«

David J Taylor wrote:
> Blinky the Shark wrote:
> []
>> Thanks, David. When I read someone say it was much larger (than,
>> IIRC, the 1/1.6" sensor, I thought I may have misread the 2/3"
>> sensor's size somewhere. It would be nice if there was only one
>> method for designating sensor sizes, and it was more intuitive than
>> x/y method. Yes, I know where that system originated; I'm a TV
>> camera operator and used to use the old tube cameras before CCDs.
>
> Yes, it is indeed confusing. In the days of 2/3-inch and 1-inch vidicons
> and 4/3-inch (or 30mm) plumbicons, I was quite happy with the designation,
> but it's inappropriate for digital cameras. I suspect that someone in
> marketing may like the apparently bigger number of 1/2.5 compared to 1/1.7
> which they think that Joe Public may believe. After all, a bigger number
> is better, right? <G>
>
> One other confusion, by the way, is the aspect ratio. Should you measure
> sensors by horizontal, vertical or diagonal? Some cameras offer 16:9 with
> a 16:9 sensor, others by cropping a 4:3 sensor. And whilst most DSLRs are
> 3:2 aspect ratio, the 4/3 system is 4:3! Is it sensible just to use the
> diagonal?

The ratio to 1 is much more easily compared (but rarely stated):

****
** diag
16:9 = 1.77 : 1 high def widescreen video 1920x1080 2.1 MP 2202
1.6 : 1 widescreen computer monitor 1920x1200 2.3 MP 2264
3:2 = 1.5 : 1 SLR landscape 1870x1246 2.3 MP 2247
4:3 = 1.33 : 1 compact digicam / TV 1750x1312 2.3 MP 2187

Megapixels and diagonal correspond pretty well so megapixels are almost
as useful of a number as diagonal. Pixel pitch is the clearest way to
state the sensor size and no manufacturer will give you that. What they
give you is 35mm equivalent focal length and the exif gives you actual
focal length so some math involved from there. Equivalent 35mm focal
length is based on the diagonal when mixing aspect ratios:
http://www.clarkvision.com/imagedeta...ity_gain_1.gif
the horizontal axis is pixel pitch in microns for various cameras. Pixel
pitch determines the range of useful f/stops and corresponds closely to
dynamic range though there is variation in different designs and that is
shown by ISO at unity gain, the real tested vertical location of the
camera models in that chart versus the hypothetical yellow curve (if I'm
understanding correctly). For choosing a camera, a more useful chart
might show megapixels against ISO at unity gain.

> 3-tube 4.5-inch Image Orthicon cameras, anyone?
>
> Cheers,
> David
>
>


--
Paul Furman
www.edgehill.net
www.baynatives.com

all google groups messages filtered due to spam

Ï "David J Taylor"
<david-> Ýãñáøå óôï
ìÞíõìá news:81c5k.9897$ ...
> Blinky the Shark wrote:
>> David J Taylor wrote:
>>
>>> Blinky the Shark wrote:
>>> []
>>>> We were using plumbicons when I got into the biz in 1977. But
>>>> certainly I was exposed to IO tales.
>>>
>>> Makes you think how far we've come since then! I think I built my
>>> first vidicon camera before 1975....
>>
>> You *built* one?
>
> Yes, and designed some of the circuits as well. It was black and white,
> by the way. Quite a few people built cameras in those days. My metal
> work was never up to much, so the lens mounting left a lot to be desired.
> I suspect that it was actually built between 1968 and 1971.
>
.....And even in the Back to the future movie, the hero has the first
camcorder that reached the market, which was a JVC with a video tube, and
VHS-C. Just after that, the CCD came out.



--
Tzortzakakis Dimitrios
major in electrical engineering
mechanized infantry reservist
hordad AT otenet DOT gr

Allen wrote:
>
> ASAAR wrote:
> > On Sat, 14 Jun 2008 00:49:13 -0700, Blinky the Shark wrote:
> >
> >>> Canon's SD950 IS has a 1/1.7" sensor. Fuji's F50fd and F100fd
> >>> have 1/1.6" sensors. The much larger S100FS has a 2/3" sensor. All
> >> I thought the 2/3" sensor was only about 8.8mm x 6mm.
> >>
> >> If not, what are its actual dimensions?
> >
> > Those are the actual dimensions (approx.), but the sizes given by
> > manufacturers (2/3" in this case) can be misleading, since they
> > don't refer to any specific sensor measurement, but to the diameter
> > of a hypothetical glass camera tube neck that the sensor might very
> > loosely fit within. For the gory details see :
> >
> >
> > http://www.dpreview.com/learn/?/Glos...r_sizes_01.htm
> >
> Thank you, ASAAR! I'm really glad tho see that chart. Why they can't
> just say "sensor size x mm by y mm" is either above or below my level of
> comprehension.
> Allen


Hello, Allen:

That's because you're not a marketeer! It's hardly in the manufacturers'
best interests, to be entirely open and honest with the public.

This obviously applies to image sensor sizes, especially. :-J


Cordially,
John Turco <>