DX Auto-Crop

[J-see]

I have been lusting for a DX body for ages especially for macro , but now this is making me think again.

Could the issue be around the diffraction limit.
Diffraction Limited Photography: Pixel Size, Aperture and Airy Disks

It surprised me too even when the numbers at DxO indicated it already. But I had never assumed the D750 would show more detail in less pixels. It shows also they're putting us to sleep with all those pixels. While they overall might improve our shots, it's not necessarily true for everything in that shot.

It makes me wonder too about Canon. It's not that I desire one but people can't be shooting it for the price or label alone. If their sensors get more out of their lenses in comparison, their lower pixel count is almost irrelevant.

 

[gqtuazon]

If anyone else has a DX and FX and a lens that makes the difference, I'd love to see if you can repeat the results.

Until then I could be blessed with a too good D750 or a too bad D3300.

Here is my opinion, and more of a "Bottom Line Up Front" or BLUF:

Pixel peeping is critical for macro shots. Not a huge factor when shooting other subjects or scenes such as landscape, waterfalls, sports, etc.

Re-sized jpgs or RAW files can be misleading. In real life, for people like me who only prints up to 8 x 12 will not see the difference on print.

People will have various results depending on the skill level, experience and how familiar is he/she with the camera features.

I think we strayed away from the DX auto crop topic but this is good since we are all learning.

What I would like to see is to use my FX camera with a Sigma 18-35mm f1.8 DX APS-C lens in low light conditions. Now all I need is that lens.

Here are samples of the same scene using a D800E in DX mode and D7000. Nikon 70-200mm f2.8 VRII lens mounted on a tripod, VR off, matrix metering, same focus point. 1/1000 should compensate the 4 mph wind.

PS: I apologize for the nice and bright sunny scene. Not trying to brag with the nice sunny weather at 68 degrees F.

Have a great day!






Note: Time difference on the D7000 is off since I have not adjusted the daylight saving time.

 

[J-see]

I resize to check the differences at 100%. If it has an effect, it isn't a beneficial one. I also shoot in FX mode but don't know if that would make much difference.

When checking the numbers, the lens difference is 10/16 for the D7000 vs 27/36 for the D800E but since it's a zoom, it isn't necessarily the same for each range. That's 62% vs 75%. That might not outweigh the pixel loss.

The difference between the D3300 and D610 is much bigger for my 200mm macro.

 

[wornish]

I can't see any obvious difference.

Perhaps and its very minor but the D800E just wins on clarity. Were these Raw or jpg processed in camera as that makes a difference.

 

[J-see]

I can't see any obvious difference.

Perhaps and its very minor but the D800E just wins on clarity. Were these Raw or jpg processed in camera as that makes a difference.

If you look at the berries, you see better detail in the D800E.



Could be DOF related since it's a bigger shot.

 

[Geoffc]

But the D800E/D7000 test is not the same comparison because we are comparing around about 16MP for both sensors in DX mode. Therefore the better sensor of the 800E should have the edge, plus lack of AA filter. The real debate was is an 11MP image from a FX sensor better than a 24MP picture from a DX sensor. I might just go and do that sample as I'm now home alone tonight.

 

[gqtuazon]

I can't see any obvious difference.

Perhaps and its very minor but the D800E just wins on clarity. Were these Raw or jpg processed in camera as that makes a difference.

I didn't use the jpg files since the presets of each cameras are different. Both images were RAW files and converted to jpg using the same presets using Adobe CS6, resized to 1000 x 667. RAW files are almost identical in file size. D800E DX mode has a file size of 17.9 mp and the D7000 has a 18.1 mp.

 

[J-see]

My D3300 has no AA filter and if I'm not mistaken, the D750 has.

I'm curious about your results. My shots are 24Mp vs 11Mp.

Actually that should be 24Mp vs 16Mp. Since I shoot FX mode and have to scale the shot to 150% to get the same size, I'm shooting 16Mp in FX.

 

[gqtuazon]

My D3300 has no AA filter and if I'm not mistaken, the D750 has.

I'm curious about your results. My shots are 24Mp vs 11Mp.

Actually that should be 24Mp vs 16Mp. Since I shoot FX mode and have to scale the shot to 150% to get the same size, I'm shooting 16Mp in FX.

I think the main factor is the slow shutter speed that you used with your testing. The lack of Mup, remote control and how stable your tripod is makes a difference.

I wait another 3 seconds after pressing the initial shutter when using Mup before press the shutter the second time to allow any micro vibrations to clear.

Perhaps this old post of mine might shed some "light" on why you are seeing a not so sharp image without Mup on your D3300.

http://nikonites.com/photography-q-....html?highlight=tripod+gqtuazon#axzz3LQa4tfy1

 

[J-see]

I think the main factor is the slow shutter speed that you used with your testing. The lack of Mup, remote control and how stable your tripod is makes a difference.

I wait another 3 seconds after pressing the initial shutter when using Mup before press the shutter the second time to allow any micro vibrations to clear.

Perhaps this old post of mine might shed some "light" on why you are seeing a not so sharp image without Mup on your D3300.

http://nikonites.com/photography-q-....html?highlight=tripod+gqtuazon#axzz3LQa4tfy1

Sharp images aren't the problem with the D3300. I shot plenty before. The shutter/mirror isn't an issue, I did that all the time. It's ridiculously silent compared to the FF. It's the button press that causes shake but I avoid that by using the self-timer.

I'll see if I can set up my over-light to up the shutter and repeat the shots.

Btw, it's not 16Mp I shot, 50% works both ways.

 

[wornish]

Mirror slap is a factor at this magnification. Glenn makes a valid point.

 

[Geoffc]

@J-see I have now tested. The full size images are here.


My testing involved a tripod, Nikkor 105mm Macro for both tests, a D7100 and a D800. Both set to Live view to aid focussing which is the same as mirror up and both support that also. Both cameras are set to F8 for a better DOF, 1/3 sec exposure as I couldn't be bothered to set lights up and ISO 200. I shot in raw and added the same sharpening to all files. No other changes made.


One thing of interest is the very obvious reduced depth of field on the D800.

For the cropped images I exported the raw files so that the D800 crops had the same long side pixels as the D7100. The logic being that the D7100 had the best native resolution and the upsizing of the D800 just shows what it looks like presented at the same size.


I think the lack of AA filter on the D7100 is showing here although I think I could get the 800 files much sharper with proper post processing.


You mentioned that the earlier samples that I sent links to were not the same size. That's because they were full size exports and the DX version had more pixels so it was a bigger.


Have a look at these images. I'm of the same opinion that I started with that the DX is better resolution than the cropped FX and I'm using a much bigger FX sensor here so my DX crop is 15 MP which is giving FX a better chance than the D750. Now if I'd used that Macro lens to fill my 36MP FX frame it would have been a whole different story

 

[Geoffc]

I will make one other point in all of this. All of the cameras mentioned so far, D800, 800E, 750, 7000, 7100, 3300 are all mighty fine cameras that take fantastic images with nice glass attached. If I could only have my 7100 I wouldn't lose too much sleep as it beats the 800 at some things and vice versa.

 

[J-see]

I shot something similar just before although I didn't have anything as detailed as yours. The set-up and light is identical, although I added light to get more shutter. I have to manually focus the D3300 with the macro. I took at least 15 shots of these and selected the best. There's some DOF difference even when I centered focus at the same spot.

I adjusted the WB so all are identical. No sharpening besides the auto-sharpen in LR.

 

[J-see]

I will make one other point in all of this. All of the cameras mentioned so far, D800, 800E, 750, 7000, 7100, 3300 are all mighty fine cameras that take fantastic images with nice glass attached. If I could only have my 7100 I wouldn't lose too much sleep as it beats the 800 at some things and vice versa.

It's not that I think any camera is bad. I always loved my D3300 and it's not that I got a FX and now I think DX is less. There's more to a good shot than 10 pixels that are sharper. But this is just testing if it really is worth taking the DX over the FX for birding or macro. I am curious if the difference is worth sacrificing the noise advantage.

 

[J-see]

Mirror slap is a factor at this magnification. Glenn makes a valid point.

It might be but the reality is, I can't shoot the D3300 without this slap. It's always there when doing macro or shooting birds.

 

[Eyelight]

It might be but the reality is, I can't shoot the D3300 without this slap. It's always there when doing macro or shooting birds.

Mirror slap is momentary and only an issue at shutter speeds where the shutter is open while the mirror-slap shake subsides. I'm guessing the affected shutter speed range is 1/30 -2 (+/-) seconds. Longer shutter speeds are not affected because the light is not intense enough to record the movement. Shorter speeds are not affected because the movement is frozen just like camera shake is frozen.

 

[J-see]

I shot fast enough for mirror-slap to not be a problem here (I hope).

The DOF bugged me so I shot a flat surface, cam leveled and live-view focus.







I shot 15 and none was better for the D3300. In the larger crop it's reasonable ok but when nerdy close, it's a different story.

And if after looking at the larger crop you think the black part is sharper, as I did; check it:



Up close the difference is that great you'd wonder if I deliberately pushed it out of focus. Still, this is my reality with the 200mm.

 

[J-see]

I redid that last one since it doesn't look right.

I set up again, used a high enough shutter and went all the way to 1:1 to rule out any difference. I focused with the liveview. This evidently is comparing the ridiculously small.




This one the D3300 does better. I hate mixed results.

It's possible the D750 could be better. It did in the previous test. If you breath into the direction of the lens' focus ring, it makes a difference. One of the problems can be having to select a shot pre-scaled. Maybe I should first scale all FX shots to 150% and only then check.

Anyways, considering the fact I can't take sharper shots even by accident, I'd say when both perform at their best, the difference with this lens won't be big enough to drop the FX in favor of the DX for crop-shots. Even if I'd gain some pixels of sharpness with the DX, I'd about always would have more diffraction when closing down in macro and more noise compared to the FX.

Now I have to redo the Tamron using something detailed and at a big enough distance to simulate bird shots. That's where it has to make the difference. Maybe here the DX might be worth taking along.

 

[J-see]

I just tried to shoot the church with the Tam at 600mm and one of my problems becomes obvious.

I can't stabilized the tripod any more than I already did and even when shooting from inside, when using live-view to fully focus on the church roof some 100 yards from here, the image in the viewfinder dances. VR off or on doesn't make any difference. The house must pick up vibrations of the traffic out there and vibe them to my pod/cam.

At a closer range that goes unnoticed but when shooting large distances, it's all too obvious.

Too bad I can't ask the whole town to stop moving until I done my testing.