mb66energy said:not missed but misinterpreted + size sometimes doesn't matter
Well spoken, Sir! ;D
Upvote
0
Monochrome-only body?
- Thread starter jolyonralph
- Start date
Well spoken, Sir! ;D
AvTvM said:I am sure there are millions of units in pent-up demand for a black&white only Canon camera.
a B&W digital camera would be really, really stupid, Canon!
In that case, why not start up a crowd funded operation to produce one, its so easy.
If there were even half a million buyers, smaller camera companies would be all over it. Even 100,000 units!
There were about 5.8 million DSLR's shipped from Jan - Sept of 2016. How many millions of those would have been sold if they did not have color? Certainly not 20%.
Upvote
0
I know this is probably Economics 101, but a monochrome camera body isn't going to be something that people buy instead of a standard DSLR, it's something some will buy as well.
So the only calculation needs to be whether enough people will do this to make it worth the production & marketing effort.
Probably not, but I'd still like one.
So the only calculation needs to be whether enough people will do this to make it worth the production & marketing effort.
Probably not, but I'd still like one.
Upvote
0
One of my coworkers has a webcam that has somehow become monochrome-only. We are all software engineers working on videoconferencing solutions (Skype for Business), and it's a continuous source of amusement and puzzlement how an RGB device could fail in such a way. Reading this thread, I'm thinking maybe it's a feature and not a bug!
Upvote
0
<EDIT>neuroanatomist said:Don Haines said:There is a difference between a Monochrome ONLY body, and converting a colour image to simulate monochrome. You end up missing all the light that the bayer filter blocked.neuroanatomist said:mb66energy said:neuroanatomist said:Done.
NOT done - roughly 75% loss of photons due to color filters in bayer array.
Really? Seriously?!? See, I wasn't really sure why but I just knew there had to be some reason I bought a couple of $14,000 Zeiss AxioCam HRc color cameras for brightfield microscopy, and a couple of $14,000 Zeiss AxioCam HRm monochrome cameras for fluorescence microscopy where sensitivity for low intensity signals is critical. Thanks for edumacating me!
Love your humour.... just got it.... ignore what follows!
The same holds for astrophotography.... you see VERY!!!!! similar colour and B/W versions of the same camera for just that reason....
</EDIT>
Ok... Seriously..... can you explain to me why I am wrong?
A normal colour camera has a bayer filter. The bayer filter blocks light. For each block of 4 pixels, two will only see green light, one will only see blue light, and one will only see red llight. If we assume an even distribution of light across the visible spectrum, then for each pixel, 2/3rds of the light will be blocked.
It would follow that if you removed the bayer filter, then each pixel would receive 3 times as much light. Surely having 3 times the signal to deal with is preferable to setting a processing flag after an image with 1/3 as much light is taken?
Upvote
0
AvTvM - you may need an even more obvious way to mark your comment as humour....lMt Spokane Photography said:AvTvM said:I am sure there are millions of units in pent-up demand for a black&white only Canon camera.
a B&W digital camera would be really, really stupid, Canon!
In that case, why not start up a crowd funded operation to produce one, its so easy.
If there were even half a million buyers, smaller camera companies would be all over it. Even 100,000 units!
There were about 5.8 million DSLR's shipped from Jan - Sept of 2016. How many millions of those would have been sold if they did not have color? Certainly not 20%.
Upvote
0
Don Haines said:AvTvM - you may need an even more obvious way to mark your comment as humour....lMt Spokane Photography said:AvTvM said:I am sure there are millions of units in pent-up demand for a black&white only Canon camera.
a B&W digital camera would be really, really stupid, Canon!
In that case, why not start up a crowd funded operation to produce one, its so easy.
If there were even half a million buyers, smaller camera companies would be all over it. Even 100,000 units!
There were about 5.8 million DSLR's shipped from Jan - Sept of 2016. How many millions of those would have been sold if they did not have color? Certainly not 20%.
Yep. It's almost like it's Emoticon Blindness Awareness Week with this thread.
AvTvM doesn't think there is a market here, people.
- A
Upvote
0
Don Haines said:
You are wrong beacuse you're spelling color with a 'u' and failing to capitalize the eponymous Bayer filter.
Upvote
0
jolyonralph said:
Anything with a Foveon sensor will give you monochrome capabilities and no resolution loss due to a Bayer filter.
Something from the Merrill or Quattro sensor generations are most current. There are differences.
The Merrill SD1M and current production SDQ mirrorless have a removable "hot mirror" that opens the door to IR photography.
One is limited to either Sigma lenses or M42 for easy adaptation.
A single well exposed image from either of those sensors will give a 50mp 35mm format camera a run for its money.
Upvote
0
Like just about everything else, it's a trade-off....unfocused said:
A colour camera gives you more spectral information, but at the cost of sensitivity...
An interesting twist on things is a monochrome sensor and the use of external filters... This way you get to choose what is important to you. B+W photographers have been using coloured lens filters for a long long time.... Astrophotographers do the same, but with very specialized filters....
Upvote
0
danski0224 said:
Except that the Foevon sensors, including the one in the Quattro, have a poor reputation for noise.
Upvote
0
neuroanatomist said:
I looked that up: http://www.dspguide.com/ch25/1.htm
I won't pretend to understand all of it, but the first two sentences make sense.
A Foveon sensor does not have any interpolation like a Bayer sensor. This is critical along things like high contrast edges.
A Foveon sensor will resolve more detail.
A "50mp" Bayer sensor really only has either 16.6mp (50/3- RGB) or 12.5mp (50/4- RGGB) of image data, and these sensors are in 35mm format cameras. The Merrill generation is 46mp using Bayer math and it is APS-C.
Here is some more on edge artifacts: http://foveon.com/files/Color_Alias_White_Paper_FinalHiRes.pdf
Upvote
0
jolyonralph said:
If you work at low ISO, that is not necessarily true.
High ISO images, say 1600, often transfer to monochrome very well even if the color looks bad or noisy.
Yes, many will say that the best place to be is 50, 100 or 200 ISO. This is really no different from using a Canon 1DsIII at 100 ISO only, which seems acceptable to many
No, the Foveon sensor will not "work" as well as a Canon or Sony above 1600 ish.
But, if one cannot afford a Leica M, the little Sigma DP Merrill 2 or 3 can provide stunning color and monochrome images with some effort. They look like point and shoot cameras, but they aren't
The SD1M DSLR has the same sensor, but lacks any live view capabilities.Some strip the Quattro blue layer using Raw Digger and get monochrome that way, the top layer is 19mp. The Merrill sensor and earlier is 1:1:1 RGB.
The SDQ with the 30mm lens kit is a killer deal for what it can do with some effort.
Upvote
0
danski0224 said:
First off, I was wrong – I somehow thought the Sigma Foveons were FF, but given that they're APS-C I would say you're right that they rival a 50 MP FF Bayer sensor for spatial resolution. But not for overall image quality.
Color interpolation does slightly reduce spatial resolution, but no where near the factor of 3-4 as you suggest. Rather, it's about a factor of 1.2-1.5, the former for typical images and the latter for high contrast monochromatic features. I know this based on empirical testing with microscopic resolution targets, colored beads and fluorescent particles (quantum dots) of known size, using the Zeiss AxioCam HRc that I winkingly mentioned above. The color version of the camera has a Bayer array and can do color interpolation, but it also can do co-site sampling, which is where a piezoelectric motor moves the sensor exactly one pixel distance, first in the X then in the Y. That way, each pixel area of the sample is recorded successively in all three color channels (only works with static subjects, obviously, but that's the norm in microscopy).
Your other logical error is having your cake and eating it, too. You can't both divide the MP count of the Bayer sensor (by your too-large factor of 3), and at the same time multiply the MP count of the Foveon. The Merrill is a 15 MP APS-C sensor. That determines the maximal spatial resolution, Sigma calling it a 46 MP sensor is marketing-speak. Comparing to 24 MP APS-C sensor with a CFA, interpolation will drop the effective MP count to 16-20 MP. Pretty much in the ballpark. Comparing to a 50 MP FF Bayer sensor, the larger pixels would result in a lower absolute spatial resolution (effectively ~13-16 MP count), but again in the ballpark as far as equivalent MP count.
However, when looking at overall IQ and comparing at a fixed output size (which is how you should compare different sensors), the greater enlargement needed for the Foveon APS-C sensor would cost you sharpness, and the larger light-gathering area of the FF sensor would mean lower image noise. So for IQ, the Foveon will rival a 20-24 MP APS-C sensor, but will fall well short of a 50 MP FF sensor.
Upvote
0
