Thanks for the comparison! I'd argue that the only meaningful analysis that can be done on 16- and 32-image photos would be strictly at 100% on the higher res photo in order to show how much more detail there is, inch for inch, than the comparatively blurry single shot photo zoomed in beyond 100%. 16- and 32-image pixel shifted photos are for printing. Even a 200 megapixel image, viewed on a 4K screen, is never going to appear as anything better than a 4K (3840x2160) image ;-)
Thanks for comparison to the pixel shift between Z8 & Zf. As I heard a lot of Nikon users said that, in term of the manual focus performance with manual lenses, Zf has better usage experience than Z8. Will you consider to make an other deep comparison on manual focus with manual lense to Z8 and Zf?
Ironically, I was testing the Zf's pixel shift this past weekend and comparing it to Adobe's new Super Resolution feature in Lightroom/Camera RAW, and was quite surprised to discover (like some other testers), that there is little if any meaningful difference between the results when it comes to increasing resolution. That's how good Adobe's new AI interpolation has become. It's to the point now where I wouldn't hesitate to use a 24MP camera, even if I knew I wanted to make enormous enlargements later. The trick is getting one's exposure and focus dialed in accurately so that you start with a good quality file. If you do that, you're golden. We're reaching a point now where the megapixel count of a camera is of little significance anymore, with dynamic range probably being the most important remaining element.
Surely if you've shooting landscape/seascapes and each shot is a longer exposure even a few seconds when you want blurry clouds and water you can give yourself a nice long exposure effect without the use of 10 stop or higher filters .
Pixel shift shooting table top tests - sounds like a new tongue twister 😂. To be honest the effects look so subtle, I doubt I’ll make much use of it but it’s an interesting idea.
I reckon an ideal use for this pixel shift is scanning medium format film - done under controlled conditions, doesn’t move and you get nice big files of your big negs - e.g. 96MP for 6x9 and I would guess ~60MP for 6x6. If I get some spare cash I want one of these ZFs!
Thanks for the interesting comparison. I don't shoot still lifes but I could see trying this on indoor flower macro images. The slightly better color gradation and sharpness could be useful especially for larger images and prints.
The noise level improvement would have become more noticeable if you had shot the original image at a high enough ISO where noise became apparent. The averaging out of the noise in the 8 and 32 pixel shift images would have shown definite improvements. Also, the image resolution is being quadrupled in the 16 and 32 pixel shift images compared to the single shot image because it doubles both the horizontal and vertical pixel count. You won’t necessarily see a big difference in the images side by side when shown at the same size. The benefit is when you zoom in to 200% or more on the original image. The pixel shift image will stay sharp and smooth while the single image starts to look pixelated so it’s useful when you want to print large and keep it clean and sharp. Macro shots show incredible detail on the Z 8 with 16 and 32 pixel shift. It’s super-macro, like having a 4X macro lens without reducing the depth of field even further. Now if Nikon could combine pixel shift shooting with focus shift, that would be an amazing in-camera feature.
So The Z6III 30mp BSI Stacked Sensor Expeed 7 processor might look a lot better with Pixel Shift or what about the Z7III 60mp 1 photo shot? I know I'm Dreaming. Cheers!
I do have a question to all pixelshifters , owner of a sony a7riii with pixelshift and also pentax k1ii and kf also with pixelshift ,k1 even with a mode where movement gets corrected up to a point. So my question is , do your cameras can do pixelshift with flash as my sony ? The pentax can’t,how is it for the other brands?
Hello both, Yes good test and we all need to test these new function out. But what I am finding using Enance in Adobe CC is a bit more useful if nothing more you can do this on moving subjects. Just a thought.
Hasselblad is one of the few manufacturers that did Multi-shoot well. They build high-precision hardware for the task. It's far better but super expensive. The mirrorless cameras basically do semi semi-random blend. It works but it's far from the real deal. The sad part is that the H is basically discontinued.
I'm a bit confused. It looks like the pixel shift manipulation was done in software not in camera. Also, do you think the Z9 will have this feature in the future?
RUclips makes it difficult for us to see the differences. QQQ: Is the colour more accurate with the stacked image? E.g., @10:12 the yellowish item on the bookshelf (to the right of the red book) is more prominent, and colours look more saturated in the 32 stacked image. That'd be important for product or scientific shots. Thx.
I sure hope they work on doing this in camera like my Olympus cameras. But I really do enjoy using my ZF otherwise. But I don’t see myself using this in its current form
Pixel shift has it’s place, could this technology be used to increase bit depth, particularly when you are converting a colour photos to black & white?
great comparison… can you guys compare what the uoscaled original version looks like compared to the 32 image stack (useful for larger prints?)…it would be also useful to see how this compares to, say, a gifapixel upscaling 😊
Exactly. While I might use pixel-shift simply for better color rendition the most obvious reason to use it is for larger prints (or better quality cropped prints). What you tested here really does not show if there is any improvement there. Also, where is pixel-shift for my Z9 flagship camera??? It is annoying that the Z8 has it but Z9 does not. Please Nikon, fix this mistake!
I really hope YT compression has affected this comparison and not general delusion, because from where i'm sat with my keen eye, and 100% P3 gamut monitor...there is no difference (if anything, the standard photos are better)
I guess that it's interesting technology, but so far it reminds me of the emporer having a "new suit of clothes". The results look to only be relevant to very anal pixel-peepers shooting fixed environments. I shoot many thousands of frames (mostly movement, but some landscape) with my Z8 and Z9, and I'm certainly not convinced that Pixel Shift is a technology that I will employ in my use case.
Thanks for the comparison! I'd argue that the only meaningful analysis that can be done on 16- and 32-image photos would be strictly at 100% on the higher res photo in order to show how much more detail there is, inch for inch, than the comparatively blurry single shot photo zoomed in beyond 100%. 16- and 32-image pixel shifted photos are for printing. Even a 200 megapixel image, viewed on a 4K screen, is never going to appear as anything better than a 4K (3840x2160) image ;-)
frankly, this video shows me pixel shift is useless .... i appreciate the video -thanks!
Took delivery of my new Z 8 last Tuesday,and now have it set up so will be trying the pixel shift feature very soon. Thanks for the tips!
Thanks for comparison to the pixel shift between Z8 & Zf.
As I heard a lot of Nikon users said that, in term of the manual focus performance with manual lenses, Zf has better usage experience than Z8.
Will you consider to make an other deep comparison on manual focus with manual lense to Z8 and Zf?
I wonder if the limitations of the laptop screen may be affecting this comparison - would a print show the difference better?
Did you guys even test the Z8 or did I miss something.
Ironically, I was testing the Zf's pixel shift this past weekend and comparing it to Adobe's new Super Resolution feature in Lightroom/Camera RAW, and was quite surprised to discover (like some other testers), that there is little if any meaningful difference between the results when it comes to increasing resolution. That's how good Adobe's new AI interpolation has become. It's to the point now where I wouldn't hesitate to use a 24MP camera, even if I knew I wanted to make enormous enlargements later. The trick is getting one's exposure and focus dialed in accurately so that you start with a good quality file. If you do that, you're golden.
We're reaching a point now where the megapixel count of a camera is of little significance anymore, with dynamic range probably being the most important remaining element.
Surely if you've shooting landscape/seascapes and each shot is a longer exposure even a few seconds when you want blurry clouds and water you can give yourself a nice long exposure effect without the use of 10 stop or higher filters .
It's useful for moire reduction if taking photos of fabric etc..
Pixel shift shooting table top tests - sounds like a new tongue twister 😂. To be honest the effects look so subtle, I doubt I’ll make much use of it but it’s an interesting idea.
Well explained. When is it coming to the Z9?
I reckon an ideal use for this pixel shift is scanning medium format film - done under controlled conditions, doesn’t move and you get nice big files of your big negs - e.g. 96MP for 6x9 and I would guess ~60MP for 6x6. If I get some spare cash I want one of these ZFs!
Well I don't know about the color but I do notice more dust on the camera. Love the Video! Cheers!
Thanks guys. Would love to see a low light comparison. I guess the differences will be more apparent.
Thanks for the interesting comparison. I don't shoot still lifes but I could see trying this on indoor flower macro images. The slightly better color gradation and sharpness could be useful especially for larger images and prints.
Interesting video. You have confirmed my bias..not worth the file size for regular viewing on a 5k screen.🧐
The noise level improvement would have become more noticeable if you had shot the original image at a high enough ISO where noise became apparent. The averaging out of the noise in the 8 and 32 pixel shift images would have shown definite improvements. Also, the image resolution is being quadrupled in the 16 and 32 pixel shift images compared to the single shot image because it doubles both the horizontal and vertical pixel count. You won’t necessarily see a big difference in the images side by side when shown at the same size. The benefit is when you zoom in to 200% or more on the original image. The pixel shift image will stay sharp and smooth while the single image starts to look pixelated so it’s useful when you want to print large and keep it clean and sharp. Macro shots show incredible detail on the Z 8 with 16 and 32 pixel shift. It’s super-macro, like having a 4X macro lens without reducing the depth of field even further. Now if Nikon could combine pixel shift shooting with focus shift, that would be an amazing in-camera feature.
So The Z6III 30mp BSI Stacked Sensor Expeed 7 processor might look a lot better with Pixel Shift or what about the Z7III 60mp 1 photo shot? I know I'm Dreaming. Cheers!
They have cancelled the mkiii
I do have a question to all pixelshifters , owner of a sony a7riii with pixelshift and also pentax k1ii and kf also with pixelshift ,k1 even with a mode where movement gets corrected up to a point.
So my question is , do your cameras can do pixelshift with flash as my sony ? The pentax can’t,how is it for the other brands?
Hello both, Yes good test and we all need to test these new function out. But what I am finding using Enance in Adobe CC is a bit more useful if nothing more you can do this on moving subjects. Just a thought.
Would this be considered as an alternative to long exposure (with 32 shots and some delays between them)?
bella zf, bella simpatica signora
Hasselblad is one of the few manufacturers that did Multi-shoot well. They build high-precision hardware for the task. It's far better but super expensive. The mirrorless cameras basically do semi semi-random blend. It works but it's far from the real deal. The sad part is that the H is basically discontinued.
I'm a bit confused. It looks like the pixel shift manipulation was done in software not in camera. Also, do you think the Z9 will have this feature in the future?
Thanks guys! If you do this again for the Z8 etc, could you find find some more detailed subjects and please go much higher on the iso, thank you.
informative content
RUclips makes it difficult for us to see the differences. QQQ: Is the colour more accurate with the stacked image? E.g., @10:12 the yellowish item on the bookshelf (to the right of the red book) is more prominent, and colours look more saturated in the 32 stacked image. That'd be important for product or scientific shots. Thx.
I sure hope they work on doing this in camera like my Olympus cameras. But I really do enjoy using my ZF otherwise. But I don’t see myself using this in its current form
Pixel shift has it’s place, could this technology be used to increase bit depth, particularly when you are converting a colour photos to black & white?
great comparison… can you guys compare what the uoscaled original version looks like compared to the 32 image stack (useful for larger prints?)…it would be also useful to see how this compares to, say, a gifapixel upscaling 😊
I was about to ask the same. That's the whole point of having more resolution
Exactly. While I might use pixel-shift simply for better color rendition the most obvious reason to use it is for larger prints (or better quality cropped prints). What you tested here really does not show if there is any improvement there. Also, where is pixel-shift for my Z9 flagship camera??? It is annoying that the Z8 has it but Z9 does not. Please Nikon, fix this mistake!
no to the music
Thanks!
Thank you very much! 🙏🏼
If you really have to pixel peep that much to see any difference then its over rated, but hey might be useful in some instances
This is great video
I really hope YT compression has affected this comparison and not general delusion, because from where i'm sat with my keen eye, and 100% P3 gamut monitor...there is no difference (if anything, the standard photos are better)
Splitting hairs- not a noticeable difference. With Lightroom AI Denoise and a bit of sharpening applied is my approach.
not convincing. So I better stick with one shot (no hassle with movement in the pic) and use AI noise reduction in LR
I would imagine stable shots would be taken in base ISO, not 1100.
4 adverts in the 1st minute...what??
I guess that it's interesting technology, but so far it reminds me of the emporer having a "new suit of clothes". The results look to only be relevant to very anal pixel-peepers shooting fixed environments. I shoot many thousands of frames (mostly movement, but some landscape) with my Z8 and Z9, and I'm certainly not convinced that Pixel Shift is a technology that I will employ in my use case.
Spot on 👌