Re: The new Nikon Z8 vs the Sony A7RV and potential A9III
hiepphotog wrote: 1bwana1 wrote: hiepphotog wrote: 1bwana1 wrote: docusync wrote: saaketham wrote:
I'm currently in the process of trading it for a Sony a7RV with a FM-er who's going the other way.
Plan to test the a7R5 as a second body to accompany my a1.
I personally can't wait until the A7R5's improved IBIS and AF ML chip are merged with a stacked sensor so the A1 II will be born. I guess it's not happening this year, so I need to be patient and control my GAS
1bwana1 wrote:
The Plena is a nice lens that should offer a recognizable rendering character in certain circumstances where bokeh balls figure prominently. It may be the best Christmas lights lens available today. So if you are a mall Santa photographer it is a great choice. It will not offer better sharpness in normal usage for a lens in this FL. It is also not a fundamentally new technology. Sigma used a similar approach of enlarging the image circle well beyond the FF sensor size in their ART series. This reduces visible light fall off, and in combination with lots of aperture blades makes for rounder bokeh balls at the edges.
I'm not an optical engineer but I think cat-eye bokeh is unavoidable based on this:
Yes, as you mentioned it can be improved by enlarging both entrance and throat diameters, but not eliminated entirely (unless it's a really, really huge mount? )
Light doesn't travel through a lens in a straight line. It exits the lens at angles around the circumference . So, it is not just about the throat diameter. It is the shape of the lens elements that project the light into a wider circle. There are full frame lenses with pretty much zero vignetting, very round bokeh balls, super sharpness, and negligible distortion inside the APSC circle. Now throw in 11 aperture blades and you are a bokeh ball king! Just look at the MTF charts of full frame lenses at the APSC circle boundaries. They are much better than at the full frame edges. Now imagine that Nikon designs a lens that has a projection at the sensor that is actually medium formant in size, but the camera only uses the circumference of the full frame sensor size. It is much like cropping a full frame at the APSC.
Below is a link to a review of the Sony 135mm GM. Look at the MTF charts. The APSC circle would be at about the 13.5 line. As you can see at that distance from the center IQ is pretty much identical to the center, and avoids the fall off at the FF edges. Leica makes a number of full frame M lenses, which have a very small diameters, where at the APSC circle the performance is identical to the center, and bokeh balls will be pretty much perfectly round still.
No maging or fundamentally new technology required. Just a different set of priorities.
In the image below is a Leica lens where the lights make bokeh balls that in this frame don't go all the way out to the full frame edges. But they do go out further than an APSC edge. As you can see they are smooth, and round. For this lens any cats eye distortion is only seen at the very edges and even then not so much. The Leica mount is tiny compared to the Z mount.
IMO, that would downplay Nikon engineers' efforts for this Plena. They did not just make a ginormous lens like the Sigma 105/1.4 to achieve minimal optical vignetting, but managed to do so with a lens not much bigger and heavier than the equivalent Canon and Sony. I am mainly curious to see what kind of compromises they made to achieve this. Sony, to an extent, also has been doing this. Their optics are smaller than others but they kept the optical vignetting in check (if not better than Canon).
I don't see that as downplaying what Nikon has done. You don't have to make the lens much bigger to accomplish this. That is why I posted the Leica made image with the small diameter lens projecting on a FF sensor to demonstrate the concept. You must admit that those are nice round smooth bokeh balls. Sigma was open about doing this exact thing with some of their designs. It also doesn't diminish the advantage that the large Z mount has when taking this approach. It may be easier or possible to do it better with the Z mount. I think the Plena represents a well chosen and engineered set of design compromises and optimizations. It also represents some very good manufacturing processes to polish the ASPH elements in a way to reduce the onion rings in the bokeh. They were able to do this while holding weight and size to a manageable level. Nothing diminishing in that.
Do you think that Nikon discovered some unique math or unique glass properties in lens design to achieve the IQ and rendering that the Plena produces. What would that be? I think that it is not a coincidence that the behaviors that the Plena exhibits are all things that are optimized by expanding the image circle, and using a piece more from the center of that circle. It has been demonstrated many times.
Having 60 mpx in my Leica M11 means I can crop down to the APSC or smaller size and still have plenty of resolution. That means I can purposely shoot wide and crop to the center of the image where aberrations are not present. It is a technique I have used on occasion.
Do you also think that shutterless/electronic shutter in very fast stacked sensor cameras is a development of Nikon's? It has been standard with other brands for over 7 years already. Just packaged and marketed differently by Nikon. It does have costs in capabilities as well. I encountered one just this week. I am able to sync strobes at double the shutter speed without having to resort to HSS with mechanical shutter than I am with electronic shutter. Fortunately with my A1 I have the choice to use mechanical or electronic shutter. This week I took advantage of the mechanical option. The Nikon Z8/9 cameras lost this ability when they deleted the mechanical shutter.
I don't believe Nikon simply designed this Plena to cover a bigger image circle to achieve this low optical vignetting. One of the Plena prototypes has 100+mm front element. This would be consistent with your theory. The same focal length and aperture covering a bigger image circle must be bigger. Keep in mind that FF has about 63% bigger area than APS-C so to achieve what you said would require a really big lens (roughly 26% bigger diameter). The final version has the same 82mm front filter size as Sony and Canon.
I do believe Nikon used special types of glasses and an unconventional design. SR element is something Sony does not have (at least Sony has not advertised to have anything equivalent). And given the 82mm front, the glass elements must collectively bend the peripheral light rays more than the equivalent Sony/Canon so those rays wouldn't be blocked, resulting in smaller than optical vignetting. The Plena does have way more exotic elements (6) than Canon (3) and Sony (3).
Honestly, I never expected Nikon to come out with this design since Nikon has been sacrificing size and weight to achieve minimal focus breathing in their lenses. If this was designed like the Sigma 105/1.4, it would be within my expectations. So I have been thinking a lot about it .
But a lens is not a system so I have not added a Nikon yet.
The use of SR lens design is consistant with what I am saying. It allows light that is being bent more
Than usual to be better corrected for the chromatc aberrations that this induces.
This helps keep diameter, thus size and weight down. Also consistant with the larger front element prototype that was rejected.
It is all based on Snells Law. The more you bend light the more aberrations you create. The more aberrations the more corrective elements you need to address them. That is also consistant with what we know about the Plena so far. I am sure we will learn more in the future.
None of this diminishes the great work Nikon did in making this lens.
Once again I will post a demo image. This time from the Sony 135gm so more relevant to the Plena. AS you can see the bokeh balls are very good overall. But there is definitely cats eye going on out toward the edges. If you cropped even outside the APSC boundaries the ball are pretty much distortion free.
I have provided a link to the image because it is from Camera Labs test and I don't want to post the image directly
Re: The new Nikon Z8 vs the Sony A7RV and potential A9III
hiepphotog wrote: 1bwana1 wrote: hiepphotog wrote: 1bwana1 wrote: docusync wrote: saaketham wrote:
I'm currently in the process of trading it for a Sony a7RV with a FM-er who's going the other way.
Plan to test the a7R5 as a second body to accompany my a1.
I personally can't wait until the A7R5's improved IBIS and AF ML chip are merged with a stacked sensor so the A1 II will be born. I guess it's not happening this year, so I need to be patient and control my GAS
1bwana1 wrote:
The Plena is a nice lens that should offer a recognizable rendering character in certain circumstances where bokeh balls figure prominently. It may be the best Christmas lights lens available today. So if you are a mall Santa photographer it is a great choice. It will not offer better sharpness in normal usage for a lens in this FL. It is also not a fundamentally new technology. Sigma used a similar approach of enlarging the image circle well beyond the FF sensor size in their ART series. This reduces visible light fall off, and in combination with lots of aperture blades makes for rounder bokeh balls at the edges.
I'm not an optical engineer but I think cat-eye bokeh is unavoidable based on this:
Yes, as you mentioned it can be improved by enlarging both entrance and throat diameters, but not eliminated entirely (unless it's a really, really huge mount? )
Light doesn't travel through a lens in a straight line. It exits the lens at angles around the circumference . So, it is not just about the throat diameter. It is the shape of the lens elements that project the light into a wider circle. There are full frame lenses with pretty much zero vignetting, very round bokeh balls, super sharpness, and negligible distortion inside the APSC circle. Now throw in 11 aperture blades and you are a bokeh ball king! Just look at the MTF charts of full frame lenses at the APSC circle boundaries. They are much better than at the full frame edges. Now imagine that Nikon designs a lens that has a projection at the sensor that is actually medium formant in size, but the camera only uses the circumference of the full frame sensor size. It is much like cropping a full frame at the APSC.
Below is a link to a review of the Sony 135mm GM. Look at the MTF charts. The APSC circle would be at about the 13.5 line. As you can see at that distance from the center IQ is pretty much identical to the center, and avoids the fall off at the FF edges. Leica makes a number of full frame M lenses, which have a very small diameters, where at the APSC circle the performance is identical to the center, and bokeh balls will be pretty much perfectly round still.
No maging or fundamentally new technology required. Just a different set of priorities.
In the image below is a Leica lens where the lights make bokeh balls that in this frame don't go all the way out to the full frame edges. But they do go out further than an APSC edge. As you can see they are smooth, and round. For this lens any cats eye distortion is only seen at the very edges and even then not so much. The Leica mount is tiny compared to the Z mount.
IMO, that would downplay Nikon engineers' efforts for this Plena. They did not just make a ginormous lens like the Sigma 105/1.4 to achieve minimal optical vignetting, but managed to do so with a lens not much bigger and heavier than the equivalent Canon and Sony. I am mainly curious to see what kind of compromises they made to achieve this. Sony, to an extent, also has been doing this. Their optics are smaller than others but they kept the optical vignetting in check (if not better than Canon).
I don't see that as downplaying what Nikon has done. You don't have to make the lens much bigger to accomplish this. That is why I posted the Leica made image with the small diameter lens projecting on a FF sensor to demonstrate the concept. You must admit that those are nice round smooth bokeh balls. Sigma was open about doing this exact thing with some of their designs. It also doesn't diminish the advantage that the large Z mount has when taking this approach. It may be easier or possible to do it better with the Z mount. I think the Plena represents a well chosen and engineered set of design compromises and optimizations. It also represents some very good manufacturing processes to polish the ASPH elements in a way to reduce the onion rings in the bokeh. They were able to do this while holding weight and size to a manageable level. Nothing diminishing in that.
Do you think that Nikon discovered some unique math or unique glass properties in lens design to achieve the IQ and rendering that the Plena produces. What would that be? I think that it is not a coincidence that the behaviors that the Plena exhibits are all things that are optimized by expanding the image circle, and using a piece more from the center of that circle. It has been demonstrated many times.
Having 60 mpx in my Leica M11 means I can crop down to the APSC or smaller size and still have plenty of resolution. That means I can purposely shoot wide and crop to the center of the image where aberrations are not present. It is a technique I have used on occasion.
Do you also think that shutterless/electronic shutter in very fast stacked sensor cameras is a development of Nikon's? It has been standard with other brands for over 7 years already. Just packaged and marketed differently by Nikon. It does have costs in capabilities as well. I encountered one just this week. I am able to sync strobes at double the shutter speed without having to resort to HSS with mechanical shutter than I am with electronic shutter. Fortunately with my A1 I have the choice to use mechanical or electronic shutter. This week I took advantage of the mechanical option. The Nikon Z8/9 cameras lost this ability when they deleted the mechanical shutter.
I don't believe Nikon simply designed this Plena to cover a bigger image circle to achieve this low optical vignetting. One of the Plena prototypes has 100+mm front element. This would be consistent with your theory. The same focal length and aperture covering a bigger image circle must be bigger. Keep in mind that FF has about 63% bigger area than APS-C so to achieve what you said would require a really big lens (roughly 26% bigger diameter). The final version has the same 82mm front filter size as Sony and Canon.
I do believe Nikon used special types of glasses and an unconventional design. SR element is something Sony does not have (at least Sony has not advertised to have anything equivalent). And given the 82mm front, the glass elements must collectively bend the peripheral light rays more than the equivalent Sony/Canon so those rays wouldn't be blocked, resulting in smaller than optical vignetting. The Plena does have way more exotic elements (6) than Canon (3) and Sony (3).
Honestly, I never expected Nikon to come out with this design since Nikon has been sacrificing size and weight to achieve minimal focus breathing in their lenses. If this was designed like the Sigma 105/1.4, it would be within my expectations. So I have been thinking a lot about it .
But a lens is not a system so I have not added a Nikon yet.
The use of SR lens design is consistant with what I am saying. It allows light that is being bent more
Than usual to be better corrected for the chromatc aberrations that this induces.
This helps keep diameter, thus size and weight down. Also consistant with the larger front element prototype that was rejected.
It is all based on Snells Law. The more you bend light the more aberrations you create. The more aberrations the more corrective elements you need to address them. That is also consistant with what we know about the Plena so far. I am sure we will learn more in the future.
None of this diminishes the great work Nikon did in making this lens.
Once again I will posta demo image. This time from the Sony 135gm so more relevant to the Plena. AS you can see the bokeh balls are very good overall. But there is definitely cats eye going on out toward the edges. If you cropped even outside the APSC boundaries the ball are pretty much distortion free.
Re: The new Nikon Z8 vs the Sony A7RV and potential A9III
hiepphotog wrote: 1bwana1 wrote: hiepphotog wrote: 1bwana1 wrote: docusync wrote: saaketham wrote:
I'm currently in the process of trading it for a Sony a7RV with a FM-er who's going the other way.
Plan to test the a7R5 as a second body to accompany my a1.
I personally can't wait until the A7R5's improved IBIS and AF ML chip are merged with a stacked sensor so the A1 II will be born. I guess it's not happening this year, so I need to be patient and control my GAS
1bwana1 wrote:
The Plena is a nice lens that should offer a recognizable rendering character in certain circumstances where bokeh balls figure prominently. It may be the best Christmas lights lens available today. So if you are a mall Santa photographer it is a great choice. It will not offer better sharpness in normal usage for a lens in this FL. It is also not a fundamentally new technology. Sigma used a similar approach of enlarging the image circle well beyond the FF sensor size in their ART series. This reduces visible light fall off, and in combination with lots of aperture blades makes for rounder bokeh balls at the edges.
I'm not an optical engineer but I think cat-eye bokeh is unavoidable based on this:
Yes, as you mentioned it can be improved by enlarging both entrance and throat diameters, but not eliminated entirely (unless it's a really, really huge mount? )
Light doesn't travel through a lens in a straight line. It exits the lens at angles around the circumference . So, it is not just about the throat diameter. It is the shape of the lens elements that project the light into a wider circle. There are full frame lenses with pretty much zero vignetting, very round bokeh balls, super sharpness, and negligible distortion inside the APSC circle. Now throw in 11 aperture blades and you are a bokeh ball king! Just look at the MTF charts of full frame lenses at the APSC circle boundaries. They are much better than at the full frame edges. Now imagine that Nikon designs a lens that has a projection at the sensor that is actually medium formant in size, but the camera only uses the circumference of the full frame sensor size. It is much like cropping a full frame at the APSC.
Below is a link to a review of the Sony 135mm GM. Look at the MTF charts. The APSC circle would be at about the 13.5 line. As you can see at that distance from the center IQ is pretty much identical to the center, and avoids the fall off at the FF edges. Leica makes a number of full frame M lenses, which have a very small diameters, where at the APSC circle the performance is identical to the center, and bokeh balls will be pretty much perfectly round still.
No maging or fundamentally new technology required. Just a different set of priorities.
In the image below is a Leica lens where the lights make bokeh balls that in this frame don't go all the way out to the full frame edges. But they do go out further than an APSC edge. As you can see they are smooth, and round. For this lens any cats eye distortion is only seen at the very edges and even then not so much. The Leica mount is tiny compared to the Z mount.
IMO, that would downplay Nikon engineers' efforts for this Plena. They did not just make a ginormous lens like the Sigma 105/1.4 to achieve minimal optical vignetting, but managed to do so with a lens not much bigger and heavier than the equivalent Canon and Sony. I am mainly curious to see what kind of compromises they made to achieve this. Sony, to an extent, also has been doing this. Their optics are smaller than others but they kept the optical vignetting in check (if not better than Canon).
I don't see that as downplaying what Nikon has done. You don't have to make the lens much bigger to accomplish this. That is why I posted the Leica made image with the small diameter lens projecting on a FF sensor to demonstrate the concept. You must admit that those are nice round smooth bokeh balls. Sigma was open about doing this exact thing with some of their designs. It also doesn't diminish the advantage that the large Z mount has when taking this approach. It may be easier or possible to do it better with the Z mount. I think the Plena represents a well chosen and engineered set of design compromises and optimizations. It also represents some very good manufacturing processes to polish the ASPH elements in a way to reduce the onion rings in the bokeh. They were able to do this while holding weight and size to a manageable level. Nothing diminishing in that.
Do you think that Nikon discovered some unique math or unique glass properties in lens design to achieve the IQ and rendering that the Plena produces. What would that be? I think that it is not a coincidence that the behaviors that the Plena exhibits are all things that are optimized by expanding the image circle, and using a piece more from the center of that circle. It has been demonstrated many times.
Having 60 mpx in my Leica M11 means I can crop down to the APSC or smaller size and still have plenty of resolution. That means I can purposely shoot wide and crop to the center of the image where aberrations are not present. It is a technique I have used on occasion.
Do you also think that shutterless/electronic shutter in very fast stacked sensor cameras is a development of Nikon's? It has been standard with other brands for over 7 years already. Just packaged and marketed differently by Nikon. It does have costs in capabilities as well. I encountered one just this week. I am able to sync strobes at double the shutter speed without having to resort to HSS with mechanical shutter than I am with electronic shutter. Fortunately with my A1 I have the choice to use mechanical or electronic shutter. This week I took advantage of the mechanical option. The Nikon Z8/9 cameras lost this ability when they deleted the mechanical shutter.
I don't believe Nikon simply designed this Plena to cover a bigger image circle to achieve this low optical vignetting. One of the Plena prototypes has 100+mm front element. This would be consistent with your theory. The same focal length and aperture covering a bigger image circle must be bigger. Keep in mind that FF has about 63% bigger area than APS-C so to achieve what you said would require a really big lens (roughly 26% bigger diameter). The final version has the same 82mm front filter size as Sony and Canon.
I do believe Nikon used special types of glasses and an unconventional design. SR element is something Sony does not have (at least Sony has not advertised to have anything equivalent). And given the 82mm front, the glass elements must collectively bend the peripheral light rays more than the equivalent Sony/Canon so those rays wouldn't be blocked, resulting in smaller than optical vignetting. The Plena does have way more exotic elements (6) than Canon (3) and Sony (3).
Honestly, I never expected Nikon to come out with this design since Nikon has been sacrificing size and weight to achieve minimal focus breathing in their lenses. If this was designed like the Sigma 105/1.4, it would be within my expectations. So I have been thinking a lot about it .
But a lens is not a system so I have not added a Nikon yet.
The use of SR lens design is consistant with what I am saying. It allows light that is being bent more
Than usual to be better corrected for the chromatc aberrations that this induces.
This helps keep diameter, thus size and weight down. Also consistant with the larger front element prototype that was rejected.
It is all based on Snells Law. The more you bend light the more aberrations you create. The more aberrations the more corrective elements you need to address them. That is also consistant with what we know about the Plena so far. I am sure we will learn more in the future.
None of this diminishes the great work Nikon did in making this lens.
Oct 04, 2023 at 08:25 AM
Previous versions of 1bwana1's message #16357693 « The new Nikon Z8 vs the Sony A7RV and potential A9III »
.