Problems at IP-audio-connections between 
can be solved by the following actions:
| • | Increasing the IP audio delay on the decoder side |
| • | Adapt transfer rate and duplex mode |
| • | Adjust Type of service |
| • | Adapt packet size |
1. Increasing the IP audio delay on the decoder side
The IP audio delay can be set from 0 ms (default) up to 5000 ms at menu item Settings/Network. Increasing the audio delay means that first all received audio data is buffered before it is decoded. Such buffering helps to prevent audio interruptions and distortions caused by jitters.
The bigger the receive buffer the more reliable the audio IP transmissions is. On the other hand the audio delay increases.
2. Adapt transfer rate and duplex mode
Sometimes interruptions or distortions of IP audio transmissions are caused by improper auto sensing of the network equipment (switch, hub , router etc.). In this case the transfer rate and the duplex mode should be changed ‘auto’ to the settings of the network. This can be at menu item Settings/Network.
Example:
With some network switches it is useful to change
| • | transfer rate from ‘auto’ to ‘100 Mbit/s’ |
| • | duplex mode from ‘auto’ to ‘half’ |
3. Adjust Type of Service
supports ToS feature (Type of Service). Basic info and how to set the ToS is described in application note 20 on Mayah Website (www.mayah.com/content/download/pdfs/appnotes/centauri/a_n_e_020.pdf).
4. Adapt packetsize
Sometimes interruptions and distortions of IP transmissions are caused by the fact that the upload and/or download capacity is too small. Therefore it should be considered the really needed bit rate is higher than the bit rate of encoder due to the IP overhead. At UDP transmissions this IP overhead does not fluctuate and can be calculated exactly. More details about UDP/IP overhead calculation can be found in chapter IP Overhead.
Usually the bigger packet size is the smaller the IP overhead is.
However, there is one restriction:
If the payload (audio data) plus the IP overhead is bigger than max. possible packet size value allowed by the network card (i.e. 1,514 Byte at electrical Ethernet card), then the rest of the payload is written to next packet. This can cause very big IP overheads.
Therefore the enables to set the IP packet size mode to variable and fix.
4.1 IP packet size mode
The IP packet size mode can be set at menu item Settings/Network to (provided system software version 2.2.0.0 or later is used ):
| • | variable (default) This means that the payload is rounded to the next full frame or audio block Advantage: No fractional audio frames or blocks in the IP packets Disadvantage: With algorithms with no fixed frame length (e.g. AAC) it can happen that the max. possible packet size is exceeded. |
| • | fixed This means that IP payload is exactly the same as the packetsize value set in menu item Settings/Network (i.e. it is not rounded to the next full audio frame or audio block). Advantage: Even with algorithms with variable frame length the packet size is fixed. Disadvantage: IP packet can have fractional audio frames or audio blocks. |
| • | sync. Same as fixed but additionally packets are transmitted in synchronized intervals. Advantage: Same as with fixed but additionally adaption to network equipment which has got problems with unsynchronized packets. Disadvantage: Same as with fix. |