're obviously biased.

Fine, do the tests yourself then. Download the raw files from both cameras at corresponsing ISO's and bring into LR or ACR and disable all NR and sharpening. That's all I did. Don't believe me, then here are others who did the same. The D800 maitains superior detail at all ISOs.

http://forums.dpreview.com/forums/read.asp?forum=1032&message=41071194

Let's wait DPR official review. Nevertheless maybe we should compare at their original size or per-pixel level.

Here are DPR lab samples and I can see 5D3 is noticeable better in these two high ISOs.

Yeahm you're looking at OOC JPEGs, which are obviously going to favor Canon.

Sure D800 ones look bigger but 5D3 hold better finer details and resolve the textual details better.

JPWGs yes, but as Slideshow Bob demonstrated, the raws tell as different story.

What special really that something 5D3 even 5D2 cannot do in controlled studio?

Then show us an example.

Reply   Reply with quote   Complain
John Sheehy
Forum ProPosts: 16,055
Like?
Re: Fred Miranda comparison confirms DXO results. 5D III blown away.
In reply to tony field, Apr 27, 2012

tony field wrote:

Why bother with a high pixel count camera if you do not use it as such.

Many reasons. A lower MP camera (or lower output image pixel dimensions) does not fare as well resampling to arbitrary sizes, rotating, correcting CA, or correcting perspective or geometric distortions. The best data to start with is always data that is oversampled. No matter what you do with it, you have the opportunity to do it right. Too many comparisons of different MP images purposely avoid scenarios where it makes the most difference; comparing out-of-camera is only useful if all you ever do is display images out-of-camera without any editing or corrections.

You want to see the difference between the high count and low count images. On the other hand, if all you do is display for web or maybe make an 8x10 print on occasion, who cares what the pixel count is.

I do. It is always better to start out with oversampled data. Starting with anything less leads to inferior quality. You should have enough pixels that a 100% crop is very soft on a 100 PPI monitor, if you want quality data.

-- hide signature --

John

Reply   Reply with quote   Complain
ultimitsu
Veteran MemberPosts: 5,577
Like?
Re: But that's not how it works
In reply to stephenmelvin, Apr 27, 2012

stephenmelvin wrote:

Nope, you scale the higher res camera down. It will still possess an advantage, since it starts with a lot more information.
That's how DXO does it for part of their test.

DXO does it because for DXO's test they dont print big and dont care about extra resolution. but for a test that concerns resolution, downsizing the larger one is plain stupid.

So, for example, you want to print an 8x12 on a printer that's 300 ppi. That's about 8 megapixels. Guess what? The RIP throws the extra resolution away.

This is why the D800's resolution advantage doesn't show up until the prints get really big.

but what is really big? 8 x 12 uses about 10 mp at 300 DPI, 16 x 24 would use 40 mp at 300 DPI, so D800 would be the minimum to print a decent 16 x 24 at 300 DPI, 5d3 would fall well short. 16 x 24 is big but not that big.

on top of that many many shots get cropped, especially for sports and wildlife. so it is just nonsensical to say resolution advantage is not useful until very large prints.

Reply   Reply with quote   Complain
John Sheehy
Forum ProPosts: 16,055
Like?
Re: Old news
In reply to stephenmelvin, Apr 27, 2012

stephenmelvin wrote:

Thing is, the DR advantage of the D800 goes away rapidly as the ISO goes up. It's about 2 1/2 stops at ISO 100, but by ISO 160, it's under two stops (9.12 vs 10.84). At ISO 320, the advantage is about a stop (8.89 vs 9.95) and by ISO 1280, they're practically the same.

You're overlooking the real meaning, and concentrating on trivia. DR is not a direct issue at high ISOs, per se. High-ISO DR is a red herring, when the absolute read noise is basically the same at all ISOs. That is because high ISOs are basically just a routine of clipping away the sensor's highest tones.

The overwhelming fact is that if you shoot at an exposure index of ISO 3200 from an ISO setting of 100 on both a D800, and a Canon DSLR, you get a reasonable ISO 3200 image after processing on the D800, and worthless garbage on the Canon, with 6x as much read noise.

-- hide signature --

John

Reply   Reply with quote   Complain
Press Correspondent
Veteran MemberPosts: 3,342Gear list
Like?
Just one?
In reply to carlk, Apr 27, 2012

Anyone disagrees that 14-24 is the ONLY Nikkor lens to better Canon? Or am I on wrong forum?

 Press Correspondent's gear list:Press Correspondent's gear list
Olympus XZ-1 Canon EOS 5D Mark III Canon EF 50mm f/1.4 USM Canon EF 17-40mm f/4.0L USM Canon EF 70-200mm f/4.0L USM +11 more
Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads