D300 v D700 resolution test

Volusian

Volusian

Messages
1,919
Edit My Images
No
Photo test scenario ....

Did a comparison test between the D700 and the D300 today. Both cameras set to ISO 200, 12 bit RAW file, using built in Flash and same lens being the Nikon 70-200 f/2.8 VR. Most other settings set the same.

Both took a photo of bottle of fairly liquid at around 7ft away at 70mm setting. Moved back slightly with the D300 as it equates to 105mm so both pictures were seeing the same amount of image.

Took both pictures and zoomed in to 300% in Camera Raw and compared the 2 images ...

Should the 2 images show similar clairty of image at 300% or should the D700 being FF be somewhat clearer / sharper ?? .. or vice versa even ..

Is the above a fair test of both cameras abilities as regards comparing image resolution ?

Your thoughts please on what i should expect to get from this test before i reveal my findings .. and has anyone esle done this same test .. ?

thanks ..
 
I would expect, due to the larger pixel size, that the D700 would have better 'definition' on a pixel level examination. It would also have shallower DOF giving the perception of a softer image away from the plane of focus.
 
I would expect to see for consistant resolution from corner to corner across the frame with the D300.

I would expect to see better resolution on the centre of the frame with the D700, but more resolution falloff in the corners due to the resolution falloff of the lens (as you are now really using the corners)

Hopefully you'd stopped down a reasonable amount so that the DOF doesn't come into play.
 
Look forward to the results, I have this lens and a 300, soon to be a 700 fingers crossed.
Mind you I am quite pleased with the results I am getting with the 90.
 
I wouldn't expect much of a resolution difference, assuming you a framing identically.

Actually the D700 will have slightly less acuity than the D300 I suspect.

(I've not tested this personally)
 
not altogether sure if the test was fair and accurate ... but did it to the best of my ability .. noticed for a while now that when closely cropped in on the D700 i seem to lose quite a bit of clarlty ... which i wouldnt really expect from the D700 ... the 2 images below are jpegs from the camera raw images when cropped in at 300% .. Test A picture is the D700 and Test B is the D300

The original files sizes were 18mb for the D700 and 10mb for the D300 in Raw format.

I will perform some more tests to see if the findings are consistant ...

D700 test pic
TestA.jpg


D300 test pic
TestB.jpg
 
I have a landscape shot framed almost identically, and apart from the much higher noise on the D300 (even at base ISO), I can't see any resolution diff per se.

Although something horrible wrong, as neither the D300 or D700 are anywhere near as sharp as I would expect given the lens. Did you shoot this handheld, or did you shoot it on a tripod with VR left on?
 
hand held with VR on ... bright kitchen area .. early afternoon. I also did a similar test with a fully typed up A4 sheet of paper. Zoomed into around 300% and the D300 was clearer with more legible text ...

I will do another test outside later on tripod later and compare again ...
 
Would it not be better to shoot on a tripod, mirror lock-up, VR off and using a remote trigger?
 
I would agree that something doesn't look right there, ignoring which one is the D700/D300. The word 'original' in the first image looks larger than in the second, was the angle these pics were taken at the same?

Definately needs to be done on a tripod with VR off, I'd also try another lens.

Were these RAW files processed in LR2 by any chance?
 
sdb123 said:
I was just thinking the same. The word 'original' in the D300 shot in particular has some ghosting...
Click to expand...

So does the D700 one.
 
These sort of tests are very hard to do well.

Ideally you'd use the same focal length - something very sharp (I'd use a Nikkor 55mm f/2.8 AI-S macro), manually focus, then move the D700 forwards (or D300 backwards) to frame it the same. Using a zoom to re-frame has problems (is sharpness consistant across the range), so I'd use a prime.

Have to use M-Up, cable release with a decent shutter speed, ISO200, same WB on both.

Personally I think its too much effort to do it properly, but pointless doing it uncontrolled.
 
puddleduck said:
Ideally you'd use the same focal length - something very sharp (I'd use a Nikkor 55mm f/2.8 AI-S macro), manually focus, then move the D700 forwards (or D300 backwards) to frame it the same. Using a zoom to re-frame has problems (is sharpness consistant across the range), so I'd use a prime.
Click to expand...

I disagree. For me, for these tests to be in anyway meaningful you have to leave the camera in the same place and use the same focal length and crop the the D700 image to match that of the D300. Changing the position of the camera defeats the purpose - all too often your position as a photographer is limited by your environment, changing your position alter isn't an option. Changing your lens may be, but to have an idea of which format is best suited for you you need to keep the distance constant IMVHO.
 
NorthernNikon said:
Changing the position of the camera defeats the purpose - all too often your position as a photographer is limited by your environment, changing your position alter isn't an option. Changing your lens may be, but to have an idea of which format is best suited for you you need to keep the distance constant IMVHO.
Click to expand...

Well I dis-agree with your dis-agreement :)

The "purpose" is to test camera resolution, there is no photographic merit or environment issues here. This is a resolution test, not art.

IMHO you MUST not use a zoom lens for this - otherwise you are testing the lens too (does it perform the same at 105mm as it does at 70mm for example? If not, you are not testing camera resolution in isolation)
 
puddleduck said:
Well I dis-agree with your dis-agreement :)

The "purpose" is to test camera resolution, there is no photographic merit or environment issues here. This is a resolution test, not art.

IMHO you MUST not use a zoom lens for this - otherwise you are testing the lens too (does it perform the same at 105mm as it does at 70mm for example? If not, you are not testing camera resolution in isolation)
Click to expand...

But you're not testing resolution to produce the same image, which you would be if you cropped the D700 image. The only practical reason to test the resolution between two formats is to see whether cropping the larger format will give you as good or a better image than using the smaller one. Moving the camera produces a different picture and BTW, I never advocated using a zoom.
 
I don't understand why you are talking about cropping?

I'm talking about framing the image the same. Simply put the D300 on a tripod, and aim at a test target. Take the shot. Put the D700 on the tripod, move the tripod forward so its framed the same as the D300. This isn't cropping..

This is the only way to test this properly IMHO. Same lens. Same test target. Framing accomplished by moving the camera (or on a macro rail)
 
I hate to say it but this is a real world test, it's hand held with a zoom, the way probably most people tend to shoot. I am rather supprised at the difference though. Wayne
 
Why would you expect to see a significant resolution difference between these the D300 and D700? If you want to see that, pitch the D300 against a D3x. The test here is as much of the lens as anything else.

There are lots of theoretical differences between crop and full frame cameras, but they are mostly slight. The kind of differences that often neutralised when comparing different camera models, and slightly different processing parameters.

For me, the difference that really sticks out is noise (as puddleduck pointed out in post #9). If the test was done at ISO1600 or ISO3200, I think the D700 would show significantly better detail in darker tones.
 
Too much pixel peeping on here, just get out and use the bloody cameras - they're both good bits of kit!

:D <<<<Big smilie
 
Not sure if the OP has seen this thread created by cannockwolf, not resolution based but an ISO comparison between the D300 and the D700.
 
swanseamale47 said:
I hate to say it but this is a real world test, it's hand held with a zoom, the way probably most people tend to shoot.
Click to expand...

Dunno... I'd have tried to get the shutter speed up somewhat for a nice sharp photo...

1/60th?

Oh and VR sucks like a big sucky thing. Try the same test without it, I expect it will actually be sharper...
 
well still not convinced i can get the D700 sharper than the D300 ... i will do some more tests with a 50mm f/1.8 nikkor this time on a tripod and see what the differences are then ..
 
puddleduck said:
I don't understand why you are talking about cropping?

I'm talking about framing the image the same. Simply put the D300 on a tripod, and aim at a test target. Take the shot. Put the D700 on the tripod, move the tripod forward so its framed the same as the D300. This isn't cropping..

This is the only way to test this properly IMHO. Same lens. Same test target. Framing accomplished by moving the camera (or on a macro rail)
Click to expand...

I'm talking about cropping as in the only meaning of cropping I know, cropping the image in order the bring the FX image to the same field of view as the DX one. There's no point in moving the camera because you're then taking a totally different picture, so what's the point in comparing it?
 
Can't say I'm surprised with the results. The D700 is probably surprised to be working at such low ISO settings. After all, surely it's normal shooting mode is ISO6400?
 
yeah D700 is rubbish, get rid of it


(sound of piggy bank smashing in background)

fuji s5 anyone? :D
 
Mr_Sukebe said:
Can't say I'm surprised with the results. The D700 is probably surprised to be working at such low ISO settings. After all, surely it's normal shooting mode is ISO6400?
Click to expand...


Hit the nail on the head;)
 
If you are trying to compare the resolution of each CAMERA, surely you need the same lens, same subject, same conditions, same subject distance; to produce the same IMAGE at the focal plane of each camera. Then you can compare how each camera handles and reproduces that same image. OK, the D700 will produce a wider field of view than the D300, so you can only realistically compare the part produced by the D300.
But at least you are comparing eggs with eggs!
 
I don't understand how you can compare it if they are at the same distance - with the wider FOV of the D700 then the resolution is obviously going to be lower than the D300 due to the pixel density?

Surely, you should frame it the same then compare to see what's sharper, the D700 at Distance X, or D300 at distance 1.5 x?

I would have thought that the D700 will have a shallower depth of field per given aperture, and possibly a better resolution for a similarly framed image (in the same way DSLR trumps a similar res P+S for DoF and acuity?)
 
Well, that just shows how difficult it is to compare the resolution of 2 different cameras with different size sensors!
Depth of field and acuity are determined by the lens, not the camera body. Hence my reasoning for using the same lens, so DoF would be the same for the D700 and the D300.
It might seem logical to shoot the same picture with each camera, but that would require
each camera to have a different lens, which would cloud the issue.
If you used the same lens, then camera to subject distance would have to be different in each case, which would also cloud the issue, since the lens performance at different distances would affect the comparison.
So, if you shoot with the same lens, at the same distance, same exposure settings of course, you get a D300 frame to compare with the same part in the D700 frame (ie you crop the D700 frame). You also have to print the pics (or view them on screen) so that
the D300 picture is the same size as the DX crop of the D700 FX picture.
This is getting bloody complicated! -- which is where I came in.
 
The D300 and D700 are different and while I can see what the OP is trying to do, in reality it is a flawed test because you cannot use the same lens for the same shot with both cameras - well, you can if you change the zoom, but that changes everything anyway.

I would be interested to see what happens if you take the same pic from the same spot, using an FX 24-70 2.8 on the D700, and then a DX 17-55 2.8 on the D300. That is how it should be in reality, especially as the DX lens is new and optimised for crop digital (ie should be computed to deliver a little more resolution, due to the increased enlargement required).

All that aside, comparing these two cameras on resolution is kind of pointless anyway - they are always going to be so close. And the OP's original test pics are seriously flawed with camera shake.

As has been said previously, the difference between the D300 and D700, is noise. And here (again) is the link that proves it very convincingly http://www.talkphotography.co.uk/forums/showthread.php?t=96739
 
It's not that hard to do a decent job...

* Tripod
* Same lens - prime
* Same settings - everything off
* Frame the shot the same (use your feet)
* Remote/shutter release

If you're shooting the D700 and D300 using the same camera to subject distance, the test is essentially pointless :¬)

Then again, this test is quite pointless anyway, as in the real world the D700 and D300 specialise in different areas of photography. Who cares which is slightly sharper at 300%? If you're printing that large, you should have bought a 5D MKII, D3X or PhaseOne :D

Edit: Or Genuine Fractals!
 
You must log in or register to reply here.
Back
Top