p.14 #1 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
KarmaKramer wrote:
Oye...the technical jibber jabber... just take photos already! Are you a photographer or an analyst?!
How does actual T-stop differ from what most everyone else here brings up like rendering, sharpness, corner sharpness, color, vignetting, handling, build, af noise, ..... you name it?
Mar 04, 2020 at 12:07 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.14 #2 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
arneh wrote:
Maybe I am. But that's why I was interested to hear if anyone actually know the T-stop of the 24. I admit I don't have much data for T-stops. But from looking at dxomark which do measure it, all of Sony's F1.4 lenses they have measured have T-stops of 1.5 or 1.6, while all Sony F1.8 lenses have a 1.8 T-stop.
So I'm not convinced the difference is "almost one stop" in light gathering ability.
Without being too pedantic (or maybe just a bit too pedantic ), you do realize that the difference you are noting could very well be rounding error. You do realize that an f number of 1.4 isn't exactly 1.4000, in fact if a lens with an f number of 1.4 is exactly one stop more than a lens of f/1.00000, then it's exact f number is 1.4142 (i.e., the square root of 2). Similarly a lens with an f number of 1.8, is unlikely to be exactly 1.8000, and if it is exactly a 2/3rds of a stop slower than f/1.4142 then it would be 1.7818. Of course all that extra precision is ridiculous because the manufacture of lenses aren't that precise. The one decimal point is probably as precise as the measurement can and needs to be. What that means is the difference in .1 may very well not be meaningful and could well be rounding error. So you should not make a big deal about 1.4 going to 1.5 and 1.8 not going to 1.9, that could simply be rounding and note in this case if there was greater precision even 1.41 is closer to 1.5 than 1.78 is to 1.9, so let's say both a 1.41 lens and a 1.78 lens lose 0.05 transmission that would round to 1.5 and 1.8 even though they lost similar amounts of light.
So saying an f/1.8 has 2/3rd of a stop smaller aperture or lets in 2/3rds of a stop less light than an f/1.4 lens makes sense. It isn't precise, but it is close to that. To make the case that it was closer to a half of stop, you would have to be able to have more precise measurements and note where somebody misreported something or made an error in rounding. You just can't get to half a stop from 2/3rds without for the aperture without some sort of error and even for the light transmission that would be huge difference that would need to be measured more precisely than the numbers that DXO provides.
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.14 #3 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
pdmphoto wrote:
How does actual T-stop differ from what most everyone else here brings up like rendering, sharpness, corner sharpness, color, vignetting, handling, build, af noise, ..... you name it?
T-stop is simply the amount of light per unit of time that goes through the lens. It is basically unrelated to the other factors you mentions.
Since I don't read Chinese I can't see what their methodology is, but that suggests the 24 loses a half stop from the nominal aperture and the 35 f/1.8 loses about a third of a stop. Might well be true, but those numbers are both bigger than I have typically seen with most lenses. I am also a bit skeptical that they can reliably measure to the second decimal point, but again it is hard to evaluate when I can't read their methodology.
p.14 #6 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Steve Spencer wrote:
Since I don't read Chinese I can't see what their methodology is, but that suggests the 24 loses a half stop from the nominal aperture and the 35 f/1.8 loses about a third of a stop. Might well be true, but those numbers are both bigger than I have typically seen with most lenses. I am also a bit skeptical that they can reliably measure to the second decimal point, but again it is hard to evaluate when I can't read their methodology.
I can't find their methodology anywhere either. Also I am pretty sure they only test single sample. Take their numbers with a huge grain of salt.
p.14 #7 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Fred Miranda wrote:
The 2/3rd stop difference (almost 1 stop) was based on their maximum apertures.
I see your point though and when I get my hands on the 20/1.8G, I will shoot them under identical lighting in manual exposure mode so we can have a better understanding of their T-stop difference.
p.14 #9 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Steve Spencer wrote:
T-stop is simply the amount of light per unit of time that goes through the lens. It is basically unrelated to the other factors you mentions.
If you shoot a lot wide open, actual T-stop can matter more than the small differences between some of those other factors.
p.14 #10 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
It's a fine line IMO. Obviously a 5mm difference at 100m is a lot smaller than a 5mm difference at 20mm, but there are a few instances where a lens image quality has triumphed focal length preferences for me.
Best example is the Voigtlander 40/1.2 - after trying the Sony 35/2.8 and ZA 35/1.4 and falling for the 35mm focal length, I picked up the Voigtlander 40 because of the FM thread. While I didn't like the effective 43mm at first, and picked up a ZM 35/1.4 and RX1RII after the Voigtlander, it's the CV 40/1.2 that has withstood the test of time and keeps me coming back because of handling and image quality. Never thought I'd be okay without a 35mm lens, but the Voigtlander 40 is making a case for it.
Another example that I was considering was trying to pick the 24 GM over my Loxia 21 (or the new 20/1.8) because of beautiful rendering and AF. However, the FL difference is large enough for me to keep coming back to my Loxia 21 or consider the Sony 20/1.8.
chez wrote:
Biggest difference between he 20 and 24...well one is a 20mm lens, the other 24mm. Pick your focal length, then the lens of that focal length that meets your requirements. Never just pick the highest image quality lens irrespective of focal length. There s a very significant difference between a 24mm image and a 20mm image.
p.14 #11 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Steve Spencer wrote:
Without being too pedantic (or maybe just a bit too pedantic ), you do realize that the difference you are noting could very well be rounding error.
Sure, but that's why I ask if anyone has actual measured numbers for the 24 mm (I assume no one has measured the 20 mm yet as it's so new). Someone else quoted a Chinese test, which is a good start (thanks!), but I don't know how accurate those numbers are.
But why I brought it up in the first place is that the general trend I've seen with T-stops is that the difference between F-stop and T-stop is bigger with large aperture lenses compared to smaller apertures. So one often does not get as much more light as it appears by buying a brighter lens. If you buy the larger aperture lens for the shallower DoF then it doesn't matter much, but if it's for light gathering then it's interesting to know how big the difference actually is.
p.14 #12 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
My uneducated understanding is that faster lenses are generally more complex, often have more elements, and are subject to more light loss. So you may get f/1.4 blur....but not f/1.4 transmission. Which is effectively what you’re describing.
I could be way off base, but I know someone will correct me if required.
-Tim
arneh wrote:
Sure, but that's why I ask if anyone has actual measured numbers for the 24 mm (I assume no one has measured the 20 mm yet as it's so new). Someone else quoted a Chinese test, which is a good start (thanks!), but I don't know how accurate those numbers are.
But why I brought it up in the first place is that the general trend I've seen with T-stops is that the difference between F-stop and T-stop is bigger with large aperture lenses compared to smaller apertures. So one often does not get as much more light as it appears by buying a brighter lens. If you buy the larger aperture lens for the shallower DoF then it doesn't matter much, but if it's for light gathering then it's interesting to know how big the difference actually is.
But I would love to see more actual numbers!...Show more →
Mar 04, 2020 at 03:25 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.14 #13 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
pdmphoto wrote:
If you shoot a lot wide open, actual T-stop can matter more than the small differences between some of those other factors.
For some applications (e.g., video) that is absolutely true, but for other applications personally I find all the other things you mentioned much more important. I can turn the ISO up a third of a stop almost all the time in my shooting and the slight decrease in transmission matters little.
Mar 04, 2020 at 04:26 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.14 #14 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
arneh wrote:
Sure, but that's why I ask if anyone has actual measured numbers for the 24 mm (I assume no one has measured the 20 mm yet as it's so new). Someone else quoted a Chinese test, which is a good start (thanks!), but I don't know how accurate those numbers are.
But why I brought it up in the first place is that the general trend I've seen with T-stops is that the difference between F-stop and T-stop is bigger with large aperture lenses compared to smaller apertures. So one often does not get as much more light as it appears by buying a brighter lens. If you buy the larger aperture lens for the shallower DoF then it doesn't matter much, but if it's for light gathering then it's interesting to know how big the difference actually is.
But I would love to see more actual numbers!...Show more →
What we need to know is not just whether the Chinese test was accurate, but also how precise it was. Is the second decimal basically a random number or do they reliably get the same measurement when test the same lens more than once and multiple versions of the same lens. As was suggested above I think we should take these results with a big grain of salt until we know more about the methodology. From what we do know (they appear to test only one lens) we can infer that the procedure very well may not be precise to the second decimal that they report.
Also, your observation that there is a trend that with bigger apertures there is a bigger difference between f-stops and T-stops I could be just random fluctuations and rounding errors. It is an interesting hypothesis, but I think we can both agree that the data needed to test it would require a methodology with precision to the second decimal place of the f-stop/T-stop and that tests multiple sample of any given lens to account for variability within any model of lens. I don't think anything close to that sort of test is available, so for now perhaps we can agree that you idea is interesting but untested.
p.14 #15 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Can we compare the relative T-stops of two lenses by measuring the exposure of RAW files shooting in a controlled lighting condition, fixing the ISO and the shutter speed?
p.14 #16 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Fred Miranda wrote:
My 20/1.8G loan is coming from B&H Photo and they still say this lens will start shipping on Friday.
Looks like a bunch of us will get this lens at the same time.
Also, are T-stops measured based on the total amount of illumination on the sensor surface or just the central peak value?
Mar 04, 2020 at 04:53 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.14 #17 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
moyang_mm wrote:
Can we compare the relative T-stops of two lenses by measuring the exposure of RAW files shooting in a controlled lighting condition, fixing the ISO and the shutter speed?
Not to anything like the second decimal of an f-stop/T-stop.
), you do realize that the difference you are noting could very well be rounding error. You do realize that an f number of 1.4 isn't exactly 1.4000, in fact if a lens with an f number of 1.4 is exactly one stop more than a lens of f/1.00000, then it's exact f number is 1.4142 (i.e., the square root of 2). Similarly a lens with an f number of 1.8, is unlikely to be exactly 1.8000, and if it is exactly a 2/3rds of a stop slower than f/1.4142 then it would be 1.7818. Of course all that extra precision is ridiculous because the manufacture of lenses aren't that precise. The one decimal point is probably as precise as the measurement can and needs to be. What that means is the difference in .1 may very well not be meaningful and could well be rounding error. So you should not make a big deal about 1.4 going to 1.5 and 1.8 not going to 1.9, that could simply be rounding and note in this case if there was greater precision even 1.41 is closer to 1.5 than 1.78 is to 1.9, so let's say both a 1.41 lens and a 1.78 lens lose 0.05 transmission that would round to 1.5 and 1.8 even though they lost similar amounts of light.