The EEE-caused incompatibility packet loss is not so intensive to create dropouts as TCP will is able to quickly figure out lost packet and will request it again. You also won’t see this in RPi interface errors counter until you look into nic registers , or use something like tshark to look for retransmissions.
If you try to look further, these irregular events causing nic in streamer to work more, or just in irregular pattern, it is producing timing and energy consumption/radiaton changes which for most applications are non-issue because of buffering and working in digital domain where is enough reserve to distinguish logical 0/1, but if you are galvanically connected to Digital to ANALOG Converter and/or in it’s close proximity these things matter in my experience.
The transmitter sends a packet from a memory buffer, the receiver gets that packet and puts it into another memory buffer, and the two packets are identical, regardless of what happened in the transport. How can there be anything closer to ideal that this? Digital transports have been doing this for ages now, whether it’s over a 6ft Ethernet patch cable or over the ocean. That’s why we moved to digital, to do away with analog losses.
Haven’t these issue been resolved already with good engineering? And if they were still causing problems, wouldn’t they be measurable?
I don’t want to make it personal, but I would really like to understand how an engineer, trained in critical thinking, abandons logic and theory (and puts it in quotes) so easily, without even trying to get some objective evidence, when everything that has been mentioned (EMI, RFI and timing/jitter) could be objectively measured with a few hundred dollars and free software ( ARTA Software (artalabs.hr)), and then goes online and spreads the “wisdom”? What hope is there for the non-experts to get to the facts?
5 Likes
Bill_Janssen
(Wigwam wool socks now on asymmetrical isolation feet!)
26
The problem is all of the fancy bad engineering manufacturers use to try to differentiate their products. Things like Diretta and I2S connections. The main result being the re-introduction of DAC problems that were solved many years ago.
Ethernet is galvanically isolated though, but in any case if requesting a retransmission for one packet out of thousands caused a NIC to work in an appreciably different manner (which still would not affect any well-engineered DAC), there would be something that can be measured and heard. But even with the aforementioned RPi3 the only sonic artifacts that has ever been heard any any test worth talking about were the dropouts.
@Rudiger_Rommer-Collm I have a little hifi shop with a focus on streaming. I am doing side by side comparisons with my customers. What they can tell you: there is a substantial difference in sound when you compare a WiiM Pro / Pro Plus / Bluesound Node / Lumin U2 Mini / Auralic Aries / Innuos Pulse Mini / Pulse / Eversolo 8 / etc. to each other (using as transport, same DAC, cable, Amp, Speakers). You will never be able to understand the differences by reading a description in a magazine or blog. You can only hear it yourself.
Descriptions in magazines or blogs are just as useful as sighted comparisons in hi-fi shops: not at all. What would be a lot more useful would be measurements. Also very useful would be double blinds tests, i.e. where neither the seller nor the buyer knows what’s playing.
My left and right eye don’t fully agree on the color of the same object. But if you measure the chromaticity coordinates of two objects and they are within human color acuity, I’ll know that if two people say they have different colors, the differences are due to each individual’s particularities, so I won’t go telling everyone the objects look differently.
What I can say so far is that the Wiim sounds different than my MacBook Pro when used as a core and streamer (data via WiFi). The Wiim sounds clearer, but also brighter to my ears. What would really appeal to me about another streamer is that the Wiim’s coax digital output isn’t quite as solid. I think a more robust version would be nicer. But I’m not yet sure how much that’s worth to me, especially since my ears definitely have deficits.
The next step will definitely be to move the LAN into the listening room, then I’ll see what happens next.
This means that my digital cable doesn‘t fix exactly on the digital coax output of the Wiim. There is a little bit shakiness between the output and the plug.
there is a big difference between inputs on a dac, it doesn’t matter what brand of dac you have. USB is the worst, followed by toslink and spdif mainly because of the limitations on whats supported, AES, ethernet, then the best so far is i2s.
USB has a lot of jitter unless you get 1 of hundreds of devices to make it sound better.
Stereophile just did a review on the Berkeley Alpha USB device to cleanup the jitter from USB. This comes from the Berkeley website:
The Berkeley Audio Design Alpha USB is a high-performance outboard USB to AES-EBU converter and re-clocking component. When using a computer as a music server, jitter becomes an audible problem. The Alpha USB is an asynchronous High Speed USB to digital audio interface that extracts the highest possible audio quality from computer audio sources. The Alpha USB features a High Speed USB 2.0 input data connection and switch selected audio output signal type — either coaxial SPDIF using a BNC connector balanced AES using an XLR connector.
“The Berkeley Audio Design Alpha USB is a breakthrough product that not only overcomes the limitations of the USB interface, but provides a state-of-the-art method of getting audio out of a computer.”
You’ve been reading the Berkeley sales material, lol. Digital audio isn’t tricky these days, eg, the PCM2707C (Texas Instruments) turns USB data into I2S audio data, and happily converts the worse to the best in your list. It’s a single chip.
Of course they would write it, they want to sell it. Never mind that there is no jitter on USB. There is on I2S but we are supposed to ignore it, because vendors charge more for it.
And also very wrong. I do not expect him to bite the hand that feeds him, but we do not need to listen to thinly disguised ads either.
Funny how in the end the “argument” from the “cables matter! muh eahrs are golden!” crowd always degenerates into “Ah is rich, ah spendz much moneys on cablez. You filthy poor peasant never heard muh cablez!”
One can only wish that HiFi dealers weren’t trying to sell you all the high margin, low usefulness crap like cables and $50K DACs…
