can 2 different lenses with same specs have different properties?
 

I am comparing a used Canon 17-55 f2.8 IS with a Tamron 17-50mm 2.8 VC. I would think that they would have basically the same properties but today, when shooting outside after the "blizzard" dumped 14 inches on my house, the sun came out. I didn't have a neutral density filter to fit both lenses, so as a fair comparison I shot without one on either lens. I set the iso to 200 and the shutter to 1/80 (I know it's not ideal but I needed to limit light more).

I started with the Tamron and found I had to stop down really far, well past f22 I think. It finally exposed properly but it was a little soft. I think that lens was in the f30 range.

I mounted the Canon 17-55 and made sure the settings were the same. To my surprise the Canon could not go any lower than f22....that is its limit. However, at f22 the image was almost exposed properly....maybe just a tad hot. Regardless, it was majorly differnt exposure from the Tamron.

On the Tamron at f22 it was majorly overexposed.

So my question is....is it possible that 2 lenses shooting at identical focal lengths, Aperture, shutter and ISO have different light gathering properties? They are rated the same but I do not know if these ratings are consistent or if they are like wattage on car stereo where a 10 watt radio is sometimes audibly louder than a 100watt radio.

Anyway, I am trying to decide which lens to keep and which to return...and if the Canon is actually letting in less light at f22 than the Tamron at f22, that may lean me towards the Canon...even thought it is slightly more expensive.

Any advise?

Yes lenses can have different properties. Especially when they are different manufacturers. Neither should have looked very sharp when closed down to such small apertures. They might look more similar with less light. You also should try a test where you can expose subjects properly between f4 and 5.6 to evaluate the lenses under normal conditions. You also want to see how they perform at the widest settings as well. The better lens might be the one which looks sharp at widest settings.

Lens design varies dramatically between different manufacturers. It's like anything else really, Beaujolais Nouveaux is red wine, but it aint Petruse:)

I realize that all lenses aren't completely equal which is why prices can be so different. But I am asking how accurate specifications are. I would think that all things being equal as far as specs, that siimilar lenses would have similar lighting gathering properites at the same f-stop.

So you are saying that I should not expect that on a APS-C sensor camera that if I switch 2-3 different lenses and have them all at say, f11, and I ensure that ISO and shutter and all other camera setting are identical that there is a good chance that the exposure will not be the same? I thought that from a mathmatical point of view that wasn't possible.

I completely understand that all glass is not equal and that some lenses will be sharper than other, some will flare different and some will had CA issues, but I just assumed that F-stop rating on camera for the same sensor format would be be consistent. If not, then how can one have any faith in buying lenses based on their aperture ratings?

Regardless, because this was not a scientific test I did, I am going to mount up both lenses again today and compare exposure at different fstops to see if this is real or if I just made a mistake yesterday.

I would have thought f1.8 is f1.8 accross the board. Not different on differerent brands of lenses.
An interesting observation Marty.
Is f-stop not set to a specific measurement of aperture diameter within the lense?


The best way would be to evaluate it with some more lenses if one could.

Well, a really quick test would seem to prove that I didn't know what I was talking about! Comparing images between lenses just shot at the same settings shows very similar exposures. The only thing I can think at this point is the overall ambient light might have changed between lenses yesterday. It was sunny and to my eyes nothing changed....but possibly some light clouds passed in front of the sun and decreased the light by a few stops but my naked eyes couldn't perceive it.

Either way, when testing it appears that from f22 to f30 is only a 1/3 of a stoop anyway! So not a big deal if there was a difference in the first place.

Sorry to start something over nothing.

Marty,

Due to the different coatings, groups of lens, width, spacing etc you will find these discrepancies. I know no one on here EVER puts their camera in Auto, most have superglued or soldered their dials to M, but is a quick way to evaluate these differences.

When no one is looking turn the dial to "A" and fire away and note the settings. Change lenses and shoot the same thing and note the differences. See what your camera changes to create the same image. I write them down because I can't remember anything.

You can also do this with the lenses in your kit to see what you would have to do to to maintain a certain look when changing lens during a shoot.

I know the quickest way to get lambasted is to say you did anyting with the dial turned to auto but the folks that designed these cameras know a thing or two about proper exposure. At the very least, it is a great way to see how your camera sees the differences of your lenses.

Dave

Alcohol Tobacco and Firearms should be a STORE, not a government agency.

An f/stop is simply the focal length divided by the diameter of the aperture. All lenses absorb, refract and reflect light to some degree this is why cinema lenses use T-stop. T stands for transmission and is a calibrated measure of the actual amount of light that passes through a particular lens. With T-stops you can swap lenses and maintain accurate focus.

Interesting Liam. Often wondered about that.
So basically the answer to this is that there can be a difference in Fstop between lenses (however small).
But t-stops with cinema lenses is more uniform accross the board.

Yes, there can be a difference in the amount of light that hits the sensor between different lenses at the same f/stop. But worth noting, we're only talking a half stop here and a stop there - nothing super major.

I've always found F-stops to be less than an objective metric.

If your Tamron is more light sensitive, you should try a comparison under low light conditions. Maybe the Tamron will again be more light sensitive and outperform the Canon. Easy to throw on a ND filter outdoors, but low light capability is either something the lens has or doesn't.

Marty, I think the most likely reason why f/22 on the Tamron did not match f/22 on the Canon is aperture activation variance. This mechanical flaw causes the lens to use a slightly different f-number than the one on the setting. So when you thought you were setting it to f/22, it actually stopped down to f/20 "by accident", resulting in an image that is too bright. On many lenses this occurs even just one stop down from wide open, and gets even worse at slower f-numbers, so f/22 will be the most inaccurate. A similar thing happens with exposure duration. I haven't tested the Canon 17-55 for this problem, but my Tamron 17-50 f/2.8 (non-VC) definitely has it: I have to disconnect the lens electronics to work around this issue for timelapse.

It's normal. At that f-number, the diffraction is so strong that it visibly impedes contrast even at normal print sizes (8x10). Very few still photographers ever want to stop down even slower that that, so the lens manufacturers don't bother with the extra effort to build it.

Yes, but the difference can only be due to transmissivity. On the two lenses you compared, it's impossible for the difference in lens transmissivity to be more than 1/3 stop. It's far more likely that it's less than one tenth of a stop, if that.

Also keep in mind that Canon does some extra stuff with the signal in some circumstances, though in this case I don't think it applies. For example, at f/2.8 and faster, Canon brightens the image to compensate for sensor angle of response. And automatic vignetting correction for Canon-brand lenses.