F stop full frame equivalents is something I never bothered with. Not on apsc or m43. I can't be bothered.
I see people do it all the time, "you know, that 2.8 lens youre using is a 5.6 in full frame"
If I share a photo shot with a m43 camera and I'm asked about aperture, I simply reply f1.7
I don't care about what a given aperture is in terms of bokeh equivalents. I'm just not that Much of a bokeh junky.
1.4 is 1.4 across all sensor sizes that I own. I'm aware of the different looks, it just doesn't matter to me.
Don't think of it when it comes to light gathering either. If I have a 1.4 lens for m43 and a 1.4 lens for my full frame, There's no math to do there that changes how ill decide on which system to bring.
What I do think is what existing light will be like and how high my iso will likely be
All that being said, there is a good enough amount of people who do care about dof equivalents and I dont think fuji should ignore thar. I think it'd be smart of them to offer some 1.8 zooms, like the sigma 18-35 1.8 or sigma 50-100 1.8. The 50-100 wasn't as popular but the 1835 was extremely popular
gdanmitchell wrote:
This notion has, unfortunately been bloated into meaning something that it does not really mean.
Yes, more photons fall on a larger area than a small one when all else is equal. But you still shoot f/11 at the same shutter speed whether your sensor is small or large.
To illustrate the trickiness of these comparison, consider rainfall. Let’s say that it rains 1” in an hour. If you look at a 10’ x 10’ area, the rate of rainfall is no different than if you look at a 100’ x 100’ area — it is 1” per hour either way.
Likewise, there’s no more light on a given area of your sensor (let’s say 1mm x 1mm) if the sensor is larger.
Technically, once you get into things like the dimensions of individual photo sites you see that this can affect noise and dynamic range potentially. But that’s not a “more light” issue — it is a “sensor bucket size” issue.
Consequently, as I mentioned earlier, folks getting something like the 16-55mm f/2.8 lens are, by and large, treating it the same way they would treat a lens with a comparable angle of view range on full frame — which is (very) roughly a 24-70mm f/2.8 lens. They will use them in more or less the same circumstances and for similar kinds of photographs. If they’d choose f/2.8 on one in a given situation, they almost certainly would choose f/2.8 with the other in the same situation.
(There is one exception. Because the effect of diffraction blur becomes apparent almost one stop sooner as you stop down on APS-C by comparison to FF, they might be moe likely to avoid the smallest apertures on the smaller sensor system.)
At about this point you typically see some forum creature enter the scene and write: “But your f/2.8 APS-C lens is really a f/4 lens!” (I’ve noticed that these are typically people who don’t actually use such APS-C lens, yet are utterly confident of the power of their arguement! ;-) Sigh. )
When I go into the field with my 16-55 f/2.8, my 50-140 f/2.8 (or, heck, just about any APS-C lens), I basically never think Oh, no! I’m really only getting f/4! Horrors!” ;-)
I have in mind doing a little experiment with a pair of systems, FF and APS-C. Let’s see if I can find some time…...Show more →
Well yes and no. From an exposure perspective f2.8 is f2.8 like you said, whether it is APS-C or FF. However the APS-C would have higher noise because there would be less total photons per pixel due to its smaller area.
Lets step back and think why we use a faster aperture. We use it for lower noise in low light conditions, or for faster shutter speed, or for shallower DOF. For shallower DOF we know that f2.8 on APS-C is similar to f4 on FF. For low light, due to one stop noise advantage of FF, you can use f4 on FF and use one stop higher ISO and get similar results as f2.8 on APS-C. Similarly for faster shutter speed again, you can use an f4 on FF and use one stop higher ISO and get similar results as f2.8 on APS-C.
So I think it is fairly reasonable to say that an f4 zoom on FF is equivalent to an f2.8 zoom on APS-C. As someone who owns and uses both APS-C and FF systems this also matches my actual experience.
Digital cameras are expected to standardize ISO exposure settings so aperture and shutter time exposure is the same regardless of the camera’s throughput (normalizes pixel and lens aperture physical size). What changes are the internal gain settings which leads to different noise levels at same exposure settings. In practice, not all brands do this well which can lead to changes in exposure settings to achieve same results. I see variances on the order of 1/3 to 1/2 stop.
gear-nut wrote:
Z9 with 50mm lens at f1.8 or XH2 with 35mm lens at f1.8? (EXIF stripped.)
PS: There are actually two hints as to which *IF* you know what to look for
Definitely difficult to tell in this scene. The lighting conditions seem bright enough for noise differences to be imperceptible. I’m not familiar with the geometry of the scene so I can’t really tell the difference in the depth of focus. Finally, I don’t own either camera so can’t identify any specific artifacts or bokeh character.
At any rate, the bottom line is there are technical performance differences between different camera sensor formats and lenses but for most conditions these aren’t going to be noticeable—specially for modern high quality systems from Fuji, Nikon, Canon etc. But that’s hardly revelatory, is it? There’s a reason Fred’s reviews tend to be comparative, at 100% pixel zoom, tilted along diagonal axis, and at scene conditions that push the camera to exhibit aberrations….
curious80 wrote:
Well yes and no. From an exposure perspective f2.8 is f2.8 like you said, whether it is APS-C or FF. However the APS-C would have higher noise because there would be less total photons per pixel due to its smaller area.
Lets step back and think why we use a faster aperture. We use it for lower noise in low light conditions, or for faster shutter speed, or for shallower DOF. For shallower DOF we know that f2.8 on APS-C is similar to f4 on FF. For low light, due to one stop noise advantage of FF, you can use f4 on FF and use one stop higher ISO and get similar results as f2.8 on APS-C. Similarly for faster shutter speed again, you can use an f4 on FF and use one stop higher ISO and get similar results as f2.8 on APS-C.
So I think it is fairly reasonable to say that an f4 zoom on FF is equivalent to an f2.8 zoom on APS-C. As someone who owns and uses both APS-C and FF systems this also matches my actual experience.
Nope. Photons per pixel are identical if pixel size and design is identical, regardless of the number of them *within the lens’ IC.* As to smaller pixels being noisier has recently been debunked with associated improvements in sensor technology; the fuji 26 and 40 mp apsc sensors are essentially equivalent as re noise; the nikon Z8/9 are slightly better than the Z7 at the same net pixel counts.
Erich6_ wrote:
Definitely difficult to tell in this scene. The lighting conditions seem bright enough for noise differences to be imperceptible. I’m not familiar with the geometry of the scene so I can’t really tell the difference in the depth of focus. Finally, I don’t own either camera so can’t identify any specific artifacts or bokeh character.
At any rate, the bottom line is there are technical performance differences between different camera sensor formats and lenses but for most conditions these aren’t going to be noticeable—specially for modern high quality systems from Fuji, Nikon, Canon etc. But that’s hardly revelatory, is it? There’s a reason Fred’s reviews tend to be comparative, at 100% pixel zoom, tilted along diagonal axis, and at scene conditions that push the camera to exhibit aberrations…. ...Show more →
Yes, we seem to agree here — the actual perceivable differences are effectively down to measurebating minutia, and usually only detectable at 100% view level. Another thing I’ve noticed is most people only detect differences in normally presented images when exif data is provided 🤣
gear-nut wrote:
Nope. Photons per pixel are identical if pixel size and design is identical, regardless of the number of them *within the lens’ IC.* As to smaller pixels being noisier has recently been debunked with associated improvements in sensor technology; the fuji 26 and 40 mp apsc sensors are essentially equivalent as re noise; the nikon Z8/9 are slightly better than the Z7 at the same net pixel counts.
You are confusing things a bit - at a pixel level the 40MP APS-C sensor does have more noise than 26MP APS-C sensor. The reason we say that the 40MP and 26MP APS-C sensors have similar noise is because at the same viewing size the noise would be similar.
To understand that lets take example of a 48MP sensor and a 12MP sensor of the same size. An individual 48MP pixel would have higher noise than an individual 12MP pixel, because its receiving 1/4th the photons. However if you then downscale the 48MP pixel image to a 12MP image, now each of these rescaled 12MP pixels is combining information from 4 smaller pixels and the result is that the noise is now similar to the pixel from the 12MP sensor. I hope that clarifies things.
For similar reasons an APS-C sensor does have just over one stop of noise disadvantage over FF sensor given similar sensor technology - i think that part is fairly well understood by most and I hope we don't have to argue over that. And like I mentioned in previous post given this one stop noise difference and one stop DOF difference, it is very reasonable to say that an f4 zoom on FF gives similar performance as f2.8 on APS-C. I really don't think there is much to disagree with here honestly - its pretty clear and straightforward
gear-nut wrote:
Yes, we seem to agree here — the actual perceivable differences are effectively down to measurebating minutia, and usually only detectable at 100% view level. Another thing I’ve noticed is most people only detect differences in normally presented images when exif data is provided 🤣
Indeed.
Folks, including the measurebators, forget some important stuff. First, those little technical differences (e.g. - how much noise, DR, etc.) typically only come into play at the fringes. In most cases, pretty much any decent camera can produce an excellent image.
And they seem to forget that the performance boundaries are not hard stops – rather they are roll-offs. For example, at the boundaries of low light and, say, deep shadows where it is true that the larger format can produce a bit less noise or a bit more DR, it isn’t like the smaller option simply ceases to work. It probably has a little more noise than the FF version, and in 99 out of 100 cases once you adjust for that in post they both still look good.
And given that fact, we have to balance that incremental performance roll-off against what may be, at least for some of us, other advantages of using a smaller system for some kinds of photography.
Anyway, back to the primary topic: I’m still going to be intrigued to see if Fujifilm actually manages to reduce the size/weight of the new 16-55 f/2.8 by that 30% margin and maintain or even improve the IQ. If so, I may be interested!
A smaller and lighter 16-55mm f2.8 would definitely be interesting. I also passed on the current version due to size and weight. The 30% reduction in size, if true, would probably be a combination of reduction in diameter and length. Perhaps something similar to Sony E 16-55mm f2.8 which is roughly 30% smaller in volume and 25% lighter compared to the current Fuji and has the reputation of being a solid performer (I have not used it myself - not a fan of Sony APS-C bodies).
gear-nut wrote:
Nope. Photons per pixel are identical if pixel size and design is identical, regardless of the number of them *within the lens’ IC.* As to smaller pixels being noisier has recently been debunked with associated improvements in sensor technology; the fuji 26 and 40 mp apsc sensors are essentially equivalent as re noise; the nikon Z8/9 are slightly better than the Z7 at the same net pixel counts.
"...IF pixel size and design is identical..."
Well sure, but how many different sized sensors exist with the same pixel design?
Also, the smaller pixel being debunked is a strong case of Yes and No. If you are talking about how we view images, YES, I agree. But punched in, it's just not the case as you can see from the DP Review scene screenshot. There definitely is more noise produced by the smaller pixels on the A7riv - but in most practical applications it doesn't ever manifest.
jakelindsay wrote:
"...IF pixel size and design is identical..."
Well sure, but how many different sized sensors exist with the same pixel design?
Also, the smaller pixel being debunked is a strong case of Yes and No. If you are talking about how we view images, YES, I agree. But punched in, it's just not the case as you can see from the DP Review scene screenshot. There definitely is more noise produced by the smaller pixels on the A7riv - but in most practical applications it doesn't ever manifest.
Yes to your beloved Sony, but now you’re in the Fuji forum and we’re discussing an X platform lens. So now do the research on Fuji 26 and 40 mp sensors. While there is a difference, most seasoned testers agree it’s insignificant.
gear-nut wrote:
Yes to your beloved Sony, but now you’re in the Fuji forum and we’re discussing an X platform lens. So now do the research on Fuji 26 and 40 mp sensors. While there is a difference, most seasoned testers agree it’s insignificant.
The 40MP Fuji sensor's pixels are 1.5x smaller than 26MP Fuji sensor, which would suggest that it would be worse by about 0.6 stop at pixel level and this matches the actual results below:
As expected at pixel level the ISO 3200 of X-T5 has more noise than ISO 3200 of X-T4 but slightly better than ISO 6400 of X-T4. This is exactly what the physics predicts and this is what you see.
Now whether that impacts your photography in a meaningful way is a different question. As I said I prefer using my Fuji APS-C system over my Sony FF system most of the time and don't find it limiting. However you have to distinguish between objective facts and subjective preferences. The objective fact is that FF sensor has a one stop noise advantage over APS-C, the subjective preference is that for my photographic use cases, that difference is not important enough to offset the advantages of my APS-C system. Anyways I have given you very clear explanation based on facts, its up to you to take it or leave it. I am done with this topic and looking forward to learning about the new 16-55mm f2.8 when it gets released. Have fun.
gear-nut wrote:
Yes to your beloved Sony, but now you’re in the Fuji forum and we’re discussing an X platform lens. So now do the research on Fuji 26 and 40 mp sensors. While there is a difference, most seasoned testers agree it’s insignificant.
My comparisons have been between the 24MP on my former XPro2 and the 40MP on my current XT5. If there’s a difference (and on the test bench there might be for all I know), in my own tests and my own photography it sure as heck isn’t visible to me. The 40MP sensor, at least in my XT5, is a fine performer.
Also, from the Irony Department, I am entertained when the response to the point that a lot of this is measurebating is… even more measurebating. ;-)
And IME, the over-measurbaters rarely use their cameras to shoot artistic images, and rather spend their time comparing minutiae on test targets. To be clear, I have zero issues with them doing that and sharing the results, I just take exception to the over significance they sometimes place on an aspect that’s is virtually irrelevant in typical imaging.
curious80 wrote:
Awesome 👍 can we talk about the lens now.
:-)
In between the distractions, some of us have been trying to! To recap my thoughts:
1. So far, it is just a rumor and details (of the rumor!) are still quite sketchy. There’s not a lot to go on, but…
2. The current 16-55mm f/2.8 is a solid, functional, useful lens. I use it a fair amount — though most of my Fujifilm photography is done with small primes. To the extent that the current lens has downsides, they might include…
3. No image stabilization, though this is less important in cameras with IBIS, which is increasingly likely to come standard in most cameras. But also…
4. For a system that bases a fair amount of its appeal on being a smaller and lighter alternative to full frame, the 16-55 f/2.8 is not exactly a tiny lens. In fact, for some folks — including those of us who like the lens — that might be its biggest minus. For that reason…
5. A smaller and lighter version of a lens with the same functionality and image quality holds some real appeal. What remains to be seen includes…
6. whether the lens (assuming the rumors pan out) has at least the same image quality as the current 16-55, and…
7. To what extent it relies on software correction to compensate for potential lens issues that sometimes show up in these smaller lenses, and…
8. Will it live up to th rumor and actually be a full 30% smaller/lighter. And…
9. What other functional changes might come with the new lens? Added IS? Different close-focus distance? And so on, but also…
10. How will it be priced? (And, with Fujifilm, we also have to ask: will it actually be available?!)
Depending on the answers to all of those questions, the rumored lens could be appealing. If the price is OK and the performance is like that of the existing 16-55 the lens is likely to be popular. For me and others who have the current lens the decision is a bit trickier — will it be enough of an improvement to warrant a switch?
