
Jack, FWIW: Forum rule #18: http://www.dpreview.com/forums/forumrules I doubt anyone would be too worried by 13 sub 800x? images. I haven't played much with galleries, so I don't know if that ...

Claiming " without chromatic aberration " is something of a stretch if you want to use it for available light photography. If you look at the original paper , aberrations are terrible at ...

Glen The meaning and connotations tend to be contextdependent, which makes it difficult to pin down. After a couple of minutes with Google, I found this page from the Guardian with some useful ...

I should wait until they actually put it on glass before you get rid of your current lenses. It's on silicon at the moment, and only works at infrared wavelengths. The Scientific American article ...

It would be more interesting if the average k was anything else. You have 4 identical subpixels, and constrained the sum to be exactly 4. If instead you constrain the sum to be 3, the average ...

Definitely No, but I agree I didn't answer your question directly. I am still trying to figure out what insight you get from these calculations, or where they take us. Can you help us out here? Ch ...

No. The distribution for mean = 4 is broader, so the probability density at the mean is lower. Using the formula in Eric Fossum's post: :: mean  1   2   3   4   5   6 

The photoelectron count is always Poisson distributed, but this tends to Gaussian when the mean is large. If we have a large number of uniformly illuminated pixels, we don't necessarily need the ...

Jack, I don't believe truncation to zero is a problem for the case described by Joofa. It is certainly a concern if you also have read noise, subtract the mean dark signal, and then truncate ...

So, to paraphrase: (1) Take groups of 4 small pixels, each collecting Poisson distributed electrons with mean 1.

Certainly since 1983. Much longer if you are talking more generally about the MKS or SI systems of measurement . In 1983 the metre was officially redefined by the 17th CGPM as " the length of the ...

I have done the same often enough to sympathise. In any event, a similar conclusion presented with a more quantitative rationale is almost always helpful. Regards

With shot noise, an important point to remember is that the square root relationship between standard deviation and signal level applies to the population mean number of electrons collected. The ...

No experience with Nikon files. Exiftool may do the trick. According to Phil Harvey's documentation, parts of the makernotes may be encrypted, but the decryption algorithm is known. I don't know if ...

Jim, Your measurements suggest that the effect on read noise is small, but there could be a larger selfheating effect at shutter speeds slower than 1/2000. One mechanism for temperaturedependent ...

I should wait until they actually put it on glass before you get rid of your current lenses. It's on silicon at the moment, and only works at infrared wavelengths. "No CA" is arguable something ...

Thanks for the link , but I couldn't find a direct answer the OP's original question "why is it called four thirds". There is a rather oblique reference to size here : From context "4/3 type" ...

Bearing in mind your comments about selective quoting If we quote more of the very same text ( my italics ): From the beginning of the same Wikipedia section : And below, complementing rather than ...

Also reported in this thread Paper: http://arxiv.org/ftp/arxiv/papers/1411/1411.3966.pdf Focal length varies strongly between the 3 design wavelengths with a significant level of scatter, even at ...

From the same Wikipedia article: Cheers
Activity older than 12 months is not displayed.

alanr0 has not added any gear yet.
Total messages 
1297 
Threads started 
0 
Last post 
2 weeks ago 
