EtherREGEN - Formal Video Imaging and Audio Review

[pkane2001]

 

On 4/10/2020 at 9:16 PM, DarqueKnight said:

Introduction

Figure 1. EtherREGEN A and B port sides. Photo from UpTone Audio website.

 

An audible improvement in clarity, detail, and spatiality was achieved by placing a Cisco SG110D-08 Ethernet switch ($100 list price, $36 street price) between my two channel stereo system's digital music player and NAS, thereby isolating them from the wireless access point they were previously connected to. Based on those results, I began looking for a formal solution for a noise-reduced, audio grade Ethernet switch. That search lead me to the UpTone Audio's EtherREGEN audiophile Ethernet switch ($649). I was intrigued by the concept of an Ethernet switch designed specifically for computer audio, rather than a modified data communication switch. Ethernet is an inherently low noise transmission protocol that provides excellent noise filtering for data communications. However, the electrical noise that get past standard Ethernet noise cancelling techniques can negatively impact the performance of high resolution computer audio and computer video systems.

 

Figure 2. Cisco SG110D-08 Ethernet switch under evaluation in May of 2019. The Cisco switch's performance was improved by replacing its switch mode power supply with a TeraDak DC-30 linear power supply. A blue filter was placed over the TeraDak LPS's LED to tone down its brightness.

 

Basically, the EtherREGEN is an Ethernet switch with filtering apparatus that addresses noise modes that affect audio (and video) quality. Video performance is parenthetically mentioned  because the EtherREGEN is marketed strictly as a computer audio device. The EtherREGEN's design is based on an "Active Differential Isolation Moat ("ADIM"), which filters noise between the input "A" ports and the single "superclean" "B" output port.

 

Over a six week period. On the day of the EtherREGEN's arrival, after a warm up of three hours, a "sneak peek", "quickie" listening session was conducted before taking it to the home theater system. I was very impressed by what was heard.

 

The top center of the EtherREGEN’s case reached a constant temperature of 115 degrees Fahrenheit when idle. The temperature ranged from 116 to 126 degrees Fahrenheit when working. Temperatures were measured by a Fluke model 561 infrared thermometer.

 

UpTone Audio recommends a break in period of 100 hours. I heard an improvement in clarity, detail, bass performance, and tactile sensation after 24 hours of playback. I did not perceive further improvements in these areas with additional hours of playback.

 

A lot of information pertaining to the environment and circumstances of this evaluation are provided. Results are summarized in the "Discussion of Video Results" and "Discussion of Audio Results" sections.

 

Video Trials

The home theater system is configured as follows:

 

Video files are served from a Synology DS918+ NAS to an Oppo UDP-205 universal disc player. The video display is an LG OLED65E8PUA 65” OLED television. The television has not been professionally calibrated, but has been tweaked using recommended settings found online. The DS918+ and UDP-205 were connected to an Amped Wireless AC 2600 Athena-EX range extender via generic “Cable Matters” brand (Amazon) Cat6a Ethernet cables. The range extender also provides wireless connectivity to my home’s 5 GHz wireless network. The Oppo player is connected to the OLED TV via an AudioQuest Cinnamon HDMI cable. The Oppo’s stock internal switch mode power supply was replaced with an upgrade Oppomod linear power supply. The home theater system is served by two dedicated 20 amp AC circuits. All electronics are fed ultra pure AC power by a PS Audio P5 AC regenerator and a PS Audio P10 AC regenerator.

 

Experimental Setup

A Nikon D800 36.3 megapixel digital camera was placed 49” from the center of the TV screen. Multiple pictures of the paused first scene from the ultra high definition Blu-ray disc “Alien Covenant” were taken with various equipment configurations. The first scene of the movie is a tight close up of the character David’s eye at 00:01:17. I often use this scene to evaluate differences in digital equipment performance. The lens was a Nikkor 24mm-70mm zoom lens. All pictures were taken at ISO 1000, with an exposure of 1/100th second at F3.2.

 

Figure 3. Camera setup

 

In one experiment, picture quality differences from the following six equipment configurations were evaluated (Ethernet cables in each case were Amazon Cable Matters Cat6a):

 

1. DS918+ NAS with its stock switch mode power supply; NAS and Oppo player connected to the Amped Wireless range extender.

 

2. DS918+ NAS with a TeraDAK DC-200VA linear power supply; NAS and Oppo player connected to the Amped Wireless range extender.

 

3. NAS and Oppo player connected to a Cisco SG110D-08 8-port Ethernet switch, with the Cisco’s stock switch mode power supply. An Ethernet cable was connected from one of the Cisco’s ports to one of the Athena range extender’s ports.

 

4. NAS and Oppo player connected to a Cisco SG110D-08 8-port Ethernet switch, with the Cisco’s stock switch mode power supply replaced by a TeraDak DC-30 linear power supply. An Ethernet cable was connected from one of the Cisco’s ports to one of the Athena range extender’s ports.

 

5. NAS and Oppo player connected to The EtherREGEN switch, with the EtherREGEN’s stock switch mode power supply. An Ethernet cable was connected from one of the EtherREGEN’s ports to one of the Athena range extender’s ports. The NAS was connected to one of the A ports. The Oppo player was connected to the B port.

 

6. NAS and Oppo player connected to The EtherREGEN switch, with the EtherREGEN’s stock switch mode power supply replaced by a TeraDak DC-30 linear power supply. An Ethernet cable was connected from one of the EtherREGEN’s ports to one of the Athena range extender’s ports. The NAS was connected to one of the A ports. The Oppo player was connected to the B port.

 

 

Figure 4. Left - picture with NAS and stock switch mode power supply. Right - picture with NAS and TeraDak DC-200VA linear power supply.

 

There were drastic differences in detail, color saturation, and clarity between pictures with the NAS's stock switch mode power supply and replacement upgrade linear power supply.

 

Figure 5. Left - picture with Cisco switch and its stock switch mode power supply. Right - picture with Cisco switch and TeraDak DC-30 linear power supply.

 

Figure 6. Left - picture with EtherREGEN switch and its stock switch mode power supply. Right - picture with EtherREGEN switch and TeraDak DC-30 linear power supply.

 

In these small, lower resolution pictures, the differences, particularly the perception of depth, among the images in figures 5 and 6 are not as apparent as when looking at the video display or at the original, higher resolution pictures. Examining enlargements at pixel levels provides more insight. Figures 8 through 10 show the differences in a section of the rim of the eye’s iris. The specific section is shown in figure 7.

Figure 7. Section of pixel level comparisons for equipment configurations 1 - 6.

 

Figure 8. Left - NAS pixel level detail with stock switch mode power supply. Right - NAS pixel level detail with TeraDak DC-200VA linear power supply.

 

Figure 9. Left - Cisco switch pixel level detail with stock switch mode power supply. Right - Cisco switch pixel level detail with TeraDak DC-30 linear power supply.

 

Figure 10. Left - EtherREGEN switch pixel level detail with stock switch mode power supply. Right - EtherREGEN switch pixel level detail with TeraDak DC-30 linear power supply.

 

Figure 11. Worst to best: Left - Pixel level detail, NAS with stock switch mode power supply and Oppo player connected to wireless range extender. Right, NAS with TeraDak DC-200VA linear power supply and Oppo player connected to EtherREGEN with TeraDak DC-30 linear power supply. The overall result of using equipment with increasingly better electrical noise abatement was a picture with increasing detail, depth, sharpness, and color accuracy.

 

Ethernet Cable Trials in Home Theater System

Prior to this experiment, two Ethernet cable trials were conducted in my two channel stereo system. In March of 2019, the five Ethernet cables shown in figure 12 were evaluated with the cables plugged into a wireless range extender.  In May of 2019, the experiment was repeated with the placement of the Cisco switch between the digital source components and the wireless range extender. In both trials, no difference in any aspect of stereophonic performance was heard with a change of Ethernet cable. Improvements in clarity, detail, and spatiality were heard with the Cisco switch.

 

Figure 12. In March and May of 2019, five Ethernet cables were evaluated, with no perceived difference in audible or tactile sensation among them. Left to right: AudioQuest Cinnamon Cat7 ($100 - 5 feet), Bluejeans Cable Cat6a ($18.50 - five feet), Amazon Cable Matters Cat6a ($3.40 - 5 feet), Synology stock Cat5e cable that came with the NAS ($0 - five feet), Belkin Cat5e ($1.87 - five feet).

 

Ethernet cable trials were conducted again in March of 2020 with the five cables shown in figure 13. An Amazon Cable Matters Cat6a cable ran from the NAS to one of the EtherREGEN's "A" ports. Another Amazon Cable Matters Cat6a cable ran from the wireless range extender to one of the EtherREGEN's "A" ports. The cable between the EtherREGEN's "B" port and the Oppo's Ethernet port was the only change. The NAS and EtherREGEN stock switch mode power supplies were used in the Ethernet cable trials.

 

Figure 13. Top to bottom: Revelation Audio Labs Prophecy CryoSilver Cat8+ ($599 1.25 m), AudioQuest Vodka Cat7 ($450 1.5 m), AudioQuest Cinnamon Cat7 ($100 1.5 m), Blue Jeans Certified Cat6a ($18.50 5 ft), Amazon Cable Matters Cat6a  ($3.40 5 ft).

 

Figure 14. Pixel level pictures, left - Amazon Cable Matters Cat6a, right - Blue Jeans Certified Cat6a.

 

Figure 15. Pixel level pictures, left - AudioQuest Cinnamon Cat7, right - AudioQuest Vodka Cat7.

 

Figure 16. Pixel level pictures, worst to best, left - Amazon Cable Matters Cat6a, right - Revelation Audio Labs Prophecy CryoSilver Cat8+.

 

Figure 17. Pixel level pictures, side by side. Left to right: Amazon Cable Matters Cat6a, Blue Jeans Certified Cat6a, AudioQuest Cinnamon Cat7, AudioQuest Vodka Cat7, Revelation Audio Labs Prophecy CryoSilver Cat8+.

 

Figure 18. Were it not for the EtherREGEN's 100 Mbps output limitation, I would have ordered a second unit for my home theater system.

 

The EtherREGEN had great performance with DVD and Blu-ray source material. On 4K material that exceeded the 100 Mbps limitation, the audio would drop out, but the video would continue unscathed. The audio would return immediately after the bit rate fell below 100 Mbps. The intro theme of "Quantum of Solace" spent a significant amount of time above 100 Mbps, which meant frequent and annoying audio dropouts. But of course, I was using the EtherREGEN for a purpose it wasn't designed for.

 

Discussion of Video Results

Picture quality, particularly the perception of depth improved significantly as I went from the wireless range extender, to the Cisco switch, to the EtherREGEN. Improvements in color rendition, fine detail, and dynamic range were easily discernible. Improvements in these areas were extended when the switch mode power supplies of the Synology DS918+ NAS and EtherREGEN were replaced with linear power supplies. In the audio area, the clarity and detail of dialog was closer to the sound of actual people speaking in front of me. The rumble and growl of low frequency effects was more finely detailed, and there was greater tactile sensation from the front stage. On Blu-ray movies with well-recorded surround tracks, I found myself noticing more information and details in the side and rear channels. A thicker, more immersive sound stage was realized with multichannel music.

 

Further improvements, in both home theater audio and video performance, were realized with better Ethernet cables. The exception to this was the AudioQuest Cinnamon cable. It outperformed the Blue Jeans cable in audio performance, but the Blue Jeans cable outperformed the Cinnamon in video performance. Compare figure 14-right to figure 15-left and the second and third pictures of figure 17.

 

Figure 19. The Cisco SG110D-08 Ethernet switch was moved from my two channel system to the home theater system. Its stock switch mode power supply was replaced with a TeraDak DC-30 linear power supply.

 

Audio Trials

The digital front end of my two channel stereo system consists of a Bryston BDP-3 digital player, Synology DS718+ NAS, dCS Puccini word clock (master clock), and dCS Debussy DAC. The BDP-3 does not have a word clock input. However, the Puccini word clock is able to lock on to and synch the BDP-3's output via USB cable connection. Therefore, both the digital player and DAC are locked to the same master clock.

 

Most music files are in DSD64 format. All vinyl records were converted to DSD64 format with a PS Audio NuWave Phono analog to digital converter and VinylStudio software. Three 20 amp AC circuits serve the two channel system: one each for the Pass Labs X600.5 monoblock power amps, and one for the source components. Each AC line is terminated with a PS Audio Soloist SE in-wall passive power conditioner. The power amplifiers are plugged into the wall. The source components are powered by a PS Audio P10 AC regenerator. The NAS's stock switch mode power supply was replaced with a TeraDak DC-150VA linear power supply. The EtherREGEN's stock switch mode power supply was replaced with a TeraDak DC-30 linear power supply.

 

Critical listening evaluation of the EtherREGEN commenced after over four weeks and over 500 hours of playback had been logged.

 

Figure 20. Block diagram of two channel stereo system digital front end.

 

Figure 21. Revelation Audio Labs split leg USB 2.0 cable. After the USB connection between the BDP-3 and the U-Clock has been set up, the power signal is typically not required any longer. Separating the audio signal wires and power signal wires allows the power leg, and its associated electrical noise, to be disconnected.

 

Stereophonic Performance Evaluation Methodology

My critical listening methodology is based on the sensory science concept of descriptive analysis. It consists of:

1. Repeated listening to the same familiar, well-recorded, songs.

2. Mapping the placement and relative size of sounds, voices, and instruments

(sound images) in the sound stage.

3. Making notes of the quality and characteristics of the sound images within the sound stage (image weight, tactile sensation, sound level, clarity, detail, articulation, etc.).

4. Comparing before and after listening notes and sound stage maps and noting differences, if any.

 

Figure 22. Lateral sound stage chart.

 

Figure 23. Aerial sound stage chart.

 

Figure 24. Example of completed sound stage charts.

 

Figure 25. View of two channel stereo system from listening position.

 

Figure 26. The room is 21 feet by 17 feet, with speakers along the long wall. The listening position is 12 feet from the front plane of the speakers. The digital player is controlled via its Internet interface by the docked tablet computer on the table.

 

Figure 27. Amplification and AC regenerator.

 

Discussion of Audio Results

Even with standard computer-grade Ethernet cables, the noise isolation provided by placing the audio components on a separate, dedicated Ethernet switch provided improvements in stereophonic holography, more ambient information retrieval, and deeper, more articulate, and more tactile bass. This was true with the Cisco SG110D-08 switch, and is true to a much greater extent with the EtherREGEN switch.

 

With the progression from the Cable Matters cable, to the Blue Jeans, to the AQ Cinnamon, to the AQ Vodka, to the RAL, increasing clarity, detail, and sound stage depth were heard. The qualities that separated the RAL from the other four was the RAL’s greater conveyance of tactile sensation, greater sense of three-dimensional space, and greater image weight. Unlike the other Ethernet cables reviewed, the RAL cable initially caused a small loss in image weight, clarity, and detail. Those losses had disappeared after 48 hours of playback. RAL recommends a break in period of 250 hours prior to critical listening. Further improvements in image weight, bass performance, clarity at the rear and sides of the sound stage, and tactile sensation were realized when the Blue Jeans Ethernet cable between the NAS and the EtherREGEN was replaced with the AQ Vodka cable. Since the RAL outperformed the Vodka, a second RAL Ethernet cable was purchased.

 

On Gregg Karukas' "Healing Song" ("Heatwave" CD), Gregg Karukas' piano notes acquired a more percussive quality and there was more sustain and decay of the notes. Kirk Whalum's tenor saxophone notes acquired an enhanced clarity and enhanced "reedy" quality with more vibrato. There was more detail in the background percussion instruments.

 

On E.S. Posthumous' "Isunova Pi" ("Cartographer" CD), the thunderous bass notes at the beginning were more thunderous, with greater vibration against my legs and chest. The drum beats were placed about 2 feet further back in the sound stage, yet produced more tactile sensation. There was more vibration from bass notes coming through the seat and armrests. The choir had greater detail and there was more separation between the choir and the other sound images.

 

On Sheila E and The E. Train's "Bahia" ("Heaven" CD), the bird chirps at the beginning were "faster", more detailed, and the echoes of the chirps were more defined. The electric bass had more weight and growl. The electric piano notes had more weight. The clarity and detail of vocals was enhanced as evidenced by more chest and breathing sounds.

 

On George Howard's "Diane's Blues" ("Attitude Adjustment" CD), electric bass notes acquired more rumble, growl, and articulation. The principle instrument, the soprano saxophone, had more space around it and the notes had more of a "reedy" edge. Drum notes were more articulate and produced more tactile sensation.

 

On Isaac Hayes' "Cafe Regio" ("Shaft" soundtrack CD), I was drawn to the sound of the tambourine while in another room. This is a song of average recording quality that I do not use for critical evaluation. It is part of one of the playlists used during the break in period. The tambourine had greater space around it, sounded louder, and the sounds of the individual metal jingles ("zills"), and the natural decay of those metal jingles, was very clearly defined.

 

On Sonny Rollins' bebop classic "Strode Rode" ("Saxophone Colossus" SACD), this marked the first time I heard true depth in a mono recording. As with previous listenings of this song, all the sound images were confined to the center of the sound stage. The difference now was that, rather than everything being jumbled together in a "big ball" in the center of the sound stage, Rollins' tenor saxophone was in front. Behind him was Tommy Flanagan's piano, which was wider than the other three sound images. Behind the piano were Max Roach's drums. Behind the drums was Doug Watkins' double bass. All instruments had increased image weight. Even though the musicians were all playing at a fast pace, individual notes were clearly articulated within individual images and were clearly distinct from the other instruments.

 

Figure 28. Snakebitten...twice. One RAL Ethernet cable was great. Two were fantastic. I didn't mind the snake bites though, because the snakes enhanced the production of Such Good Sound.

 

The EtherREGEN's manual states (page 7):

 

"If you are going to hear a difference with an Ethernet cable, it will more likely be with the 'B'-side cable. This is also a good one to keep short."

 

Unfortunately, for my long-suffering wallet, I heard a difference between Ethernet cables on both the "A" and "B" sides.

 

Associated Equipment - Home Theater System

LG OLED65E8PUA OLED TV

Oppo UDP-205 Blu-ray Player With Oppomod Linear Power Supply Upgrade

Bryston BDP-1 Digital Music Player

Synology DS918+ NAS With TeraDak DC-200VA Linear Power Supply

Yamaha CX-A5100 Surround Sound Preamp/Processor

Adcom GFA-565se Power Amp (325 Watts Per Channel/4 Ohms)/Front Speakers

Adcom GFA-565se Power Amp (325 Watts Per Channel/4 Ohms)/Center Channel Speaker

Adcom GFA-565se Power Amp (325 Watts Per Channel/4 Ohms)/Surround Speakers

Polk Audio LSiM 706C Center Channel Speaker

Polk Audio LSiM 707 Front Speakers

Polk Audio LSiM 702 F/X Surround Speakers

SVS PB12 Ultra/2 Subwoofer

PS Audio P5 AC Regenerator (Plasma TV, Blu-ray Player, Tivo, Digital Music Player)

PS Audio P10 AC Regenerator (Preamp/Processor, Power Amplifiers)

PS Audio AC-5 Power Cords For AC Regenerators

PS Audio Statement SC Power Cords For All Electronics

Revelation Audio Labs Prophecy Cryo-Silver Dual Conduit USB Cable (BDP-1 To UDP-205)

AudioQuest Cinnamon HDMI Cables (UDP-205 To CX-A5100, CX-A5100 To OLED TV)

Douglas Connection DCF-92 Speaker Cables, Front, Center Speakers

DSR Silverline (Amp To Wall), Monster UL/CL3 (In-Wall) Speaker Cables, Surround Speakers

Blue Jeans LC-1 In-Wall Subwoofer Cable

Blue Jeans Cat6a Ethernet Cables For NAS And Oppo Blu-ray Player

Cisco SG110D-08 Ethernet Switch For NAS And Oppo Blu-ray Player With TeraDak DC-30 Linear Power Supply

Signal Cable Silver Resolution XLR Interconnects (Preamp To Power Amps)

Two Dedicated 20 Amp AC Circuits Terminated With PS Audio Soloist SE In-Wall Passive Power Conditioners

Salamander Synergy Quad 30 Audio Credenza

Dell Venue Pro 11 7140 Docked Tablet Computer For System Control

Logitech Harmony One Remote

 

Associated Equipment - Two Channel Stereo System

Bryston BDP-3 Digital Player

dCS Puccini U-Clock word clock

dCS Debussy DAC

Synology DS718+ Network Attached Storage With TeraDak DC-150VA Linear Power Supply

Pass Labs XP-30 Three-Chassis Line Level Preamplifier

Pass Labs X600.5 Monoblock Power Amplifiers (1200 Watts Per Channel/4 ohms)

PS Audio PowerBase Isolation Platforms For DAC, BDP-3, U-Clock, And P10 AC Regenerator

22 Pound Granite Slabs For Coupling Source Components To Isolation Platforms

Black Diamond Racing Isolation Mini Pits And Mk IV Cones

AudioQuest Sky Gen2 XLR Interconnects

AudioQuest Everest Speaker Cables

Revelation Audio Labs Prophecy CryoSilver Digital Coaxial Cable - RCA connectors

Revelation Audio Labs Prophecy CryoSilver Digital Coaxial Cable - BNC connectors

Revelation Audio Labs Prophecy CryoSilver USB 2.0 Dual Leg Cable

PS Audio PerfectWave AC-12 Power Cords

PS Audio PerfectWave P-10 AC Regenerator

Polk Audio SDA SRS 1.2TL Loudspeakers (Heavily Modified)

Salamander Synergy Triple 30 Audio Credenza

Dell Venue Pro 11 7140 Docked Tablet Computer For System Control

Logitech Harmony One Remote

 

The photographs of the screen show a likely focus shift or possibly different frames, as well as, exposure differences that have nothing to do with ethernet. A more careful experiment is needed, with much better stabilization of the equipment (TV and camera), an exact frame fix, as well as completely fixed focus and exposure.

[pkane2001]

5 hours ago, DarqueKnight said:

All the Ethernet cable trial photos were taken at the same exposure and same fixed focus. See response to plissken above. In which photographs are you seeing a difference in focus?

 

This is comparing two photos used in Figure 11, Worst to best. As I said, possibly different frame, or frame in transition, not necessarily focus difference. Obvious exposure and color balance differences, though.

 

Quote

Figure 11. Worst to best: Left - Pixel level detail, NAS with stock switch mode power supply and Oppo player connected to wireless range extender. Right, NAS with TeraDak DC-200VA linear power supply and Oppo player connected to EtherREGEN with TeraDak DC-30 linear power supply. The overall result of using equipment with increasingly better electrical noise abatement was a picture with increasing detail, depth, sharpness, and color accuracy.

 

Make sure your browser can show loop GIF image animations and click on the image to see the larger version:

 

[pkane2001]

7 minutes ago, DarqueKnight said:

 

There is no difference in exposure. The TV put out different amounts of light depending on the power supply attached to the NAS. The camera settings remained constant. The color balance also changed when the NAS's power supply was changed. All the pictures were cropped and then resized to the same size, so there are small aspect differences due to the cropping and resizing. That is why the two white flecks and the two eye lashes alternately move toward and away from each other in the circled portion of a screen capture from your looped video.

 

 

You are missing all the other parts of the eye moving and changing relative positions and sizes between the two shots. It's much more than the two flecks you circled.  These are just not comparable exposures. You need much better controls on your 'measuring' equipment before declaring that the reason for these differences is the ethernet switch. 

[pkane2001]

1 minute ago, DarqueKnight said:

 

You missed my point with regard to my comment about the circled area.

Thank you for your advice and for sharing your opinions and insights.

56 minutes ago, DarqueKnight said:

All the pictures were cropped and then resized to the same size, so there are small aspect differences due to the cropping and resizing. That is why the two white flecks and the two eye lashes alternately move toward and away from each other in the circled portion of a screen capture from your looped video.

 

No. Resizing an image, doesn't warp it. The two images are warped in relation to each other. These are most likely different frames, or less likely the result of a focus or focal length shift.