[Pavel]

Here is the link to nikon APS-C: http://www.dxomark.com/index.php/en/Lens...ric-Scores. I have particular trouble with T-stop. I read the explanation and I understand the concept, but I still do not understand the mechanics how the number is derived and whether higher number is better than lower number or the reverse. Similarly with CA, I understand what CA is, but the usual explanation (the diameter of the purple/green halo) does not seem to fit. Thanks

P

[Pavel]

Hmmm - 28 views and no takers?

[NT73]

I viewed. But I don't speak klingon, It is a bit above me Pavel.

[Toad]

I'd tell you if I understood it....really!

[matthew]

Pavel, you made it much farther into the explanations than I did: I hit that massive graph with so many close little marks and shut down. I'll have another go at it when I have more time - i.e., tomorrow at work - but it seems like a wall of noise so far.

[Irma]

Pavel, I read your post yesterday, saw the graphics and I didn't understand anything. I thought tomorrow with fresh head...Today in the morning, I did it again, and I don't really know what all this is about... sorry....

[Kombisaurus]

There's a famous saying that states "There are lies, damn lies, and statistics"...
In photography, there are lies, damn lies, and DXO...

While I actually think their testing is valid and the results useful in the hands of a small number of people who know what they are doing, the number of conflicting and wrong conclusions that many people (ie trolls and DPReview users ) draw from DXO data is astounding. A person could pick just about any combination of cameras and find DXO data that "proves" both cameras are better than each other.
I'm not having a go at you Pavel - I don't consider you a troll (or a DPReview user!). Just saying the data is very easy to misinterpret.

As for your question though... sorry but the link didn't work for me. It returned a "Module Not Found" error page.

[Pavel]

Thank you guys- now I do not feel so bad. I have been reading technical materials not necessarily in my field and I am generally good with technical plots, but this really eluded me (and still does). If I ever find a decent explanation foir this, I will post here.

[enviroguy]

So, Pavel, now you know. Or, not really.

[Pavel]

I was incensed enough to write to DXO.

[shuttertalk]

Haha... Ich spreche kein Deutsch...

My interpretation of the T-stop is that they're just measuring light transmission and testing whether the manufacturer's claimed f-stop / aperture is really what it is. i.e. the canon 85mm f/1.8 is supposed to be f/1.8 but their measured "T-stop" is actually more like 2.2

http://www.dxomark.com/index.php/en/Lens...amera)/483 (click on the transmission tab)

The 2nd graph shows the difference between the claimed f-stop and their measured "t-stop". In this case, there's a difference of "0.4".

It would stand to reason that you would want the 2nd number to be as low as possible.


Hope that helps...

[Pavel]

This is very helpful Julian. The difference between stated max aperture and t-stop is the amount of light lost due to scattered light (flare). Thus the flare is presumably 0.4 f-stop at maximum aperture. The smaller the number the less flair tendency there is. I think you pretty much got it Julian. Real good.

Pavel

[shuttertalk]

NP Pavel - happy to help!