Sabaj D5 vs. SMSL M500 vs. Topping DX7 Pro: measurement comparison

[NielsMayer]

A better test would be a 50 Hz square wave. Basically half an impulse response that repeats over and over again. Probably the only way to properly characterize a DAC in the frequency and phase domains.

Or a series of 32 , 320, 3200, and 32000 tone tests attempting to construct a 50 Hz square wave from its precisely phase-matched Fourier components (i.e. to make the edges square up to the bandwidth of the device (say 40khz to be safe a 96k sample rates).

I bet the noise floor rises disproportionately with the number of tones. I could look up the articles and quote them just to have ppl make grunting noises in response so what's the point.

Be careful your "science" is not a semmelweis reflex: "Mob behavior found among primates and larval hominids on undeveloped planets, in which a discovery of important scientific fact is punished". (Timothy Leary).

 

[BDWoody]

Be careful your "science" is not a semmelweis reflex: "Mob behavior found among primates and larval hominids on undeveloped planets, in which a discovery of important scientific fact is punished". (Timothy Leary).

 

[NielsMayer]

No, I would not worry whatsoever.

Thank g*d our electronics are designed in India or China where people learn the math and engineering like i did and create products, instead of spewing doctrinaire nonsense and being blowhards on some random internet website...

From the marketing of a device reviewed here claiming 1/2 picosecond jitter (while y'all be wanking off to devices with femtosecond clocks -- apparently all that huge expense and engineering was for all for show right??) -- no its because without a precise clock your high sample rates and bit depths are just garbage.

"Optical had to go… by definition it has a jitter of 4ns. That’s 4000ps."

In case no-one here has characterized jitter in a digital system (I have and got random
failure rates down in millions of shipping boards) it is a cumulative process. 4 ns on the receiving end; 4ns on transmitting end. And thats just the LED and photoreceptor . At 192 k you can hear 11ps of jitter and the optical interface, independent of the rest of the jittery chain just added 8ps on its own.

 

[MediumRare]

At 192 k you can hear 11ps of jitter and the optical interface, independent of the rest of the jittery chain just added 8ps on its own.

Will this test work to check the audibility of jitter? https://hydrogenaud.io/index.php?topic=106076.0

 

[TC!!]

Thank g*d our electronics are designed in India or China where people learn the math and engineering like i did and create products, instead of spewing doctrinaire nonsense and being blowhards on some random internet website...

From the marketing of a device reviewed here claiming 1/2 picosecond jitter (while y'all be wanking off to devices with femtosecond clocks -- apparently all that huge expense and engineering was for all for show right??) -- no its because without a precise clock your high sample rates and bit depths are just garbage.

"Optical had to go… by definition it has a jitter of 4ns. That’s 4000ps."

In case no-one here has characterized jitter in a digital system (I have and got random
failure rates down in millions of shipping boards) it is a cumulative process. 4 ns on the receiving end; 4ns on transmitting end. And thats just the LED and photoreceptor . At 192 k you can hear 11ps of jitter and the optical interface, independent of the rest of the jittery chain just added 8ps on its own.

"wanking off" - you're British, right?

 

[Shoaibexpert]

Thank g*d our electronics are designed in India or China where people learn the math and engineering like i did and create products, instead of spewing doctrinaire nonsense and being blowhards on some random internet website...

From the marketing of a device reviewed here claiming 1/2 picosecond jitter (while y'all be wanking off to devices with femtosecond clocks -- apparently all that huge expense and engineering was for all for show right??) -- no its because without a precise clock your high sample rates and bit depths are just garbage.

"Optical had to go… by definition it has a jitter of 4ns. That’s 4000ps."

In case no-one here has characterized jitter in a digital system (I have and got random
failure rates down in millions of shipping boards) it is a cumulative process. 4 ns on the receiving end; 4ns on transmitting end. And thats just the LED and photoreceptor . At 192 k you can hear 11ps of jitter and the optical interface, independent of the rest of the jittery chain just added 8ps on its own.

So you're saying optical jitter by definition is audible (4ns)...but doesn't the femto second clock if the DAC reclock and correct it? Also, seems you do consider the DACs being compared here to have inaudible jitter (femtoseconds)? I mean If you agree then what is the argument about...other than who is the champion of auditory science here?!

 

[Veri]

So you're saying optical jitter by definition is audible (4ns)...but doesn't the femto second clock if the DAC reclock and correct it?

It does... AKM, Sabre and Cirrus Logic, they will all process the digital input in such a way that jitter isn't a problem. What can be a problem is when your device loses the 'lock' on your input which can happen with particularly jittery sources.

But to say you will hear that jitter? Lol no. You will either hear music or nothing, if the connection breaks up. Claiming anything else is head-fi hocus pocus, especially with nothing to substantiate such a claim ...

 

[Shoaibexpert]

It does... AKM, Sabre and Cirrus Logic, they will all process the digital input in such a way that jitter isn't a problem. What can be a problem is when your device loses the 'lock' on your input which can happen with particularly jittery sources.

But to say you will hear that jitter? Lol no. You will either hear music or nothing, if the connection breaks up. Claiming anything else is head-fi hocus pocus, especially with nothing to substantiate such a claim ...

Jitter is an anomaly within the time dimension...can it be expressed in Dbs like most reviews on ASR (I know it must be but just don't know the link). What I would expect is that the sequence and time difference of bits processed by the DAC would result in faster or slower playback of music...at femtoseconds differences (yeah! I am a layman when it comes to these technical things...but I'm Keen to learn )

 

[Veri]

Jitter is an anomaly within the time dimension...can it be expressed in Dbs like most reviews on ASR (I know it must be but just don't know the link). What I would expect is that the sequence and time difference of bits processed by the DAC would result in faster or slower playback of music...at femtoseconds differences (yeah! I am a layman when it comes to these technical things...but I'm Keen to learn )

No you're right. It's just that of all DACs measured so far, very few did not pass the 'jitter' test, as in audible artifacting.
Generally it's a matter of having coaxical/optical connection issues, or not.

 

[Shoaibexpert]

No you're right. It's just that of all DACs measured so far, very few did not pass the 'jitter' test, as in audible artifacting.
Generally it's a matter of having coaxical/optical connection issues, or not.

Thanks. Is audible artifacts measurements a reliable testvfir jitter then?

 

[Veri]

Thanks. Is audible artifacts measurements a reliable testvfir jitter then?

You're not going to hear jitter unless something is seriously broken. You can look at some examples if you're curious
http://archimago.blogspot.com/2018/08/demo-musings-lets-listen-to-some-jitter.html
http://www.sereneaudio.com/blog/what-does-jitter-sound-like

 

[Shoaibexpert]

You're not going to hear jitter unless something is seriously broken. You can look at some examples if you're curious
http://archimago.blogspot.com/2018/08/demo-musings-lets-listen-to-some-jitter.html
http://www.sereneaudio.com/blog/what-does-jitter-sound-like

Maybe a tangent but I was wondering why Topping DX7PRO used the 3 Accusilicon chips to manage jitter...these cost around $100 alone...whilst M500 has far cheaper components...I mean... surely Topping engineers are seeing or hearing something we are not coz both DACs result in similar audible artifacts as measured on ASR and these Chinese companies won't increase their cost of production unless it makes a real difference in performance (something we can't see).

 

[RickSanchez]

You're not going to hear jitter unless something is seriously broken. You can look at some examples if you're curious
http://archimago.blogspot.com/2018/08/demo-musings-lets-listen-to-some-jitter.html
http://www.sereneaudio.com/blog/what-does-jitter-sound-like

Thanks for posting these articles. I also viewed (briefly) the study that was linked in Archimago's post: "Theoretical and Audible Effects of Jitter on Digital Audio Quality" (Eric Benjamin & Benjamin Gannon, 1998)

I was hoping it would be really insightful, but the fact that the study is 20+ years I think makes it far too out-of-date to be relevant given the advances in DAC chips and DAC implementations that we see on the market today. As spelled out in section 3.0:

• "As reported by Adams[3], the intrinsic susceptibility of DACs to jitter varies according to their basic operational principles. His analysis divides DACs into three types: current division, delta-sigma with continuous-time reconstruction filter, and delta-sigma with switched capacitor reconstruction filter. The analysis shows that current division DACs and delta-sigma DACs with switched capacitor filters have similar sensitivity to jitter. At the time of the writing of this paper the authors are not aware that any manufacturers of delta-sigma DACs are using continuous time reconstruction filters and no such DACs were available for testing or listening. For that reason this class of DAC is not investigated in this paper."

 

[Veri]

surely Topping engineers are seeing or hearing something we are not coz both DACs result in similar audible artifacts as measured on ASR and these Chinese companies won't increase their cost of production unless it makes a real difference in performance (something we can't see).

Haha. I see your point but these manufacturers are competing for the cleanest output. One could even argue why buy a $600 DAC at all. You see these DACs nearing -150dB or even -160dB in digital jitter. Is a value this low necessary? Hardly. But it is very clean. From a practical standpoint this is very much in 'measurebating' territory.

Also note Amir's annotation with many lower priced DACs like the Topping DX3. Some spikes visible, not an audible issue.

 

[SIY]

Thanks for posting these articles. I also viewed (briefly) the study that was linked in Archimago's post: "Theoretical and Audible Effects of Jitter on Digital Audio Quality" (Eric Benjamin & Benjamin Gannon, 1998)

I was hoping it would be really insightful, but the fact that the study is 20+ years I think makes it far too out-of-date to be relevant given the advances in DAC chips and DAC implementations that we see on the market today.

Unfortunately, no advances in ears since then. So we can definitely see improvements in measurements, but we had already hit audible transparency. Note how extreme the jitter had to be to hit audibility- and that was on test tones designed to highlight the issue. Just about any engineered DAC is orders of magnitude better, even in 1998.

 

[dkinric]

Jitter is an anomaly within the time dimension...

Can you please cite you source - which Star Trek episode was this?
Spoiler... most of them

 

[Shoaibexpert]

You're not going to hear jitter unless something is seriously broken. You can look at some examples if you're curious
http://archimago.blogspot.com/2018/08/demo-musings-lets-listen-to-some-jitter.html
http://www.sereneaudio.com/blog/what-does-jitter-sound-like

After reading and listening to the samples ...im in the camp that you possibly cant hear jitter on modern DACs...Period!

 

[BDWoody]

It's remarkable what actually testing yourself can do! So few bother...they'd prefer to argue...

 

[NielsMayer]

So you're saying optical jitter by definition is audible (4ns)...but doesn't the femto second clock if the DAC reclock and correct it? Also, seems you do consider the DACs being compared here to have inaudible jitter (femtoseconds)? I mean If you agree then what is the argument about...other than who is the champion of auditory science here?!

The clocks on the DAC are for the USB section. If you're concerned about jitter, use the USB interface, as that is what was measured to have low jitter.

IMHO and contrary to the perpetual disinfo you get from this site, all the other inputs do not "reclock" -- jitter in, garbage out. Reclocking is in and of itself potentially problematic, although there are claims that a "dual PLL" design fixes such issues. I have not seen a consumer DAC with this feature other than the ADI-2 "steadyclock" (TBD -- does it suffer from the problems alluded to in the paper about dual PLL reclocking i posted previously).\

The ADI-2 DAC supports sample rates between 44.1 kHz and 768 kHz. Furthermore, RME's SteadyClock FS guarantees exceptional performance in all clock modes. Thanks to a highly efficient jitter suppression, the DA-conversion always operates on highest sonic level, being completely independent from the quality of the incoming clock signal.

Does the DX7Pro /M500/etc have a "steadyclock" -- if it did it would have an FPGA and a bunch of other crap and Topping or SMSL would have pretty marketing pictures and statements about it as a feature.

The inclusion of the IIS input is because that form of interface was designed for low jitter required by the higher bitrates it can carry.
The audible jitter at 44ps shown for 44.1k/16 bits decades ago needs to be proportionally scaled per bit depth 16->24, or 16-32 (do the math, that's much much smaller, aka why we worry about "femptosecond clocks") and bitrate 44.1k->384k to be state of the art.

The Coax and AES/EBU intefaces are next in terms of low-jitter. Sometimes AES/EBU can add jitter, despite it's potential of reducing noise and therefore that form of induced jitter. But there are two differential inputs whose hysteresis curves (these are digital inputs made out of transistors inside the chips) double the potential jitter sources at the border between analog and digital signal design).

The optical is the most jittery of all of them all due to the physical properties of photodiodes and receptors themselves, and thus suitable for hooking up to your TV or computer to watch youtube cat videos.

 

[Arvind]

I am curious to see comparison of unbalanced SINAD performance of all these three DACs.