@almarg After following those links I just ordered a few CAT7 cables to compare to the CAT5 I’ve been using between NAS, Ethernet hub, and microRendu. I suspect some of the variations observed between Ethernet cables are system dependent. Moreover, the entire chain of digital being has evolved since those threads. @douglas_schroeder What EN cables have you tried that make THE difference? |
Doug, thanks for the missionary work in this area. It may the missing link in CPU audio. Puns intended, of course. Best, Dave |
I just received from Amazon a 50' Theo & Cleo CAT7 and a five pack of 3' PiperCables TM Pro Series CAT7. The Pipers were advertised as tested on a Fluke meter, which at least implies better than no-name quality. These replaced generic CAT5 between a QNAP NAS, a Cisco EN switch, and a microRendu into a Esoteric K-01X. Wow, a meaningful upgrade! Larger stage, increased resolution, weightier LF foundation, super quiet background. I can't recall a better $50 spend. |
Au contraire, the difference that I’m hearing between EN "looms" <$50 is remarkable and refreshingly cost-effective. I hope that Doug’s forthcoming review of extravagantly priced EN cables includes comparisons to some of the cost-effective shielded CAT7 and CAT8 mainstream brands mentioned favorably above and to optically isolated EN transport. |
(Sigh) For this thread it would be nice to confine the "great cable debate" to Ethernet-- which is at least a fresh context for cable discussions. To paraphrase The Kinks, other Kable Kontroveries are sufficiently covered in the archives. |
@jinjuku Two nice hi-res files and lots of fun listening. I could hazard a guess as to which sounds the best, but for a meaningful test kept to one variable we need to compare identical songs across the two cables. Any possibility of your replicating that scenario? |
@unreceivedogma No foul. I’m a cable skeptic myself. However, I’m more skeptical about the high price of commercial cables than about differences in sound between cables. All DIY cables in my system. |
@jinjuku Can your test files be downloaded for playback? If so, let us know how to find them. |
@jinjuku I’m afraid that your approach violates the basics of a proper ABX test. As ClarityCap’s OEM Sales Manager, I’m privy to research that we did with focus groups ABX testing coupling capacitors. This was done in conjunction with the graduate-level psychoacoustics dept. at a UK University. It’s critical to present listeners with a consistent sample and to allow them the leisure of comparing samples under their own control without the dissonance of toggling arbitrarily between samples. Redo the recordings and I’ll be happy to participate. |
That test is more Where's Waldo than ABX. Post two files of the same music, each recorded with a different Ethernet cable, and the opportunity to compare them at leisure. |
Doug, if you listen to Jinjuku’s two posted hi-res files you will understand that he knows how to make an excellent recording from ADC to DAC to ADC. If the files are played back through your system, that’s all that’s important here. |
@jinjuku The main problem with your test(if I understand it correctly) is that it is a multivariable experiment. You swap cables 7-10 times at regular intervals throughout one track. Unfortunately the music is changing on top of the cable changes-- confusing the listener by precluding direct comparisons. What you need to do is to repeat the track(or a segment of that track) with the same cable across the duration of each sample. Then all that the listener must do is to identify and appraise the deviant segment. And the odds at 7:1 are still in favor of the house... |
Gambling introduces one more variable. While some "claimants" may be stimulated to listen closely, others may become distracted and confused by stress. I hypothesize that the latter case characterizes most audiophiles... The best thing would be to record seven samples, post the files, and gather stats across a broad number of listeners to the downloads. That's how Fremer does it on Analog Planet, which is good enough. |
The only problem I have with Douglas's review is that his system passes Ethernet through household AC wiring. I haven't attempted that method or studied the academics, but it could be the weak link that invalidates findings with boutique EN cables at the end points. |
@douglas_schroeder I haven’t been able to find much discussion for good or bad on high-end forums about IOP. The point is occasionally made that it can adversely affect certain components(a Burmester was mentioned). Others report or speculate that Ethernet over an AC circuit underscores the importance of good power conditioning to address high frequency noise. Others report affects of noise from appliances, motors, etc. running on the same circuit or across the mains. I agree that if you use IOP and are still able to hear differences between wires --and your computer audio system is also performing at an historic high-- then your cable comparison is valuable. However, one could raise a question as to the extent of interaction between the end cables and the IOP. Would the outcome have been different if these cables were used in a system without IOP? My CPU setup (QNAP i5 NAS> SoTM-modified D-Link hub> SOtM SMS-200ultra> SOtM USBultra > USB> Esoteric K-01X), is wired based more upon principles accepted for audio than for computer networking. All music devices and the network are inside the audio room on their own bridged subnet connected by short Ethernet cables. I found that this subnet approach improved upon a 50’ Ethernet to an internet router in another room. Performance issues may have been caused by the long cable run, the cable itself(generic CAT7) or perhaps by the lack of isolation from noise and/or OSI Layer 3 activity propagated by the internet router. I’ll add that the SoTM-modified hub contributes to the "upgrade": that hub and the two other SoTM components on the Ethernet, are all slaved to a common master clock. Obviously computer audio is a minefield of interdependent variables, of which ethernet cabling is one. |
