If anyone has experience with:
—“audiophile” ethernet switch alone
—etherREGEN alone
—combination of audiophile” ethernet switch and etherREGEN
I would be grateful to hear about it. Both approaches seem to try to eliminate the propagation of noise. People seem to be pleased with each approach alone, but it isn’t clear to me whether both in combination would be audibly better.
I've got an EtherREGEN and I like it very much. It brought a significant improvement to my digital streaming front end utilizing both copper Ethernet and optical fiber over the combination of using an el cheapo generic fiber media convertor (FMC) and an audiophile-grade one, the Sonore OpticalModule (aka "OM"). Installing the OM last summer brought big improvements over using el cheapo, generic, FMCs because of the quality of its power supply, and the crystal oscillator clock is
much higher quality than those found in generic FMCs. Adding the ER into this network demonstrably improved the sound quality and presentation even more, with a notable reduction in the noise floor, increased spatiality in the presentation, improved imaging, more accurate and lifelike rendering of instrumental tones and harmonics, and voices in vocals. Full-scale classical music has improved presenation of the symphony hall and its acoustics, as placement of individual instruments in the orchestra.
All of the other switches referenced above are modifications of existing commercial or consumer-grade Ethernet switches and there a number of factors why they are sub-optimal to one extent or another for high-end audio applications, not the least of which is no management of high-impedance leakage currents and poor managment of clock phase noise. The AQVox is simply a modded D-link switch. For example, it does not have the isolation transformers in the RJ45 jacks specially designed and implement for use in audio applications. The SOtM Ethernet switch was designed for audio applicaions, but it is about 2X the price of ER, and it doesn't manage high-impedance leakage currents the way ER does. The EtherREGEN is a bespoke Ethernet switch designed from the ground up specifically for audio applications by John Swenson, who spent a career of 30 years working for Broadcom and Cisco in designing Ethernet solutions for network applications & products. If you look up my thread "EtherREGEN on the way" you'll find a lot more content on what differentiates this Ethernet switch from modded generic models.
As for running a combination of two or more Ethernet switches together "in serial", the reason this can provide an improvement in audio quality is due to a additive reduction of clock phase noise. When "stacking" devices using higher-quality crystal oscillators (clocks), one obtains a successive reduction in jitter/clock phase-noise traveling on ground planes and causing variation in clock edges, both overall and inside the chips. This why the differential isolation and reclocking works as well as it does in EtherREGEN. The other generic switches above do not do this because they were not designed to do this. In IT applications, these noise factors don't matter.
So, while its possible to obtain some improvement by "stacking" two switches in serial, as SOtM has demonstrated at audio shows, the benefits realized depend on the quality of the devices in the "stack". Personally, I've found using an Optical Module, which uses the same high-quality Crystek CCHD-575 oscillator as the EtherREGEN, provides a notable, but not major, increase in audio quality over just the EtherREGEN alone. Because of the very widely varying nature of configurations and devices used in networked digital audio systems, YMMV. Folks just have to try various implementations and see if they work for them. Using an EtherREGEN is a safe bet as there is a no-questions asked 30-day return policy. You can try it in your own configuration, and if you find it provides no benefit, obtain a full refund.
Hope this helps, and let me know if there's any other questions you have.
:congrats:
