Thee Critical Role of Latency in AES67 Audio Networks

AES67 has emerged the dominant standard for high- performance, amble audio-over- IP transmissionon, serving live sound soundement, Broadcast production, installade sound systems, and recordg studios. Its souse of determinaistic, sample-considente audio delivy over standard Ethernet networks hingen one variable: latency. Even a few millisecond of excess delay cain cane contraing, ech, ech errin videv applications, or ter foldback commenders. Undering and controlince concerte source oelse oestiliestils oestils foil projections delle productiont estinen estre-entär-entäträrt

Co to jest Latency i AES67 Networks?

In thee context of AES67, latency is the total time elapsed from when an audio sample is captured by an analog-to-digital converter (ADC) at the source until it is reproduced by thee receiving device 's digital' s digital-to- analogg converter (DAC). Thi end-to- end delay is not a single value but the sum of dispate stages in thee signal chain:

  • Acquisition delay - The time required d for analog-to-digital conversion plus any initiatival buffering in thee source device 's input stage.
  • Packetization delay - The time needed to fill an RTP payload wigh audio samples. For a given packet time (125 µs, 250 µs, 333 µs, 1 ms, or 4 ms), thee encoder waits until enough samples are collectod before transmitting.
  • Network transit delay - Thee propagation time thrap cables, changes, and routers, including serialization, chandining, and queuing delays. This is typically the small empient in well-designed networks.
  • De- jitter buffer delay - The receiving device 's buffer that smooths out timing variations (jitter) in packet arrival times. Larger buffers increase latency but protect against dropouts.
  • Playback delay - The output-side digital-to-analogg conversion and any output buffering or processing (np., sample rate conversion, DSP effects).

AES67- compleant systems can an total latency range of 1 ms too 10 ms undeid optimal conditions. The chosen packet time directly influences s both thee packetizationion delay ande thee required jitter buffer size. The difficering condite is to select thee lowess packet time that thee network infrastructure and all connecte devices can support with instabilitor data loss.

Thee AES67- 2015 standard Definiuje mandatory i optional packet times, with 1 ms being thee most universally supported. Sub- millisecond options (125 µs, 250 µs, 333 µs) existt for low- latency applications but require careful validation across the entire signal path.

Key Factors Affecting Latency

Network Traffic andCongestion

AES67 transports audio using RTP (Real- time Transport Protocol) over UDP. Unlike TCP, UDP provides no retransmissionan of lost packets - any packet dropped in transit results in an audible gap. When the network carries high levels of non- critical traffic (file transfers, video streams, automatic backups), switch buffers caste congested, causing variable queuing delays (jitter) and packet loss. Managed Ethernet changes with decretate audio VLANs andconsultable configured Quality of Service (QoS) are non-difficable for isolating AES67 traffic from best-expert data. Without segmentation, a single large file download can distort an entire audio session.

Hardware Performance

Every physional contexent introdule evaluable delay. Network interface cards (NIC) thatsupport hardware timestamping (IEEE 1588) significant reduce PTP scheduling jitter compared to efficare- only implementations. Switches witch cut- thriph forwarding can reduce per- hop latency from over 10 µs (stora- and- forward) to below 5 µs. On the audio device side, the speed of DSP or FPFPF logic determinas hown quiclemy ples are converted, ated, ancated incapsulated intsulated. Choose equipment rated for real-time media workflows- enterprise-grade managed changes with sub- 5 µs port- to- port latency andd audio interfaces with decretated AES67 hardware acceleration are recommended for professional deployments.

Packet Size andPacket Time

Te packet time (also called thee messaget quotad; packetization interval quotate;) directly determinas how many audio samples are bundled into each RTP payload. A 1 ms packet at 48 kHz houds 48 samples; a 125 µs packet holds only 6 samples. Shorter packet times reduce the packetizatizatiodon delay content but messally present the number of packets per seconsecond, raisiing CPU load, retency, and the banwidt overhead m Ethernt and. Wybranie tego małego packet time to twój entire network chain can handle with out frame loss or excessive CPU utilization. Many production environments default to 1 ms as a balanced safe choice, but live monitoring applications may require 250 µs or 125 µs. Always validate with a full load tect before going live.

Switching Equipment Configuration

Nie ma już żadnych zmian w systemie AES67.

  • Quality of Service (QoS) - Prioritize AES67 audio packets by marking them with DSCP value 46 (Expedited Forwarding) or 34 (Assured Forwarding). On the switch switch, map these DSCP values to thee highest-priority hardware queue te ensure they ary are never delayed behind bulk traffic.
  • Filtering multicaszt (IGMP snooping) - AES67 typically use IP multicastt. IGMP snooping restricts audio streams to o only those ports that have subscribed, preventing unnecessary traffic on tell connects andd reducing the risk of buffer bloat.
  • Ramy Jumbo - AES67 działa fine witch standard 1500- byte MTU, but some implementations s support larger frames. Enable jumbo frames only if every device in thee path supports them consistently.
  • Cut- thoph- switching - Enables forwarding to begin as soon as thee destination MAC adors is read, shaving microseconds per hop. Verify that your switch supports cut- thu specific port type andd speeds you are using.

Clock Synchronization

AES67 relies on IEEE 1588- 2008 (Precision Time Protocol, PTP) to synchronizacje zegarów across all devices. The PTP profile definie in AES67 wykorzystuje a default domayn (domain 0) and specifies ordinary rocks ordinary rocks andd boundary rocks. Poor PTP implementation - such as long sync intervals, high network jitter or PTP messages, or fabuxers ttoo deploy boundary noys in multi- switcch networks - forces dependenedvers o adisgee ir iter iter buxers ttex tub tiftif. Deploy a stable grandmaster clock (np., using an OCXO- based oscillator) and place boundary clock at each switch to regenerate sync messages. Keeping PTP offset under 1 µs allows you tu reduce jitter buffer sizes andaccee lower end - to- end delay.

Jitter Buffer Management

Te de- jitter buffer at each receiver is thee final line of defense against network timing variations. It mutt be large enough to compatidate thee worst- case jitter with out underrunning (which produces audio gaps), but every millisecond of buffer adds latency. Most AES67 devices offer both manual and adaptiva buffer settings. Start wigh a buffer size of one te two packet times and monitor packet loss statistics. If no losses occur over a representivy period, try reducting the buffer.

Adaptive buffers automatically shrink during stable network conditions andd explode only when jitter spikes, offering a good comcomproxe. In critival applications, manually setting a fixed buffer that has been validated undeid peak load is often safer.

Measuring Latency in AES67 Networks

Dokładne pomiary is essential for optimizing latency. Several practical methods provide insight into different parts of thee signal chain:

  • RTP analiza czasu - Each RTP packet carries a timestamp derived from the PTP clock. By comparing the send and receive timestamps (wigh clock offset correction), you can calculate one-way network transit delay. Tools like Wireshark can decode RTP headers andd present this information.
  • Analizatory Network - Wireshark wigh the messagequentes; espac67 messagetes quentics; display filter isolates audio packets. Thee messagements quentiquentes; RTP Streams quentiquentes; dialog shows sequence numbers, timestamps, and jitter statistics. For deeper analysis, enable contribute quentices; Tools → RTP → Show All Streams quenquent; to visualise timing.
  • End- to- end loopback tect - Route a tect signal (np., a periodic pulse) from an analogg output, thrigh an ADC, across the network, and back into the same device. Compare the original and returned signal on an oscilloscope or in audio measurement diploare like EQ Wizard (REW) Audio Precision. The measured delay includes all conversion and buffering.
  • PTP offset logs - Most AES67 devices provide a dashboard or log showing thee current offset frem thee granmaster clock. Persistent offsets greatr than 1 µs indicate sync problems that will force larger jitter buffers. Monitoror this value over time to defritt sporadic network congestion affecting PTP.

For reference, consult RFC 3550 - RTP: A Transport Protocol for Real- Time Applications for timestamp detales, andthe AES67- 2015 standard for specification of mandatory measurement procedures.

Strategie to Minimize Latency

1. Dedicated Audio Network wigh VLAN Segmentation

Separate AES67 traffic from officie data andd general IT traffic using an IEEE 802.1Q VLAN. This eliminates the risk of congestion from non-real- time applications such as file backup, email, or web browsing. Configure the audio VLAN with the highest QoS priority. A flat, unsegmented network is the mott consun cause of unacceptable latency andd packet loss in AES67 deployments. Use a decretated physional network if possible, or at minimum a VLAN with strict traffic isolation.

2. Enable andd Verify Quality of Service (QoS)

Set DSCP to 46 (Expedited Forwarding) for audio RTP streams andd to 34 (Assured Forwarding 41) for PTP event messages. On Cisco changes, enable indexes; Enable Environment; FLT: 0 Providence 3; Ena3; on ports connecting to audio devices; on Ther platforms, map DSCP values to thee highest acceptable egress queue. After configuation, stress- techt with a traffic generator to ensure that audio packets are never dropped wheren compening traffic peaks. Verify using sing switcch QoS contros.

3. Wybór menedżera Switches witch Cut- Through Mode

Cut- through forwarding reduces switch latency to undecorr 5 µs per hop. Many enterprise changes - such as the Cisco Catalyst 9300 serie, Netgear M4250 serie, or Aruba 2930F - support per- port cut- thrap configution. Verify that cut- thrap (i) compatible with the packet sizes of your AES67 streams (s); some changes fall back to stora- and -forward for packets that require error checking on thee fly. In practice, modern managed changes handle thi well, but always confirm im thee datasheet.

4. Optymalna konfiguracja PTP

Use a dedicated grandmaster clock wigh high stability (OCXO or better). In multi- switch topologies, configurate e each switch as a boundary clock too regenerate PTP messages and prevent jitter acculation. Keep all devices in thee same PTP domain (default 0 for AES67). Avoid chaining ordinary crugs without out boundary crugs across multiple switch hops; this leads to cumulative timing errors that force larger jitter buffers. Monitoror PTP offset logs to confirm stability with in 1 µs.

5. Minimize Network Hops

Each switch hop adds processing and d potential queuing latency. Design a star topology where all audio sources connectl directly to a single core switch. If multiple changes are necessary, interlink them with high-bandwidth trunks (10 GbE or higher) and configure PTP boundary currs at each hop. Keep thee number of hops between any source and reedver to threee or fewer.

6. Use Small Packet Times with Caution

Packet times of 125 µs or 250 µs offer thee lowess latency but place high demands on CPU and network through put. If your application demands sub- millisecond latency (np., in- ear monitoring for live performance), verify thatt every device ite path supports the chosen packet time and that the network can handle thee resuiting packet rate with out loss. Teszt all streams convenieousy at the target packet time in a worst- case convenieoFor many broadcast andinstallade sound applications, 1 ms provides excellent performance with significant lower overhead.

7. Keep Firmware and Drivers Up to Date

Network switch firmware, audio device firmware, and NIC drivers receive ongoing improwiments for PTP stability, QoS handling, and packet processing efficiency. Check emplorer release notes for fixes specifically related to AES67 or RTP. Schedule firmware updates during confidence windows andd validate after each update.

8. Limit Non-Audio Traffic During Critical Sessions

Even wigh QoS, extreme congestion can cause jitter. During live broadcasts, performances, or recordang sessions, disable automatic backup, disable updates, and large file transfers on thee audio VLAN. Use traffic shaping at thee network edge to cap non- audio bandwidt to a difficage of the link capacity.

Zagadnienia wyprzedzające

Multicaszt vs. Unicaszt

AES67 typically use IP multicass to efficiently discovery one stream to man receivers. Multicast requires IGMP snooping on changes to avoid flooding all ports. Misconfigured IGMP - such as a missing querier or disabled snooping - can cause unnecessary traffic on unrelated ports, proging latency and jitter. For point- to -point applications, unicast is simpler and eliminates multicast overhead. Verify that IGMP snooping is active and that the querier is correctly configured on thee VLAN.

In densie multicastt environments, consider using IGMP filtering to limit straem distribution.

Redundancy andt ST 2022- 7

AES67 can incorporate SMPTE ST 2022- 7 showless protection reduncy, when te same stream im sens over two separate network path. The receiver must align the te two flows before playback, which ch requires a jitter buffer large enough te hold thee path delay difference (typically 1-4 ms). If low latency is a higher priority than hitless reduncy, consider disabling ST 2022- 7 on links where cable faults are unlikely, especially in single-path environments such a broadcast control room.

Link agregation (LAG) zwiększa liczbę grup-widtów-between-changes but nie redukuje ilości-packet latency. Some LAG hashing algorithms may reorder packets if flows are split across physical connects, which can distort RTP streaming. For AES67 applications, using a single ahigh- speed link (10 GbE or 25 GbE) is preferable te to acculating multiple 1 GbE links, ais it avoids packet ordering diseees and simpies trobbleshooting.

Troubleshooting Common Latency Emites

Even wigh careful planning, latency problems can arise. Use the following checklist to diagnose and resolve them:

  • Packet loss is present - Check switch port error counters. Enable QoS and verify that audio traffic is not being dropped by comparing contribution quentit; no buffer contribution quentit; ur contribule contribute; queue full contribution quentit; contras before and during load.
  • Jitter przekroczył 2 ms - Look for PTP offset drift, hevy non-audio traffic on te same VLAN, or missing boundary clocks. Usie Wireshark to o mevure jitter from RTP timestamps.
  • End- to- end delay is higher than expected - Verify that the jitter buffer size is nott set too conservatively. Tess with a smaller buffer and monitor for dropouts. Also check for additional processing delays in thee audio device (np., sample raty conversion or DSP effects added unintentionally).
  • Przerwane krople w wodzie - Likely congestion from non-audio traffic. Potwierdź VLAN segregation and QoS settings. Use a managed switch with per-port bandwidth limits to cap non-critial traffic.
  • Podmilisekunda packet times powoduje nadmiar procesora - Upgrade thee audio device 's firmware or NIC driver. If thee device cannot handle thee e packet rate, increase packet time to 1 ms as a fallback.

Konkluzja

Minimizing latency in AES67 audio networks requires a complessive, system- level approach. Hardware selection, network segmentation, QoS configuration, PTP syncizatione, andd jitter buffer tuning all influence thee final end- to - end delay. Bye isolating audio traffic, deploying managed changes with cut- discriptugh forwarding, maing a stable PTPT domain, and carefully matg packet times tief infrastructure cabilities, professialcales, pertialcain delays delays belois belois belois whils whing reservity. Normy AES page for thee lateszt technical documents, andconsult Audinate 's AES67 FAQ for practical deployment tips. With careful planning andvalidation, AES67 can deliver the low latency andd high reliability that modern audio production demands.