Dispelling the handheld blur myth.... D800 + 105 VR

Started May 27, 2012 | Discussions
Great British Landscapes
Great British Landscapes Regular Member • Posts: 321
Re: Dispelling the handheld blur myth.... D800 + 105 VR

itzhak_ wrote:

Yes I looked at the full res version as well. Not sure if I get the point you're trying to make here, but wouldn't you say the downsized version is less blurry than the 100% one?

Where do you see blur? Don't mistake blur with out of focus elements of the image, or on the fly itself. To my eye there is absolutely no blur in this image - looks rock solid to me!

itzhak_ Regular Member • Posts: 169
Re: Dispelling the handheld blur myth.... D800 + 105 VR

I'm sorry, I really don't see what you're getting at. Let's say that there is no motion blur whatsoever in this specific image. Does that prove that when viewing a higher MP image at 100% you don't need higher shutter speeds in general than if you had a lower MP image?

I've stated that I think this image is great in terms of sharpness even at 100%, but that doesn't really disprove the myth because it could be that the shutter speed, VR, and handholding abilities of the OP could handle the increased demand.

Either way, if the point of this discussion is to disprove a myth, then we should talk about the principles involved rather than one specific good example. I don't think anyone has claimed that you can't produce sharp images with the D800 viewed at 100%, but rather that it would be somewhat more demanding than if you had fewer MP.

Great British Landscapes wrote:

itzhak_ wrote:

Yes I looked at the full res version as well. Not sure if I get the point you're trying to make here, but wouldn't you say the downsized version is less blurry than the 100% one?

Where do you see blur? Don't mistake blur with out of focus elements of the image, or on the fly itself. To my eye there is absolutely no blur in this image - looks rock solid to me!

-- hide signature --
Andre Affleck Senior Member • Posts: 2,362
Re: Dispelling the handheld blur myth.... D800 + 105 VR

inasir1971 wrote:

FYI - a VR FX lens that can resolve the D800/E within the 'standard range' which I take as 35-70 is not available (the 24-120 f4 VR cannot resolve the D800 across the frame - I had one so I know, and we don't know about the Tamron). So yes with FX you are going to be using more primes and non-VR lenses.

Actually, my statement was more tongue in cheek, but now that you mention it, doesn't the 24-120 f4 VR only suffer at the long end and only with large apertures? It's a bit of an overgeneralization to say the D800 can't resolve it.

What's your point?

My point is that effects of motion blur is the same, regardless of resolution (in general). The reduction in sharpness (or resolving power) is proportionately the same. People get fooled into thinking that hi res cameras suffer more loss because of the magnification affect of hi res sensors, but it's not true.

If a 36MP camera looses 20% resolution due to motion blur (down to 29MP), then a 24MP sensor would also loose 20% (down to 19MP) for that same motion blur. The loss becomes less proportionate at the extreme ends of resolution, but in general, lower res cameras are not somehow immune to motion blur. See my first 2 posts here for examples:

http://forums.dpreview.com/forums/read.asp?forum=1021&message=41652749

Andre Affleck Senior Member • Posts: 2,362
Re: Dispelling the handheld blur myth.... D800 + 105 VR

itzhak_ wrote:

I'm sorry, I really don't see what you're getting at. Let's say that there is no motion blur whatsoever in this specific image. Does that prove that when viewing a higher MP image at 100% you don't need higher shutter speeds in general than if you had a lower MP image?

I've stated that I think this image is great in terms of sharpness even at 100%, but that doesn't really disprove the myth because it could be that the shutter speed, VR, and handholding abilities of the OP could handle the increased demand.

Either way, if the point of this discussion is to disprove a myth, then we should talk about the principles involved rather than one specific good example. I don't think anyone has claimed that you can't produce sharp images with the D800 viewed at 100%, but rather that it would be somewhat more demanding than if you had fewer MP.

See this post for a controlled test showing the affects of motion blur on high vs. low res sensors:
http://forums.dpreview.com/...forums/read.asp?forum=1021&message=41652749

itzhak_ Regular Member • Posts: 169
Re: Dispelling the handheld blur myth.... D800 + 105 VR

If both cameras register the same amount motion blur proportionately to each camera’s resolution (i.e. higher resolution = more blurred pixels), then motion blur will be more evident on the higher MP image if both images are viewed at 100%. The same would apply if your compared a 36 MP image and a downsized 12 MP image from the same camera. I agree that it has nothing to do with what camera you use, but rather how large you view the image, which is the point I have been trying to make all along.

If someone bought the D800 with the intention of using the full 36 MP image, then I think this someone would benefit from being aware of this fact. In "normal" use, however, I'm sure there is nothing to worry about, as it is after all better to have the extra pixels and not use them than it is to not have them at all (all else being equal, of course).

Andre Affleck wrote:

itzhak_ wrote:

I'm sorry, I really don't see what you're getting at. Let's say that there is no motion blur whatsoever in this specific image. Does that prove that when viewing a higher MP image at 100% you don't need higher shutter speeds in general than if you had a lower MP image?

I've stated that I think this image is great in terms of sharpness even at 100%, but that doesn't really disprove the myth because it could be that the shutter speed, VR, and handholding abilities of the OP could handle the increased demand.

Either way, if the point of this discussion is to disprove a myth, then we should talk about the principles involved rather than one specific good example. I don't think anyone has claimed that you can't produce sharp images with the D800 viewed at 100%, but rather that it would be somewhat more demanding than if you had fewer MP.

See this post for a controlled test showing the affects of motion blur on high vs. low res sensors:
http://forums.dpreview.com/...forums/read.asp?forum=1021&message=41652749

-- hide signature --
Andre Affleck Senior Member • Posts: 2,362
Re: Dispelling the handheld blur myth.... D800 + 105 VR

itzhak_ wrote:

If both cameras register the same amount motion blur proportionately to each camera’s resolution (i.e. higher resolution = more blurred pixels), then motion blur will be more evident on the higher MP image if both images are viewed at 100%.

The motion blur would be more evident, but so will the details. The ratio of blur to resolving power will be the same, and so the percentage of acuity loss would be the same as well. Therefore, whatever improvement in technique you do to recover that loss for the higher res camera should also be done to recover the same loss to the lower res camera.

I agree that it has nothing to do with what camera you use, but rather how large you view the image, which is the point I have been trying to make all along.

Yes, but I think your point was still that you would need better technique for the high res camera. The 100% crop makes you think there is more loss for the res camera, but its the same loss. Better technique would reduce the loss further for the camera with even more resolution!

RichardBedford Regular Member • Posts: 104
Re: Dispelling the handheld blur myth.... D800 + 105 VR

russbarnes wrote:

I'm very happy with it indeed. It got a mention on my website here: http://russbarnes.co.uk/glass

GAP IMAGES wrote:

Russ,

Great images with the D700 and the D800!

Nice shots with the Zeiss 21mm ... how are you liking that lens with the D800?

GAP

OldScotch Contributing Member • Posts: 682
Much better evidence

That's a lot more convincing, thank you.

chris maytag Regular Member • Posts: 264
Re: Dispelling the handheld blur myth.... D800 + 105 VR

russbarnes wrote:

I'm posting this because I'm tired of reading the FUD from people

Yep. My 800E does pretty well with fabrics, and video, and low light, too. Whodathunkit, based on what we hear?

Laslo Varadi
Laslo Varadi Veteran Member • Posts: 5,390
Re: Nikon 105mm VR is on the recommended list

I would trust Nikon as the authority on this.
--
Laslo
http://www.digitalexpressionsphotography.com

 Laslo Varadi's gear list:Laslo Varadi's gear list
Sony a9 Nikon D850 Nikon AF-S Nikkor 24-120mm F4G ED VR Nikon AF-S Nikkor 28-300mm f/3.5-5.6G ED VR Nikon AF-S Nikkor 50mm f/1.8G +10 more
John Sheehy Forum Pro • Posts: 21,739
Re: Dispelling the handheld blur myth.... D800 + 105 VR

NRG wrote:

...if the light levels fall low enough, at a certain point, shooting with the D800 will require better (read: steadier) hand held shooting techniques than shooting under the same conditions with the same settings on say a D700 for example - as we are talking about a 36 megapixels vs 12 in this case). As htiek mentioned, it boils down to mathematics.

Referring to mathematics as a proof, without showing the math, is just a lot of hot air. There isn't even any proof that the math is applied correctly.

The D800 can only require higher shutter speeds than another full-frame camera as far as resolution is concerned is if the mirror slaps harder, and/or there is some kind of relative vibration between the lens and the sensor. The latter would require a minimum shutter speed regardless of focal length or subject magnification. None of this has anything to do with pixel counts or pixel density. It affects the pre-binned analog image.

Therefore, it's unfair to make a claim that handheld blur is a myth regarding the D800. It isn't. It's just that under certain lighting conditions, camera shake won't be an issue.

It is a myth, because its validity relies on another myth - the myth that pixel contrast is the measure of image resolution, regardless of the number of pixels composing the image.

The final product is the result of a combination of blurs; diffraction, lens aberration, motion blur, anti-aliasing, AND blur due to large pixel size, a very important blur which eludes people who fall for the myth.

-- hide signature --

John

John Sheehy Forum Pro • Posts: 21,739
Re: Russ - I agree but ...

Fred Mueller wrote:

I can also see just a touch of blur in your shot (nearly vertical - from 11:30 to 5:30)...

If it were from a 12MP image, you might not have been able to tell the direction of the blur, as the larger pixels would have blurred the image more than the motion blur.

The higher pixel density creates a less blurred version of the image, even if the motion component is more visible as such.

You can simulate the phenomenon in Photoshop by taking a sharp crop of an image or artificial graphic, and applying a motion blur to it (or a curved one with the convolution filter), and then running the Pixelate|Mosaic filter in preview mode, and trying different "pixel" sizes. Nothing is clearer than the original; big pixels just blur everything more, making it harder to "pick out" motion blur for the purposes of identifying it. Now, if you can identify it, you can theoretically correct it by de-convolving the image with the right software; the higher the pixel density, the more accurately it can be corrected, with the least amount of increased image noise.

-- hide signature --

John

John Sheehy Forum Pro • Posts: 21,739
Re: Russ - I agree but ...

ortega wrote:

i have the D800 and an old 105mm Micro AF-D

from my experiments I do got sharper results when i shoot at a faster shutter speed than if i don't

That's a given; the actual point of contention is that some people believe that the higher pixel density of the D800 makes it harder to get a sharp image. This is not true at all, at least not because of the pixel density. What is true is that it is harder to get high pixel-to-pixel contrast, as it should be, but pixel contrast is not what determines image resolution; that is a product of pixel contrast and pixel count.

-- hide signature --

John

John Sheehy Forum Pro • Posts: 21,739
Re: Dispelling the handheld blur myth.... D800 + 105 VR

Andre Affleck wrote:

inasir1971 wrote:

What's your point?

My point is that effects of motion blur is the same, regardless of resolution (in general).

The motion blur is the same. The effect of motion blur is different, if only because the blur can drop the maximum IQ of the denser sensor more, even if the absolute IQ is better for the denser sensor.

The reduction in sharpness (or resolving power) is proportionately the same. People get fooled into thinking that hi res cameras suffer more loss because of the magnification affect of hi res sensors, but it's not true.

If a 36MP camera looses 20% resolution due to motion blur (down to 29MP), then a 24MP sensor would also loose 20% (down to 19MP) for that same motion blur. The loss becomes less proportionate at the extreme ends of resolution, but in general, lower res cameras are not somehow immune to motion blur.

Not true; motion blur does hurt the denser sensors' IQ more, compared to its potential; the confusion comes in when it is assumed that the point of comparison normalizes the potentials.

It's simple when you look at it the right way. The denser sensor has higher potential. It is easier to erode that higher potential with poor technique, mediocre lenses, and high diffraction. However, the same technique, lens, and diffraction will always yield a better resolution with higher pixel density, all other things like mirror slap and AA filter strength (as measured in pixels), remaining the same. You could say that since the denser sensor causes less blur (image blur, not the irrelevant pixel blur), that you can actually get the same quality with a denser sensor with slightly more diffraction, or with a slightly slower shutter speed (especially if the blur is mostly circular).

-- hide signature --

John

Keyboard shortcuts:
FForum MMy threads