"The 8c work best with a simple unmanaged switch¹. I'll include a Netgear GS105 with the shipment. If you like, I'll also include a miniDSP UMIK-1 measurement microphone. Our dealers usually install the 8c for their clients which includes setting up RoomMatching. When we sell direct, we do the same either at the client's location or remotely via TeamViewer and video conferencing software. You explained why you didn't want the 8c to be delivered and installed by me. However, it'd be a good idea to accept remote setup support. This maximizes the chance of smooth sailing. Are you open to that?" This being a purchase-included service, I obviously agreed. My DHL Freight shipment from the Netherlands held a 4-day sleepover in Aachen. Once it awoke from slumber, another three days had it at our Dublin facility. By late Monday, DHL's website had it out for delivery. With the holiday season in full swing, more delays were inevitable but by Friday that week I was in business. With two 10m Ethernet links I wired the 8c up to our unmanaged switch and perched them on British Track Audio stands. By typing 'http://8c-xxxx.local' into my music iMac's Safari browser—xxxx representing the serial number—each speaker was discoverable on the network though still invisible to Audirvana. I emailed Martijn receipt confirmation and asked for the initial TeamViewer session to configure RoomMatching as he would for any paying client. "At the moment the 8c only supports Roon and Spotify. We're working on Audirvana and other services. If you wish to use BACCH, you need Roon or Spotify." That was interesting as BACCH was this review's whole purpose yet I do neither. I'd hoped to access the speakers via UPnP from inside Audirvana as I'd done with an earlier Lumin and Silent Angel server. Alas, no dice. This was still early days for the BACCH'd 8c.
__________________________

¹ "Stutters in streaming audio is usually caused by funny behaviour of a network switch. The thing that makes streaming to the 8c dependent on the switch whereas other Roon Ready devices might work fine on the same switch is that you're streaming to two separate 8c which are both 'masters'. They both have their own DSP and audio clock and communicate with each other via Ethernet to synchronize those clocks. According to our developers, some switches run all manner of power-domain optimization in the background which may mess up time-sensitive communication on the LAN. In more high-end switches the user can typically configure these settings. But it's generally more cost-effective and surely easier to just use a simple unmanaged switch from a reputable company. We generally recommend the NetGear GS105 because it's widely available, not too expensive and just works."

In the left Spotify window we see one speaker selected. The other panes show two views of the speaker's browser window.

So I reactivated my lapsed Spotify Premium subscription. In its 'connect to device' window, I now spotted both D&D. Selecting one made no sound and I had no clue how to select them both. I'd wait for Martijn's intervention to show me the ropes and how to set each speaker's browser adjustments correctly. In the meantime I installed the most current RoomEQWizard freeware onto my music iMac. Our Zoom connection and TeamViewer access to my local area network would happen over my desktop's Win10/64 workstation since I don't do Martijn's preferred Google Meet. Via REW and the included USB microphone, he would measure the room at my seat, input his corrections to subsequently store on each speaker. Using two Ethernet feeds in parallel, he runs two different clocks to presumably have clever means to sync them. With this gig I was definitely on unfamiliar footing. I was frightfully fidgety over potentially fumbling a ferocious fortissimo for lack of physical volume control. The Kii Three came with a big rotating knob on its hardwired remote. Now I'd be down to a jittery software slider or equivalent virtualized interface? I already envisioned myself lurching for 'pause' or 'stop' should inadvertent full blast beckon. Or did I suffer an overactive imagination? I'd know soon enough. The 0.5dB-stepped browser volume above certainly looked safe enough. I just hoped that those unsightly panes didn't need to remain open during actual listening.

At the end of our 2½-hour Zoom/TeamViewer marathon I finally had sound over analog XLR and digital AES/EBU but not with 320kbps Spotify Premium over RJ45. The two live connections had proven inexplicably recalcitrant to pass signal. When they finally did which Martijn and his tech also saw from their end, they had no idea why. At that point of remote intervention, they'd not done anything else to make it so. They also saw how Spotify only sent meta not music data to their speakers. To figure out why, fix it and let me experience low-res BACCH, we booked a follow-up session to loop in a second tech unavailable for the first. That's because Roon had refused me opening a temporary account, denied Martijn's log-in credentials for his own account then those of his tech. Just then BACCH remained testily off limits. The Irish network gremlins were out in force. Yet the network did pass REW's test tones. Martijn could dial in sub 180Hz room correction whilst having me place his microphone in 21 different clearly communicated positions. Beyond Room EQ there are endless opportunities for additional notch to shelving filters. We can apply voicing tweaks across the bandwidth and make the 8c's frequency response anything we fancy.

This doesn't include all that's sonically relevant as the thinking reader will appreciate. There's more to playback than just frequency response. I learnt that each 8c hosts its own Linux computer. Those run a constant background game of ping not pong to sync their clocks over the network. I learnt that the virtual volume controls of Roon and Spotify lock to the 8c's browser control. That automatically starts at -30dB then progresses in 0.5dB even 0.1dB steps. Actual attenuation happens on 57-bit math inside each speaker just before D/A conversion. For listening I used my Singxer SU-6 bridge's AES/EBU output into the right speaker. From there another AES/EBU cable linked to the left speaker whose own loop-out XLR shorted via the included plug. D&D's setup window had already assigned the correct channel to each box. Now I bypassed Audirvana's optional software volume and used the combined 8c control from my Apple keyboard as shown next. Those command codes travelled over my 10m Ethernet links to each speaker whilst signal transmitted over AES3.

On my Apple keyboard, these two controls became my 0.5dB volume ± commands.

Just the facts, sir. That was pretty much my instant reaction. The Dutchies were superbly clean, focused, spatially sorted and subjectively linear. Martijn's room EQ had only focused on reducing the main room resonance peaks at 35Hz and 70Hz. Being most bass-capable sealed box speakers, his 8c in free space rode those peaks like a surfer carves swells. It's precisely why my usual Qualio IQ hybrid open baffles are already down 6dB/100Hz to hand over 4th order to our dual 15" Ripole sub and let its asymmetrical cancellation pattern seriously minimize those peaks. What the 8c didn't do now was the IQ's greater textural elasticity which derives from its dipole radiation above 600Hz deliberately activating the ambient field; nor its related tonal infusion. I also suspect that in general the high-feedback class D power modules of the 8c exert more damping across its vocal midrange and upper bands than our class A/B 2.5MHz DC-coupled monos do over the open-backed 6" Satori mids and Mundorf AMT of our Polish speakers.