As I said, we could go back and forth on this, in the manner of third-graders, and apparently you want to do that. I have no great desire to teach third grade. But if we are not seeing what is clearly in front of all of our noses – commodity items like the AKM and ESS chips or the Hypex amplifier boards – I could offer one piece of general advice: there’s a lot of fluff still being generated by marketing buzz for high-margin luxury goods, and it’s perpetuated shibboleths about audio and engineering that are wildly obsolete. It’s possible – I don’t really know – that we all may need to cleanse from our observational skills some of these cobwebs that keep us from seeing the things we’re looking at.
You are a lucky guy in that you think that any implementation of AKM and ESS DAC chips sound great. You are even luckier than you think Hypex Class D amps sound great. You don’t need to shop much or spend much to be a happy listener.
To my ears, there is a lot more to a DAC than the DAC chip it uses. I have yet to hear a Hypex amp that does not give me a splitting headache in an hour or less.
What you need to stop doing is assuming everyone is you. I don’t hear what you hear and you clearly don’t hear what I hear…
I don’t think I said that. I said that they were commodity devices which makes it easy for a competent engineer to build a great DAC. Sure, incompetents can use them incompetently.
I don’t think they sound great; I think they are great amps partly because they have no sound. Transparent, with good power and very low distortion. They simply amplify the sound from a great source.
Look, I’m sorry if some infirmity or quirk does not allow you to enjoy the great sound that these commodity devices provide – I know how that goes; I get a headache from Shirley MacLaine movies that others enjoy. But that does not make them any less commodities. If you are looking for them and cannot find them, well, there are multiple hypotheses as to why that may be the case, and only one of them is that they are not present.
As you say, everyone hears differently. A lot of commodity devices are engineered to an 80/20 rule, targeting most consumers, but not all consumers.
2 Likes
Yes, I have checked it with a data monitor. The data from Tidal comes in bulk. Most of the times not a complete song but in bulk of a minute up to three munutes of audio. If you want to show the waveform you have to analyze the whole track, which is possible but in case of a temporarely busey network or a not so fast internet connection this would lead to a delay in starting playback which we don’t want.
If you are playing directly from the core computer it is easy to check. Start playing a song, pull out the ethernet plug and see how long it keeps on playing. That’s the buffer length.
And if you do not hear a difference when pulling the plug there will be no cable or reclocker in the world that’s going to make a difference here.
1 Like
Thank you to everyone who voted in this survey and who commented. I have been intrigued by how the Roon community thinks about what matters in the digital domain to achieve good, better, best sound. The votes are virtually tied, which perhaps explains why there’s so much debate between these two points of view.
I recognize that providing only 2 choices, apologies for any double entendre, has its limitations. To clarify my intent in posing this question, I agree with the point of view that the storage and transmission of digital data is extraordinary robust. If it were not so, we could not entrust virtually ever form of record keeping and communication that affects our daily lives to the digital domain.
For example, I have 7TB of music files, most are high res, backed up on multiple drives, and I’ve never experienced an audible problem associated with data integrity.
Digital however is a man made creation to facilitate the storage and transmission of data. Music, another man made creation, however, is an analog phenomenon. Our ears hear in the analog domain.
One can logically assume the conversation of digitally encoded music into an analog waveform, a DAC can be designed in an infinite variety of ways, should be the only variable in how a data file sounds.
Yet it has been my experience changing how a digital file is delivered to a DAC can make very noticeable, audible differences. For example moving Roon Core from a PC to a Nucleus made a big change for the better. More subtle changes to power supplies, cabling, network switches, etc have made noticeable improvements in how my system sounds.
I can’t prove why, nor measure what changed. But if I can hear a meaningful improvement that fits my sense of cost / benefit, then I’m happy. So I’m always interested to hear what others have tried that works for them. By experimenting with new ideas the sound of my digital system has improved significantly over the past few years.
And with the availability of high resolution music files, I think we are in an exciting time where we have the ability to experience very realistic sound at a much more affordable price.
So count me in the everything matters camp. And if you don’t agree that’s perfectly ok.
5 Likes
Analog music storage is a man made creation, however music is an acoustic phenomenon. Our ears hear in the acoustical domain not in electrical waveforms.
4 Likes
I guess you didn’t look in the right place, or maybe you’re just ignoring measurements. This site has measurements for quite a few popular DACs and amps. You can find a lot of excellent DACs and amps in the few hundred dollar range.
Audio Electronics Review and Measurements Index | Audio Science Review (ASR) Forum
1 Like
Again, ASR is not a place I go to to find out about how things sound. Nor do I go there for measurements as Amir does not do a good job of that either.
If you want to buy your audio equipment based on measurements, you go right ahead, I buy audio equipment based on how it sounds. I look at the measurements too but they don’t really factor into my buying decisions. I trust what I hear.
I have custom made tube line stage and tube amp for which I don’t even have measurements. I don’t care because they sound incredible!
3 Likes
So you like tube distortion. I don’t care much about distortion of any kind.
2 Likes
Whatever you want to think…
1 Like
It’s not about what I think really. If we’re talking about personal preferences, we can’t and should not agree. If we want to agree on something though, it should be based on facts and science. We’re talking about bits here, and science says bits are bits, all the way from storage to internal DAC buffers. Then, if we take transparency as an objective way to evaluate and compare DACs (the only objective way I can think of), we have to look at measurements, and based on that, good DACs are a commodity. Everything that comes after the DAC is not bits anymore, so it’s not relevant anymore. If you like tubes, use tube amps, or any other sound effects for that matter.
1 Like
Except you are only thinking about the bits. There are a bunch of electronics involved getting those bits from the storage device all the way to whatever parts are used in the DAC to do the digital to analog conversion.
And no, good DACs are not a commodity. DAC chips are a commodity.
2 Likes
I have and will buy based on measurements, if only to keep manufacturers honest. I know audio/food analogies are usually bad (McDonald’s anyone?), but to me, this is similar to checking the ingredients before buying food: if it has bad stuff in it or ‘taste enhancers’ (e.g. MSG), I don’t really care if it tastes good because it’s either not healthy or not natural. Same way, I wouldn’t care if audio gear sounded good but wasn’t transparent because I want to listen to recordings, not DACs or amps. Once the measurement aspect is cleared, I can go ahead and check the sound. We can trust our ears, but they can’t work without our brains, which, unlike audio analyzers, are known to be biased.
Also, I’d be interested to know why Amir is not doing a good job at measurements.
1 Like
That is one the main mistakes in thinking.
We know very little on how and what to measure. So measurements are only what we know at the moment.
Most measurements are done in the frequency domain, very little to none in the time domain.
Frequency domain is like: is it a pussycat or lion that I hear
Time domain is like: Is it in front of me, behind me and how far away.
Our brain can tell all that. Measurements not even close.
1 Like
Absolutely true.
Not sure I buy that. Electronics can routinely measure time differences measured in nanoseconds if not picoseconds. By contrast I’ve seen estimates of 10-20 microseconds quoted for the temporal resolution of human hearing - at least three or four orders of magnitude less precise. Or did you have something else in mind?
Absolutely false. Engineers have been building what we can call modern audio recording equipment for almost a century now. You’d think they know something about how and what to measure. Audiophile’s disdain for all the engineering that went behind these machines is simply ridiculous. We’ll never know everything, but that doesn’t mean we’re complete bozos who got lucky.
Regarding the time- and frequency-domain representations of signals, your analogy may be poetic but shows little understanding. Both these representations are equivalent and can be completely derived from one another. That is, Fourier transform (and its digital equivalents) is reversible. The reason much processing and analysis is done in the frequency domain is that it’s much more efficient or relevant there. For example, filtering is just multiplication in frequency, which is much simpler and faster than convolution in time.
Then there’s binaural audio. By measuring the HRTF (head-related transfer function), which is a frequency-domain representation, and applying it in real time to the signal that is played back, and combining that with head tracking, it is possible to simulate spatial sound (i.e. sound coming from ALL directions and ranges) with only two sound sources (stereo pair of speakers or a pair of headphones). To put it another way, measurement-based frequency-domain processing can tell your brain what to perceive. I’d be a bit more humble regarding our exquisite powers of perception.
4 Likes
There are 10 types of people in the world; those who understand digital and those who don’t.
2 Likes
You are 100% correct, they know something about how and what to measure, most definitively not everything (yet).
1 Like
Don’t hold your breath. As I said, we won’t know everything ever, but we don’t need to know everything to make things work, do we? Had we wanted to account for the gravitational pull on our rockets of every single star in the Andromeda galaxy, we would have never put a man on the moon. We just need to get to the point where what’s left to understand doesn’t make any practical difference anymore. Digital audio has been there already for a while now, and you have to come up with more than anecdotes and marketing pitches to deny that.
6 Likes