exmor VS Effio E

Show and tell for your homebrew nv projects. DIY 'Stickies' here too.
User avatar
some bloke
TRADER MEMBER
Posts: 7541
Joined: 27 Jan 2012, 16:14
Location: Cornwall or Leicester

exmor VS Effio E

Post by some bloke » 06 Jan 2015, 01:08

My bro has spotted some 12000 TVL 3MP sony Exmor camera's:

http://www.ebay.co.uk/itm/171535870103

I know this SOUNDS better than the 700 TVL we use but whats really what with it?
Image

koiman97
Posts: 291
Joined: 23 Oct 2012, 21:13
Location: Oldham manchester

Re: exmor VS Effio E

Post by koiman97 » 06 Jan 2015, 01:52

Hi i have just pulled one apart and put it in a black box see how it preforms it dose gwt hot on 12v volt and cuts out so going to reg tthw voltaged to 9v let u how i get on.good clear pic but havent got a 16mm decent lens for it yet
chris

User avatar
MrNewbie
Posts: 298
Joined: 22 Dec 2012, 21:28
Location: Grafton, Ohio USA

Re: exmor VS Effio E

Post by MrNewbie » 06 Jan 2015, 04:49

That is a CMOS camera...

I have always thought that there was some rule somewhere that stated..Cmos cameras stink for NV use.
Please koiman97, post back your results. We all like the idea of 1200tvl.

The best I could get was 50-60 yards with the 2 or 3 cmos cameras I have played with.
I hope you get better results and find the holly grail of cameras..

User avatar
some bloke
TRADER MEMBER
Posts: 7541
Joined: 27 Jan 2012, 16:14
Location: Cornwall or Leicester

Re: exmor VS Effio E

Post by some bloke » 06 Jan 2015, 06:49

MrNewbie wrote:That is a CMOS camera...

I have always thought that there was some rule somewhere that stated..Cmos cameras stink for NV use.
Please koiman97, post back your results. We all like the idea of 1200tvl.

The best I could get was 50-60 yards with the 2 or 3 cmos cameras I have played with.
I hope you get better results and find the holly grail of cameras..
Well yea but no but yea but:

The new Photon is CMOS and that is so seriously sensitive to low light that we need to block a load of light intake out most times with an iris. I think I read somthing about small pixels taking in less light - but maybe the Photon has big pixels.

I'm sure Bruce will be along soon.
Image

terry1001
Posts: 981
Joined: 27 Aug 2012, 09:30
Location: Ipswich, Suffolk

Re: exmor VS Effio E

Post by terry1001 » 06 Jan 2015, 20:54

As far as I recall the issue with early cmos cameras was that they used more power than a similar ccd one. Many of the best cameras now available (DSLRs etc) use cmos sensors. Currently cmos cctv cameras do seem to offer higher resolutions than ccd which seem to be stuck at about 700tvl.
Generally people don't seem to be looking at cmos units but there should be some suitable ones out there although quite a few are a bit too specialised and use hdmi or ethernet connections rather than composite video.

phoenix
BRUCE ALMIGHTY
Posts: 4865
Joined: 09 May 2012, 14:40
Location: Aberdeen

Re: exmor VS Effio E

Post by phoenix » 06 Jan 2015, 21:02

Dave,
Sorry it's taken me so long :D
There are many statements made in that advertisement
Here's my take on some of them:
960H video output- this is a relatively recently introduced analogue video recording resolution and at present is the highest recording resolution available for analogue video signals. 960H recordings have a resolution of 976x582 pixels
This is actually only the wide screen version of the existing D1 recording resolution of 720x576 pixels, but does mean that the image will be displayed in it's correct proportions when displayed on a wide screen display

1200TVL - The camera can do 1200TVL, but the 960H output signal can only do 750TVL so I don't think you'll see much difference there.
And if you want to record at 960H resolution, you'll need a 960H DVR - and there aren't many of those around yet
E700 cameras have a sensor with 976x582 pixels (that's why Sony call them 960H sensors), but the camera is not set up to produce a 960H output signal. The E700 outputs a D1 signal (720x582 pixels) and that is 540TVL

Sony Exmor IMX238LQJ sensor - this is a 1.3 megapixel 1/3 CMOS sensor which they claim (correctly) is twice a sensitive as the previous generation of Exmor sensors. it has a sensitivity of 1344mv at F5.6
However, the ExView HAD CCD sensor in the E700 (ICX673AKA) has a sensitivity of 2450mv at F5.6, so the Exmor is much less sensitive - which is not surprising since the pixels in the Exmor sensor have only half the surface area of the pixels in the ExView sensor

3 megapixel lens - they make M12 lenses by the million and then grade them according to how close they are to the exact dimensions they should be. These lenses produce a circle of light very slightly larger than the diagonal dimension of the sensor (6mm) and most of the distortion produced by a lens with less than perfect dimensions will occur at the edges of the image. Megapixel lenses will be closer to the exact design dimensions of the lens and accordingly will produce less distortion at the edges. Edge distortion is supposedly more objectionable in an HD image.

I really don't think the camera would cut the mustard in a scopeless build without chucking a load of IR at it, and you would need a screen with higher resolution than the existing 800x480 jobbies to see the improvement in the camera image

Cheers

Bruce
LAND ROVER - THE WORLD'S WORST 4X4 BY FAR

User avatar
some bloke
TRADER MEMBER
Posts: 7541
Joined: 27 Jan 2012, 16:14
Location: Cornwall or Leicester

Re: exmor VS Effio E

Post by some bloke » 07 Jan 2015, 00:36

Thank you Bruce - I just knew I'd get an in depth comparison - and then some from you me old beauty. :thumbup:

I'll copy and paste that into an email to my bro thanks.
Image

User avatar
MrNewbie
Posts: 298
Joined: 22 Dec 2012, 21:28
Location: Grafton, Ohio USA

Re: exmor VS Effio E

Post by MrNewbie » 07 Jan 2015, 05:20

phoenix wrote:Dave,
Sorry it's taken me so long :D
There are many statements made in that advertisement
Here's my take on some of them:
960H video output- this is a relatively recently introduced analogue video recording resolution and at present is the highest recording resolution available for analogue video signals. 960H recordings have a resolution of 976x582 pixels
This is actually only the wide screen version of the existing D1 recording resolution of 720x576 pixels, but does mean that the image will be displayed in it's correct proportions when displayed on a wide screen display

1200TVL - The camera can do 1200TVL, but the 960H output signal can only do 750TVL so I don't think you'll see much difference there.
And if you want to record at 960H resolution, you'll need a 960H DVR - and there aren't many of those around yet
E700 cameras have a sensor with 976x582 pixels (that's why Sony call them 960H sensors), but the camera is not set up to produce a 960H output signal. The E700 outputs a D1 signal (720x582 pixels) and that is 540TVL

Sony Exmor IMX238LQJ sensor - this is a 1.3 megapixel 1/3 CMOS sensor which they claim (correctly) is twice a sensitive as the previous generation of Exmor sensors. it has a sensitivity of 1344mv at F5.6
However, the ExView HAD CCD sensor in the E700 (ICX673AKA) has a sensitivity of 2450mv at F5.6, so the Exmor is much less sensitive - which is not surprising since the pixels in the Exmor sensor have only half the surface area of the pixels in the ExView sensor

3 megapixel lens - they make M12 lenses by the million and then grade them according to how close they are to the exact dimensions they should be. These lenses produce a circle of light very slightly larger than the diagonal dimension of the sensor (6mm) and most of the distortion produced by a lens with less than perfect dimensions will occur at the edges of the image. Megapixel lenses will be closer to the exact design dimensions of the lens and accordingly will produce less distortion at the edges. Edge distortion is supposedly more objectionable in an HD image.

I really don't think the camera would cut the mustard in a scopeless build without chucking a load of IR at it, and you would need a screen with higher resolution than the existing 800x480 jobbies to see the improvement in the camera image

Cheers

Bruce
You obviously know a bit about these cameras, sensors and their chipsets. Frankly more than I want to, that's the point! You agree that cmos stink. Yet ATN-X and I guess the Photon both use CMOS cameras to get decent results.

What do they know that we do not?
How are they doing it?

phoenix
BRUCE ALMIGHTY
Posts: 4865
Joined: 09 May 2012, 14:40
Location: Aberdeen

Re: exmor VS Effio E

Post by phoenix » 07 Jan 2015, 06:13

I'm not saying that CMOS stinks - it's a lot better relative to CCD than it used to be, and if you want HD sensors, it's much cheaper to make them with CMOS than with CCD.
Also, for daytime use when there's plenty of light around, you wouldn't know the difference between a CMOS and CCD sensor of the same size and same number of pixels.
For our NV applications with low visible light or illumination from low power IR sources, the the best CCD sensors (ExView HAD etc) are still significantly better than CMOS.
Remember that the sensor market is not driven by what the NV community wants, but rather by the much larger cctv market - and that market wants HD images.
HD images mean CMOS sensors so we'll see more and more cameras with CMOS sensors being used in NV applications whether we want them or not.
If we use HD CMOS sensors then it will mean we will need to use more IR to illuminate the target.
In the case of the Photon, the CMOS sensor in that camera is not even close to HD (0.3MP) so it's pixels are large and therefore much more sensitive to light than the pixels in an HD camera with the same size sensor (1/3) but which has 1.3Mp

Cheers

Bruce
LAND ROVER - THE WORLD'S WORST 4X4 BY FAR

User avatar
Torchwood
Posts: 1601
Joined: 04 Oct 2012, 23:25
Location: Barnet, london

Re: exmor VS Effio E

Post by Torchwood » 07 Jan 2015, 08:07

Xsight uses a separate processors to process the hd signals, it's a digital signal. Like a mini computer, that's how you get gps, wifi, compass and so on. Very much like a mobile phone, we can not use the phone as a screen because it's digital. We use analog signals without any processing. Until the day that we are able to have a processor build into our units, we have to do we analog signals, which is not a bad thing at all, as Bruce said it quite nicely.


Toni
Image

Post Reply