Started Feb 3, 2013 | Questions thread
Shop cameras & lenses ▾
Jack Hogan
Veteran MemberPosts: 4,923Gear list
The SLT system costs 1 Stop in SNR performance - not 1/3
In reply to Pitbullo, Feb 4, 2013

Pitbullo wrote:

Ron Poelman wrote:

Pitbullo wrote:

The translucent mirror really does steal some light, and to me that is an issue when it comes to upgrading later on.

1/3 of a stop is a deal breaker, really ?

When I see comparisons at high ISO (especially FF) I always get the impression that the SLT is performing quite noticeably worse than the Canikons.

Trying to compare apples to apples (say two 24MP full frame cameras with same-generation sensors made by the same manufacturer, like the A99 and the D600) the SLTversion is one stop worse as far as SNR is concerned, as one can easily see in DxOmark's SNR graphs. From their data:

My elaboration of DxOmark full SNR fitted data for the A99

My elaboration of DxOmark full SNR fitted data for the D600

Accordingly, the SLT system has an effective Absolute QE* of around 8% vs the non SLT system of 16%, the same one stop difference.

Where does the 1/3 of a stop mentioned above come from?

* Effective average Absolute QE takes into consideration the effect of any filter sitting on top of the sensor, mainly the CFA and the SLT if present, when calculating photon-electron coversion efficiency.

Reply   Reply with quote   Complain
Post (hide subjects)Posted by
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow