Crystal Ball Required... Locked

Started Jul 9, 2013 | Discussions
This thread is locked.
SushiEater
Senior MemberPosts: 2,002Gear list
Re: Crystal Ball Required...
In reply to qianp2k, Jul 13, 2013

Here is what D800 can easily do and 5D3 can't without banding and excessive noise.

Original.

Processed. Small picture does not do it justice. And there is something wrong with DPR displaying pictures because on my screen in PS6 the same picture has a lot more contrast.

 SushiEater's gear list:SushiEater's gear list
Nikon D800E Nikon D810 Nikon AF-S Nikkor 16-35mm f/4G ED VR Sigma 70-200mm F2.8 EX DG OS HSM Nikon AF-S Nikkor 24-120mm f/4G ED VR +10 more
maljo@inreach.com
Veteran MemberPosts: 6,680
5D3 or D800? Neither will limit your photography
In reply to King Redshirt, Jul 13, 2013

These are two terrific cameras.  Pick either one and go have a blast!

maljo

Monterey sunset, tripod, 70-200 f2.8 zoom at f8 and 200 mm, modern DSLR

qianp2k
Forum ProPosts: 10,350Gear list
Re: Crystal Ball Required...
In reply to SushiEater, Jul 13, 2013

SushiEater wrote:

No, I am not saying that D800 has better AF. Both cameras are about equal after FW update on D800.

What I am saying is that in order to shoot a person/actor being interviewed (not posing for you) in a very low light you need to get a shutter speed of at least 1/160 and take a lot of shots to get few sharp.I would love to get more like 1/250-1/320.

OK based on what you said above and your another post below, I don't see why Nikon cameras have particular advantages in these scenes at high ISO.  Actually based on two lenses you described, Canon 70-200L/2.8 IS II is sharper than Nikon counterpart, 5DIII AF in low light w/o using assistant light, is as good as or could be slightly better than D800.  If you could afford another camera, then get Canon 300L/2.8 IS II instead of Sigma 120-300/2.8 OS, that again is better than Nikon 300G/2.8 VR.

These are not just paparazzi shots. I usually pre-focus and wait until I get some kind of expression on the face.

5DIII has 61 AF pt, 45-cross-pt among them (faster lenses have more cross-pt available to choose from) so you can select a cross-pt and 5D3 can shoot faster in burst rate.

Even a simple sudden laugh can ruin the picture. I was struggling with 5D2 and 7D. With D800 there is a lot less struggling because I can underexpose (thus getting the higher shutter speed) and bring it back without banding.

5D2 and 7D are old cameras and 7D is crop camera so not at the same level of D800.  But 5D3 is a different story.  In high ISOs, 5D3 actually has higher DR than D800.  Technically there is no difference from perspective of noises/grain between you expose correctly (on your subject, the interviewed person) at one ISO stop and you underexposed (in order to use faster shutter) and then push 2-stop back.  Canon cameras have no more banding than Nikon in high ISOs as Canon cameras have higher DR than Nikon's ones in high ISOs that confirmed in DXOMark sensor's tests.

I'd say 5D3 will handle as good as D800 in your described scenes.

-- 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
qianp2k
Forum ProPosts: 10,350Gear list
Re: Crystal Ball Required...
In reply to SushiEater, Jul 13, 2013

SushiEater wrote:

Here is what D800 can easily do and 5D3 can't without banding and excessive noise.

this is a typical method of extreme ETTL (expose on left) that you exposed on highlight and then push a severely unexposed photo many stops back, not a good method in general.  nobody disputes by doing the same this way, Nikon wins hands down.  But I will not do this way.

Original.

Resulted a severely underexposed photo with your main subject (80% of scene) buried in deep dark shadow.

Processed. Small picture does not do it justice. And there is something wrong with DPR displaying pictures because on my screen in PS6 the same picture has a lot more contrast.

by extreme shadow lifting we see typical consequence of side effects - surreal look, distorted color tonality, excessive noises/grains that I already can see with such small size, would be very obvious if you export large size.  This is a scene you should use better technique - GND or reverse GND filter along the horizontal line.  Sure the right top small mountain area will be a bit darker but that doesn't have much details in your photo anyway.  However the vast 80% of your lower part areas will have much more details, cleaner and will have better color tonality if you expose correctly.  Another unimpressive sample of showing bad technique.  If you want show us D800 shadow pulling ability - yes it's impressive and we have seen enough such deep dark shadow pulling games, but not impressive from perspective of final photo's IQ.

-- 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
Mako2011
Mako2011 MOD
Forum ProPosts: 16,065
Re: Crystal Ball Required...
In reply to SushiEater, Jul 13, 2013

SushiEater wrote:

Here is what D800 can easily do and 5D3 can't without banding and excessive noise.

Original.

Processed. Small picture does not do it justice. And there is something wrong with DPR displaying pictures because on my screen in PS6 the same picture has a lot more contrast.

Pretty cool example. You can to well with the 5DmkIII but do indeed have to resort to aggressive demanding and NR. Topaz has improved a bit in this are. But can't show detail that was not captured.   I like your take on the sky!  Pretty rad.

-- hide signature --

My opinions are my own and not those of DPR or its administration. They carry no 'special' value (except to me and Lacie of course)

SushiEater
Senior MemberPosts: 2,002Gear list
Re: Crystal Ball Required...
In reply to qianp2k, Jul 14, 2013

qianp2k wrote:

SushiEater wrote:

No, I am not saying that D800 has better AF. Both cameras are about equal after FW update on D800.

What I am saying is that in order to shoot a person/actor being interviewed (not posing for you) in a very low light you need to get a shutter speed of at least 1/160 and take a lot of shots to get few sharp.I would love to get more like 1/250-1/320.

OK based on what you said above and your another post below, I don't see why Nikon cameras have particular advantages in these scenes at high ISO. Actually based on two lenses you described, Canon 70-200L/2.8 IS II is sharper than Nikon counterpart, 5DIII AF in low light w/o using assistant light, is as good as or could be slightly better than D800. If you could afford another camera, then get Canon 300L/2.8 IS II instead of Sigma 120-300/2.8 OS, that again is better than Nikon 300G/2.8 VR.

Please explain where exactly did I mention focusing or lens sharpness problem?
Why in the world would I sell very excellent zoom lens for a prime which is not any better? And please don't start me on that. The problem is to stop subject movement and the only way to do it is with high shutter speed that 5D3 can't provide because I have to use proper exposure and I can't.

These are not just paparazzi shots. I usually pre-focus and wait until I get some kind of expression on the face.

5DIII has 61 AF pt, 45-cross-pt among them (faster lenses have more cross-pt available to choose from) so you can select a cross-pt and 5D3 can shoot faster in burst rate.

5D3 can have 150 cross points how is that going to help if focusing is not a problem with either camera.

Even a simple sudden laugh can ruin the picture. I was struggling with 5D2 and 7D. With D800 there is a lot less struggling because I can underexpose (thus getting the higher shutter speed) and bring it back without banding.

5D2 and 7D are old cameras and 7D is crop camera so not at the same level of D800. But 5D3 is a different story.

Not true at all. 5D3 is about 1/2 better in RAW than 5D2 at best. Search here, many people ran that test a year ago. But again it has nothing to do with underexposure and subsequent banding.

In high ISOs, 5D3 actually has higher DR than D800.

First of all it doesn't. Second of all, banding is not really related with DR. And third, DR has nothing to do with it because I am not recovering highlights so I never reach full potential of either cameras. I can underexpose D800 is much as want and it will never show banding. Just take a look in the other post below. How much underexposure do you think that is?

Technically there is no difference from perspective of noises/grain between you expose correctly (on your subject, the interviewed person) at one ISO stop and you underexposed (in order to use faster shutter) and then push 2-stop back. Canon cameras have no more banding than Nikon in high ISOs as Canon cameras have higher DR than Nikon's ones in high ISOs that confirmed in DXOMark sensor's tests.

Really. In what universe? Actually DXO was measured at really high ISO, not at 3200. In either case I don't shoot higher than 3200 anyway.

I'd say 5D3 will handle as good as D800 in your described scenes.

-- hide signature --
 SushiEater's gear list:SushiEater's gear list
Nikon D800E Nikon D810 Nikon AF-S Nikkor 16-35mm f/4G ED VR Sigma 70-200mm F2.8 EX DG OS HSM Nikon AF-S Nikkor 24-120mm f/4G ED VR +10 more
SushiEater
Senior MemberPosts: 2,002Gear list
Re: Crystal Ball Required...
In reply to qianp2k, Jul 14, 2013

I could use a stronger word to describe your condition but I am sure now that you are CRAZY and certainly don't have a clue and experience. The whole idea behind this shot was not to overblown the sky at 30 second exposure to get a misty look from water. And you don't see much grain or noise, you are just saying it. In fact all I did is converted from RAW without any processing for noise or banding for that matter.

Here is a very small 100% crop at full resolution. There is no noise, period.

Post your 5D3 shot in this situation and lets see what it looks like.

 SushiEater's gear list:SushiEater's gear list
Nikon D800E Nikon D810 Nikon AF-S Nikkor 16-35mm f/4G ED VR Sigma 70-200mm F2.8 EX DG OS HSM Nikon AF-S Nikkor 24-120mm f/4G ED VR +10 more
Andrew53
Regular MemberPosts: 280Gear list
Re: Crystal Ball Required...
In reply to King Redshirt, Jul 14, 2013

I have or had all three of these cameras. I am a long time Canon user.

1. Canon 5D Mark III

2. Nikon D800

5. Nikon D600

D600 - I had endless problems with white balance, which is why I eventually sold the camera. The dust issue is/was real and unacceptable in my opinion. IQ was probably the best of all three of theses cameras by a small margin. AF is OK but not as good as the D800. I think they have fixed the white balance and if they fixed the dust then this camera would move up to number 3.

D800 - Excellent camera with excellent IQ. 36mpx rarely adds any value unless you are on a tripod with an excellent lens and perfect technique. However, the extra megapixes do not make the IQ any worse. They do however make image processing significantly slower. For a wedding photographer who does thousands of images in a day this can be a big issue. For personal use the camera is excellent. Shooting bank system is just about the dumbest thing I have seen in 45 years of working with electronic and camera stuff.

Canon 5D Mark III. IQ is very close to the D600 and D800. The Nikons have slightly better IQ but focus errors and lenses will make more difference. Excellent AF - better than either of the Nikons. Better line up of lenses. L-series lenses have consistent quality, consistent contrast, consistent colours. The silent shutter mode is very nice. For non intrusive photography the only thing better is a leaf shutter or a point and shoot with electronic shutter. Must be used to understand just how much quieter this is than the D800 or D600.

My 2cents from owning and using all three cameras. I have kept the D800 for the 14-24 and 28-300VR. For the rest I prefer Canon lenses.

 Andrew53's gear list:Andrew53's gear list
Fujifilm FinePix X100 Sony RX100 Fujifilm X100S Canon EOS 5D Mark II Leica M9 +65 more
qianp2k
Forum ProPosts: 10,350Gear list
Re: Crystal Ball Required...
In reply to SushiEater, Jul 14, 2013

SushiEater wrote:

I could use a stronger word to describe your condition but I am sure now that you are CRAZY and certainly don't have a clue and experience.

What you're talking about. Please be calm and discuss/debate with merits not thru emotion please.

The whole idea behind this shot was not to overblown the sky at 30 second exposure to get a misty look from water.

Still can do and you can shoot ISO 100 instead with a GND filter or a reverse GND filter. You then expose on your 80% of front scene that will be much detailed, cleaner and sharper. The bright sky is protected by GND filter that you can fully recover.

And you don't see much grain or noise, you are just saying it. In fact all I did is converted from RAW without any processing for noise or banding for that matter.

Here is a very small 100% crop at full resolution. There is no noise, period.

this is the brightest area in your 80% of lower part of scene. Sure you can apply NR to reduce noises but that effectively smear details on already lost details from severe underexposure. They are not very detailed, not very sharp. Please show those very deep dark shadows at 100% size on sides of rock for example.

Post your 5D3 shot in this situation and lets see what it looks like.

But the point is that I will NOT shoot in your way as I said. I will use a GND or reverse GND filter on horizontal, expose correctly at ISO 100 instead that could even expose longer as I expose on front dark area not on sky that protected by GND or reverse GND filter. You saw my BIF samples in page 2 of this thread. Now please show us how your D800 can take better photos if you expose on sky then pull a deep dark birds 3-6 stops from shadow. I am waiting to see.

http://www.dpreview.com/forums/thread/3514360?page=2

'Exposure on highlight then push deep dark shadows in extreme' is not a good technique in general, and will not replace traditional good techniques.

This is nothing to do with 5D3 vs D800.  Even 5D3 had 14-stop DR or I shoot with D800, I STILL will not shoot in your way but on the way I descried above, period.

-- 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
qianp2k
Forum ProPosts: 10,350Gear list
Re: Crystal Ball Required...
In reply to SushiEater, Jul 14, 2013

SushiEater wrote:

qianp2k wrote:

SushiEater wrote:

No, I am not saying that D800 has better AF. Both cameras are about equal after FW update on D800.

What I am saying is that in order to shoot a person/actor being interviewed (not posing for you) in a very low light you need to get a shutter speed of at least 1/160 and take a lot of shots to get few sharp.I would love to get more like 1/250-1/320.

OK based on what you said above and your another post below, I don't see why Nikon cameras have particular advantages in these scenes at high ISO. Actually based on two lenses you described, Canon 70-200L/2.8 IS II is sharper than Nikon counterpart, 5DIII AF in low light w/o using assistant light, is as good as or could be slightly better than D800. If you could afford another camera, then get Canon 300L/2.8 IS II instead of Sigma 120-300/2.8 OS, that again is better than Nikon 300G/2.8 VR.

Please explain where exactly did I mention focusing or lens sharpness problem?

My points are that 5D3 AF lock-on speed is likely a bit faster and Canon lenses are better that both help to increase AF hit rate at specific shutter speed.

Why in the world would I sell very excellent zoom lens for a prime which is not any better? And please don't start me on that.

Sigma 120-300/2.8 OS is not as good as Canon 300L/2.8 IS II as latter not only sharper but AF lock on faster that would help you in your job.  I also believe two cameras not one camera solution.

The problem is to stop subject movement and the only way to do it is with high shutter speed that 5D3 can't provide because I have to use proper exposure and I can't.

Again my point is that 5D3 with better Canon lenses will help you to improve hit rate.  But still technically no difference between you shoot at ISO 3200 then push 2-stop back and at ISO 12800 or slightly lower.

These are not just paparazzi shots. I usually pre-focus and wait until I get some kind of expression on the face.

5DIII has 61 AF pt, 45-cross-pt among them (faster lenses have more cross-pt available to choose from) so you can select a cross-pt and 5D3 can shoot faster in burst rate.

5D3 can have 150 cross points how is that going to help if focusing is not a problem with either camera.

From what I have read, 5D AF lock-on speed is faster, shooting fast, better in AI-servo mode and more responsible overall that will increase your hit rate

Even a simple sudden laugh can ruin the picture. I was struggling with 5D2 and 7D. With D800 there is a lot less struggling because I can underexpose (thus getting the higher shutter speed) and bring it back without banding.

5D2 and 7D are old cameras and 7D is crop camera so not at the same level of D800. But 5D3 is a different story.

Not true at all. 5D3 is about 1/2 better in RAW than 5D2 at best. Search here, many people ran that test a year ago. But again it has nothing to do with underexposure and subsequent banding.

D800 doesn't have better high ISO than 5D3 and even doesn't have better DR than 5D3 in high ISO.

In high ISOs, 5D3 actually has higher DR than D800.

First of all it doesn't.

yes, according to DXOMark, 5D3 DR even out D800 at ISO 2000, and then is slightly better in high ISO, higher ISO, better from 5D3 although not huge difference.

Second of all, banding is not really related with DR.

related but Canon cameras don't have more banding than Nikon in high ISOs, certainly not at ISO 3200.

And third, DR has nothing to do with it because I am not recovering highlights so I never reach full potential of either cameras. I can underexpose D800 is much as want and it will never show banding. Just take a look in the other post below. How much underexposure do you think that is?

At ISO 3200, whatever you can do with D800 you also can do with 5D3.

Technically there is no difference from perspective of noises/grain between you expose correctly (on your subject, the interviewed person) at one ISO stop and you underexposed (in order to use faster shutter) and then push 2-stop back. Canon cameras have no more banding than Nikon in high ISOs as Canon cameras have higher DR than Nikon's ones in high ISOs that confirmed in DXOMark sensor's tests.

Really. In what universe? Actually DXO was measured at really high ISO, not at 3200. In either case I don't shoot higher than 3200 anyway.

5D3 is no less than D800 at ISO 3200 from either DR or SNR perspective according to DXOMark.

I'd say 5D3 will handle as good as D800 in your described scenes.

-- hide signature --
-- 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
qianp2k
Forum ProPosts: 10,350Gear list
Re: 5D3 or D800? Neither will limit your photography
In reply to maljo@inreach.com, Jul 14, 2013

maljo@inreach.com wrote:

These are two terrific cameras. Pick either one and go have a blast!

Agreed, both can take excellent photos if you do correctly or both can take crappy photos if you don't do correctly.

It's an excellent photo, and I don't see anyone would need to open shadow in this photo

See a good photographer knows to use tools, good techniques and post processing to take his photos, and obviously he doesn't have to shoot with Nikon D800

http://www.dpreview.com/articles/7249705584/shooting-godafoss-erez-marom-takes-us-behind-the-picture

maljo

Monterey sunset, tripod, 70-200 f2.8 zoom at f8 and 200 mm, modern DSLR

-- 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
SushiEater
Senior MemberPosts: 2,002Gear list
Re: Crystal Ball Required...
In reply to qianp2k, Jul 14, 2013

My points are that 5D3 AF lock-on speed is likely a bit faster and Canon lenses are better that both help to increase AF hit rate at specific shutter speed.

Your point is meaningless because there is no problem with AF. And faster (if it was faster to begin with) has nothing to do with shutter speed. You are sounding like an amateur now just to prove something that is not a problem.

Why in the world would I sell very excellent zoom lens for a prime which is not any better? And please don't start me on that.

Sigma 120-300/2.8 OS is not as good as Canon 300L/2.8 IS II as latter not only sharper but AF lock on faster that would help you in your job. I also believe two cameras not one camera solution.

That is very debatable in the first place but why are you even talking about it since it is not the problem?

The problem is to stop subject movement and the only way to do it is with high shutter speed that 5D3 can't provide because I have to use proper exposure and I can't.

Again my point is that 5D3 with better Canon lenses will help you to improve hit rate. But still technically no difference between you shoot at ISO 3200 then push 2-stop back and at ISO 12800 or slightly lower.

How many times do I have to tell you that hit rate is not a problem?

And yes there is big difference between shooting Nikon at 3200 with pushing and Canon at 12800. There is a big difference just to start using 5D3 at 12800 to begin with.

These are not just paparazzi shots. I usually pre-focus and wait until I get some kind of expression on the face.

5DIII has 61 AF pt, 45-cross-pt among them (faster lenses have more cross-pt available to choose from) so you can select a cross-pt and 5D3 can shoot faster in burst rate.

5D3 can have 150 cross points how is that going to help if focusing is not a problem with either camera.

From what I have read, 5D AF lock-on speed is faster, shooting fast, better in AI-servo mode and more responsible overall that will increase your hit rate

Again, hit rate is not a problem. Sudden subject movement is.

Even a simple sudden laugh can ruin the picture. I was struggling with 5D2 and 7D. With D800 there is a lot less struggling because I can underexpose (thus getting the higher shutter speed) and bring it back without banding.

5D2 and 7D are old cameras and 7D is crop camera so not at the same level of D800. But 5D3 is a different story.

Not true at all. 5D3 is about 1/2 better in RAW than 5D2 at best. Search here, many people ran that test a year ago. But again it has nothing to do with underexposure and subsequent banding.

D800 doesn't have better high ISO than 5D3 and even doesn't have better DR than 5D3 in high ISO.

In high ISOs, 5D3 actually has higher DR than D800.

First of all it doesn't.

yes, according to DXOMark, 5D3 DR even out D800 at ISO 2000, and then is slightly better in high ISO, higher ISO, better from 5D3 although not huge difference.

Second of all, banding is not really related with DR.

related but Canon cameras don't have more banding than Nikon in high ISOs, certainly not at ISO 3200.

Of course Canon cameras have banding, even at normal exposure. There are several posts here about it, just search.

And third, DR has nothing to do with it because I am not recovering highlights so I never reach full potential of either cameras. I can underexpose D800 is much as want and it will never show banding. Just take a look in the other post below. How much underexposure do you think that is?

At ISO 3200, whatever you can do with D800 you also can do with 5D3.

Not true. I can recover at least 2 stops of underexposure from D800 but can't from 5D3.

Technically there is no difference from perspective of noises/grain between you expose correctly (on your subject, the interviewed person) at one ISO stop and you underexposed (in order to use faster shutter) and then push 2-stop back. Canon cameras have no more banding than Nikon in high ISOs as Canon cameras have higher DR than Nikon's ones in high ISOs that confirmed in DXOMark sensor's tests.

Really. In what universe? Actually DXO was measured at really high ISO, not at 3200. In either case I don't shoot higher than 3200 anyway.

5D3 is no less than D800 at ISO 3200 from either DR or SNR perspective according to DXOMark.

I don't care about DXO, they change their minds every five seconds. Besides even at high ISO the difference is minor. How often do you shoot at 12800 or higher?

I care what I can do with 5D3 and D800. 5D3 can't even recover from ISO 100. Just a quick search.

http://www.fredmiranda.com/5DIII-D800/index_controlled-tests.html

http://horshack.smugmug.com/photos/i-7hpFr5V/1/O/i-7hpFr5V.jpg

And if you believe DXO so much how about DXO sensor rating? D800e-96, 5D3-81!!!! Now that is HUGE difference.

No matter how you cut it D800 IQ is so much better than 5D3.

And that idiot who compared identical lenses on both cameras and claimed that Canon can get close to Nikon is totally wrong because he did not put good identical lenses on both cameras.The emphasis on Identical here. All he proved that some Canon lenses might be better than Nikon.

I'd say 5D3 will handle as good as D800 in your described scenes.

Until you actually try it you can say whatever you want but it is not necessary the truth.

The interesting thing is that you are only talking and have not posted a single example.

-- hide signature --
-- hide signature --
 SushiEater's gear list:SushiEater's gear list
Nikon D800E Nikon D810 Nikon AF-S Nikkor 16-35mm f/4G ED VR Sigma 70-200mm F2.8 EX DG OS HSM Nikon AF-S Nikkor 24-120mm f/4G ED VR +10 more
SushiEater
Senior MemberPosts: 2,002Gear list
Re: Crystal Ball Required...
In reply to qianp2k, Jul 14, 2013

qianp2k wrote:

SushiEater wrote:

I could use a stronger word to describe your condition but I am sure now that you are CRAZY and certainly don't have a clue and experience.

What you're talking about. Please be calm and discuss/debate with merits not thru emotion please.

I am not emotional at all, I am just stating the obvious.

The whole idea behind this shot was not to overblown the sky at 30 second exposure to get a misty look from water.

Still can do and you can shoot ISO 100 instead with a GND filter or a reverse GND filter. You then expose on your 80% of front scene that will be much detailed, cleaner and sharper. The bright sky is protected by GND filter that you can fully recover.

I can do a lot of difference things but this thread is not about what I could do or what I did. I could have brought powerful light and paint those rocks with light.

But this thread is about if can Canon 5D3 recover from underexposure. And the answer is, it can't. As simple as that.

And you don't see much grain or noise, you are just saying it. In fact all I did is converted from RAW without any processing for noise or banding for that matter.

Here is a very small 100% crop at full resolution. There is no noise, period.

this is the brightest area in your 80% of lower part of scene. Sure you can apply NR to reduce noises but that effectively smear details on already lost details from severe underexposure. They are not very detailed, not very sharp. Please show those very deep dark shadows at 100% size on sides of rock for example.

But I didn't. This crop was neither sharpened or noise reduced. And the brightest part of the scene still remains the sky. And you want me to show details on the rock that was half a mile away even more underexposed? Seriously? I want it to remain a silhouette so I don't care about details on it.

Post your 5D3 shot in this situation and lets see what it looks like.

But the point is that I will NOT shoot in your way as I said. I will use a GND or reverse GND filter on horizontal, expose correctly at ISO 100 instead that could even expose longer as I expose on front dark area not on sky that protected by GND or reverse GND filter.

No, the point is the same as above.

You saw my BIF samples in page 2 of this thread. Now please show us how your D800 can take better photos if you expose on sky then pull a deep dark birds 3-6 stops from shadow. I am waiting to see.

http://www.dpreview.com/forums/thread/3514360?page=2

Why in the world would I do that then it is not required?

'Exposure on highlight then push deep dark shadows in extreme' is not a good technique in general, and will not replace traditional good techniques.

That is your opinion and not mine. Or so using GND is? I wanted high contrast between background and the sky and GND would not give me that look. That was the look I was going for. You don't like it? It is your problem. But considering that I sold 2 of them at $250 a pop I would say that I hit it right on the head. Besides you are constantly going off topic. We are not criticizing my photo we are talking about 5D3 ability to recover from underexposure.

This is nothing to do with 5D3 vs D800. Even 5D3 had 14-stop DR or I shoot with D800, I STILL will not shoot in your way but on the way I descried above, period.

Again, your opinion and not mine.

-- hide signature --
 SushiEater's gear list:SushiEater's gear list
Nikon D800E Nikon D810 Nikon AF-S Nikkor 16-35mm f/4G ED VR Sigma 70-200mm F2.8 EX DG OS HSM Nikon AF-S Nikkor 24-120mm f/4G ED VR +10 more
DtEW
Senior MemberPosts: 1,854Gear list
Re: Crystal Ball Required...
In reply to SushiEater, Jul 14, 2013

SushiEater wrote:

I could use a stronger word to describe your condition but I am sure now that you are CRAZY and certainly don't have a clue and experience. The whole idea behind this shot was not to overblown the sky at 30 second exposure to get a misty look from water.

I think in your claimed massive amount of experience, you missed that the D800's "ISO50" (LO1) is actually an expansion, i.e. digitally pulled from ISO100. See here:

http://www.dpreview.com/reviews/nikon-d800-d800e/19

If the "whole idea behind this shot was not to overblown (sic) the sky at 30 second exposure to get a misty look from water," then congratulations... you choice of "ISO50" made the sky that much easier to blow out by shrinking your highlight range.

And you don't see much grain or noise, you are just saying it. In fact all I did is converted from RAW without any processing for noise or banding for that matter.

Here is a very small 100% crop at full resolution. There is no noise, period.

Post your 5D3 shot in this situation and lets see what it looks like.

Actually, a 5DMKIII user (not to mention a competent D800 user) at ISO100 and an ND2 filter would do this shot much better than your attempt.  Could've probably even gone to 1 minute for much smoother water, since the reduction in highlight range going from the D800's ISO100 to LO1 ("ISO50") is about 1 entire stop.

I make no claims for the relative capabilities for the two cameras, just that your "experience" claim is somewhat dubious.

 DtEW's gear list:DtEW's gear list
Canon PowerShot G11 Canon EOS 6D Sony a6000 Canon EF 100mm f/2.0 USM Canon EF 75-300mm f/4.0-5.6 IS USM +18 more
The Davinator
Forum ProPosts: 13,267Gear list
Re: Crystal Ball Required...
In reply to DtEW, Jul 14, 2013

DtEW wrote:

SushiEater wrote:

I could use a stronger word to describe your condition but I am sure now that you are CRAZY and certainly don't have a clue and experience. The whole idea behind this shot was not to overblown the sky at 30 second exposure to get a misty look from water.

I think in your claimed massive amount of experience, you missed that the D800's "ISO50" (LO1) is actually an expansion, i.e. digitally pulled from ISO100. See here:

http://www.dpreview.com/reviews/nikon-d800-d800e/19

If the "whole idea behind this shot was not to overblown (sic) the sky at 30 second exposure to get a misty look from water," then congratulations... you choice of "ISO50" made the sky that much easier to blow out by shrinking your highlight range.

And you don't see much grain or noise, you are just saying it. In fact all I did is converted from RAW without any processing for noise or banding for that matter.

Here is a very small 100% crop at full resolution. There is no noise, period.

Post your 5D3 shot in this situation and lets see what it looks like.

Actually, a 5DMKIII user (not to mention a competent D800 user) at ISO100 and an ND2 filter would do this shot much better than your attempt. Could've probably even gone to 1 minute for much smoother water, since the reduction in highlight range going from the D800's ISO100 to LO1 ("ISO50") is about 1 entire stop.

I make no claims for the relative capabilities for the two cameras, just that your "experience" claim is somewhat dubious.

As the 5D3 has the same issue at iso 50, your point is moot.

 The Davinator's gear list:The Davinator's gear list
Canon PowerShot G3 Canon PowerShot SX150 IS Canon EOS D30 Canon EOS 10D Nikon D2X +18 more
SushiEater
Senior MemberPosts: 2,002Gear list
Re: Crystal Ball Required...
In reply to DtEW, Jul 14, 2013

I just love people who through theories around without anything to back them up.

But just for you so experienced, if sky is blown (in this light) it looks ugly yellowish and there is nothing you can do to bring it back. At ISO 100 it would not be a 30 second shot it would be a 15 seconds shot.

I am also not a god so I could not stop the time Sun was rapidly setting.

But again, this is off topic. The only reason I posted this picture is to show how much D800 can recover where in this situation 5D3 would have horrible banding.

 SushiEater's gear list:SushiEater's gear list
Nikon D800E Nikon D810 Nikon AF-S Nikkor 16-35mm f/4G ED VR Sigma 70-200mm F2.8 EX DG OS HSM Nikon AF-S Nikkor 24-120mm f/4G ED VR +10 more
qianp2k
Forum ProPosts: 10,350Gear list
Re: Crystal Ball Required...
In reply to SushiEater, Jul 14, 2013

SushiEater wrote:

My points are that 5D3 AF lock-on speed is likely a bit faster and Canon lenses are better that both help to increase AF hit rate at specific shutter speed.

Your point is meaningless because there is no problem with AF. And faster (if it was faster to begin with) has nothing to do with shutter speed. You are sounding like an amateur now just to prove something that is not a problem.

I am talking about overall AF response time and AF lock-on speed. From what I have heard 5D3 edges out D800 in both areas that help you to capture the moment and improve AF hit rate.

Why in the world would I sell very excellent zoom lens for a prime which is not any better? And please don't start me on that.

Sigma 120-300/2.8 OS is not as good as Canon 300L/2.8 IS II as latter not only sharper but AF lock on faster that would help you in your job. I also believe two cameras not one camera solution.

That is very debatable in the first place but why are you even talking about it since it is not the problem?

It will improve your AF lock-on speed and will be sharper if you use two cameras.

The problem is to stop subject movement and the only way to do it is with high shutter speed that 5D3 can't provide because I have to use proper exposure and I can't.

Again my point is that 5D3 with better Canon lenses will help you to improve hit rate. But still technically no difference between you shoot at ISO 3200 then push 2-stop back and at ISO 12800 or slightly lower.

How many times do I have to tell you that hit rate is not a problem?

Did you ever talk blurry? I bet many of some of those blurry not caused by shutter speed but a small lag of AF lock-on speed in panning mode.

And yes there is big difference between shooting Nikon at 3200 with pushing and Canon at 12800.

No, 5D3 doesn't have more banding issue than D800 at ISO 3200.

There is a big difference just to start using 5D3 at 12800 to begin with.

Technically no difference between you shoot at ISO 3200, then increase exposure 2 stops in software (I am not talking tone curve adjustment but moving exposure bar) and you shoot at ISO 12800. Otherwise if both shoot at ISO 3200 I don't see why D800 has an obvious advantage.

These are not just paparazzi shots. I usually pre-focus and wait until I get some kind of expression on the face.

5DIII has 61 AF pt, 45-cross-pt among them (faster lenses have more cross-pt available to choose from) so you can select a cross-pt and 5D3 can shoot faster in burst rate.

5D3 can have 150 cross points how is that going to help if focusing is not a problem with either camera.

From what I have read, 5D AF lock-on speed is faster, shooting fast, better in AI-servo mode and more responsible overall that will increase your hit rate

Again, hit rate is not a problem. Sudden subject movement is.

Again it's partially related to AF lock-on speed and hope you get. How many PJs use Canon cameras as we see in TV these days?

Even a simple sudden laugh can ruin the picture. I was struggling with 5D2 and 7D. With D800 there is a lot less struggling because I can underexpose (thus getting the higher shutter speed) and bring it back without banding.

5D2 and 7D are old cameras and 7D is crop camera so not at the same level of D800. But 5D3 is a different story.

Not true at all. 5D3 is about 1/2 better in RAW than 5D2 at best. Search here, many people ran that test a year ago. But again it has nothing to do with underexposure and subsequent banding.

D800 doesn't have better high ISO than 5D3 and even doesn't have better DR than 5D3 in high ISO.

In high ISOs, 5D3 actually has higher DR than D800.

First of all it doesn't.

yes, according to DXOMark, 5D3 DR even out D800 at ISO 2000, and then is slightly better in high ISO, higher ISO, better from 5D3 although not huge difference.

Second of all, banding is not really related with DR.

related but Canon cameras don't have more banding than Nikon in high ISOs, certainly not at ISO 3200.

Of course Canon cameras have banding, even at normal exposure. There are several posts here about it, just search.

All cameras have banding more or less. Canon banding is ONLY obviously when you severely underexpose and then push many stops especially in base ISO.

And third, DR has nothing to do with it because I am not recovering highlights so I never reach full potential of either cameras. I can underexpose D800 is much as want and it will never show banding. Just take a look in the other post below. How much underexposure do you think that is?

At ISO 3200, whatever you can do with D800 you also can do with 5D3.

Not true. I can recover at least 2 stops of underexposure from D800 but can't from 5D3.

I don't think so at ISO 3200. You need to provide scientific samples taken side by side at ISO 3200 with comparable lenses and let us see.

Technically there is no difference from perspective of noises/grain between you expose correctly (on your subject, the interviewed person) at one ISO stop and you underexposed (in order to use faster shutter) and then push 2-stop back. Canon cameras have no more banding than Nikon in high ISOs as Canon cameras have higher DR than Nikon's ones in high ISOs that confirmed in DXOMark sensor's tests.

Really. In what universe? Actually DXO was measured at really high ISO, not at 3200. In either case I don't shoot higher than 3200 anyway.

5D3 is no less than D800 at ISO 3200 from either DR or SNR perspective according to DXOMark.

I don't care about DXO, they change their minds every five seconds. Besides even at high ISO the difference is minor. How often do you shoot at 12800 or higher?

the question is for you actually, as you seem need to shoot at ISO 12800.

I care what I can do with 5D3 and D800. 5D3 can't even recover from ISO 100. Just a quick search.

Rubbish. D800's advantage is ONLY obvious in your shooting style that you severely underexposed photos, and then push shadows many stops back at base ISO. I'd not shoot in that way as I said.

http://www.fredmiranda.com/5DIII-D800/index_controlled-tests.html

a) it's mainly for his balance purpose in his report otherwise overwhelmingly he prefers 5D3 and he used 5D3 most times in real world shooting; b) Fred himself said it's better for 5D3 to overexpose a little bit upto +1EV (or ETTR technique that I used); c) it's just for demo purpose so he didn't further process.

http://horshack.smugmug.com/photos/i-7hpFr5V/1/O/i-7hpFr5V.jpg

And if you believe DXO so much how about DXO sensor rating? D800e-96, 5D3-81!!!! Now that is HUGE difference.

a) DXOMark summary total score is highly subjective and vastly overweighed on specific factor such as shadow DR; b) if we check detailed test data except 2-stop DR advantage in shadow areas at base ISO, 5D3 sensor stands well with D800 (such as SNR) in other areas; c) Sensors alone don't take photos but must together with lenses, then the gap is quickly narrowed.

No matter how you cut it D800 IQ is so much better than 5D3.

Much better? Don't think so as confirmed in DXO system (lens on cameras) tests.

http://www.dxomark.com/index.php/Publications/DxOMark-Reviews/Which-lenses-should-you-choose-for-your-Canon-EOS-5D-Mark-III/Canon-EOS-5D-Mark-III-vs.-Nikon-D800-Competition-is-closer-than-expected

24-70L II on 5D3 vs 24-70G on D800 18 vs 15 mpix

70-200L/2.8 IS II on 5D3 vs 70-200G/2.8 VR on D800 21 vs 20 mpix

Whoops, Canon two F2.8 zoom on 5D3 still beat Nikon counterparts on D800 despite 14mp deficit.

And that idiot who compared identical lenses on both cameras and claimed that Canon can get close to Nikon is totally wrong because he did not put good identical lenses on both cameras.The emphasis on Identical here. All he proved that some Canon lenses might be better than Nikon.

I'd say 5D3 will handle as good as D800 in your described scenes.

Until you actually try it you can say whatever you want but it is not necessary the truth.

Many PJs already using 5DIII everyday and I still see more white lenses on Canon on TV these days 

The interesting thing is that you are only talking and have not posted a single example.

I did in Page 2 of this thread, and I am still waiting your BIF photos by exposing on sky and then pulling deep dark birds many stops for comparisons such as large size to show us such technique has not side effects.

http://www.dpreview.com/forums/thread/3514360?page=2

 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
qianp2k
Forum ProPosts: 10,350Gear list
Re: Crystal Ball Required...
In reply to SushiEater, Jul 14, 2013

SushiEater wrote:

qianp2k wrote:

SushiEater wrote:

I could use a stronger word to describe your condition but I am sure now that you are CRAZY and certainly don't have a clue and experience.

What you're talking about. Please be calm and discuss/debate with merits not thru emotion please.

I am not emotional at all, I am just stating the obvious.

I am not sure who is crazy, not me certainly. I am not sure what's obvious? If you talking about shooting in extreme ETTL and pulling dark shadow from both cameras, yes but I don't play in your game as I will not shoot in your way.

The whole idea behind this shot was not to overblown the sky at 30 second exposure to get a misty look from water.

Still can do and you can shoot ISO 100 instead with a GND filter or a reverse GND filter. You then expose on your 80% of front scene that will be much detailed, cleaner and sharper. The bright sky is protected by GND filter that you can fully recover.

I can do a lot of difference things but this thread is not about what I could do or what I did. I could have brought powerful light and paint those rocks with light.

If you believe the way you did as in this sample fine go ahead. But there is a better way that can generate better IQ and hope you listen. Your way is not the only way.

But this thread is about if can Canon 5D3 recover from underexposure. And the answer is, it can't. As simple as that.

No this thread is not about this if you read OP. Nobody disputes D800 can recover a severely unexposed photo better. I am only disputing there is another way and likely can generate better photos.

And you don't see much grain or noise, you are just saying it. In fact all I did is converted from RAW without any processing for noise or banding for that matter.

Here is a very small 100% crop at full resolution. There is no noise, period.

this is the brightest area in your 80% of lower part of scene. Sure you can apply NR to reduce noises but that effectively smear details on already lost details from severe underexposure. They are not very detailed, not very sharp. Please show those very deep dark shadows at 100% size on sides of rock for example.

But I didn't. This crop was neither sharpened or noise reduced. And the brightest part of the scene still remains the sky. And you want me to show details on the rock that was half a mile away even more underexposed? Seriously? I want it to remain a silhouette so I don't care about details on it.

However I seriously don't see much details on this either. If you expose correctly rather had to push shadows many stops, you'd ended with noticeable more details and sharper photo. D800 ISO 800 or ISO 1600 (equivalent after you pushing shadow) cannot be as good as ISO 100 or 200 anyway.

Post your 5D3 shot in this situation and lets see what it looks like.

But the point is that I will NOT shoot in your way as I said. I will use a GND or reverse GND filter on horizontal, expose correctly at ISO 100 instead that could even expose longer as I expose on front dark area not on sky that protected by GND or reverse GND filter.

No, the point is the same as above.

That scene is a perfect sample a GND or a reverse GND should be used as there is a clear horizontal lien there.

You saw my BIF samples in page 2 of this thread. Now please show us how your D800 can take better photos if you expose on sky then pull a deep dark birds 3-6 stops from shadow. I am waiting to see.

http://www.dpreview.com/forums/thread/3514360?page=2

Why in the world would I do that then it is not required?

Just to prove why a severe ETTL (after pushing deep dark shadows many stops) still is as good as, as detailed, as sharp as if you expose correctly (as I did). NO WAY.

'Exposure on highlight then push deep dark shadows in extreme' is not a good technique in general, and will not replace traditional good techniques.

That is your opinion and not mine. Or so using GND is? I wanted high contrast between background and the sky and GND would not give me that look. That was the look I was going for. You don't like it? It is your problem. But considering that I sold 2 of them at $250 a pop I would say that I hit it right on the head. Besides you are constantly going off topic.

See, you resort to the same old tacit, your pictures sold or not, or how much sold nothing to do with techniques. If you can adjust contrast or colors easily in software if you use GND. Your photo actually has typical surreal look after extreme shadow lifting. But post processing can also create such HDR look photo easily.

We are not criticizing my photo we are talking about 5D3 ability to recover from underexposure.

Gee, again I am not disputing D800 has better underexpose recovery and nobody actually disputing that. I am ONLY disputing there is a better technique and better way to shoot photos, and I believe you could get a better IQ if you use GND/reverse-GND filter and expose correctly on 80% of front areas at ISO 100 rather ISO 50 (that sacrifices highlight).

This is nothing to do with 5D3 vs D800. Even 5D3 had 14-stop DR or I shoot with D800, I STILL will not shoot in your way but on the way I descried above, period.

Again, your opinion and not mine.

Sure I am just not agreeing your shooting method, so not playing in the same game as yours.

-- 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
qianp2k
Forum ProPosts: 10,350Gear list
Re: Crystal Ball Required...
In reply to DtEW, Jul 14, 2013

DtEW wrote:

SushiEater wrote:

I could use a stronger word to describe your condition but I am sure now that you are CRAZY and certainly don't have a clue and experience. The whole idea behind this shot was not to overblown the sky at 30 second exposure to get a misty look from water.

I think in your claimed massive amount of experience, you missed that the D800's "ISO50" (LO1) is actually an expansion, i.e. digitally pulled from ISO100. See here:

http://www.dpreview.com/reviews/nikon-d800-d800e/19

If the "whole idea behind this shot was not to overblown (sic) the sky at 30 second exposure to get a misty look from water," then congratulations... you choice of "ISO50" made the sky that much easier to blow out by shrinking your highlight range.

And you don't see much grain or noise, you are just saying it. In fact all I did is converted from RAW without any processing for noise or banding for that matter.

Here is a very small 100% crop at full resolution. There is no noise, period.

Post your 5D3 shot in this situation and lets see what it looks like.

Actually, a 5DMKIII user (not to mention a competent D800 user) at ISO100 and an ND2 filter would do this shot much better than your attempt. Could've probably even gone to 1 minute for much smoother water, since the reduction in highlight range going from the D800's ISO100 to LO1 ("ISO50") is about 1 entire stop.

Exactly, no need to shoot at ISO 50 if he exposed on front darker areas not on bright sky.  This is a typical scene should use a filter as there is clear horizontal line there.  I personally would use a 10-stop Lee Bigstopper that water will be very smooth and could add another GND for sky if necessary.

I make no claims for the relative capabilities for the two cameras, just that your "experience" claim is somewhat dubious.

-- 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
SushiEater
Senior MemberPosts: 2,002Gear list
Re: Crystal Ball Required...
In reply to qianp2k, Jul 14, 2013

If you believe the way you did as in this sample fine go ahead. But there is a better way that can generate better IQ and hope you listen. Your way is not the only way.

Probably not the only way but you have not posted a single sample!!!!!

So far I see from you nothing but a talk. You know a picture worth a thousand words. I think you just hit the limit. Now post something similar and see if you can do better.

 SushiEater's gear list:SushiEater's gear list
Nikon D800E Nikon D810 Nikon AF-S Nikkor 16-35mm f/4G ED VR Sigma 70-200mm F2.8 EX DG OS HSM Nikon AF-S Nikkor 24-120mm f/4G ED VR +10 more
Keyboard shortcuts:
FForum MMy threads