This is easy enough to test, but I wanted to get some feedback on what approach is better in terms of sound quality. I recently dove into computer audio with Roon and a microRendu (mR) w/LPS and have been thrilled with the result. Since my computer with attached HD’s are in the listening room I moved my core off the iMac and onto a SonicTransporter (ST) so I can turn the iMac completely off eliminating any noise. I use a Netgear gigabit switch after my Airport Time Capsule router with the ST plugged into this switch and then a 14M Ethernet cable over to the mR. All cabling is tested Blue Jean Cable cat 6.
I’m wondering if moving the ST and switch nearby the audio rig so there’s just 1M between the switch and the mR would improve sound quality? Does anyone have real world experience they’d like to share? With typical analog signals you usually want reasonably short cables and even shorter for USB. The guideline for Ethernet data is up to 100M. Common sense would say that shorter than 14M would be better, but perhaps with a small audio file over cat 6 it simply makes no difference whether it is a 1 or 14M cable.
Yes, and I can assure you that you will not be disappointed with the converters.
Very important, install the converters directly for streamer / player / computer …
DO NOT use a network switch after the converters.
Is the length of the cat 6 cable important in my application after the final optical conversion to ethernet? If electrical isolation is the key, then I only need a short piece of fiber to effectively isolate.
Meridian Audio discovered that RF, picked up by the Ethernet cable, could get into its product (ID40 card) and was unambiguously audible. In fact, they heard the difference first and it took them ages to realize that the audio quality improved when the Ethernet cable was removed. They then worked around the problem, initially in the Anniversary 808 by using RF suppressors on the cables, and subsequently with the development of the ID41 card which has much better E/M isolation.
So please don’t state as categoric fact that Ethernet cables don’t make a difference: they can, but not in the way that you imagined, and longer cables might pick up more RF. That’s not to say that you need to spend a fortune to resolve the problem – far from it – but well made and double-screened (screened twisted pair plus screened sheath) Ethernet cables are well worth considering.
Back to my original question if 14M of cat6 was pushing it for signal integrity and I heard for both of you no, it is not an issue for this length. It then shifted into an isolation discussion and then into cable quality. I have no skin in those arguments – just curious about the isolation angle since it was brought up.
Thanks for your input, but let’s leave it on the topic you both agree on – 14M of cat6 is not going to compromise signal.
Then that was a startlingly poor piece of electrical design in its original form. You’ll know very well that this has been comprehensively tested on squeezeboxes using Audiodiffmaker and there is zero effect from removing the ethernet cable from a player and playing out the buffer.
OMG, how can I force TIDAL to change the 14 million meters of cable and to remove the thousand of network switches on the way from their servers to my computer?
As I read this MUST be done by TIDAL, otherwise my sound quality is inferior.
… sorry for the sarcasm, but everytime I read such recommendation for ethernet or any other digital paket based transmission I want to bang my head to the next wall.
@anon55914447 - Mark, I’ve gained respect for your posts both here and on hifiwigwam, so my aim is not to pick a fight but merely to report my findings. Also, my electronics knowledge ended at A level so I am sure that others can provide more plausible explanations than I.
I found out about this on a French hifi site a year or so ago and it was then picked up by computeraudiophile and other places. I had already established that playing audio files locally rather than across the network sounded ‘better’ and that with no connection to the network, ‘better’ improved further. I had dismissed doing without a network because all my music lives on a nas and I couldn’t be ar*ed with all the synching of various HDs I’d have to do in order to make sure my collection was up to date. However, having perceived the improvement above I decided to give the optical connection a go. I didn’t even bother to try with the stock power supplies so I can’t comment on the difference but used a couple of Paul Hynes PSUs that I bought from ebay. The difference was surprisingly impressive and very much a positive one. I thought the sound was already pretty good but with the ‘isolation’ in place the sound had more focus and air. So much so that I bought another pair to see if further isolation might bring further improvements as it apparently had to others. I didn’t hear any improvement with the second pair so when I get around to it I’ll pop them on ebay.
I do think that your bringing up of the audiodiffmaker experiment is missing the point somewhat, in the same way that the misguided individual who carried out the trial in the first place was missing the point. As I understand it, no one is arguing that the bits are changing. No one is arguing that the sound is changing at a data level - this would indeed have been shown using ADM. I assume that the improvement has something to do with reduction of noise across the GND connection or perhaps the packets’ timing were being affected in some manner that affected the sound, somewhat akin to re-clocking a USB signal or changing the clock/oscillator in a CD player or DAC. As I say, speculation only.