Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Oct 9, 2017 22:06:58 GMT
Hi folks,
I've recently purchased an IMX224 camera. Something odd has only just dawned upon me with the quoted sensor size. It states that the diagonal size is 1/3", and then in brackets it says (6mm). Now, 6.35mm is 1/4", and 8.5mm is 1/3".
So what is the story here??? What have I missed?
I'm noting similar size misrepresentations with other sensor specs.
Alex.
|
|
|
Post by davy on Oct 9, 2017 22:45:58 GMT
Lol,, I dunno,,, maybe just average it out, Or they all do fishing as a hobby,, know what there like at exaggerating how big the fish they catch 😂
|
|
|
Post by ChrisV on Oct 9, 2017 23:05:39 GMT
I think it all goes back to the good old days of valves and tubes for TV cameras where they had rough classes of sizes 1/3, 1/2 2/3" etc. So they just lump them into the nearest class There's a whole bunch of sensors that are called 1/3, eg the 224 and 290 which actually are quite different in size. The quoted size for the 224 is 6.09cm diagonal. www.sony-semicon.co.jp/products_en/IS/sensor0/img/product/cmos/IMX224_225LQR_Flyer.pdfGet your ruler out Alex. Mind you it does make for a smaller FOV.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Oct 9, 2017 23:27:29 GMT
Good God, valves and tubes times??? Even my little analogue LN300 camera was correctly labelled at 1/4". I was never looking at the imperial measurements, only the metric. Thanks Chris. But it really seems to me that someone is just taking the piss out of everyone else. Imperial rulers do measure to a finer scale than 1/3"... How can such misleading nomenclature still exist, for flaming digital technology? REALLY! Do I need to peddle the dynamo to power the camera too? One thing is class type, I can understand that. Another thing is quoting the ACTUAL size specs, which is a totally different thing. I have to say that even the metric dimensions quoted by different sellers for different sensors leaves A LOT to be desired. Some sensor sizes are quoted as being flaming identical when they are totally different items!!!!! Look at the sensor specs table on this link. Freaking ridiculous! bit.ly/2y7IrEVWho would have thought that I'd have to do SO MUCH digging to get to the true specs of ANYTHING. Does not inspire confidence, does it... Odd. Very odd.
|
|
|
Post by howie1 on Oct 10, 2017 0:40:35 GMT
We Humans are a strange bunch aren't we! Talk about making something crazy complex and confusing instead of doing something straight forward and simple!! I mean jeez ... ... the 1" collection tube (sensor) of early TV camera's measured 1" diagonally across (25.4mm). But on every sensor ever made right up to and including todays sensors, the pixels around the edge are not actually used to collect photons but are in fact used to gather dark current and other stuff which then gets applied to / adjusts all the signals coming from those actual photon gathering pixels - those actual photon gathering ones are called 'effective' pixels. So it turned out (way back in those early TV broadcast days) that the 4:3 ratio diagonal of 'effective pixels' on that 1" sensor tube was 16mm. So from that day on the 'Rule' was 1" = 16mm. So a 6mm sensor is converted to 'inches' using that 1"=16mm 'Rule' ... or rounded up or down to the nearest fraction of the so-called inch! IE 6mm/16mm = 0.375. And the closest inch fraction is 1/3" so they called it a 1/3 sensor! And a 4/3 sensor is one which is approx 1 and 1/3 times 16mm which is about 22mm and so on and so forth! Why the heck didn't they just call a 6mm sensor a 6mm sensor? A 22mm sensor a 22mm sensor? And lastly ... it doesnt totally change the FOV that much ... but ... if you are pedantic about these things (and judging by what I've read up on most AP'ers are because they go nuts over arc secs per pixel!), then when you use an online FOV calculator, or Stellarium FOV calc, the pixel count in horizontal and vertical which you enter should strictly be the H and V 'effective' pixels in order to be closer to the actual FOV and arcsecs/pixel, compared to entering the 'total' H and V number of pixels - which, as above, is the 'effective' pixels which gather photons PLUS the dark current/noise etc gathering pixels around the edge of the 'effective' pixels region. Gawd ... since researching bunch of stuff as now getting into AP rather than EAA, I've turned into like those CN posters who pontificates and pours out useless knowledge at the drop of a hat!! Shoot me now!
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Oct 10, 2017 1:50:08 GMT
Thanks Howie.
What a fraud! One that would make an accountant proud!
* I have a pumpkin, but blood oranges command a better price. I know! A blood orange is a plant, my pumpkin IS a plant, I'll just call my pumpkin a blood orange and charge the blood orange price! Oh joy! *
So why the heck don't they do the same when quoting in metric???
A dumb example of "it was done that way 30 years ago, so we will continue doing this today", despite technology changing things.
What a dumb, misleading situation. No doubt my rant won't change things, but it needs to be called for what it is.
|
|
|
Post by Dragon Man on Oct 10, 2017 15:15:22 GMT
|
|