|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
61.239.75.90
In Reply to: RE: Sorry fmak, Ryelands, etc. posted by Mercman on November 15, 2014 at 07:49:16
Hi,
Hate to break the news.
There is no magic bullet, system wide, reliable for Computer/Audio System interactions. There are just so many variables.
As an example, depending on context in the system a 99 USD product we sell can be the most significant upgrade, no matter what, or it can make the sound worse.
The key is to try things in system, tweak around with anything you can get at, there is a synergy somewhere. We are working making it less than shooting craps in the dark, but it is not easy.
Right now my system does best with a Mercury cable and nothing else, except extreme and hardcore mains treatments which are not commercial (and will be illegal [not electrical code] in some countries).
Having DAC and Amplifier integrated in a single box and without an earth anywhere (but with good isolation) does help killing some of the usual crap in more complex systems. A big mains filter (think bigger than most mains distributions) between audio (integrated plus Sub) and PC/Plasma TV/Sat Box systems does help a lot.
That huge CLCLC Filter has a corner around 1kHz and is good to past 100MHz... They don't make 'em like that anymore.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
Follow Ups:
I think you hit the bull's-eye with the phrase "complex systems". If it's complex there are too many ways for things to go wrong, so simpler is better. All the unnecessary equipment needed for listening to digital music can be unplugged completely (power and signal) and if battery powered switched off. Equipment that can't be switched off such as cell phones can be Faraday caged (e.g. in a microwave oven that's unplugged). All circuit breakers other than the one needed to power the audio system can be pulled. Then one has a manageable problem, i.e. a computer, DAC, amplifier and speakers.
Tweak this simple system first and get it to sound good. If necessary, then the missing components can be added back in, one at a time. If you can't get the simple system to sound good then at least there are a small number of places to look.
Unfortunately, so long as reviewers continue to use all of these measures in giving expensive equipment good reviews, we will continue to get products that are unstable and require tweaking. This represents a fundamental error in testing methodology. Products should be evaluated under worst-case conditions, not best case conditions. This is the only way to get valid results. Later, if one wants to actually use these products one can use them with confidence that they will perform even better. This applies to all types of products, not just audio.(With some products such as automobiles, some of the worst case testing can be actually dangerous and illegal if done on public roads, but is still necessary for proper reviewing.)
WARNING: when done testing don't forget to restore circuit breakers, remove cell phones from the microwave, reconnect necessary appliances, etc... Failure to do so can have serious domestic consequences or worse. Been there. Done that. :-(
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
I have come to realise that one of the worst sounding practice is to power P4 in conjunction with any other 12 Vdc devices from a linear power supply! Common sense really.
"It's a complex system"
NONSENSE!!!!!!!!!!!!!!!!!!!!!!!!!
It's one-port digital-in and one-port analog-out. At rediculous low transmisson rates.
The WWW I'd call a complex system. And I still receive bits as they were sent 10thousands of miles away at magnitudes of higher throughput (global scale).
Man, we're in the digital domain (upstream) here.
As long the DAC receives bits in a readable way (non-corrupted streams), it'll be the DACs (audio interface - to be exact) job to eliminate any other distortion that's been added upstream. No matter if we talk EMI/RFI/noise/groundloops/power issues/data jitter asf. asf. The issues are known for more then a decade by now.
Thorsten IMO just tries to defend the shortcomings of his units.
Though he's not doing better/worse then the competition.
iFi tried to get some issues under control by using batteries, using USB filters, FemtoClocks and more!!! Obviously that all is still not sufficiant.
I'm in here since 2007. We/You still discuss the same subjects all the time over and over.
And you guys still try to tweak your upstream environment.
Some at least, should have realized by now, that it'll never work unless you'll get a DAC that's able to get the mess under control.
Computer based audio still feels like gambling for best sound.
Enjoy.
Hi,
> NONSENSE!!!!!!!!!!!!!!!!!!!!!!!!!
>
> It's one-port digital-in and one-port analog-out. At
> rediculous low transmisson rates.
That is what the DAC itself is. Once you connect to a source (even a lowly DVD/BlueRay player via SPDIF) and a sink (say a pair of active speakers - for the absolute minimum), the overall system has attained massively greater complexity.
Attach a screen to the source (or use WiFi to connect a remote Display/UI) and complexity goes up more.
Another issue is that gear must conform with agency requirements (be it CE, FCC, IEC and so on) that are formulated entirely without any reference to (or exemptions for) Audio Quality.
Often the resulting designs from any but the most crazy designers that conform to these agency requirements tend to involve design approaches that are the polar opposite of what Audio Quality requires.
Today I had occasion to show one of my senior engineers in detail exactly how grounds are arranged in a product and how they use the metal chassis as secondary ground system at Radio Frequencies of the "plane" type while retaining classic star grounding for Audio.
Up to now he had always wondered why I don't just pour one big ground plane on the PCB (like everyone else). Definitely makes passing EMC easier. Also poison for audio quality.
So, wherever possible I keep audio grounds as "Audio style" and use suitable means to link the audio ground at RF ONLY to the case, which for RF forms aground-plane and/or faraday cage.
Trust me when I say it is not "simple" or "easy" if you ignore agency requirements (e.g. you DIY your own gear), if you cannot ignore them the difficulties go up by the square of the different(and conflicting) codes involved.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
It's about power and grounding and various forms of parasitic coupling. This is all EE stuff. If you're a computer programmer you will miss all of these details. If you believe that bits are just bits, then you are wasting your time reading this forum.
This is not simple stuff. The "transmission rates" may be low in the digital domain, but they are not low when you consider digital noise which goes 1000 times higher in frequency than any "audio" signal. This noise is radiated and coupled ("every wire is an antenna") and enters low level amplification sections via feedback loops. At this point it intermodulates with the audio signal proper, producing spurious effects in the audible range.
It doesn't take lots of parts to make a complex system, provided there is non-linearity. Consider the Three Body Problem in Classical Mechanics. A little Googling turned up these slides with pretty pictures. Strange things happen when one has an unstable system.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
I am an EE btw.Perhaps you should read what I wrote.
The audio interface has to take care on all the "known" distortions.
The designer has to make sure that close to nothing impacts the analog output quality of his device.
Most of the challenges are known to the vast majority of designers and EEs (and me too).
The real problem:
Usually budget, knowledge and market situation prevents from doing things right - doing it right from an enduser perspective.
Manufacturers usually run the "It's-good-enough-for-them-and-the-competition---We-need-some-space-left-to-sell-the-next-upgrade" mode.
(iFi nano-> micro-> mini)Everything else are stupid excuses - blame the others for mediocre gear performance. The funny thing is. It works. Stupid users (that includes me) tweak their upstream environment, invest hundreds/thousands of $ and endless hours in questionable tweaks... ...and guess what!?!?
As a matter of fact: The system will sound better after applying many of these great tweaks. And it finally proves: The Audio interface is that good that it shows any of these changes in terms of a "better sound experience".
That'll make the successful enduser tweaker a hero!!!
(Some of these even become official reviewers.)Poor mans conclusion: The very competent EE (manufacturer) gave the right directions: Build a new house and your own powerplant - live in a Faraday cage - Restructure and reinvent the WWW and computer industry - just to get a 300$ device an environment that it really deserves.
That sounds like a reasonable solution of all our problems to me.Less knowledgable people - especially in forums like AA or CA - don't even realize how much they get fooled all the time.
Enjoy.
Edits: 11/19/14 11/19/14 11/19/14 11/19/14 11/19/14
don't necessarily know much about audio, which includes acoustics, psychology and human emotion
Some of my PhD electrical colleagues are just textbook believers when it comes to audio ie straight wire with gain, bits are bits and all that nonsense. Most don't even think that capacitor quality makes any difference; to them it is just two strokes on a circuit diagram.
Do your PhD electrical colleagues design audio equipment for the audiophile niche? If not, I wouldn't expect them to know the nuances of electronic components as they are applied to audio equipment.
However, there are many EEs who do work in the audio field and do understand.
And then there are acoustics trained engineers and PhD's who know nothing about electronics or computers. We see some of that here with random tweaks that have no basis in understanding the electronic components or material properties that make up those components.
"Hate to break the news."
"There is no magic bullet, system wide, reliable for Computer/Audio System interactions. There are just so many variables."
I totally agree Thorsten. I stated this in my review of these products. And yes, I like the Mercury cable.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: