SharpCap - free Astro Webcam Capture Software

[pianorealm]

Thanks Robin!

[pianorealm]

Robin.

Thanks for all your help sorting the problem. As I said, turns out Sharpcap was absolutely fine......just my dodgy hardware confusing me!

Sharpcap, Basler  and Pylon now running great.

Howard

[badhex]

Hi Robin

I've been starting out with imaging using a modified Lifecam Cinema HD and have a few questions about using it with SharpCap. I know that there are a fair few people using them so I'm hoping I won't be the first to ask these questions.

I'm using the most recent version on SharpCap which is 2.0.862 AFAIK but away from my imaging laptop at the moment.

- The Lifecam does not have an option for gain. Is this the driver at fault? If so, do you happen to know if there are alternative drivers which may help?

- Is there any way to force enable default settings options for webcams, or is everything driver related (I would assume so)?

- When reducing the resolution (to get better FPS in this case), I expected to see a smaller overall image, but with the same size image of the object (Jupiter, so it was very obvious) - i.e. just a smaller area of the CCD being used - however, the whole image became smaller, including Jupiter itself. This is counter intuitive to my understanding of how this should work, again is this driver related or am I missing something?

Any light you can shed on the above questions would be much appreciated!

Many thanks

Joe

[rwg]

Hi Joe,

As far as I have been able to work out, most webcams these days implement (in hardware) something called 'USB Video Class' which means that they can work without having to have specific drivers. Of course the manufacturers then add drivers too to 'customize' the experience. Quite often we'd interpret 'customize' as 'make worse'... Anyway, the Lifecam cinema doesn't implement a gain control at the hardware USB Video Class level, which means that no driver is going to add one for you. That being said, the brightness control on the LifeCam Cinema seems to work a bit like a gain control - if you turn it up and the exposure down you can definitely see the noise rise.

Zoom is the key to the reduced resolution working the way you expect - turn it up to the maxium and the smaller resolution works like a crop on the main picture instead of a scaling down of it.

Finally, you may get better results by manually changing the driver for the camera from the lifecam cinema driver to the 'USB Video Device' driver. This gets rid of most of the 'improvements' that MS have tried to implement in the Lifecam Cinema driver and gets you closer to the underlying hardware capabilities of the camera. For instance doing this gets the 'MJPG' colour space back that was available long ago before one or another driver upgrades killed it. I'm fairly sure that MJPG is the way the camera compresses the frames to send to the PC by default, the Lifecam Cinema driver then turns this into YUY2 for you, which means bigger files but no gain in quality as it has already been MJPG compressed...

The way to switch the driver is to go to the device manager, find the device (in the Imaging Devices category), right click, choose 'Update Driver Software', choose 'Browse My Computer for Driver Software', then 'Let me pick from a list', then choose 'USB Video Device'.

cheers,

Robin

[PeterCPC]

I have an Orion G3 colour camera and the software that came with it is just awful. I wanted to use Sharpcap but it could not recognise the camera. Of course the Orion camera is designed as a single/multiple shot camera not taking avi.

Pity Sharpcap can't be used.

Peter

[rwg]

I have an Orion G3 colour camera and the software that came with it is just awful. I wanted to use Sharpcap but it could not recognise the camera. Of course the Orion camera is designed as a single/multiple shot camera not taking avi.

Pity Sharpcap can't be used.

Peter

Hi Peter,

try the ASCOM driver for the Orion G3 to see if that works at all in SharpCap. If it does and you are interested in helping me develop full SharpCap support for this camera (i.e. I write the code, you test it!!!) then let me know.

cheers,

Robin

[badhex]

Hi Robin

Thanks for your feedback. I'm a sysadmin by trade so only too aware of the supposed 'improvements' made by proprietary software and drivers!

I have read that switching drivers also lets you select 1280x720 @ 30fps using YUY2, which is not possible with the MS one - you have to either use MPJG, or drop the res to 640x480 to get 30fps. Not got round to that yet, I guess I was hoping against hope that gain was not hardware level and that it might appear as well.

Thanks for the heads up on the zoom - so if zoom crops the sensor, but the video is still output at the same res, it must be upscaling the pixels, which is presumably pointless for our needs as all we are getting is bigger files. Do you know what area, pixels or percentage wise, of the CCD is being used at max zoom (I guess this is hardware dependent)? If I'm working this out correctly the optimum would be to set the resolution to that size - e.g. if zoom at max uses 320x240 on the CCD, I should set the res to 320x240 too for 1:1 scaling. Is it just a case of playing with the settings to see what gives me the correct output?

One more question on the colourspace/compression. So, from what you said am I to understand that the frames are ALREADY compressed using MJPG by the hardware, and there's nothing you can do about it? If so that's kinda rubbish. Is that true of a lot of webcams you've worked with or another Lifecam specific deficiency?

Many thanks for your replies already, and any more info you can help with.

Joe

[rwg]

Hi Joe,

with the zoom, it only does anything when the resolution has been dropped, so for example

* at 1280x800 resolution zoom does nothing you will get the same picture whether set to 1 or 10

* at 320x240 resolution, zoom=1, you get a 320x240 reduced image of all of the camera frame

* at 320x240 resoltuion, zoom=10, you get a 320x240 crop of part of the camera frame (use the pan/tilt box below the zoom to choose the area)

1280x720 @30fps YUY2 is an illusion. USB 2 doesn't have enough bandwidth for it - YUY2 is 2 bytes per pixel, so the bandwidth is 1280*720*30*2*8 = 442 Megabits/second.

USB 2.0 is theoretically 480Mb/s, but at least 10-15% goes on overhead - the official maximum throughput is 426Mb/sec (http://electronics.stackexchange.com/questions/24700/why-are-usb-devices-slower-than-480-mbit-s) but many implementations don't hit that level by a good 20%.

My conclusion from this and the fact that the 'USB Video Device' driver only gives 30fps in MJPG is that the lifecam driver is doing MJPG -> YUY2 in software on the PC - at least when you select the high frame rates. Of course that's the worst thing to do as you already have JPG compression artifacts in each frame to which you are adding colour space compression artifacts when you go to YUY2. All you get is the illusion of YUY2 at 30fps and bigger AVI files.

cheers,

Robin

[badhex]

Ah, of course. Aware of USB2.0 actual usable bandwidth but don't know huge amounts about video compression and relative data bandwidth needed for different methods - never put two and two together to realise that it wouldn't *actually* be possible to chuck that much data down the bus!

The zoom/pan/tilt settings are awesome, a great feature - actual nice, usable software for once! Who said you couldn't have form and function?! Thanks Robin!

[rwg]

Yet another new version of SharpCap is now available - version 2.10.

* New feature - Bahtinov mask focus assistant. This will detect the 3 line pattern from a Bahtinov mask and show an indication of how near focus you are
* You can now type values in the box that shows the value of a camera property - Just type a number and press enter.
* If an exposure control shows its value in ms or s in the value box, you can type in 33ms or 1.5s and SharpCap will understand it.
* New histogram, capable of showing all 3 channels at once for a colour camera
* Further performance improvements for ASI/ZWO cameras
* Fix for capture sequences not working when 'single frame' selected
* Improved focuser control UI - should have more sensible step defaults and less drawing issues
* Speed improvements starting QHY cameras
* Custom resolutions available for ASI/ZWO cameras (click on Custom... in the list)
* Several bug fixes from crash reports - thank you for taking the time to send in bug reports.

Early adopters can find it on the SharpCap downloads page, otherwise it will go live via auto-update in a week or so.

cheers,

Robin

[jambouk]

Sounds good. I like the bahtinov idea, i have used a program valled bahtinov grabber, but if it's built in that's event better.

Maybe you could build in a collimation checking routine too, and that would help check and adjust the collimation before starting an imaging run.

Great though, look forward to trying it.

James

Sent from my iPhone using Tapatalk

[rwg]

Sounds good. I like the bahtinov idea, i have used a program valled bahtinov grabber, but if it's built in that's event better.

Maybe you could build in a collimation checking routine too, and that would help check and adjust the collimation before starting an imaging run.

Bahtinov Grabber was the inspiration for integrating that funcitonality into SharpCap - it's a very cool program but I found it kind of fiddly having a second program to juggle on the screen so thought it would work more smoothly inside Sharpcap. Finding lines in an image is a fairly well explored machine vision problem and can be done using something called a Hough transform, so it wasn't too much trouble to imlement. At some point in the future I will add an autofocus option to it too.

Not quite sure how SharpCap could check collimation though - I know that there is a thing called a 'Duncan mask' for SCTs that shows a pattern that changes depending on collimation, but I don't even have a newt, so newtonian collimation and how you might detect it in an image is a mystery to me

cheers,

Robin

[jambouk]

I have seen some software which analyses the pattern of an out of focus star for SCTs and Maks, and looks for the degree of concentricity (if that is a word); if this was built into sharp cap, the user could use a camera to look at a star, take it out of focus, increase the gain/exposure so the rings were brighter, then overlay the target from the software and as you alter the collimation screws in live view it would tell you if the rings were getting worse or better; like bahtinov grabber but for SCT collimation.

Just a thought for future considerations.

Jd

Sent from my iPhone using Tapatalk

[jambouk]

Here is the software I was thinking of, a transparent set of rings which can be altered to fit over an out of focus star and used as a template to alter the collimation:

http://stargazerslounge.com/topic/136286-als-collaid-sct-collimating-assistant/

James

[rwg]

Hi James,

thanks for that link - I think that can be automated so that the software an detect the circles for you - there is a technique called a 'Hough Circle Transformation' that detects circles in an image and a black image with white circles in is pretty much perfect for it.

Right now, I'm busy trying to update the UI of SharpCap to improve it substantially for a new 2.5 or 3.0 version, but I will definitely have a look at the collimation idea as I find collimation a bit of a trial too at times.

cheers,

Robin

[jambouk]

No worries. Thanks for all the hard work you put into sharpcap. Version 2 seems much more stable on my laptop which is a blessing.

Regards

James

Sent from my iPhone using Tapatalk

[joelsfallon]

Hey, ~1 month sharpcap user here.

Just recently I've been having some issues with some of my images filmed with sharpcap, there appear to be concentric ring artifacts in my planetary images.

I remember getting this at least one or twice within 2 weeks, but it usually sorts itself out somehow, until just this week, they are constantly occurring.

I have updated sharpcap in the past week, other than that I can't think as to why this is happening, is anyone else experiencing this? These rings appear when modifying wavelets in Registax 6.

Examples taken with SharpCap:

http://i.imgur.com/TqLAWL3.png

http://i.imgur.com/NUmHALV.png

Example taken with FireCap: 

http://i.imgur.com/mTUb5QH.png

AMcapture also gives me clearer results, so I assume it's just sharpcap that I'm having this problem with. Any ideas?

[rwg]

Hey, ~1 month sharpcap user here.

Just recently I've been having some issues with some of my images filmed with sharpcap, there appear to be concentric ring artifacts in my planetary images.

I remember getting this at least one or twice within 2 weeks, but it usually sorts itself out somehow, until just this week, they are constantly occurring.

I have updated sharpcap in the past week, other than that I can't think as to why this is happening, is anyone else experiencing this? These rings appear when modifying wavelets in Registax 6.

Examples taken with SharpCap:

http://i.imgur.com/TqLAWL3.png

http://i.imgur.com/NUmHALV.png

Example taken with FireCap: 

http://i.imgur.com/mTUb5QH.png

AMcapture also gives me clearer results, so I assume it's just sharpcap that I'm having this problem with. Any ideas?

Could just be down to different camera settings - on some cameras low gains give you a banded effect on the captured images which would be turned into the sort of rings you see by the wavelet sharpening. In general, SharpCap doesn't do any processing to the images - it just takes the image provided by the camera and sends it to the output file.

Let me know what camera you are using and what settings you captured at and I might be able to give some more specific advice.

cheers,

Robin

[joelsfallon]

Thanks for the reply, 

as for the camera I used:  ZWO ASI120MC

You are correct when it comes to low gains, all videos were taken with about 20% gain, 0% gamma, 480x320 resolution - everything else at default

I will attempt another planet shot tonight with higher gain to see if it makes a difference. 

As a side note: I cleaned the primary lens, mirror diaganol, barlow lens & sensor cover with a lens cleaning kit prior to taking any of the videos, so I assume that dew was not the cause.

[joelsfallon]

Thanks for the reply, 

as for the camera I used:  ZWO ASI120MC

You are correct when it comes to low gains, all videos were taken with about 20% gain, 0% gamma, 480x320 resolution - everything else at default

I will attempt another planet shot tonight with higher gain to see if it makes a difference. 

As a side note: I cleaned the primary lens, mirror diaganol, barlow lens & sensor cover with a lens cleaning kit prior to taking any of the videos, so I assume that dew was not the cause.

My apologies for not getting back sooner, I've had nothing but cloud cover recently, and now I'm busy for the weekend. I'l do my best to get some results up monday/tues

[jambouk]

My version (2.1.915.0) keeps telling me there is an update, 2.10.915, but the update just won't install and i can't see it on the website - is there a but with the update service, or is it my PC?

JD

[rwg]

Oops, sorry about that - a mistake in the update definition file online meant that it kept trying to load what it thought was a new update but was in fact the same version.

now fixed I think.

cheers,

Robin

[jambouk]

Thanks

Sent from my iPhone using Tapatalk

[knobby]

Thanks for the reply,

as for the camera I used: ZWO ASI120MC

You are correct when it comes to low gains, all videos were taken with about 20% gain, 0% gamma, 480x320 resolution - everything else at default

I will attempt another planet shot tonight with higher gain to see if it makes a difference.

As a side note: I cleaned the primary lens, mirror diaganol, barlow lens & sensor cover with a lens cleaning kit prior to taking any of the videos, so I assume that dew was not the cause.

Asi recommend gamma at 50...

Maybe try that next time

[jambouk]

I never knew about the gamma at 50; i'll give that a go too.

Thanks.

Sent from my iPhone using Tapatalk