Is there anything known about the accuracy of timestamps returned by the Gracenote services? For example, if I take the starting moment of a TV program according to the eyeQ EPG, add the values obtained from the Gracenote Entourage app, how close does this come to the actual moment in time the sample was fingerprint?

Kind regards,


Hi,That should get you within 1-3 seconds of the actual moment you fingerprinted. 

Thanks for your answer cweichen. I actually did what I described in this first post here. I obtained a few of the described timestamps (EPG+Entourage match offset) and compared these with my local timestamps (the moment the fingerprinting command was issued). Next, I compared both timestamps with each other and subtracted them from each other. After doing this a few times (same EPG information and channel, samples obtained in the given order), I obtained the following time differences (sec:ms).03:14200:19400:15903:15300:20503:12000:16203:19300:20203:18803:19103:13000:23500:20100:19803:20603:261 As can be seen here, there is a pattern in these differences. About half of them range between 03:130 - 03:261 while the other half ranges between 00:120 - 00:235. Clearly there is a pattern here, namely caused by a more or less fixed difference of 3 seconds. My question is: where does this 3 second difference come from? Is this caused by the Acr.GnAcr.DoManualLookup() method I'm using (the moment I record my local timestamp is the moment I call this method) or is this server side related? Note: I saw the sticky about stackoverflow. But I doubt anyone from the community will be able to answer this question and it is also a follow up question on this forum topic, so therefore I post it here. If this is a problem I can post it on stackoverflow aswell.