sb0 changed the topic of #m-labs to: ARTIQ, Migen, MiSoC, Mixxeo & other M-Labs projects :: fka #milkymist :: Logs http://irclog.whitequark.org/m-labs
<whitequark> sb0: btw I made symlinks to ttyUSB_kc705 and *aux
<GitHub> [artiq] sbourdeauducq commented on issue #712: The throughput should be around 1MB/s (like the previous runtime), not 100k/s. https://github.com/m-labs/artiq/issues/712#issuecomment-294073954
<GitHub> [artiq] whitequark commented on issue #712: Yes. Please multiply all numbers I mentioned previously (except 25 kbps) by a factor of 10, I misremembered, and could not find the logs as it was a while ago. https://github.com/m-labs/artiq/issues/712#issuecomment-294074267
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
futarisIRCcloud has joined #m-labs
klickverbot has quit [Ping timeout: 252 seconds]
ohsix has quit [Ping timeout: 260 seconds]
ohsix has joined #m-labs
mumptai has joined #m-labs
hobbes- has quit [Quit: ZNC - http://znc.in]
hobbes- has joined #m-labs
<GitHub> [artiq] sbourdeauducq commented on issue #719: In addition to what Robert said: the applet doesn't know the master address when embedded, all the data goes through the dashboard using the IPC pipe. And when the applet is used in the browser, there is no master.... https://github.com/m-labs/artiq/issues/719#issuecomment-294136973
<GitHub> [artiq] sbourdeauducq commented on issue #719: In addition to what Robert said: the applet doesn't know the master address when embedded, all the data goes through the dashboard using the IPC pipe. And when the applet is used in the browser, there is no master.... https://github.com/m-labs/artiq/issues/719#issuecomment-294136973
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
<GitHub> [artiq] dhslichter commented on issue #712: So is this due in part to the reimplementation of the runtime in Rust? https://github.com/m-labs/artiq/issues/712#issuecomment-294174966
<GitHub> [artiq] dhslichter commented on issue #712: So is this due in part to the reimplementation of the runtime in Rust? https://github.com/m-labs/artiq/issues/712#issuecomment-294174966
klickverbot has joined #m-labs
klickverbot has quit [Ping timeout: 258 seconds]
klickverbot has joined #m-labs
klickverbot has quit [Ping timeout: 260 seconds]
klickverbot has joined #m-labs
<GitHub> [artiq] whitequark commented on issue #712: > So is this due in part to the reimplementation of the runtime in Rust? Does that slow the soft CPUs?... https://github.com/m-labs/artiq/issues/712#issuecomment-294226559
mumptai has quit [Quit: Verlassend]
klickverbot has quit [Ping timeout: 268 seconds]
<GitHub> [artiq] dhslichter commented on issue #712: Ack. But to put some numbers on it, a single RTIO event is ~10 bytes, so a longer DMA sequence with 10,000 pulses (20,000 events) would take ~200 ms to send even at 1 MB/s, so increasing the speed beyond that level (as discussed above, by making various possible design modifications) is definitely something to put in the "future improvements" column. https://github.com/m-labs/artiq/issues/712#issuecommen
futarisIRCcloud has joined #m-labs
<whitequark> sb0: well, this DMA batching is problematic
<whitequark> it can only be done for rtio_output, not rtio_output_wide
digshadow has joined #m-labs
digshadow has left #m-labs [#m-labs]