What makes a webcam a good webcam?

[M00NMonkey]

Hi All,

Ok, so the clouds still aren't playing ball and are stoping me getting out and experimenting with my kit, so I thought Id do a bit more research on things. Mainly, what makes a webcam good for astrophotography?

Now, I dont really know anything about the workings of webcams, I have read threads on here and have seen results from different kit, some are better than others, but why is this?

There seems to be 2 types of chip- CCD and CMOS. And it would seem that most (dedicated) astrocams are CCD. is there a reason for this?

I have researched about this stuff but just get bogged down with too much data and in the end I still find that I dont really understand why one should be better than the other. Price can also vary dramatically. But, knowing the original cost of the SPC900 was around a fiver, and this has outstanding results, price isn't a reason. (I know they are about 50 now, but thats just lack of supply and greed).

No one seems to have found a replacement for this (spc900) webcam. Is this because of mod-ability? (im making up words know ) or down to this CMOS chip?

Basically, I dont want to have to read up on the entire life history of webcams and understand every quantum jitter of the workings of them. But I am keen to understand them on a basic level and see if there are general rules for reasons why some things work, others dont.

Hope this all makes sense!?

Clear Skies

Jez

[Cath]

I think the SPC900 just happen to there at the right time, so was the one everyone concentrated on at the time. Their are other webcams that use exactly the same sensor chip, no doubt their are webcams that have better sensor chips.

It's a bit like the beta-max/vhs, blue-ray/HD DVD wars I think.

Above all, what you want is the highest signal to noise ratio (the lower the noise level the better) and dynamic range (the ability to cope with both very dark and very bright parts of an image at the same time).

[knobby]

Should make a good thread

[knobby]

CCD chips are better at low light work... But more expensive to manufacture.

I believe that the CCD sensor and the electronics are separate on the board, which allows long exposure modding, whereas the CMOS type are kind of 'all in one ' so its a lot harder to mod ( also cheaper to manufacture) which is probably why they are now like rocking horse poo

[M00NMonkey]

Above all, what you want is the highest signal to noise ratio (the lower the noise level the better) and dynamic range (the ability to cope with both very dark and very bright parts of an image at the same time).

Hi Cath, Thanks for reply.

Is there a name for this said ratio? When I get the specs of webcams there is so much data I am not really sure what I am looking at. Im not dumb (honest ) its just I havent learnt about this yet. A naivety I wish to rectify

CCD chips are better at low light work... But more expensive to manufacture.

I believe that the CCD sensor and the electronics are separate on the board, which allows long exposure modding, whereas the CMOS type are kind of 'all in one ' so its a lot harder to mod ( also cheaper to manufacture) which is probably why they are now like rocking horse poo

Hi Knobby, yes I am hoping this will be a good thread

It did seem, from the research Iv done so far, that the CCD was the better of the two and typically are harder to find nowadays. Maybe an astro-company has brought out the supplier of the cheap webcam CCD chips and are using them in their own branded cameras, and just charging extortionate prices for, what was once, a cheap CCD chip??. This is just the conspiracy-esque part of my brain trying to butt in on my logical thinking!!

Also, I dont know if you were aware, but rocking horse poo is becoming even more scarce, as it seems Findus mistake almost anything for beef these days, and they have all been ground up into a lasagne!!

So, I digress, has anyone had any luck using a CMOS based webcam? or is it really not worth trying out a route down this road?

Thanks

Jez

[JamesF]

The SPC900 and other webcams based on the same sensor were more of a happy accident than anything else I think. Some of them happened to be easy to mod and the firmware and drivers gave sufficient control to make the conversion to astro imaging cameras feasible. Historically CMOS sensors haven't been as sensitive as CCDs and that also worked in their favour.

The consumer webcam market however seems to be heading down a path towards greater simplicity of management (ie. the camera firmware does as much of the control as possible) and high resolution HD sensors. These are often far from ideal goals from the point of astro imaging. Zoom lenses and similar features are also being added that make modding a more involved process. It's probably also a fair assumption from a webcam designers point of view that the target is going to be reasonably well lit, so low light performance perhaps isn't a priority.

People have had some success with CMOS webcams in the past, but my own experience has suggested that they perform better when you have lots of aperture. This might be personal bias, or it could be that CMOS sensors have been less sensitive.

The picture (as it were is changing however. The ASI120MM and ASI120MC cameras are CMOS as far as I'm aware and if you look at some of the images being posted here they're very impressive (ok, not mine, but I'm still learning .

So, I'd not get too hung up on CCD vs. CMOS for planetary imaging. I think it's more important to look at the other features of any potential camera, such as whether you have access to the gain/gamma/white balance/exposure time controls and what the sensitivity of the sensor is.

James

[M00NMonkey]

The SPC900 and other webcams based on the same sensor were more of a happy accident than anything else I think. Some of them happened to be easy to mod and the firmware and drivers gave sufficient control to make the conversion to astro imaging cameras feasible. Historically CMOS sensors haven't been as sensitive as CCDs and that also worked in their favour.

The consumer webcam market however seems to be heading down a path towards greater simplicity of management (ie. the camera firmware does as much of the control as possible) and high resolution HD sensors. These are often far from ideal goals from the point of astro imaging. Zoom lenses and similar features are also being added that make modding a more involved process. It's probably also a fair assumption from a webcam designers point of view that the target is going to be reasonably well lit, so low light performance perhaps isn't a priority.

People have had some success with CMOS webcams in the past, but my own experience has suggested that they perform better when you have lots of aperture. This might be personal bias, or it could be that CMOS sensors have been less sensitive.

The picture (as it were is changing however. The ASI120MM and ASI120MC cameras are CMOS as far as I'm aware and if you look at some of the images being posted here they're very impressive (ok, not mine, but I'm still learning .

So, I'd not get too hung up on CCD vs. CMOS for planetary imaging. I think it's more important to look at the other features of any potential camera, such as whether you have access to the gain/gamma/white balance/exposure time controls and what the sensitivity of the sensor is.

James

Once again you are a fountain of knowledge, James! Thanks for the info.

Well, its sounds like theres hope for the CMOS as, like you say, the pictures say it all. But, as Knobby suggested, the CMOS is more of an "ALL_IN_ONE" package which, to me, would hint at the whole unit being purpose built. (i.e the ASI120MM designed for AP). And this would make modding one not designed for a certain job, difficult. (i.e the cheap CMOS's not being fit for AP)

One thing I am a little confused over is MP (mega pixels). I was under the impression that the more MP you have the clearer the image. So a higher MP rating would make a better picture. But is it like this......A rubbish chip with low sensitivity and a high MP rating will give you a high resolution image of rubbish quality. And one with a low MP but a high sensitivity chip will give a lower resolution of a high quality image. ? (if that makes sense?)

Cheers

Jez

[JamesF]

One thing I am a little confused over is MP (mega pixels). I was under the impression that the more MP you have the clearer the image. So a higher MP rating would make a better picture. But is it like this......A rubbish chip with low sensitivity and a high MP rating will give you a high resolution image of rubbish quality. And one with a low MP but a high sensitivity chip will give a lower resolution of a high quality image. ? (if that makes sense?)

There area a number of factors at play here that don't work the same way as for terrestrial imaging/photography.

Firstly, the telescope itself has a fixed resolution which is a function of its aperture. There's no benefit in having camera pixels that are significantly smaller than the smallest thing your telescope can resolve. Added to that the fact that we're sitting here on planet Earth under a nice warm (sometimes) fluffy (always, seems to be) blanket of an atmosphere means that resolution has an absolute limit which I think is often suggested to be one arcsecond in excellent seeing, but more like two arcseconds on average. So there's probably not much point having a camera that has pixels so small that you get fifteen pixels per arcsecond if the smallest thing you can see with it is thirty pixels wide.

Also. imagine you want to image a target where some faint area corresponds to a block of four pixels on your high resolution camera, or a single pixel on a camera with pixels twice as big in each direction. If in the time you're creating one frame a single photon from that faint area arrives at your camera sensor then with the first camera the photo can only hit one of the pixels, so you end up with only one quarter of the faint area represented in your final image, whereas with the second camera that one photon will cause the entire area corresponding to that part of the target to be lit, so in a sense that lower resolution camera more accurately represents what you were imaging. Obviously there's a "fix" for this, which is to collect those photons for longer, but you can't afford to do that for too long with planetary imaging because the atmosphere will distort the image further.

High resolution cameras may also make physical trade-offs resulting in a certain percentage of photons not actually hitting any pixel but just randomly bumping into some other non-sensitive part of the electronics. Potentially that non-sensitive area increases overall as the size of the pixels decreases.

So the process becomes a balancing act between the seeing, the telescope, the pixel size and the target as well as other things I've probably forgotten to mention meaning that high resolution sensors will not always do the best job and in fact will sometimes do a worse job than lower resolution sensors.

James

[M00NMonkey]

Cheers, James. Thats helped tie together some other loose threads of knowledge I have acquired from books, too. Makes a lot of sense now.

[burger]

I am still on the lookout for a sensibly priced 900 webcam.

I now have 5 others all with CCD chips all listed on this site

I have taken them all to bits and fitted the nose piece but not attempted the long exposure mod(out of my ability)

WEB-CAMERAS

logitech 3000

creative labs NX ULTRA

2 X D-Link 350+ cannot get to work on XP think it needs win98 only

Cisco/ rebadged logitech 3000 ( cheapest £2.99 inc postage)

Philips 675

All were under £8.00 each on ebay inc postage

Not had a decent chance to try them all out but in daylight they all look pretty sharp using sharpcap.

Cracking pictures of upturned bucket in the farmers field opposite.

Everyone seems to bid on the 900's and no one on the others, now they may not be as good but for a total novice will suit me until I find a 900

[M00NMonkey]

I am still on the lookout for a sensibly priced 900 webcam.

I now have 5 others all with CCD chips all listed on this site

I have taken them all to bits and fitted the nose piece but not attempted the long exposure mod(out of my ability)

WEB-CAMERAS

logitech 3000

creative labs NX ULTRA

2 X D-Link 350+ cannot get to work on XP think it needs win98 only

Cisco/ rebadged logitech 3000 ( cheapest £2.99 inc postage)

Philips 675

All were under £8.00 each on ebay inc postage

Not had a decent chance to try them all out but in daylight they all look pretty sharp using sharpcap.

Cracking pictures of upturned bucket in the farmers field opposite.

Everyone seems to bid on the 900's and no one on the others, now they may not be as good but for a total novice will suit me until I find a 900

Hi Burger,

Yeah it would appear that the whole world and his dog are bidding on the 900's.

Thats great your testing out a variety of webcams. Id be very interested to hear of any feedback and images of any results you have with them.

Keep us posted!!

Cheers

Jez

[burger]

Hi Burger,

Yeah it would appear that the whole world and his dog are bidding on the 900's.

Thats great your testing out a variety of webcams. Id be very interested to hear of any feedback and images of any results you have with them.

Keep us posted!!

Cheers

Jez

Yeah Jez

So would I,

had the telescope since early Jan used for 20 mins so far, this must be the highest hour/hobby ratio pass time going, at this rate it will take me a year to clock up 1hr viewing time,

Only thing I have found is each camera gives different control options in sharpcap so some will be better to control than others, just limited looking at a muddy field in daylight.

Will post some images when I get some, check back june/july time

[knobby]

Just won a Logitech 4000 for £4.70... I'll be hacking that apart one night this week, I'll do an asda cam vs logicam test CMOS vs CCD

[M00NMonkey]

Be good to see the results of those, Knobby.

I have finally got a working sharpcap (a long story all I can say is vmware/bootcamp) and hopefully will get chance to get a decent run on the xbox cam. Though my priority will be to get some subs of m42/m31, it looks like the clouds may hold of til midnight so fingers crossed.

I will post any results later this week.

[knobby]

Hopefully it will turn up tomorrow... So I can start hacking it to peices

... Carefully of course

[knobby]

Its here... Screw drivers at the ready!

Anyone know what settings would work for this one in sharp cap.

Also is it possible to find out if it has the Sony or sharp CCD... Apparently they can have either?

[M00NMonkey]

Its here... Screw drivers at the ready!

Anyone know what settings would work for this one in sharp cap.

Also is it possible to find out if it has the Sony or sharp CCD... Apparently they can have either?

No replies yet Knobby? Sorry I am unable to help with settings or CCD info.

Have you had any luck yet? clouds given way?

[knobby]

Its stripped and cleaned... CCD and ir filter were pretty grubby.

The 4000 pro has either the Sony CCD (spc900 type) or the sharp CCD, on the later models... Sadly mine has the sharp one. But still its a CCD so will hopefully be better than my Asdacam.

Its still in bits waiting for some epoxy to set before I put it back together.

Got it working on Win7 too which was a bit of a faff !

just need a cloud blower now

[M00NMonkey]

Ah, the infamous cloud blower! I am on the look out for one of them myself I swear its more cloudy ever since I brought the scope but its all relative I guess.

Good to hear everything works and nearly ready.

Clear Skies

[knobby]

by the way, the only way to see if its a sony or sharp is with a magnifying glass ! the Sharp has nicely spaced straight soldering, whereas the Sony has very untidy looking soldering,

found courtesy of this Cloudynights discussion

http://www.cloudynights.com/ubbthreads/showflat.php/Number/5255403

Its all stripped and rehomed in a nice little Maplin project box ready to go !

just give me a gap in the clouds this weekend PLEASE

[M00NMonkey]

Clouds not playing ball eh, Knobby! maybe get the test done Next weekend?!?

Regarding the sony CCD chip.....I am still not sure exactly what I am looking for regarding specs when looking at all the webcams. But one thing I did notice is that there seem to be a lot of CCTV cams that appear to have the sony CCD in. I think?.

It does look like they have limited access to controls but has anyone ever tried a CCTV cam for imaging? or is there a really obvious reason, that I am missing, why they're not used?

Cheers

Jez

[Gina]

CCTV cameras are generally very low resolution unless you spend a ton or more on them. Even then they are standard PAL (or NTSC) which is around 500 lines interlaced, so you only get 250ish lines per frame. Then you need a TV to USB adapter to use with a computer. Webcams are usually at least 640x480 even with the cheapest. They are also far better suited to AP. With most webcams you can change the resolution. For lunar AP an MS Lifecam works very well.

[M00NMonkey]

Cheers Gina.

I figured there must be a reason, thanks

Jez

[Gina]

Put simply - webcams are cheaper, easier to use and more controllable. I think that sums it up.

[M00NMonkey]

Anyone know anything about the Microsoft lifecam hd-6000?