I tested out the theory behind DxOMark's F-Stop blues article

[IglooEater]

I really didn't want to believe that dropping serious money on a 50mm f1.0 was completely pointless, so I decided to do a sort of "blind" test with my 5D Mark III.

But for me, right now, just knowing that having a 1.0 aperture does actually make a difference on a digital camera is enough.

The testshows, that the 1.0 lens transmits not much more light than the 1.4 lens. To gatter most light possible, the 1.0 is of very few value, it seems to be only about bokeh then.

Interesting that two people look at the same images and come to opposite conclusions... The one who spent $$$$ on a 50/1.0L concludes that it was money well spent, the one who didn't spend $$$$ concludes it would not be worth spending it.

'Round these parts, we call that bias.

We also call that different needs and priorities..

 

[C-A430]

IF the sensor really didn't at all register light incident at high angles (coming form the outermost part of the lens, i.e. from the parts of the lens between 1.4 and 1.0) then the bokeh should not be different either.

Well said. That also opens another question - does the same LENS on same CAMERA camera at same APERTURE give different bokeh at center and near the edges?

Doesn't vignetting difference at progressively wider apertures account for the lower amounts of light that DXO is measuring? And shouldn't the actual iso be noted in some exif if it is being changed under the hood? And finally the newest body that I saw listed in the referenced article was the 5d, just seemed odd. Granted it was 2010 would current bodies handle iso in the same way one would want to know. That is change to a higher effective iso as apertures get larger, and were in body lens corrections on and available then?

And to the original poster, where did you get those models? Voodoo? They kinda creeped me out.

Actually it is widely known that (RGB) micro-lenses on digital sensors increase vignetting (more vignetting than on film). According to rumors everyone is working on technology to bypass that, but it seems none has been implemented. Also it is believed that mirrorless will be the first to get that technology since they are most affected (flange distance?).

In meantime camera manufacturers (supposedly) add a cheat to make sure it is difficult to make a test showing that from f1.8 you are getting diminishing returns (less sharpness and more vignette but barely any difference in light gathering or bokeh) and with it they protect sales of f1.4 and f1.2 lenses.

 

[Sporgon]

I seem to remember reading somewhere that the best digital camera to gain most benefit from an f1 or 1.2 lens was the original 12.7 mp 5D :-X

No wonder the prices are going up

 

[IglooEater]

I seem to remember reading somewhere that the best digital camera to gain most benefit from an f1 or 1.2 lens was the original 12.7 mp 5D :-X

No wonder the prices are going up

I would ass-ume a larger pixel works better with a higher angle of incidence. I wonder how the A7s does.

 

[Sporgon]

I seem to remember reading somewhere that the best digital camera to gain most benefit from an f1 or 1.2 lens was the original 12.7 mp 5D :-X

No wonder the prices are going up

I would ass-ume a larger pixel works better with a higher angle of incidence. I wonder how the A7s does.

I think the article I read was before the time of the A7s. With an adapter and the longer distances from rear element to sensor needed for the EF f/1 or 1.2 I imagine it would be better than the old 5D. However with native f/1.2 lenses, should they exist, on the much shorter flange distance I am guessing that it would not be as good.

 

[funkboy]

I've always thought that the reason for buying a ultra wide aperture lens was more related to shallow depth of field than getting more light. There are known ways to design cameras and lenses to do a better job at ultra wide apertures, but the cost is very high. Still, we have seen recent patents trying to improve on the situation by using modified pixels or microlenses near the edges so they pick up more light.

Yep, IIRC Leica claims they do that on their rangefinder cameras as the wide-aperture angle of incidence is even higher than a DSLR due to the short flange-back distance.

 

[AJ]

All lenses exhibit vignetting (peripheral light fall-off) to some extent.
For a given lens, vignetting is typically worse at lower fstops.
Is this a function of optics, or the angle with which the light hits the sensor, or both?
Suppose it's optics, and the vignetting extends all the way to the center of the image. Then fstop is indeed a meaningless ratio and we should talk about tstop. Furthermore we should then recognize that tstop varies across the image (gets worse toward the corners).

 

[gwooding]

After reading this thread I was intrigued so I decided to do some tests myself.
The gear I used is
1DX MK2
7D MK2
50 f1.2.

I shot 6 different scenarios with each camera:
Shutter Speed: 1/40s f1.2 contacts not taped
Shutter Speed: 1/40s f1.2 contacts taped
Shutter Speed: 1/40s f1.8 contacts not taped (effectively 1EV underexposed)
Shutter Speed: 1/40s f1.8 contacts taped (effectively 1EV underexposed)
Shutter Speed: 1/20s f1.8 contacts not taped
Shutter Speed: 1/20s f1.8 contacts taped

I shot in RAW and loaded the images into Capture One Pro 10. I used the "Linear Response" curve. I white balanced the images using the gray card. All lens corrections were turned off. I then added a colour readout in the middle of the gray card to assist with quantifying the difference. This readout is also roughly in the center of the frame to eliminate the impact of vignetting (which can be clearly seen in how bright the corners in the f1.8 1/20s images are).

Looking at the 1DX results:
The first thing to notice is that there is a definite reduction in exposure when shooting at f1.2 and taping the lens contacts. In order to rectify this difference I had to add 0.2EV to the exposure in Capture One.
In order to match the exposure of the f1.8 1/20s I had to add 0.35EV to the exposure of the f1.2 taped image.

As a sanity check I tested out how much I needed to add to the f1.8 1/40s image to match the exposure of the 1/20s image and I found adding the expected value of 1.0EV to be correct.

I believe that we can therefore conclude that on the 1DX MK2 the 50 f1.2 loses 0.35EV wide open due to this phenomenon.

Interestingly the 7D results differ
I found that in this instance to match the correct exposure with the taped f1.2 I had to add 0.6EV (adding to 0.4EV to the 1EV underexposed f1.8 1/40s image matches the f1.2 taped exposure).

From this test it seems that the 7D reduces the advantage of a wide aperture lens even further. The question remains as to whether this is due to the higher pixel density or micro lens structure. I think if we could get someone to perform a similar test on a 5DS and something APSC but with a low pixel density we could get a better understanding.

 

[gruhl28]

After reading this thread I was intrigued so I decided to do some tests myself.
The gear I used is
1DX MK2
7D MK2
50 f1.2.

I shot 6 different scenarios with each camera:
Shutter Speed: 1/40s f1.2 contacts not taped
Shutter Speed: 1/40s f1.2 contacts taped
Shutter Speed: 1/40s f1.8 contacts not taped (effectively 1EV underexposed)
Shutter Speed: 1/40s f1.8 contacts taped (effectively 1EV underexposed)
Shutter Speed: 1/20s f1.8 contacts not taped
Shutter Speed: 1/20s f1.8 contacts taped

I shot in RAW and loaded the images into Capture One Pro 10. I used the "Linear Response" curve. I white balanced the images using the gray card. All lens corrections were turned off. I then added a colour readout in the middle of the gray card to assist with quantifying the difference. This readout is also roughly in the center of the frame to eliminate the impact of vignetting (which can be clearly seen in how bright the corners in the f1.8 1/20s images are).

Looking at the 1DX results:
The first thing to notice is that there is a definite reduction in exposure when shooting at f1.2 and taping the lens contacts. In order to rectify this difference I had to add 0.2EV to the exposure in Capture One.
In order to match the exposure of the f1.8 1/20s I had to add 0.35EV to the exposure of the f1.2 taped image.

As a sanity check I tested out how much I needed to add to the f1.8 1/40s image to match the exposure of the 1/20s image and I found adding the expected value of 1.0EV to be correct.

I believe that we can therefore conclude that on the 1DX MK2 the 50 f1.2 loses 0.35EV wide open due to this phenomenon.

Interestingly the 7D results differ
I found that in this instance to match the correct exposure with the taped f1.2 I had to add 0.6EV (adding to 0.4EV to the 1EV underexposed f1.8 1/40s image matches the f1.2 taped exposure).

From this test it seems that the 7D reduces the advantage of a wide aperture lens even further. The question remains as to whether this is due to the higher pixel density or micro lens structure. I think if we could get someone to perform a similar test on a 5DS and something APSC but with a low pixel density we could get a better understanding.

I'd lost track of this thread, surprised that no one has responded to these test results. Thanks for posting this, gwooding. These look like the most valid results so far. They seem to confirm that there is a loss of light with fast lenses, although at least with your lens/camera combinations it isn't a complete waste using a faster lens; there is still some exposure benefit.

 

[Arahn]

So this means that when I am shooting in very low light situations, let's say f1,2 ISO 25600, it would actually make sense to step down to f1,6 ISO 25600, underexpose -0,66 EV and later ramp up the image by 2/3 of a stop in post? The result then would be the same level of noise as the f1,2 image but feature even have more DOF!?

To make my trail of thought clearer*:
My Settings: f1,2 - ISO 25600 - 1/100s
Camera really shoots with: f1,2 - ISO 40000 - 1/100s (+0,66 ISO to compensate for "lost light")

New settings: f1,6 - ISO 25600 - 1/100s (-0,66EV - camera does not compensate)
+0,66EV in post f1,6 - ISO 40000 - 1/100s (Same Noise as in "real camera settings" but 0,66 f-stop more DOF)

*assuming the cameras do not compensate for the phenomenon when reaching f1,6 or smaller.


Or am I totally of track here?

 

[ajfotofilmagem]

So this means that when I am shooting in very low light situations, let's say f1,2 ISO 25600, it would actually make sense to step down to f1,6 ISO 25600, underexpose -0,66 EV and later ramp up the image by 2/3 of a stop in post? The result then would be the same level of noise as the f1,2 image but feature even have more DOF!?

To make my trail of thought clearer*:
My Settings: f1,2 - ISO 25600 - 1/100s
Camera really shoots with: f1,2 - ISO 40000 - 1/100s (+0,66 ISO to compensate for "lost light")

New settings: f1,6 - ISO 25600 - 1/100s (-0,66EV - camera does not compensate)
+0,66EV in post f1,6 - ISO 40000 - 1/100s (Same Noise as in "real camera settings" but 0,66 f-stop more DOF)

*assuming the cameras do not compensate for the phenomenon when reaching f1,6 or smaller.


Or am I totally of track here?

Not exactly. Tests show that the cameras distort the ISO value to compensate for the greater waste of extremely steep light rays on F1.2 lenses. It would be something like the correction of vignette in camera. But as with vignettes, the correction is most needed near the edges of the sensor, and in lesser quantity in the center of the image.