Puzzled by Canon vs. Olympus

I have a Canon 7D Mk 2 and an Olympus OM-D E-M5 (Mk 1) and I was interested in comparing them with a view to buying more Olympus lenses... but I'm puzzled.

Using the 7D with an EF 70-200mm f2.8 L lens set at 200mm and ISO 200, I was taking a scene. Set at f8 on Av, it gave about 1/200 sec shutter speed.

Using the OM-D with a 40-150mm f2.8 Pro lens, set at 150mm and ISO 200, same scene, set at f8 on Av, it gives a shutter speed of about 1/100 or so.

What's going on here? The 35mm focal length equivalent is roughly similar (320mm for the Canon, 300mm for the Oly) and everything else is the same, but the Oly says the light is about half what the Canon reads. Both produced perfectly good images of roughly equivalent IQ (though I'd say the Canon was a little better) but I don't understand why the difference in readings.

Can anyone explain?

Any number of things.

I'm going to go with the fact you have two different cameras with different metering algorithms as probably the biggest difference.

Rather than let the camera pick the exposure, do it manually and compare the results.

Damn, I was hoping someone would come up with something about the physics of the optics or whatever...

Yes, I'd figured the metering algorithms would mean there'd be some difference, I just didn't expect it to be over a stop. And both images look reasonably well exposed, if anything the Canon is lighter, while if I change the settings on the Olympus to match it becomes obviously under exposed.

Unfortunately I don't have a separate light meter so I can't make an independent reading for manual settings.

Tried with my 5D and a 300mm lens as well. It pretty much matches the 7D, but I'd be surprised if it hadn't.

I shall put it down as a mystery... and open the Oly up a stop more when I use it.

I see what you are getting at:

Canon @ 200mm + F8 + ISO 200 = 1/200
Olympus @ 150mm + F8 + ISO 200 = 1/100

In a somewhat recent review of the Tamron 150-600 G2 Canon mount, Dustiom Abbott compared the G1 version and noted the G2 on the same 5D MKIV body showed less light transmission through the lens than the G1.

In another review of the Tamron 24-70 G2 Dustin compared the Sigma and Canon 24-70 and also noted light transmission differences between these lenses, again same bodies.

Aperture size is directly relevant to focal length but construction of lenses IE how many elements, coatings etc. are likely to impact light transmission insome way given all things are equal; it's physics.

Not to mention cameras sensors are different.

I'm not always a big fan of Tony Northrup (partially because in the past he didn't seem to follow his own advice as spoken in the following video), but this video explains a lot of the problems with your methodology here.

https://youtu.be/MM7hA​v86odw

Shaun Liddy wrote in post #18611925
I see what you are getting at:

Canon @ 200mm + F8 + ISO 200 = 1/200
Olympus @ 150mm + F8 + ISO 200 = 1/100

In a somewhat recent review of the Tamron 150-600 G2 Canon mount, Dustiom Abbott compared the G1 version and noted the G2 on the same 5D MKIV body showed less light transmission through the lens than the G1.

In another review of the Tamron 24-70 G2 Dustin compared the Sigma and Canon 24-70 and also noted light transmission differences between these lenses, again same bodies.

Aperture size is directly relevant to focal length but construction of lenses IE how many elements, coatings etc. are likely to impact light transmission insome way given all things are equal; it's physics.

Not to mention cameras sensors are different.

Looks like there is a difference in transmission between the two lenses.. Check out DXO measurements of the two lenses. (click "measurements" then "transmission")

gjl711 wrote in post #18611935
Looks like there is a difference in transmission between the two lenses.. Check out DXO measurements of the two lenses. (click "measurements" then "transmission")

Now that certainly could go some way towards explaining it... although the actual Olympus lens I was using doesn't have a DxO score on that site, it does indicate a possible explanation. I'd never heard of "T-stops" before. Thanks for that.

I'd use manual mode for better comparisons, then you can really bring in the laws of physics instead of the 'laws' of a manufacturer's algorithms.

Alveric wrote in post #18612047
I'd use manual mode for better comparisons, then you can really bring in the laws of physics instead of the 'laws' of a manufacturer's algorithms.

Without an incident light meter, using manual mode would just mean I'd be matching up the indicators displayed by the camera - effectively I'd be doing what the aperture priority mode does anyway. No point.

But it's not that important; I'm not trying to do a detailed review or anything. I want less weight to carry around, and I am thinking of buying am OM 300mm lens (and possibly an OM-D EM-1) to use for wildlife photography, instead of the Canon 7D with a 400mm which is currently my usual. I was just trying to see if there was any huge loss of quality in an Olympus as compared with the Canon. The Oly quality seems OK, but the loss of a whole stop due to light transmission would mean slower shutter speeds, or higher ISO, which might be an issue.

But until I can actually try the Olympus combo there's no way to be sure.

Gannet77 wrote in post #18611911
Damn, I was hoping someone would come up with something about the physics of the optics or whatever...

.
But it is about the physics of the optics. . Lens construction - optic design and physics - has everything to do with the T stops, which, as you now know, is the source of the difference between the exposure settings.

So exactly what you were hoping for is exactly what happened - someone came up with something about the physics of the lens design.


.

I have gear consisting of a handful Canon and a handful of Oly bodies and lenses. I don't think that I have noticed what you have in this comparison, in fact there are times that I switch mid shoot and dial in the settings of the camera I was using into the other body without issues. That is with flash though.

One thing I have always noticed is Canon seems a lot more geared to protecting highlights by default. In my 7D and some others I end up dialing in a 1/3 -2/3 exposure comp.

Gannet77 wrote in post #18612092
Without an incident light meter, using manual mode would just mean I'd be matching up the indicators displayed by the camera - effectively I'd be doing what the aperture priority mode does anyway. No point.

But it's not that important; I'm not trying to do a detailed review or anything. I want less weight to carry around, and I am thinking of buying am OM 300mm lens (and possibly an OM-D EM-1) to use for wildlife photography, instead of the Canon 7D with a 400mm which is currently my usual. I was just trying to see if there was any huge loss of quality in an Olympus as compared with the Canon. The Oly quality seems OK, but the loss of a whole stop due to light transmission would mean slower shutter speeds, or higher ISO, which might be an issue.

But until I can actually try the Olympus combo there's no way to be sure.

You don't need a light meter for a quick head to head. All you need is daylight and Sunny 16. Then you can compare how each camera renders the same scene under the same light conditions using the same settings.

Gannet77 wrote in post #18612092
Without an incident light meter, using manual mode would just mean I'd be matching up the indicators displayed by the camera - effectively I'd be doing what the aperture priority mode does anyway. No point.

No, what is being suggested is to use manual mode so that you can select all of the same settings on each camera i.e. set them both to f/8, ISO200 and 1/200. Then you can compare the exact same point in the scene and read the RGB values. Using a tool like photoshop or lightroom, you can brighten the darker image until the RGB values match, and then you will know the exact difference in exposure.

Possible sources of the difference:

Lens light transmission.
Camera ISO error
Lens aperture error
Shutter speed timing error

JeffreyG wrote in post #18612261
No, what is being suggested is to use manual mode so that you can select all of the same settings on each camera i.e. set them both to f/8, ISO200 and 1/200. Then you can compare the exact same point in the scene and read the RGB values. Using a tool like photoshop or lightroom, you can brighten the darker image until the RGB values match, and then you will know the exact difference in exposure.

Agreed. Having the settings the same is crucial for the test. I would go so far as to say to shoot custom white balance, and raw + jpg for both cameras to see how the raw is handled by your preferred editor. This will also let you test the in camera jpg processing for each setting on each camera, even a setting such as "neutral" must apply some changes to the RAW.

just to nitpick, error might not be the right word. I think camera manufacturers deliberately choose real world ISO settings that do not match what their actual setting should be in order to gain just a little edge in image quality.

As others have said, lens light transmission probably plays a part. But if the below graph from the Photons to Photos site is accurate, 2/3 of a stop of the difference is down to how each camera interprets ISO—the 7DII is 1/3 stop below standard ISO and the Oly is a full stop below (i.e. when both cameras are set at ISO 200, the 7DII is actually at ISO 160 and the Oly is actually at ISO 100).