Equivalent f-stop on dSLR

[ST1100]

We use the term "equivalent focal length" on the crop factor in dSLRs for the smaller sensor size. Is there an equivalent 'f-stop'?

When i use a 50/1.4 on a full-frame, the image brightness is at f1.4. How about on an dSLR with a 1.6x? Is it still f1.4 or faster, or slower?

From what i understand, when an image is f1.4, every single point on the image is at 'f1.4 brightness', cropped or not. So i believe even with a 1.6x, the lens is still a f1.4 lens. Is this understanding correct?

If not, what is the correct way to calculate 'effective f-stop' on a crop?

 

[ivor]

I understand that, it would be 80mm f/1.4 instead of 50mm f/1.4

 

[cyanbloodbane]

my understanding is that a 50/f1.4 on a dslr is still a 50/f1.4. but due to the smaller ccd size, the final image appears like its taken with a 80mm (assuming ~1.5x mag. factor)

hence sunny 16 still work with dslr.

 

[Garion]

Good question, ST1100, I must admit this never crossed my mind too. My guess is that the f-stop remains the same whether cropped or uncropped. Maybe the experts here could clarify?

 

[Newman]

The 1.x multiplier affects the image view only not the f-stop. Like what cyanbloodbane said, this is due to the smaller than full frame CCD/CMOS. So images we take appear closer which is not true as they are actually cropped. Therefore your f-stop is not affected at all unless you use something like an extender.

 

[ST1100]

Hi, just to throw a curved ball in here. Consider a 50/1.4 on a 10D (1.6x) and an 80/1.4 on a full frame.

Both have the same angle of view. However, the 50/1.4 image is less bright, because it has less glass to collect light. A 50/1.4 has a diameter of 35.7mm, while an 80/1.4 has a diameter of 57mm - an area of 2.5x more. (Area = pi*r^2).

So, a 50/1.4 on a 10D is actually slower than an 80/1.4 on a full frame. In other words, a 50/1.4 on a 10D is actually an *effective* 80/2.x.

 

[jasonpgc]

Consider comparing APS format film and a 35 m format film. Both have the same ISO, only the format is different.

The light recieved by the film should be based on the intensity of light falling on each "grain" or pixel not the entire surface area of the film. A smaller film format will not affect the intensity of light recieved by the film.

I believed you have this doubt due to a confusion with extension tubes and teleconvertors.

Extension tude actually magnify the image by increasing the distance between the lens rear element and the plane intended for projection. Therefore the light recieved by the film is weaker. For example, a projector projecting on a screen 2m away and a screen 5m away with the same power output. The 5m screen image will definately appear to be dimmer.

I believe the light from the teleconvertor is suck up in a similar way. Althought it has tele lens group to double the focal length, the actually distance between the rear element of the original lens and the film has increase due the the thickness of the teleconvertor.

But for the case of a APS and 35mm format, the lens still maintain the same rear lens to film distance and aperture size, therefore only the image area that is not capture by the smaller sized APS format is lost. The intensity of the light recieved should remain the same as what the lens is designed for.

 

[Newman]

Originally posted by ST1100
Hi, just to throw a curved ball in here. Consider a 50/1.4 on a 10D (1.6x) and an 80/1.4 on a full frame.

Both have the same angle of view. However, the 50/1.4 image is less bright, because it has less glass to collect light. A 50/1.4 has a diameter of 35.7mm, while an 80/1.4 has a diameter of 57mm - an area of 2.5x more. (Area = pi*r^2).

So, a 50/1.4 on a 10D is actually slower than an 80/1.4 on a full frame. In other words, a 50/1.4 on a 10D is actually an *effective* 80/2.x.

Actually a 50/1.4 & 80/1.4 should transmit the same amount of light. And that goes the same for all f1.4 lens or any other lens with the same f number. The amount of glass is not the issue but the aperture size is.

Extension tubes increases the focal length thereby increasing the f-number.
eg. for a 50mm/f1.4 lens, aperture size is 35.7mm
When u add a 2x extension, focal length increases to 100mm but aperture size is the same so f-number=100/35.7 which is f2.8.

 

[sehsuan]

guys, i think most of us have to remember that the f-stop is determined by focal length (mm)/aperture dimension (mm)....

for any given slr lens, both are fixed constant, so that's it, the light transmitted into the camera is still the same. BUT if we factor in the sensor size, then because of lesser area (compare 1Ds and d30/d60/10d), there is less light hitting onto the cmos/ccd, and *possibly* AF might be slower, but this should only reflect in low light condition. in sunlight condition, there are more than enough photons to go into the camera to run the AF at speeds with practically no difference.

 

[Newman]

sehsuan,
D60 and 10D same CMOS size but 10D AF is faster. So we can rule out that size of sensor is an issue for light transmission. Remember, the sensor is the digicam's film. So does it mean that larger format cams have faster AF? I doubt so. AF is largely affected by optics and electronics of the cam.

 

[YSLee]

ST1100: EV values will always remain constant for a given aperture and shutter speed, regardless of format use.

sehsuan: I suggest you do some research before you post and mislead others. The reliablility of an autofocus camera depends on three things: the brightness of the lens, the AF module in place, and the AF motor in use. Nothing to do with the CCD/CMOS; prior to that, how'd you think film SLRS did their AF?

During AF, the CCD/CMOS sensor of a DSLR has absolutely no part in focusing; after all the image is being reflected via the reflex mirror to the viewfinder. A little thinking on one's part would've realised that. The AF module is under the reflex mirror; that's where the camera "sees" where and how to focus, and gives instructions to the AF motor to turn accordingly.

Once again, it would be appreciated if you could refrain from posting totally incorrect assumptions and passing them off as facts in that authoritative tone which is very misleading.

Newman: AF slows down in 645/6x6 AF cameras, then there's no AF to speak of in even larger formats!

*imagines an AF Sinar 4x5 monorail*

 

[Newman]

Originally posted by YSLee
Newman: AF slows down in 645/6x6 AF cameras, then there's no AF to speak of in even larger formats!

Hi YSLee,
Thanks for letting me know that. Actually, I also meant 645 format. Any larger, I've no knowledge whatsoever. I've only seen AF lenses for 645. Why does the AF slow down in this medium format?

 

[sehsuan]

thank you YSLee for pointing out my error.

forgot about the mirror... argh...

but the first part, about the same aperture thing, is still correct, i hope?

 

[Knighthunter]

Copy and past from Nikon Website:

Focal length and Picture angle
When Nikkor lenses originally designed for the 35mm [135] format are used on the Nikon DX Format, photographers believe that the lens' focal length is increased by a factor of 1.5. For example, a 17-55mm lens would become a 25.5-85.5mm lens. Technically, that is not correct. In fact, there is no change of focal length as you switch the lens from one format to the other. The actual change is that the picture angle of the lens is narrowed on the digital format because the digital format is smaller in relation to the lens' image circle projected on the sensor. Focal length and reproduction ratio remain constant for the comparable viewed portions of the image. It is the picture angle that changes.
This new DX Nikkor lens' true focal length is 17-55mm with a picture angle that will vary from 79° at 17mm to 28°50' at 55mm.
The combination of reduced picture angle and a smaller image circle enabled the creation of this compact, lightweight DX Nikkor lens that covers a range from wide-angle to portrait and medium telephoto shots.

Source: http://www.nikon-image.com/eng/news_release/2003/afs_dx17_55.htm

And I just can't understand how ppl got idea that CCD/CMOS is affecting lens brightness. AF is driven by the AF module below the mirror, the CCD/CMOS only works when the shutter is OPEN.

 

[Newman]

Originally posted by sehsuan
thank you YSLee for pointing out my error.

forgot about the mirror... argh...

but the first part, about the same aperture thing, is still correct, i hope?

Aperture is not fixed. It can be reduced but not increased beyond its max i.e. f1.4 -> f2.8,f5.6,etc but f1.4 cannot change to f1.0. If can, I don't need to buy f2.8 L lens liao.

 

[YSLee]

Ah, Knighthunter is correct, it's under the mirror, not in the pentaprism. Corrected my post. Sorry for the stupid error.

Newman: Well, two things that come to mind are that there is more glass to move, and that AF development for MF cameras have not been as long nor as in high demand compared to 35mm formats.

 

[Newman]

I see. Mucho gracias amigo!

 

[wacko]

Originally posted by ST1100
Hi, just to throw a curved ball in here. Consider a 50/1.4 on a 10D (1.6x) and an 80/1.4 on a full frame.

Both have the same angle of view. However, the 50/1.4 image is less bright, because it has less glass to collect light. A 50/1.4 has a diameter of 35.7mm, while an 80/1.4 has a diameter of 57mm - an area of 2.5x more. (Area = pi*r^2).

So, a 50/1.4 on a 10D is actually slower than an 80/1.4 on a full frame. In other words, a 50/1.4 on a 10D is actually an *effective* 80/2.x.

hmmmm, but if the 10D had a full frame CMOS, the angle of view would be different. from how i see it, both pics ought to be more or less equally bright cos the actual angle of view for the 50/1.4 is larger than the 80/1.4, it just so happens that the smaller than full frame CMOS in the 10D is not able to capture all available light, thus you have something sort of like auto-cropping. and this follows your "equal brightness cropped or not cropped" theory.

layman's view, sorry if it's horribly wrong... :embrass:

 

[ST1100]

Sorry wacko, i am not able to follow your train of thot.

Anyway, i've sorted out my theories. A 50/1.4 on a 10D *is* effectively an 80/1.4, bcoz a 1.4 brightness is relative to image circle and not focal length; ie a 400/1.4 and a 50/1.4 (for the 35mm format) both cast the same strength of light on every point of the image circle (ie f1.4), totally independent of sensor type or size (or presence). [If this assumption is incorrect, someone pls inform me.]

Wow, i have an 80/1.4 and a 135/1.8.

(Of course, with the bokeh quality of a 50/1.4 and 85/1.8 )

Thanx for all the contributions, guys.

Jason: to clarify, no i wasn't thinking of convertors or extensions. What started me thinking was someone in another forum who casually claimed that a 50/1.0 was slower than a true f1.0.

 

[Newman]

Originally posted by wacko
hmmmm, but if the 10D had a full frame CMOS, the angle of view would be different. from how i see it, both pics ought to be more or less equally bright cos the actual angle of view for the 50/1.4 is larger than the 80/1.4, it just so happens that the smaller than full frame CMOS in the 10D is not able to capture all available light, thus you have something sort of like auto-cropping. and this follows your "equal brightness cropped or not cropped" theory.

layman's view, sorry if it's horribly wrong... :embrass:

:mad2:

wacko:kok: