Sensor size table

Baz,

Please do join the other discussion thread.

This is an often misunderstood topic worthy a better exercise instead of stuck in a different subject.

Thanks!

ccs_hello
 
Hi,
I used to be a mathematician but I'm better now!

Therefore my only contribution to this interesting and useful discussion is somewhat less than academic.

One of the best "standard" lens ever made for a 35mm camera was the Pentax 43 mm f1.7 (Limited Edition). Its focal length was equal to the negative diagonal.

It is still sold by Pentax and is now classed as a short tele in their range of three fix focus compact lens for their DSLRs. In the days when I used this lens on a late model Pentax LX I found it a great "normal" lens. On moving to a Nikon F90X for which there was no 43mm lens I adopted the Nikon 35mm F2.0 as my normal or usual lens.

In the days I used these lenses I often made 20" X 16"prints and mounted them on even larger mounts (aspect ratio 1.25) even though the 35mm negative's aspect ratio was 1.5 and found that 35mm was even more useful than 43mm.

I'm sure you can feel this coming! - I now use an Olympus 4/3 DSLR with a recently acquired 12mm to 60mm zoom (24 to 120mm in conventionally equivalent 35mm) The image chip has an aspect ratio of 1.25 as do most of my best prints.

When reviewing the focal length used with my new pride and joy over the last 3 months I find that the most used focal length is 15mm to 25mm (30 to 50 mm in conventionally equivalent 35mm format). Perhaps I'm just old fashioned - No!

Anyway, I was told that we adopted 50mm as the 35mm format lens standard because this was the most highly developed movie film lens focal length in 1900 which was 1/2 frame format - almost APS-C.

Could it be that a "normal or standard" lens is more related to the aspect ratio of the prints we like to make or the images we like to view than to the diagonal of the image chip we use or the negative we used to use?
 
  • 4/3rd is 17.3x13mm (check the specs sheet on any Oly site they always say this), the 18x13.5 dates back to pre-DSLR era (the same way as APS used to be 24x16 in film era).
  • Canon's APS-C has had many considerable variation within the 1.6x form factor, the original EOS DRebel + 10D had a sensor of 22.7x15.1mm, that's the largest 1.6x Canon sensor AFAIK, the smallest is what you mentioned in the table 22.2x14.8, with models like 20D & 30D having intermediate sizes (22.5x...).
--
Xavo

image is the only validation and most of their life isn't real.
modified from Sam Sparro - Black and Gold.
 
I found the table helpful and depressing both at the same time, even though there were errors. The overall picture was pretty much accurate. I don't understand why Canon doesn't boot there XXXD line up to a 1.3 sensor (I'm thinking cropping and enlarging here). Some soul will say it is because it is expensive but the price would drop if more 1.3 sensors were used; there is a pile of XXXD's sold, but not so many Mark l's with the 1.3 size.

Obviously there isn't much a difference between the Nikon sensor or the Canon sensor. We would definitely have heard the Nikon guys lording it over the Canon guys that there sensor was better if the slightly larger size made any difference.
--
Rationally I have no hope, irrationally I believe in miracles.
Joni Mitchell
 
Yes, I've moved on to a DSLR...But a hot prosumer for under $800 would be my forst choice for travel photography.

Still waiting for a 2/3" prosumer... that would update the caliber camera that the Konica Minolta A2 was.Too bad a 2/3 sized sensor couldn't be left at 8mp and modernized for noise and autofucus speed.
 
  • 4/3rd is 17.3x13mm (check the specs sheet on any Oly site they
always say this), the 18x13.5 dates back to pre-DSLR era (the same
way as APS used to be 24x16 in film era).
from dpreview sensor sizes
http://www.dpreview.com/learn/?/Glossary/Camera_System/sensor_sizes_01.htm
4/3" 18.000 13.500
and



Olympus quote image area size on specification sheets not the sensor size. We think now that this is b/se the sensor is normalised to accept the different frame ratios of mFT, which allows 3x2, 4x3 and 9x16 framing. That would explain why E3 sensor which uses 10.1Mp, has 11.8Mp on the chip.

The nominal size of the chip then (which is what everyone else uses) is 18x13.5mm

--
Riley

in my home, the smoke alarm is the dinner bell (just)
 
  • 4/3rd is 17.3x13mm (check the specs sheet on any Oly site they
always say this), the 18x13.5 dates back to pre-DSLR era (the same
way as APS used to be 24x16 in film era).
from dpreview sensor sizes
http://www.dpreview.com/learn/?/Glossary/Camera_System/sensor_sizes_01.htm
4/3" 18.000 13.500
and

as u see here olympus themselves compare 18x13.5 to 24x16, and in their specs sheet they CONSISTENTLY give 17.3 x13
Olympus quote image area size on specification sheets not the sensor
size. We think now that this is b/se the sensor is normalised to
accept the different frame ratios of mFT, which allows 3x2, 4x3 and
9x16 framing. That would explain why E3 sensor which uses 10.1Mp, has
11.8Mp on the chip.
this is laughable and i am sure u yourself don't believe in it, when the 4/3 standard was defined in 2003, panasonic's idea of cameras that shoots multiple aspect ratios was not yet there or at least not envisaged for 4/3rd dslrs.

as for the 11.8MP, that's the total MP count! ALL cameras have total MP count besides the well-known effective MP count because u need "invisible" peripheral pixels outside the image in order to make accurate calculations and PP for the visible peripheral pixels on the edge of the image (Bayer array interpolation would be off at the edge pixels otherwise).
The nominal size of the chip then (which is what everyone else uses)
is 18x13.5mm
Canon use the term "image area" in their specs sheet and give after that the well-known size

so againg for everybody, the correct comparisons are:

17.3x13mm vs 22.2x14.8mm vs 23.6x15.8mm (effective sensor size of 4/3rd vs Canon APS-C vs Nikon APS-C respectively)

or

18x13.5mm vs 24x16mm (original 4/3rds [of video cameras pre-Oly's 4/3rd] vs nominal APS of film era 24x16mm).

i'm absolutely sure of what am saying so won't follow up on it if u reply
with your usual deceptive zigzagging arguments.

--
Xavo

image is the only validation and most of their life isn't real.
modified from Sam Sparro - Black and Gold.
 
sounds like you are not aware of the fuji s100 introduced last PMA (2/3" 11mp sensor). there are also countless 1/1.6", 1/1.7" cameras which are just a tiny bit smaller than 2/3" (=1/1.5").

http://www.dpreview.com/reviews/specs/Fujifilm/fujifilm_S100FS.asp
Yes, I've moved on to a DSLR...But a hot prosumer for under $800
would be my forst choice for travel photography.

Still waiting for a 2/3" prosumer... that would update the caliber
camera that the Konica Minolta A2 was.Too bad a 2/3 sized sensor
couldn't be left at 8mp and modernized for noise and autofucus speed.
--
Xavo

image is the only validation and most of their life isn't real.
modified from Sam Sparro - Black and Gold.
 
  • 4/3rd is 17.3x13mm (check the specs sheet on any Oly site they
always say this), the 18x13.5 dates back to pre-DSLR era (the same
way as APS used to be 24x16 in film era).
from dpreview sensor sizes
http://www.dpreview.com/learn/?/Glossary/Camera_System/sensor_sizes_01.htm
4/3" 18.000 13.500
and

as u see here olympus themselves compare 18x13.5 to 24x16, and in
their specs sheet they CONSISTENTLY give 17.3 x13
the diagram above comes from Olympus
http://www.olympus-esystem.com/dea/technology/four-thirds/sensor-size.html
the diagram discusses sensor size
the spec sheets for cameras discuss image area
Olympus quote image area size on specification sheets not the sensor
size. We think now that this is b/se the sensor is normalised to
accept the different frame ratios of mFT, which allows 3x2, 4x3 and
9x16 framing. That would explain why E3 sensor which uses 10.1Mp, has
11.8Mp on the chip.
this is laughable and i am sure u yourself don't believe in it, when
the 4/3 standard was defined in 2003, panasonic's idea of cameras
that shoots multiple aspect ratios was not yet there or at least not
envisaged for 4/3rd dslrs.

as for the 11.8MP, that's the total MP count! ALL cameras have total
MP count besides the well-known effective MP count because u need
"invisible" peripheral pixels outside the image in order to make
they dont have an Mp count a full 16% more than the sensor
that is more than required for dark register etc
accurate calculations and PP for the visible peripheral pixels on the
edge of the image (Bayer array interpolation would be off at the edge
pixels otherwise).
The nominal size of the chip then (which is what everyone else uses)
is 18x13.5mm
--
Riley

in my home, the smoke alarm is the dinner bell (just)
 
This is version 1.6.

The data are mostly based on the active photosites area.

When the sensors grow larger, the size definition becomes more course, unlike type 1/2.3", type 1/2.4", or type 1/2.5". Sensors within the same category get grouped together.



ccs_hello
 
Yep.

For example,
Finepix f40fd is using a type 1/1.6".
Finepix f31fd, Canon G9 are using type 1/1.7".
Canon A610, A620, G5, G6, Coolpix 5200, ... are using type 1/1.8".

ccs_hello
 
  • the APS-H as in the 1D mk III is 28.1x18.7mm (diagonal 33.75mm, area 525.47mm²). you have 28.7 instead of 28.1 in the table with the diagonal and area calculated based on that error.
  • olympus use 17.3x13mm in their specs sheets just as other manufacturers use 22.2x14.8 or 23.6x15.8 in their specs sheets and they (oly) mention 18x13.5 only in the context of talking generally about the 4/3 standard and giving a rough idea about its size compared to other formats. they compare 18x13.5 to the approximate but now obsolete size of 24x16mm of APS-C, so as i said before i think that 18x13.5 is meaningless in the context of this table which lists the imaging area of each sensor. 17.5x13.2 or so is a size mentioned in Oly E-1 literature (not the specs sheet) and it refers to an area intermediate between the total pixels area and imaging pixels area, these are not the dimensions that should fairly be compared to Canon's 22.x14.8 for example (which is the imaging/effective area of the sensor).
in your place i would keep things simple as they were in the beginning, 1 main size of each type and more details in footnotes if necessary. for example instead of 2 Canon APS-C sizes, i'd include just one (22.2x14.8) which is what most recent models are + a footnote that some (older) models were slightly larger. I would omit film sizes from the table. too many sizes unnecessarily crowd the table making it harder to follow IMO.
This is version 1.6.

The data are mostly based on the active photosites area.

When the sensors grow larger, the size definition becomes more
course, unlike type 1/2.3", type 1/2.4", or type 1/2.5". Sensors
within the same category get grouped together.



ccs_hello
--
Xavo

image is the only validation and most of their life isn't real.
modified from Sam Sparro - Black and Gold.
 
  • the APS-H as in the 1D mk III is 28.1x18.7mm (diagonal 33.75mm,
area 525.47mm²). you have 28.7 instead of 28.1 in the table with the
diagonal and area calculated based on that error.
  • olympus use 17.3x13mm in their specs sheets just as other
manufacturers use 22.2x14.8 or 23.6x15.8 in their specs sheets and
they (oly) mention 18x13.5 only in the context of talking generally
about the 4/3 standard and giving a rough idea about its size
compared to other formats. they compare 18x13.5 to the approximate
but now obsolete size of 24x16mm of APS-C, so as i said before i
think that 18x13.5 is meaningless in the context of this table which
lists the imaging area of each sensor. 17.5x13.2 or so is a size
mentioned in Oly E-1 literature (not the specs sheet) and it refers
this is where art departs reality

the kodak sensor for E1 was KAF5101 and the pdf from kodak is still around as kaf5101celongspec.pdf and available here
http://www.dta.it/support/data_sheets/kodak/kaf-5101ce.pdf
verify here
http://www.kodak.com/US/en/corp/pressReleases/pr20030624-01.shtml
this sensor is defined in size as 19.7 x 15.04mm

the E300 sensor is KAF8300, and its document is kaf-8300longspec.pdf, if you want it i can have it sent to you. You may still find it on the net. It is also 19.7x15.04mm; yet the nominal size is still 18x13.5mm. Further to this, the fov between E1 and later models like E3 is identical. Conclude the imager size is the same
to an area intermediate between the total pixels area and imaging
pixels area, these are not the dimensions that should fairly be
compared to Canon's 22.x14.8 for example (which is the
imaging/effective area of the sensor).
yet it states 'sensor size', it is a nominal value just the same

--
Riley

in my home, the smoke alarm is the dinner bell (just)
 
note that the Olympus E-1's sensor has 3 confusingly named areas:

1) Total pixels area 18x13.5mm (that's normal just as all cameras have total pixel counts and total pixel area). pixels on the edges of this area help in processing the image but dont appear in the final image

2) Active pixels area 17.4x13.1 or 17.5x13.2mm (and here lies the confusion, people confuse "active" with "effective" which in this case are not the same, other cameras normally do not have a comparable spec to this area). pixels on the edges of this area are of more importance than those of the previous total area put still don't make it to the final image. they are called dark pixels or something like that.

3) Imaging pixels area 17.3x13mm (this is comparable to what you find in other camera's specs sheets). the final image that you get from the camera whether it be E1 or any other 4/3rd camera was/is captured on this 17.3x13 area, thats what you'll always find in the specs sheets:

http://www.olympus.co.jp/en/support/imsg/digicamera/download/manual/esystem/man_e1_en.pdf (check page 42 of the PDF: "Screen size: 17.3 mm (H) x 13.0 mm (V) (0.9" x 0.5")").

http://www.olympusamerica.com/e3/specifications.asp
"Image sensor size 17.3 mm (H) x 13.0 mm (V)"

http://www.olympusamerica.com/cpg_section/product.asp?product=1386&fl=4
"Sensor Size 17.3 mm (H) x 13.0 mm (V)"

--
Xavo

image is the only validation and most of their life isn't real.
modified from Sam Sparro - Black and Gold.
 
the E300 sensor is KAF8300, and its document is kaf-8300longspec.pdf,
if you want it i can have it sent to you. You may still find it onthe net.
Still here:

http://www.kodak.com/ezpres/business/ccd/global/plugins/acrobat/en/datasheet/fullframe/KAF-8300LongSpec.pdf
It is also 19.7x15.04mm; yet the nominal size is still18x13.5mm.
Data sheet:
17.96 x 13.52 mm active area; 22.5 mm diagonal
Active pixels: 3326 x 2504 @ 5.4 micron

'Total' and 'effective' pixel counts are higher, and occupy a larger area, but 18x13.5 is the imaging area, as far as I can tell.
Further to this, the fov between E1 and later models like
E3 is identical. Conclude the imager size is the same
HTH
--
Alan Robinson
 
note that the Olympus E-1's sensor has 3 confusingly named areas:
theres nothing confusing in it
1) Total pixels area 18x13.5mm (that's normal just as all cameras
have total pixel counts and total pixel area). pixels on the edges
he means nominal, like for every other sensor
of this area help in processing the image but dont appear in the
final image
dark register lay outside, the total imager size carries those samples
2) Active pixels area 17.4x13.1 or 17.5x13.2mm (and here lies the
confusion, people confuse "active" with "effective" which in this
case are not the same, other cameras normally do not have a
comparable spec to this area). pixels on the edges of this area are
of more importance than those of the previous total area put still
don't make it to the final image. they are called dark pixels or
something like that.

3) Imaging pixels area 17.3x13mm (this is comparable to what you find
in other camera's specs sheets). the final image that you get from
the camera whether it be E1 or any other 4/3rd camera was/is captured
on this 17.3x13 area, thats what you'll always find in the specs
sheets:

http://www.olympus.co.jp/en/support/imsg/digicamera/download/manual/esystem/man_e1_en.pdf (check page 42 of the PDF: "Screen size: 17.3 mm (H) x 13.0 mm (V) (0.9" x 0.5")").

http://www.olympusamerica.com/e3/specifications.asp
"Image sensor size 17.3 mm (H) x 13.0 mm (V)"

http://www.olympusamerica.com/cpg_section/product.asp?product=1386&fl=4
"Sensor Size 17.3 mm (H) x 13.0 mm (V)"

--
Xavo

image is the only validation and most of their life isn't real.
modified from Sam Sparro - Black and Gold.
--
Riley

in my home, the smoke alarm is the dinner bell (just)
 
the E300 sensor is KAF8300, and its document is kaf-8300longspec.pdf,
if you want it i can have it sent to you. You may still find it onthe net.
Still here:

http://www.kodak.com/ezpres/business/ccd/global/plugins/acrobat/en/datasheet/fullframe/KAF-8300LongSpec.pdf
It is also 19.7x15.04mm; yet the nominal size is still18x13.5mm.
Data sheet:
17.96 x 13.52 mm active area; 22.5 mm diagonal
this is it here from kaf-8300longspec.pdf




Active pixels: 3326 x 2504 @ 5.4 micron
'Total' and 'effective' pixel counts are higher, and occupy a larger
area, but 18x13.5 is the imaging area, as far as I can tell.
yup, you gotta draw the line somewhere, but as you can see, a lot of pixels lay outside of the 18x13.5mm nominal size, thats the same for all sensors
Further to this, the fov between E1 and later models like
E3 is identical. Conclude the imager size is the same
HTH
--
Alan Robinson
--
Riley

in my home, the smoke alarm is the dinner bell (just)
 

Keyboard shortcuts

Back
Top