Jitter-bug: Difference between revisions
Jump to navigation
Jump to search
No edit summary |
No edit summary |
||
| (2 intermediate revisions by the same user not shown) | |||
| Line 1: | Line 1: | ||
Voiopmonitor sniffer 16.0 fixes jitter calculation for huge jittered calls. The jitter was always lower than it should be compared to wireshark | Voiopmonitor sniffer 16.0 fixes jitter calculation for huge jittered calls. The jitter was always lower than it should be compared to wireshark due to rounding issue in the code. The MOS scoring was not affected by the bug because another algorithm is used. | ||
Here you can see simulated jitter/loss call comparison processed by sniffer < 16.0 where avg jitter is 39 for caller and 42 for called which is wrong - also see those two charts and PDV differences | |||
[[File:jitter-bug.png]] | [[File:jitter-bug.png]] | ||
This picture is CDR processed by sniffer >= 16.0. AVG jitter is 49.1 and 50.3 which is on par with RTCP reports as well in wireshark. Also PDV are higher as it should be as well as charts looking "better" | |||
[[File:jitter-fixed.png]] | |||
Latest revision as of 01:07, 28 May 2016
Voiopmonitor sniffer 16.0 fixes jitter calculation for huge jittered calls. The jitter was always lower than it should be compared to wireshark due to rounding issue in the code. The MOS scoring was not affected by the bug because another algorithm is used.
Here you can see simulated jitter/loss call comparison processed by sniffer < 16.0 where avg jitter is 39 for caller and 42 for called which is wrong - also see those two charts and PDV differences
This picture is CDR processed by sniffer >= 16.0. AVG jitter is 49.1 and 50.3 which is on par with RTCP reports as well in wireshark. Also PDV are higher as it should be as well as charts looking "better"
