After purchasing my digital still camera (Nikon Coolpix 995) I was very curious to compare its light gather capabilities to my webcam Vesta 675 SC2.
I set both cameras to tripods abreast.
Pictures from Coolpix were taken at full resolution (2048x1536) with normal jpeg compression. Then they were resampled to VGA and QVGA resolution by Lanczos interpolation. No further processing was used. The aperture of digicam was set to 3.0 in order to achieve same conditions for both cameras. Also focal length was set to 8.8mm in order to achieve approximately the same field of view as webcam.
Pictures from Vesta were captured at 640x480 resolution.
The ambient temperature was about 25°C. All Nikon Coolpix 995 photos were taken without cooling. Vesta photos were taken with and without cooling.

For better comparison, please calibrate your monitor (set maximum possible contrast, to see details in dark areas of test images):

Comparison of Nikon Coolpix 995 digital camera versus Philips Vesta 675SC webcam

Low room light - three 60W incadescent lamps - each working at 15W triac regulation)
NIKON CoolPix 995	Philips Vesta 675 SC2	Note
F3.0, 1s, ISO 400, Incadescent	F3.0, 1/5s, Gain 100%, Incadescent	The room was illuminated by three 60W incadescent lamps - each working at 15W (triac regulation). That's why color temperature of light was shifted to red part of spectrum.
The same picture as above (for better comparison)	F3.0, 1/5s, Gain 100%, Incadescent, stacked 15 frames	The same conditions as above - only the webcam photo was created by stacking 15 frames. Regarding noise - digicam single photo is comparable to stacked webcam photo.
Dark - distant street lights
NIKON CoolPix 995	Philips Vesta 675 SC2	Note
F3.0, 60s, ISO 400, Incadescent, Noise Reduction	F3.0, 30s, Gain 100%, Incadescent, AMP OFF	Now the lights were switched off. Only light from 2 distant street lamps (80m, 100m) was penetrating through balcony doors. The green light on the Vesta's photo is green LED of power supply visible only from Vesta's position.
The same picture as above (for better comparison)	F3.0, 30s, Gain 100%, Incadescent, AMP  ON, Peltier cooled	Regarding noise - digicam single photo is comparable to cooled webcam photo. Webcam photo was taken with shorter exposure and is lighter.
The same picture as above adjusted in K3CCDTools: Levels: 0 - 50% Gamma: 2.00	The same picture as above adjusted in K3CCDTools: Levels: 10 - 60% Gamma: 2.00	The above pictures were adjusted in K3CCDTools - adjusted levels and set gamma factor. Digicam reveals its much higher dynamic range and better color interpretation (see the flowers in corner above pillow). Furthermore we must remember, that digicam was used without cooling! Digicam's internal noise reduction system (subtraction of darkframe) is very powerful!
F3.0, 60s, ISO 800, Incadescent, Noise Reduction	F3.0, 40s, Gain 100%, Incadescent, AMP OFF	The same light conditions as above, but another exposure...
F3.0, 60s, ISO 800, Incadescent, Noise Reduction	F3.0, 30s, Gain 100%, Incadescent, AMP ON, Peltier cooled
F3.0, 60s, ISO 800, Incadescent, Noise Reduction	F3.0, 40s, Gain 100%, Incadescent, AMP ON, Peltier cooled	The last two comparisons show, that F3, 60s, ISO800 exposure is more comparable to Vesta Pro 30s exposure than 40s exposure
F3.0, 60s, ISO 800, Incadescent, Noise Reduction	F3.0, 60s, Gain 100%, Incadescent, AMP ON, Peltier cooled	The last three comparisons show, that digicam's F3 / 60s / ISO800 exposure is the most comparable to Vesta Pro 30s exposure. That means, that my Vesta 675SC2 has sensivity about ISO1600 (using gain 100%).

Conclusion:
My tests show, that my Philips Vesta 675SC2 webcam is about twice more sensitive than Nikon Coolpix 995 digicam at ISO800. That means, that webcam has sensitivity about ISO1600. This is only rough estimation, I know, that webcam's characteristic is not linear.
On the other hand, Coolpix 995 has better dynamic range with much better resolution and color fidelity - so it may provide better results after some post processing. Tests in real astrophoto in future will unveil more...

Efficiency of Nikon Coolpix 995 Noise Reduction

F3.0, 60s, ISO 800, Incadescent, without Noise Reduction	DarkFrame (with the same exposure)	Note: The thermal noise in original image is higher than it is shown here. JPEG compression used here causes removing higher frequencies in the images. The CCD amplifier glow at top of image is visible.
F3.0, 60s, ISO 800, Incadescent, with Noise Reduction	F3.0, 60s, ISO 800, Incadescent, without Noise Reduction, with subtracted DarkFrame (in K3CCDTools)	Internal Noise Reduction gives image with better contrast. The comparison reveals, that internal noise reduction system is very effective. It seems (it's only my privat opinion), that it uses DarkFrame subtraction with full 12-bit images captured from CCD chip.

Computer generated images, real images, drawings and texts are property of the author and may not be reproduced or used without permission of author.

Last Update: 16.06.2002