Eugene Smiley wrote on 11-12-2007 0:22: > Jan Hoevers wrote: >> Scott Baker wrote on 10-12-2007 17:37: >>> Does anyone have any experience with one of these? >>> http://linuxdevices.com/articles/AT8083883801.html >> I'm rather amazed about the specified ±100ms accuracy for a $1000 >> device. Any ten year old PC would perform much better than that. > True, but can you put a 10 year old PC on top of a pole connected to a > surveillance camera? Why would the NTP server need to be on top of the pole with the camera? It's an ethernet device! What if a building is guarded by more camera's? Buy this device for every camera? I doubt if that is the way it is intended. It's not the way NTP is intended. > The way it's written, it's a product looking for a market not a product > filling > a market need. What keeps "CCTV (closed-circuit TV) surveillance and digital > video recording (DVR) applications" from using a non-GPS/WWV NTP server for > "sub-second accuracy?" Strange way of creating a market, turning a general device (NTP server) into a specialized device (NTP server for CCTV), with little other benefits than decreased accuracy. The low power consumption (0.9 W) is still a bit high for battery operation. Perhaps that 100 millisec is a typo, 100 microsec would be more reasonable for a GPS driven NTP server.
I expect this overpriced device won't find its way into NTP pool circles. Jan _______________________________________________ timekeepers mailing list [email protected] https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
