The night and day performance you speak of must be the frame rate, buffer performance etc. that makes the difference.
The reason I mentioned it was you said that you loved the D4 AF and was not impressed with the D800 AF which was confusing.
Nope. Yes the other factors were nice, but just the pure performance of the focus was way better.
Im not a tec guy so don't ask me to explain the whys as I simply do not know.
But I can tell you this.
On the same setup, same product moving at the same speed, same cloths, same light ect the keeper % were way better on the D4
First few passes were shot on the D800
I got about 5 or 6 sharps in about 50 frames
Next few passes were shot with the D4
I got about 80 sharps in around 100 frames.
I got the same general results 3 different times in 3 different setups.
Granted, I did not spend a ton of time trying to fine tune the settings, but I did try a few different setups but the results never really changed.
Nope. Yes the other factors were nice, but just the pure performance of the focus was way better.
Im not a tec guy so don't ask me to explain the whys as I simply do not know.
...
The likely reason is that, while the Nikon D4 and D800 so seem to share "nearly" equal specifications, as relates to the autofocus (AF) section of the camera, there are differences.
In this case the primary difference is likely the higher resolution of the D800. In an action setting with continuous AF and high frame rate, the Expeed 3 image processor (common to both cameras), has to both manage the AF section and process the image data (in addition to all other processor related duties.) It is likely that the higher resolution of the D800 saturates the processor faster, affecting other parts of the camera, including AF.
The second difference, which I am unable to explain, is that the metering seems slightly different between the 2 bodies. While they both share the same specifications of a 91,000-pixel RGB sensor, the sensors seem to vary in sensitivity:
Metering Range:
D4: -1 ± 20 EV (Matrix or center-weighted metering), 2 to 20 EV (spot metering)
D800: 0 to 20 EV (3D color matrix or center-weighted metering), 0 to 20 EV (spot metering)
Since the metering sensor is also used to assist in steering the AF during 3D tracking, it stands to reason that there will be differences in tracking speed and tracking accuracy.
First few passes were shot on the D800
I got about 5 or 6 sharps in about 50 frames
Next few passes were shot with the D4
I got about 80 sharps in around 100 frames.
I have to wonder if at 36mp, are D800 users simply seeing a soft image at 100% view because the AF locked on somewhere within the tolerance range, but not totally perfectly, and the resolution of the D800 now reveals this? In other words, is the D800 showing us tiny variations in AF accuracy that are within spec but are not visible on the D4? But surely the D7000 would have this same issue. I think my old 7D did. Viewing at 100% was often somewhat disappointing.
-Jack
An "accurate" reproduction of a scene and a good photograph are often two different things.
I have to wonder if at 36mp, are D800 users simply seeing a soft image at 100% view because the AF locked on somewhere within the tolerance range, but not totally perfectly, and the resolution of the D800 now reveals this? In other words, is the D800 showing us tiny variations in AF accuracy that are within spec but are not visible on the D4? But surely the D7000 would have this same issue. I think my old 7D did. Viewing at 100% was often somewhat disappointing.
Yes, I think that's a valid point. The higher resolution (and tighter pixel density) sensors allow a greater scrutiny of the image through higher effective magnifications of the image plane.
I doubt that it explains all of the issues that user "nw scout" is seeing, but the effect is still a consideration.
I suppose shooting some tests on the D800 in smaller resolutions would reveal if this speculation has any validity.
I think I will do that this fine weekend we have coming up.
Also, I have to agree. Viewing the 36MP files at 100% is oftentimes disappointing.
It has been provent hat the D3 has slightly better AF performance then the D700 as well and supposedly they have the same AF sytem and in this case the same sensor and all that. I do think Nikon "makes sure" that their flagship FF will always perform better then their lower cost siblings.
It has been provent hat the D3 has slightly better AF performance then the D700 as well and supposedly they have the same AF sytem and in this case the same sensor and all that. I do think Nikon "makes sure" that their flagship FF will always perform better then their lower cost siblings.
I wonder if the D3/D700 AF relationship is like the 1DX/5D3 in that they have the same sensor but different processing horsepower.
-Jack
An "accurate" reproduction of a scene and a good photograph are often two different things.
Comments
Nope. Yes the other factors were nice, but just the pure performance of the focus was way better.
Im not a tec guy so don't ask me to explain the whys as I simply do not know.
But I can tell you this.
On the same setup, same product moving at the same speed, same cloths, same light ect the keeper % were way better on the D4
First few passes were shot on the D800
I got about 5 or 6 sharps in about 50 frames
Next few passes were shot with the D4
I got about 80 sharps in around 100 frames.
I got the same general results 3 different times in 3 different setups.
Granted, I did not spend a ton of time trying to fine tune the settings, but I did try a few different setups but the results never really changed.
The likely reason is that, while the Nikon D4 and D800 so seem to share "nearly" equal specifications, as relates to the autofocus (AF) section of the camera, there are differences.
In this case the primary difference is likely the higher resolution of the D800. In an action setting with continuous AF and high frame rate, the Expeed 3 image processor (common to both cameras), has to both manage the AF section and process the image data (in addition to all other processor related duties.) It is likely that the higher resolution of the D800 saturates the processor faster, affecting other parts of the camera, including AF.
The second difference, which I am unable to explain, is that the metering seems slightly different between the 2 bodies. While they both share the same specifications of a 91,000-pixel RGB sensor, the sensors seem to vary in sensitivity:
Metering Range:
D4: -1 ± 20 EV (Matrix or center-weighted metering), 2 to 20 EV (spot metering)
D800: 0 to 20 EV (3D color matrix or center-weighted metering), 0 to 20 EV (spot metering)
http://www.nikonusa.com/Nikon-Products/Product/Digital-SLR-Cameras/25482/D4.html#tab-ProductDetail-ProductTabs-TechSpecs
http://www.nikonusa.com/Nikon-Products/Product/Digital-SLR-Cameras/25480/D800.html#tab-ProductDetail-ProductTabs-TechSpecs
Since the metering sensor is also used to assist in steering the AF during 3D tracking, it stands to reason that there will be differences in tracking speed and tracking accuracy.
Moderator of the Cameras and Accessories forums
I have to wonder if at 36mp, are D800 users simply seeing a soft image at 100% view because the AF locked on somewhere within the tolerance range, but not totally perfectly, and the resolution of the D800 now reveals this? In other words, is the D800 showing us tiny variations in AF accuracy that are within spec but are not visible on the D4? But surely the D7000 would have this same issue. I think my old 7D did. Viewing at 100% was often somewhat disappointing.
An "accurate" reproduction of a scene and a good photograph are often two different things.
Yes, I think that's a valid point. The higher resolution (and tighter pixel density) sensors allow a greater scrutiny of the image through higher effective magnifications of the image plane.
I doubt that it explains all of the issues that user "nw scout" is seeing, but the effect is still a consideration.
Moderator of the Cameras and Accessories forums
I think I will do that this fine weekend we have coming up.
Also, I have to agree. Viewing the 36MP files at 100% is oftentimes disappointing.
Monte
14-24 24-70 70-200mm (vr2)
85 and 50 1.4
45 PC and sb910 x2
http://www.danielkimphotography.com
I wonder if the D3/D700 AF relationship is like the 1DX/5D3 in that they have the same sensor but different processing horsepower.
An "accurate" reproduction of a scene and a good photograph are often two different things.