For what its worth, the Lumin T3 is starting to sound really nice; improvements in every way. Needless to say I am very happy to hear this trending in such a positive manner. Maybe this will be a keeper, but I am not ruling out the P1 as a better choice for me.
Great to hear that the T3 is getting better and better and proving to be enjoyable. Just some info for reference: your T3 will take a minimum of 400 hours to"get "on the cam" as they say in motor racing, and 500 hours to fully burn in. Listening 5 hours/day, that's about 3 1/3 months. The 500 hours burn-in requirement is consistent for many Lumin products including the P1, the T3, and the new U2.
Another interesting thing I learned today, and maybe others can weigh in on this:
Went to a local dealer yesterday to talk about cabling; options, results, costs, etc. During this discussion he asked if I had ever experimented with Ethernet cable changes. In my mind, this did not really make sense (12" of cable??). He proceeded to show me the results through a demonstration. First up the (blue) CAT5e cable straight into the streamer which happened to be a Moon streamer, but not sure of the model. Played a track that was more bass oriented jazz. Then he changed (using a basic TPI Ethernet switch).
Next up was an Audioquest Vodka cable. I was surprised to hear the bass quality go from a little muddy to really clear and tight. I'm sure there was other improvements as well, but we only listened for 30 seconds at a time to the same track at the same volume. Next up was a Nordost cable, but not sure what the model was. Another step up but this time less pronounced. I ended up leaving with the Audioquest cable. I must say I am pretty amazed at this improvement to my sound. It really positively impacted my system in every way imaginable. Don't want to bore you with the typical cliche audiophile jargon. I am now even happier with the Lumin T3. It did turn out to be a nice upgrade for me. I really can't understand how 1 foot of signal cable can have such a profound effect. Anyone have similar results?
Yes. Ethernet cables have a audibly significant impact on audio quality.
Regarding your experiences with Ethernet cables: they really make a difference, sometimes significantly so, depending on their construction. When I was implementing up my "streaming set-up" (a "remote" music server sending ripped or "streamed" music files to a network bridge/streamer), I did what is known in product development as a "competitive benchmarking" of several makes of Ethernet cables. I rented in some some Ethernet cables, and bought a few others, including the Wireworld Belden Cat 6A, Supra Cat, 8, Audioquest Cinnamon, Starlight Cat 8, and Audioquest Vodka. Of of these, I thought the Vodka was the best, it's a nice cable. Once they became available, I also tested the Shunyata Venom and Alpha Ethernet, and of all the cables, I evaluated, these were the best I evaluated.
So, my final rankings were:
Shunyata Alpha
Shunyata Venom
Audioquest Vodka
WW Starlight Cat 8
Supra Cat 8
AQ Cinnamon
Belden Cat 6A (really didn't like this cable)
If you need a quality Ethernet cable for other connections and are "budget-constrained", I'd suggest considering the Supra Cat8, just be mindful it can be a bit bright-sounding in certain set-up/positions (where the cable is being used), and on certain content.
One of the reasons that different Ethernet cables can sound different is due to their construction. Shielded Ethernet cables with the shields connected at "both ends" can conduct high-source impedance leakage current* (usually caused by switch-mode power supplies), which causes threshold jitter*, which leads to timing errors (the effects of timing errors are very audible, which is why high-end "master clocks" make such a big difference for digital music reproduction).
From John Swenson, who designed EtherREGEN:
"Clock thresholds, ground-plane noise, and receiver jitter:
So where does jitter come from? All oscillators (which generated a clock signal) have jitter, some more, some less. We classify this jitter component as “source intrinsic jitter.” Additional jitter can be added to this at the receiver of the clock, which is typically a flip-flop inside the DAC chip. The flip-flop has its own intrinsic jitter, which is called “receiver intrinsic jitter.” But there is a third class, “threshold jitter,” which is harder to conceptually understand, and threshold jitter is what will be discussed in the most depth in this paper.
Even though we tend to think of digital signals as being "binary" with just two possible states—high (1) and low (0)—changing instantaneously between them, this is not actually the case. The signal is an analog voltage. When going from a high to a low state it hits all the intervening voltages, and the change takes a finite interval of time. This is called the "ramp time,"—the time it takes to "ramp up" or “ramp down." This time interval can be very fast, but it is still not zero. A common ramp time found in modern digital audio circuits is several nanoseconds (ns). A ns is one-billionth of a second, and while that sounds fast, in the realm we are discussing it is not. This is important because every circuit receiving a clock signal has a threshold voltage. This is the voltage above which the circuit thinks the signal is a 1, and below which it thinks it is a 0. So what happens when this threshold changes? Think of a line going up from left to right, with a threshold at say 2 volts. Whenever the signal gets above 2V, the circuit sees it as going from 0 to 1. But what if the threshold changes from 2.0V to 2.1V? The signal takes a little while longer to get to 2.1 than 2.0, so the time it takes the circuit to see this change to a “1” also increases slightly. Note that the clock signal itself did not change; the change in timing was entirely a property of the receiver.
Okay, so why is the receiver changing threshold? Well the receiver itself really isn't, it stays at 2.0V. Where does the extra 0.1V come from? The voltage on the ground pin of the receiver! The 2.0V is not some physical property of the universe. It is the voltage between the ground pin and the clock input. However, if there is any voltage between the ground pin of the receiver and the ground pin of the source, this difference will be seen as an effective change in threshold, thus changing the timing the receiver sees, otherwise known as threshold jitter."
*–threshold jitter reference:
https://cdn.shopify.com/s/files/1/0660/6121/files/UpTone-J.Swenson_EtherREGEN_white_paper.pdf?v=1583429386
The other noise factor that has an audible effect is common-mode noise†, and this is why the higher-end Shunyata Ethernet cables (Alpha and higher) have an in-line common-mode noise filter. Reducing common-mode noise also plays a significant role in how good an Ethernet cable sounds.
†-Common-mode noise:
https://www.pulseelectronics.com/wp-content/uploads/2021/01/Pulse-Power-BU-Understanding-Common-Mode-Noise.pdf
