Qhwoeprktiyns Posted October 21, 2018 Share Posted October 21, 2018 Just read about this new protocol being developed in Japan: https://www.diretta.link/top.html Has anyone heard of this and understands how it works ? Link to comment
Qhwoeprktiyns Posted October 22, 2018 Author Share Posted October 22, 2018 Thanks for your contribution. Link to comment
YU Posted October 23, 2018 Share Posted October 23, 2018 Flattening target process by small send data size and short between time. Link to comment
Qhwoeprktiyns Posted October 23, 2018 Author Share Posted October 23, 2018 Thanks. So what is the theoretical benefit? Less processing on the client side? Link to comment
YU Posted October 23, 2018 Share Posted October 23, 2018 Target processing is small. Load fluctuation to target power supply is small Link to comment
Miska Posted October 23, 2018 Share Posted October 23, 2018 1 hour ago, YU said: Flattening target process by small send data size and short between time. Smaller the data block, higher the protocol overhead and CPU load... Quote Thanks. So what is the theoretical benefit? Less processing on the client side? Smaller the packet, more processing time it takes on the client side because of involved overheads. Quote Target processing is small.Load fluctuation to target power supply is small And in worst case you send packets so frequently that the packet rate falls into audio band. Like UAC's 8 kHz... Then it shows up easily in the measurements... P.S. Is this yet another protocol that doesn't work over WiFi? (like AES67/Ravenna/Dante/AVB) Signalyst - Developer of HQPlayer Pulse & Fidelity - Software Defined Amplifiers Link to comment
YU Posted October 23, 2018 Share Posted October 23, 2018 That's right. If the packet size under the MTU size, processing high. multiple packets can not be processed at once,then tasklet occur frequently. but processing per one transaction is short process time. Diretta is optimize best MTU size. Target do not manage buffer control. designed to reduce processing, use UDP etc. case 22Mhz DSD Send Sycle = 500usec size = 1494 byte 44.1Khz PCM Send Sycle = 2000usec(limit) size = 767 byte Link to comment
Miska Posted October 23, 2018 Share Posted October 23, 2018 6 minutes ago, YU said: That's right. If the packet size under the MTU size, processing high. multiple packets can not be processed at once,then tasklet occur frequently. Diretta is optimize best MTU size. Target do not manage buffer control. designed to reduce processing, use UDP etc. There are good zero-copy TCP offload engines that handle large DMA transmissions without CPU involvement... Just do full speed transmission of a file over 1 Gbps or 10 Gbps ethernet and observe which method has lowest CPU load... This is quite well optimized, since practically entire Internet relies on fast transmission of huge amounts of data over TCP. Signalyst - Developer of HQPlayer Pulse & Fidelity - Software Defined Amplifiers Link to comment
YU Posted October 23, 2018 Share Posted October 23, 2018 It has a different purpose. TCP need send ACK..... designed to suppress the effect on the target. Only use of the local network. It is not aimed at increasing the throughput. We are planning to implement FPGA and Hardware. ZeroCopy is a future task Link to comment
Qhwoeprktiyns Posted October 23, 2018 Author Share Posted October 23, 2018 I assume you are from Megatech, in Tokyo? If so, can you tell us more about the release dates? I understand there are two components: sink and synch. Are both available under Linux? Link to comment
YU Posted October 23, 2018 Share Posted October 23, 2018 Yes we had linux liblary.(company license) we will release binary on ALSA Sink and Sync after one month Link to comment
Qhwoeprktiyns Posted October 23, 2018 Author Share Posted October 23, 2018 Thanks. Perhaps you could let us know in this thread? I am sure many will be interested in testing this. Link to comment
Miska Posted October 23, 2018 Share Posted October 23, 2018 1 hour ago, YU said: case 22Mhz DSD Send Sycle = 500usec size = 1494 byte 44.1Khz PCM Send Sycle = 2000usec(limit) size = 767 byte Now the challenge with this is similar to USB Audio Class... 500 µs = 2 kHz 2000 µs = 500 Hz While USB Audio Class uses 125 µs = 8 kHz These fall into audio band and can be frequently detected from the DAC's analog output. Signalyst - Developer of HQPlayer Pulse & Fidelity - Software Defined Amplifiers Link to comment
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now