Is an external clock on a DAC necessary?
ฝัง
- เผยแพร่เมื่อ 22 ก.ค. 2024
- Audiophiles pay big bucks for external long term accuracy clocks added to their DACs. Are they necessary and if so, for what? Have a question you want to ask Paul? www.psaudio.com/ask-paul/
I am getting close to publishing my memoir! It's called 99% True and it is chock full of adventures, debauchery, struggles, heartwarming stories, triumphs and failures, great belly laughs, and a peek inside the high-end audio industry you've never known before.
I plan a few surprises for early adopters, so go to www.paulmcgowan.com and add your name to the list of interested readers. There's an entire gallery of never before seen photos too. - วิทยาศาสตร์และเทคโนโลยี
![เปิดบ้านสุดหรู "เเจ็ค the ghost" ลั่นถ้าป๋องกพลลำบากผมเลี้ยงเอง!! l [Nickynachat]](http://i.ytimg.com/vi/1xbW4vULqGc/mqdefault.jpg)
The best clock is the one sitting next to your dac. If your DAC is well designed and therefore has a good clock, you don't need an external one for those 0.01% better clock performance. Sometimes having less devices in the signal chain is better.
I love it when Paul gets confused, however he always manages to get back on track. :-)
One of the issues surrounding external clocks is that they can actually induce distortion and noise that some people like. For example, in pro audio a big clock manufacturer is Antelope. They actually advertise that they ADD specific jitter or "acoustically focused clocking" to a low jitter clock in order to create a coloration of the sound. Hardly any modern converters are going to benefit from a clock outside the mainboard, which is usually much less accurate, so what a consumer is getting is a specific flavor of euphonic distortion.
At least Paul you try too explain things and put your time in to help the industry out , and you run a good business , better than most companies that we never hear from .
Hello Paul. Thank you again for elucidating me about the "myth" of having an external clock. I recently went to Audioshow, in Lisbon, and there was a portuguese high-end external clock manufacturer saying marvelous things about his rubidium clock, when connected to an excelent transport, from Esoteric... I didn't hear the difference. I'm glad that you agree with Daniel Weiss, because I also have the same opinion.
You probabably didn't hear a difference because what you describe is an atomic clock which is accurate in the long term. In audio look for short term jitter specs.
Am I the only noticing that more and more of Paul’s videos tend to contain a subtle, and sometimes not so subtle, pitch for PS Audio gear? Personally, I liked it better when the videos talked about concepts in a more agnostic way.
Shocking ...a salesman trying to sell his products. It's always been about increasing ps audio visibility and sales. I do enjoy Paul's rants though.
gotham61 lol yeah I have noticed that also since channel started lol he got me !!
Why would it be any other way ?
I don't understand all the tech about it...just always trying to improve sound quality...someone please tell me a good external clock to use with my setup shown below...I was thinking of the Gustard C18 since my DAC & DDC Is also Gustard...
Present system- Vincent hybrid 737 -4 Ohm: 2 x 300 Watt, Gustard R26, U18, ifi zen streamer, Dali Rubicon ll, Rel T7x w/ AudioQuest Type 5 REL Acoustics High-Level Subwoofer Cable...
The clock information is sent with the signal there for if you are using a transport as a source then the dac is using the clock in the transport. The clock in the dac is not being used to process the incoming signal. Connecting an improved internal or external clock to the dac is not relevant as the timing damage or jitter has already occurred
However the closer the clock is to the dac chip the less errors you have. Some dacs automatically slave the transport or source if they are from the same company forcing it to use the clock in the dac. This is quite noticable.
However clocks like a stable quiet environment away from noisy transports and dacs. So if costs isn't an issue then having a dedicated clock and slaving both the dac and all your sources is the ultimate
that makes sense to me. it's all about real physical distance, and putting a console in one room and a hdr in the basement I can't see not having problems, and an external clock would solve it, right? Which is why the external master clock industry exploded
It's important to note here that in order to have the DAC clock used, the connection to the source (e.g. computer) has to be via USB which is asynchronous; otherwise, if you connect to the source via, say, coax or optical/toslink then the clock accuracy is determined by the source.
No mention of why one would consider slewing the output sample rate in a DAC?
I got about halfway through the vid. If your DAC is your playback device, you don't need a clock. If you're running digital components to a DAC with the available option of setting slave and master, then you don't need a clock. If you're running multiple digital components in serial and parallel, which honestly is more likely in a music studio, you'll probably need a clock.
It's not about synching issues of multiple digital devices in regular stereo setups (as it indeed is in the studio), but it's about the quality of the clock signal which influences the SQ.
Exactly! A DAC is inherently self clocking as the clock is embedded in the SPDIF/AES stream. It's called Manchester Encoding. 10mbs Ethernet used the same technology. And what many analog audio guys don't grasp is the concept of clock and phase lock. You just can't feed a DAC circuit a random clock signal. Somehow that clock must be phase locked to the digital audio stream. Otherwise you get garbage, no intelligible audio at all. DACs that use an external clock are just using that clock as a re-clock in a buffer memory. Similar to how a video signals are aligned in a modern video switcher or mixer. And here's a dirty little secret: If using a free running external clock, sooner or later you will drop samples as the buffer under or over flows. Depending on the buffer depth and the clock accuracy versus the encoded SPDIF stream, that may take hours or even days to happen, but at some point if listening continuously, you will hear a click or pop. No way around that even with a Cesium clock source. The vendors just bank on the fact that you will miss the glitch or you won't listen continuously long enough for that to happen. So if you want absolute error free playback 100%, don't use a unlocked external clock. But then who listens to audio days on end without a break?
Brilliant Paul! I feel like you let us into your office or the shop for a 10 minute audio education session. Thank you for taking your time to make these videos!
one ...more time , thanks Paul
youve made me think if the sample rate is 44khz as thats double 22khz how do we get the stereo signal from that as stereo would be 22khz x 2
In order to produce and hear frequencies up to 22khz you have to double that for sample rate.
Yeah but how does the clock get affected if you're travelling from one end of the world to the other how much will it be out by....
External? It depends whether better or worse. External clocks purpose is to synchronize different devices. Eg. in a studio. Using PLLs and good supply of clocking circuitry nowadays minimize the needs of external clocks, You can use multiplexed signals to drive your PLLs.
On the other hand, an external clock has no direct effect on the internal quality of clocking and thus sound. It can, if it is better than the internal. But I’m highly in doubt of this. Just one bad clock dealing stage after the external (or internal) clock with high noise power supply can ruin and rejigger all the high quality clocking signal. This is the reason, we use direct clocking to our DACs. The DAC is direct connected to the clock (XO), all the other parts being of higher jitter tolerance are only indirectly supplied, some even by multipliers. The clock itself is provided with ultra low noise power supply.
I have a Esoteric DAC, I found playing it without pll1 clock ( clock out) sounds better, why is it
seeing how the dCS Rossini and Vivaldi are both among the best DAC's ever made, and both come with the option of an external clock, which says everything you need to know.
Directstream Dac ignores the incoming clock, a design decision by Ted.
In audio/video *capture*, long time stability of clocks is an issue. Prosumer devices typically drift away about 1 video frame (1s/25 frame in Europe) per hour in my experience, some devices fare worse. So in long recording workflows, you’d either stop video record in every break - or use an external clock :) some devices have both ref-clock in and out so one of them could act as the master and the rest are just slaves, but there’s nothing special about the master - it is just your reference.
To me that's the main reason DACs have a clock input: to synchronize different devices. And even then it's less about being accurate and more about being inaccurate in the exact same way.
Penny Lane yup, if there are digital sources that benefit from being in sync. Playback systems can probably be designed to be less dependent of this, for example if a single DAC is responsible for all of the DACing. Designed to expect input source jitter, e.g. up sample to a higher frequency and make sure the input samples are updated in a regular pace regardless the input inconsistencies. It would not be entirely trivial, but it not leagues of complexitiy above what students have to work with in other clock domain troubles in digital design courses. The main question is of course if there are two systems that strongly benefit from being in sync - for example do you for any reason have two DACs in the system? Then a reference clock sounds like very good idea :)
Why 44.1 and not just 44?
dsp.stackexchange.com/questions/17685/why-do-we-choose-44-1-khz-as-recording-sampling-rate with a little bit of Googling!
"At the time, there existed a product in the late 1970s called the Sony F1 that was designed to record digital audio onto readily-available video tape (Betamax, not VHS). That was at 44.1 kHz (or more precisely 44.056 kHz). So this would make it easy to transfer recordings, without resampling and interpolation, from the F1 to CD or in the other direction.
My understanding of how it gets there is that the horizontal scan rate of NTSC TV was 15.750 kHz and 44.1 kHz is exactly 2.8 times that. I'm not entirely sure, but I believe what that means is that you can have three stereo sample pairs per horizontal line, and for every 5 lines, where you would normally have 15 samples, there are 14 samples plus one additional sample for some parity check or redundancy in the F1. 14 samples for 5 lines is the same as 2.8 samples per horizontal line and with 15,750 lines per second, that comes out to be 44,100 samples per second."
@ Yes the math works. But also the F1 and competing products were quite complex. The biggest issue was tape dropouts. The raw audio data was heavily encoded with error correction bits. And then the digital audio was not recorded through the horizontal and vertical blanking intervals for two reasons. The video sync pulses had to be preserved and in lower cost VTRs the head switching glitch would have caused severe data errors. As audio is continuous and video is recorded like film as discrete still frames, these digital converters had to buffer the audio in memory and write to tape only during active video. DAT took that idea even further. The rotary heads only wrap 90 degrees around the tape. The "missing time" is handled by buffer memory and faster than real time write and reads.
I think DACs need an SD card reader, battery and charger. Probably less than $30 more in production costs for a lot more value.
A "DAC" doesn't read files, it can't, not by itself. Something else internal in the chassis has to feed the DAC with I²S. That costs money too.
I have a dac amp, with a front usb port for flac playback etc. It sounds worse than the usb from my lap top, and also from a streamer. Noticeably worse, unenjoyable in fact.
it sounds worse because of the chips that are required to play the files to the DAC, A "DAC" doesnt read files, it can't, not by itself.
A file is just data , the same stuff that comes threw the USB. You might thinking of a desire to have some interface to chose which file to send.
I think you commented on the wrong video bud
Why do you think and comment that?
I use a GPSDO 10MHz reference clock oscillator for my Amateur Radio rigs and my DAC.
Is a clock really a clock for the time?😂 I thought it was something to improve the sound.
It does affect the sound and yes, its for time.
bro, sound is governed by time, like all else is
Time and a word.
a word within a word within a word...first the was the word... :)
Word.
@@InsideOfMyOwnMind stand corrected "W"
"In the beginning was the word." "Under the rocks are the words, and some of the words are theirs."
an accurate clock is very much required for precision PWM generation, such a clock can cost thousands of dollars! and can have stability down to parts per billion, in audio that is not so much important, like most of the things, audio is very forgiving in fact ;)
Yes it is necessary because the standard in audio is so bad.
No audio company really uses them, they exist as individual boxes that come with lousy LPSU's that are horrible in sound, that is why people don't think that they work. So it's kind of a big kept secret, that no one really has been able to figure out how to use.
My friend got one: Gustard u16, and he uses it for his streamer. When doing a lousy standard job, it only makes the sound about 1.5x better. But after he now has used a couple of years fully understanding how it works and locking the signal with a GPS lock on 9.9999997 mhz, and using certain cables with LPSU's that cost about 5 times more than the clock and building a custom LPSU, and pairing with an uptown audio isoregen. It now gives him about 10x the effect. So yes it is necessary once you have heard that huge difference.
Here is the sound difference:
1. Harshness down to about 1/3.
2. All music he owns is now relevant and natural even in DSD format.
3. The level in warmness and depth is so much more evolved you would not believe it.
4. And the timing is so crazy, that the sound is just rock solid, and that is even with gear that is subpar.
5. Volume before could only go up to about 50% before it got uncomfortable. Now he is at about 80%.
6. Every detail becomes available while the sound becomes more natural.
7. Detail also very importantly gets a lot better released so it sounds like you have really expensive components in your gear.
Again, you have to be really good at setting it up for it to do a big difference. This it not a plug and play product and then you are done. My friend who is one of the most hardcore audiophiles, used 1000's of hours tweaking it, and making it work right on his streamer.
There is a huge potential to be had when manufactors can figure out how to bypass the internal clock and use a really good external clock in DAC/streamer/CD/digital amps/routers.
It also makes sense when you think about it, you isolate a part so it stays more true and does not effect the rest of the components in a bad way.
But again, you have to have the skill to do it, buying it is not enough. Plus there is a difference in quality between the clock products.
well, I wouldn't trust a manufacturer that makes dacs to be the final word on whether their dac needs an external clock, or not, so I'm already skeptical of Paul and the noise he makes in this video. Not to say I'm the one who knows. Your comment makes sense, as the world becomes more and more deconstructed down to nano particles, more and more accuracy seems a reasonable pursuit. or you can say it's not necessary unless your dac, which we make, sucks, which ours doesn't, which transliterated, is what he said, in a very convoluted manner, IMho
I think dCS is proving for many years that you're wrong. 😏 Did you ever hear the difference their clocks make? It's not subtle..
Sure, a good clock makes a difference, why does it need to be external?
@@mornecoetzee735 , because it's meant as an option. The regular clocks inside dCS DAC's are already great (no one can deny they sound fantastic as-is) but the external ones are significantly better, with their own housing and PS. Although at a significant cost the upgrade is very much audible for anyone with two normal ears (I have tested this). More space, dynamics and timbral color.
@@vincentvanrooij9675 There are no clock generators needed in a basic SPDIF decoder feeding a DAC chip. The clock is extracted from the audio stream by simply counting bit transitions. If you do have a local or external oscillator it must be locked to the audio stream through a slow PLL and/or a buffer memory must also be used. The ultimate DAC clock source must be somehow be phase locked to the audio samples. There is no way around that.
@@andydelle4509 It is locked, otherwise it indeed couldn't work - you're absolutely right about that. I think you're missing the point here though. Just listen to a dCS DAC with and without an optional external dCS clock and the increased SQ when adding the clock is very easy to hear.
@@vincentvanrooij9675 Well a typical audiophile external clock is not locked per se. The DAC has a buffer and the external clock determines the output sample rate. This does work to reduce jitter quite well and is a common technology in many areas of electronics. In fact every CD transport has a buffer memory to eliminate the mechanical errors. But as I explained below, this setup will glitch if the buffer over or under runs. And this will ultimately happen as the two clock signals, the external clock and the SPDIF stream are not synchronous. As jitter implies the SPDIF clock varies up and down, this tends to even out the buffer storage. And of course everytime you swap discs or call up a new file to play, the buffer resets to half full so in the real world the over/under problem never really happens.
No
It's not rocket science. Faster clock, lower jitter, better sound. Clock upgrades are a thing for a reason. Atto and Zepto clocks? Can't wait! Sign me up!!
Of the last 20 videos, 16 questions were from outside the US and 4 were from inside US....We have question about stuff in the US too Paul.
This is so patriotic XD
Why is the nationality of the people asking the questions important to you?
Your keeping count , we should use you for our DACs. Extra transparency guaranteed 😉
@@chrisvinicombe9947 Yeah, I'm a BHK 2.0 model...
@@carstenjunge1327 It's not, I just noticed that most of the questions lately were from overseas. Thought is was an interesting stat. Giving Paul some grief.. Plus I have a question in his pipeline.... 😀
Can anyone spot the Trump supporter in these comments? haha WOW!
build back better word clocks lmao and wall out those pesky mexican jitters
@@hehunches 😂
No