Nikon: Anyone with a 50mm 1.4/1.8 and 17-55?

[Cheng]

Just had my first encounter with 'T-Stops'. In manual mode, using the same settings on a Sigma 18-50 2.8 compared to a Nikon 50mm 1.4, for each equivalent exposure, the prime is about 2/3rd stops faster than the fixed aperture zoom. So to take equivalent exposures with both lenses, I have to set the 50/1.4 at -2/3 EV. Which when using the prime as a reference, makes the Sigma a 18-50 3.5 equivalent, in my eyes anyway! Just wondering, if the 17-55 is any better? Could anyway with both take 2 photos at the same exposure, compare the difference, and tell me roughly how fast the 17-55 is compared to the prime i.e. is it roughly 17-55 2.8, 3.2 or 3.5? Cheers!

Just to clarify (as I realise the above makes little sense):

Nikon 50mm 1.4 prime , set at 50mm/2.8: 1/40 at 2.8 - 0 stops different, the Reference Value
Sigma 18-50mm 2.8 zoom, set at 50mm/2.8: 1/25 at 2.8 - 2/3rds stop lower than Reference

So this shows that even when set to the same values, the fixed aperture 2.8 zoom is not as wide at 2.8 as the prime at 2.8.

 

[Cheng]

I thought I'd just test my 18-200 VR to see how that fared - relative to the 50/1.4, it too transmit 2/3rds of a stop less light for each equivalent exposure. These were all at 50mm by the way, with as near identical framing as possible.

 

[Paulie-W]

Just tried a quick test at iso 800 f2.8 app prority 17-55 1/15 ........ 50 (1.4) 1/25. :shrug:

 

[Blue Eagle]

I thought I'd just test my 18-200 VR to see how that fared - relative to the 50/1.4, it too transmit 2/3rds of a stop less light for each equivalent exposure. These were all at 50mm by the way, with as near identical framing as possible.

Erm, the stops go: /1.4 /2 /2.8 /4 /5.6

So your f/1.4 is two stops faster than any /2.8 lens for example. I'm not sure how you are testing this, but if you keep the light (try indoor, artificially lit wall with the curtains drawn), the aperture and the iso all constant, then you should be able to see the shutter speed move. And the shutter is easy to get your head around since a doubling or halving is a movement of one stop.

 

[Cheng]

Just tried a quick test at iso 800 f2.8 app prority 17-55 1/15 ........ 50 (1.4) 1/25. :shrug:

Ahh, cheers, looks about 2/3rds to me too! Did you compare the histograms so they were similar, or just read off the viewfinder the value? If you just read it off, you might want to take a shot and compare the histograms, as sometimes the metering is very slightly different too!

F stop calculator

 

[Paulie-W]

Ahh, cheers, looks about 2/3rds to me too! Did you compare the histograms so they were similar, or just read off the viewfinder the value? If you just read it off, you might want to take a shot and compare the histograms, as sometimes the metering is very slightly different too!

F stop calculator

Historgrams look pretty similar (still havent got my head around reading them properly yet). I can post the actual shot up if you like.

 

[Cheng]

That's fine, as long as they're roughly in the same position, it sounds about right! Basically histograms are just distributions of light - the further to the right it is, the brighter the image is. So to get similar looking exposures, the right most line of the histograms (and the histogram in general) should more or less be in the same place. So it looks like upgrading to a 17-55 would not really be beneficial in terms of transmission, than the sigma! Thanks for your time!

 

[Cheng]

Erm, the stops go: /1.4 /2 /2.8 /4 /5.6

So your f/1.4 is two stops faster than any /2.8 lens for example. I'm not sure how you are testing this, but if you keep the light (try indoor, artificially lit wall with the curtains drawn), the aperture and the iso all constant, then you should be able to see the shutter speed move. And the shutter is easy to get your head around since a doubling or halving is a movement of one stop.

The stops I were mentioning were not full stops, but 1/3rd stops - after all, 1 stop is quite a big leap! These were taken entirely by artificial light...its pretty dark now a days!

 

[Paulie-W]

No worries

 

[Martyn...]

I have a Nikon 50 f1.8, and the 17-55 f2.8 which was your original question, which is about the only part of your post that I understand clearly.

I set the zoom to 50mm, I aimed at the exact same spot with the 50mm prime, and the zoom, the meter readings were the same, which is what I would expect.

17-55 (at 50mm) = 1/60 f4 ISO 1000, 1/60 f2.8 ISO 500

50mm f1.8 = 1/60 f4 ISO 1000, 1/60 f2.8 ISO 500

The conclusion is that both lenses expose the same, which is what they should do.

Just to add:-

Having read through your post again the issue you are encountering is probably down to the difference in quality of the lens elements and coatings between the Sigma and the Nikkor, the Nikkor being superior on both counts.

 

[Cheng]

I have a Nikon 50 f1.8, and the 17-55 f2.8 which was your original question, which is about the only part of your post that I understand clearly.

I set the zoom to 50mm, I aimed at the exact same spot with the 50mm prime, and the zoom, the meter readings were the same, which is what I would expect.

17-55 (at 50mm) = 1/60 f4 ISO 1000, 1/60 f2.8 ISO 500

50mm f1.8 = 1/60 f4 ISO 1000, 1/60 f2.8 ISO 500

The conclusion is that both lenses expose the same, which is what they should do.

Just to add:-

Having read through your post again the issue you are encountering is probably down to the difference in quality of the lens elements and coatings between the Sigma and the Nikkor, the Nikkor being superior on both counts.

I'd love to believe this to be true, as I've been considering buying a 17-55, but Paulie-W's experience seems to be different - he seems to experiencing a similar situation to me!

There's a possibility that the 50mm 1.4 we both used transmits more light than the 50mm 1.8 does at 2.8 which could explain why you're getting the same readings, but I think this would be negligible and am not really convinced this is the case. I do believe we need more evidence to suggest that the 17-55 at 50mm/2.8 transmits the same amount of light as a prime lens at 2.8.

Did you point the camera and just read off the LCD, or take an exposure and compare histograms? Sometimes the metering can be slightly different, so whilst it tells you that it thinks that they are the same exposures, the actual histograms (and thus exposure) is different.

Apparently zoom lenses usually lose some light compared to their prime counter parts due to the added number of elements and complexity in design, so perhaps it's wishful thinking to expect a 2.8 zoom to be the same as a 1.4 prime set to 2.8, but your lens may be proving me wrong! I hope this makes (some sort of) sense!

 

[Martyn...]

I'd love to believe this to be true, as I've been considering buying a 17-55, but Paulie-W's experience seems to be different - he seems to experiencing a similar situation to me!

There's a possibility that the 50mm 1.4 we both used transmits more light than the 50mm 1.8 does at 2.8 which could explain why you're getting the same readings, but I think this would be negligible and am not really convinced this is the case. I do believe we need more evidence to suggest that the 17-55 at 50mm/2.8 transmits the same amount of light as a prime lens at 2.8.

Did you point the camera and just read off the LCD, or take an exposure and compare histograms? Sometimes the metering can be slightly different, so whilst it tells you that it thinks that they are the same exposures, the actual histograms (and thus exposure) is different.

Apparently zoom lenses usually lose some light compared to their prime counter parts due to the added number of elements and complexity in design, so perhaps it's wishful thinking to expect a 2.8 zoom to be the same as a 1.4 prime set to 2.8, but your lens may be proving me wrong! I hope this makes (some sort of) sense!

Cheng, do not get hung up on over analysing everything, that is an engineers job, not a photographers

Just repeated the test, same result, checked the histograms, they showed a slight variance but nothing of note, probably because I could see a slight difference in framing.

To be 100% accurate the test would need more control, with regard to precise framing, and absolute control of lighting.

If you are looking for a reason not to buy the 17-55 Nikkor that's fine, but you would be missing out on a very good camera lens.

To take your test to the extreme, I need to compare my 200-400 f4 VR against my 300 f4 AF S, the former having 24 elements the latter 10.

I shall report back shortly on the above

 

[Martyn...]

300 f4 AF S and 200 - 400 VR f4 (set at 300mm)

300 f4 AF S 1/60 f4 ISO 1100, 200-400 f4 1/60 f4 ISO 1800

Histograms looked close, the test was done hand held, there was some fluctuation in the 200-400 ISO, but the figures shown were achieved several times for both lenses.

I still got similar results for the 50 to 17-55 comparison, no glaring difference.

 

[Cheng]

Martyn, sounds good, and sounds like you know what you're doing! Just wondering about Paulie's findings then - I must admit I do like the sound of the 17-55, but not sure if I can justify the extra £350 on top of a 3rd party 17-50 2.8 for the extra 2/3rd stop, when the jump to a FF D700 might be the better option!

 

[Martyn...]

Martyn, sounds good, and sounds like you know what you're doing! Just wondering about Paulie's findings then - I must admit I do like the sound of the 17-55, but not sure if I can justify the extra £350 on top of a 3rd party 17-50 2.8 for the extra 2/3rd stop, when the jump to a FF D700 might be the better option!

As I said it is impossible to be 100% without 100% control of all factors, I am away on holiday Monday, so am pushed for time, however I am sure if I repeated the test in slightly different conditions I may see a different result.

Paulie was comparing to the f1.4, I was using the f1.8 that could be a factor.

If you are not spending most of your time with a long tele on the camera chasing small birds, I would go with the D700, I may be adding one to my camera bag in the near future, I have some older lenses that would be ideal on the FF body.

 

[Arkady]

hmmm...will it actually make any difference in the real world?
After all the variables inherent in outdoor photography - subject luminance range, ambient light, reflected light, direct light...blah, blah, blah...
Just adjust exposures to each scene accordingly and stop worrying would be my advice...

So what if there's a 1/3 stop variance - use the lens that gives the best results and take some photos...

 

[Cheng]

hmmm...will it actually make any difference in the real world?
After all the variables inherent in outdoor photography - subject luminance range, ambient light, reflected light, direct light...blah, blah, blah...
Just adjust exposures to each scene accordingly and stop worrying would be my advice...

So what if there's a 1/3 stop variance - use the lens that gives the best results and take some photos...

Very true ... most of the time I'd say! However, I was thinking about venturing into some basic gig photography, and here there would be quite a big difference - was thinking if I could get away with the flexibility of a 17-50 2.8 vs a 1.4 50mm...the extra 2/3rds of a stop is the final nail in the coffin for that decision, where on DX I need all the light I can get at night! Looks like I'll just be using the 50mm 1.4 then..