Would a higher download speed (in this case, from 20mbps to 34mbps) result in a better SQ?
No.
A 24/192 stream only requires 9Mbps.
Of course more bandwidth means if someone else in the household is also on video or streaming that your stream won’t drop. If the cost is nominal go for it.
But in pure SQ terms no difference.
It’s latency to the source which causes streaming problems (drops etc.)
Modern streamers have large buffers and your current 20Mbps connection is sufficient even with hires.
A streamer needs much less than 20mb for an audio stream, so no, assuming that all of the bandwidth is available to the streamer, you don’t need more.
You should also consider latency, which is more likely to affect an audio stream from a remote server than download speeds.
Ha ha same content as my post.
We must be right
I will go against the flow.
Possibly, yes because you might be getting less than your specified download speed. This could be because your neighbours are using part of the bandwidth and there is a high contention ratio or there are other devices in your house using up some of the bandwidth.
I’d use a broadband speed checker to see what you are actually getting in terms of download speed and latency during peak times in the evening.
The stutter might not be obviously audible. If varying bandwidth causes some frames not arrive on time for very short periods then that would give rise to some distortion.
The pause might be very short. There might be several short pauses over a period of time. Bear in mind all the discussions on the impact of network loading on sound quality not to mention gold plated Ethernet connectors.
All this talk of stuttering and dropouts is missing the point. The OP has a 2nd gen Naim streamer with a generous buffer so a 20MB connection is going to be fine as long as there isn’t very significant latency which can cause a buffer to empty. That is far more likely to be a problem with the 1st gen streamers and their unusually small buffers.
Even if there was occasional stuttering it’s debatable whether it could be described as reduced sound quality as normal playback would be unaffected.
I do believe (and as an amateur I admit an element of conjecture here) that any irregularities in the timing of the data stream, possibly caused by internet latency, may cause small losses of sound quality in some systems as the streamer works a little harder to process it. Possibly remedies for this might include using a Cisco Catalyst switch which will have a quality clock which can smooth out the timing, using a 2nd gen streamer which the OP is already doing, and using a separate DAC in order to decouple the analogue audio signal from any such effects. @Simon-in-Suffolk is free to shoot me down in flames if he thinks I’m off the mark here as he has been responsible for tracking down issues like this in the past.
Hi Chris
Pretty good summary and definitely don’t want to attempt to shoot you down . I suggest probably best be careful about not confusing switch ethernet serialisation clocks with audio sample data clocks - they are chalk and cheese. Switch clocks will have a degree of phase modulation noise - and this noise energy (think a tiny micro powered FM transmitter) can couple into the streamer at infinitesimally small levels but potentially subtly modulate ground planes or sample clocks.
With respect to cloud streaming sound quality I think the more likely variables will be compression and treatment variables the cloud providers require or apply themselves to provide the LUFS level they define for their service - that is the average weighted loudness energy over a period of time of the audio signal.
Quite. That’s why I said, “A streamer needs much less than 20mb for an audio stream, so no, assuming that all of the bandwidth is available to the streamer, you don’t need more” and, “…The OP has a 2nd gen Naim streamer with a generous buffer so a 20MB connection is going to be fine as long as there isn’t very significant latency…”
It’s a little naïve to believe the mysteries of hifi sound quality can the explained/determined by dividing and multiplying a few abstract numbers together.
I use a streamer. It draws 500mA, yet it specifies a 2A power supply is required.
Go figure.
If it’s got a transformer there is an inductive load which will draw more current under certain conditions, power up for example.
The fuse is there to protect if things are obviously going wrong and it needs to make things safe.
So, no, then…
You came to the wrong place if you were hoping for a short answer! But yes, the short answer is that 20 meg is fine.
I have connected my NDX2 wirelessly, is there any chance of improvement by going wired connection, below is the speed of the internet and other details which I do not have any technical knowledge.
Those results say nothing about how strong the WiFi signal is at the NDX2. But assuming it is pretty strong, then I don’t think you would get any SQ gain from using Ethernet instead of WiFi. It would be different if you were talking about an NDX though.
As David says, it’s all about how good your wifi is. If the NDX2 is within reach of the router you could always run a temporary ethernet cable and see whether you get improvements in sound, connectivity or both. The received wisdom seems to be that if you can use ethernet, then use it.
With the Linn Klimax DS3 / Katalyst, it can only be hardwired.
Presently on 200mbs full fibre package, but usually higher than this. Even the WiFi is good using a Mesh Network.
This is the Wi-Fi results
This is the Hardwired results
Interestingly, the Wi-Fi results are better for download than the Hardwired results.
DG…
Even though I have virgin fibre 250 the streamer I use is limited to 100mbps. I think Naim limit their streamers to 100.
If your streamer Ethernet port is blinking orange it’s probably working at 100. Green would be 1000gbps.
For some reason (I can’t remember the reason) I think my router is 56mbps max when using WiFi.
Because its old?