[Nada]

• The science says that external clocks introduce jitter / noise, almost always. It's possible that this change in sonic performance is preferable (people don't always like what measures well).
• Goldenone did a review of the Gustard x26 with an ext clock (which I just found) and his measurement analysis revealed overall increased jitter with the external clock, but potentially less jitter performance in the lower frequencies. I believe, but can't be sure, the R26 is using the same femto clock synthesiser as the x26. He is using a different ext clock and the X26 dac but I hypothesise this is what is happening in my set up

Great review and analysis thanks.

The measurements of the Gustard X26 with the superb external clock in the Audio Precision is the best model we have for the R26, given the similarities in the clocking design

@GoldenOne has done a brilliant analysis and write up:

So a question that many people will have: “Is adding a 10Mhz clock worth it?”

The short answer: No, 10Mhz clocks were never intended to improve jitter performance and almost never will do so. They are a tool to solve a problem in professional environments that does not exist in home setups.

As you can see, quite notably poorer jitter performance than without the 10Mhz clock.

So why is this?
The reason why a 10Mhz clock will worsen jitter isn’t to do with the particular 10Mhz clock used (though it will have an impact of course), even if you used the most perfect 10Mhz clock in the world you’re still going to get poorer performance simply due to how 10Mhz clocks are actually implemented.
https://goldensound.audio/2022/04/14/gustard-x26-pro-measurements-f20-firmware-10mhz-clock/

 

[BShaw]

Once again a great write-up @Odezra. But darn! I’m naturally disappointed you didn’t experience the improvements I got - and still very much enjoy - in my system. I gotta say I feel bad for effectively encouraging you to spend a decent amount of money on this if it ends up bringing your system such a minor improvement. I’ve tried to acknowledge I’ve been a bit of a convert/zealot on this (and still am) and say YMMV a few times but appreciate that far from cancels out the enthusiasm I’ve expressed and my positive observations.

In terms of why the different result? I don’t claim to be any technical expert here, just a hifi enthusiast trying things, and reading as much as I can. I am aware of others in this and other threads getting very modest results and a lot of variability with cables etc. And others like me who get great results and would never give theirs up.

I can and will theorise on the reasons for the differences, but really it is just very slightly informed guesswork. So here goes: variability in phase noise of OCXOs from unit to unit (some suggestion early OCK-1s may have phase noise up to 5db better than the claimed spec of -110db. No longer the case with OCK2s getting the best OCXOs. One experienced clock user was surprised/impressed I could hear cable differences with a ‘mere’ -110db unit, so perhaps I lucked in, I’ve no way of measuring it); variation in OCK1 units’ quality of internal wiring /circuits resulting in greater internal impedance mismatches/reflections; shielding & level of internal reflectivity (return loss) of cable used (what cable did you use?); length of time OCK-1 has been left on - how long have you had it?

So only obvious controllable variables, things to try that jump out to me are: more time (eg days/week); sine rather than square output (former better for me); putting DAC in NOS/DSD Direct mode (also better); maybe a different cable but totally understand if you don’t want to go there - good money after bad etc. That’s it off the top of my head, will advise if I think of anything else.

Again, darn!

Jake

Jake-what BNC cable are you using? I was going to order one from Danny at Requisite, I use their ICs, but having doubts about twisted pair geometry for this application, also not confident about the impedance. JH labs says their 75 ohm cable will work, beautiful cable, again concerned about impedance….Beatechnik’s cable seems like it might be ok, but calls the conductor “high purity copper. Since you’ve had such positive results, I figured maybe I should follow your lead.

 

[Jake2]

Jake-what BNC cable are you using?

Hi Bshaw I'm using the Gustard C2, but it's > USD150 so I'm wary of recommending it, especially given it is on top of the investment in a clock which is no sure thing per @Odezra 's experience.

Others have found much cheaper cables (as in 1/5th the price) that they consider work well, so the C2, despite working for me - and this will surprise no one at all - is probably a bit overpriced! There's a recommended <USD20 one - nearly as 'good' as the C2 - can't recall the name offhand. Will update later.

I found this Mini-Circuits ~USD35 cable which I just had delivered from Mouser. Haven't tried yet as have been away and my router has died, so no streaming dammit, but looks as good or better built than the C2. No one recommended that cable, I selected it - as from a specialist RF cable/measurement manufacturer and had good specs in/characteristics that from what I've read are relevant to use as a clock cable (well shielded, very low 'return loss' spec re internal reflections at 10mhz).

So no need to spend as much as I did.

 

[BShaw]

Hi Bshaw I'm using the Gustard C2, but it's > USD150 so I'm wary of recommending it, especially given it is on top of the investment in a clock which is no sure thing per @Odezra 's experience.

Others have found much cheaper cables (as in 1/5th the price) that they consider work well, so the C2, despite working for me - and this will surprise no one at all - is probably a bit overpriced! There's a recommended <USD20 one - nearly as 'good' as the C2 - can't recall the name offhand. Will update later.

I found this Mini-Circuits ~USD35 cable which I just had delivered from Mouser. Haven't tried yet as have been away and my router has died, so no streaming dammit, but looks as good or better built than the C2. No one recommended that cable, I selected it - as from a specialist RF cable/measurement manufacturer and had good specs in/characteristics that from what I've read are relevant to use as a clock cable (well shielded, very low 'return loss' spec re internal reflections at 10mhz).

So no need to spend as much as I did.

Thanks Jake, that one is "copper clad steel". I'm used to audio cables of other sorts, it's been surprising to find a lot of the BNC cables made with an aluminum, steel, or TP copper conductors, and some of them like the Gustard C2 are definitely overpriced iMO for what they are. But maybe someone can quench my curiosity about why the conductor material doesn't seem as important as with other cables we use in our systems? Maybe things like insertion loss and reflections are the most critical, and the signal is just a narrow (or single) frequency? I'm p*ssing in the dark here, have no knowledge about clock cables. Any experts? I tried searching several ways, couldn't find a relative thread....

 

[Odezra]

Once again a great write-up @Odezra. But darn! I’m naturally disappointed you didn’t experience the improvements I got - and still very much enjoy - in my system. I gotta say I feel bad for effectively encouraging you to spend a decent amount of money on this if it ends up bringing your system such a minor improvement. I’ve tried to acknowledge I’ve been a bit of a convert/zealot on this (and still am) and say YMMV a few times but appreciate that far from cancels out the enthusiasm I’ve expressed and my positive observations.

In terms of why the different result? I don’t claim to be any technical expert here, just a hifi enthusiast trying things, and reading as much as I can. I am aware of others in this and other threads getting very modest results and a lot of variability with cables etc. And others like me who get great results and would never give theirs up.

I can and will theorise on the reasons for the differences, but really it is just very slightly informed guesswork. So here goes: variability in phase noise of OCXOs from unit to unit (some suggestion early OCK-1s may have phase noise up to 5db better than the claimed spec of -110db. No longer the case with OCK2s getting the best OCXOs. One experienced clock user was surprised/impressed I could hear cable differences with a ‘mere’ -110db unit, so perhaps I lucked in, I’ve no way of measuring it); variation in OCK1 units’ quality of internal wiring /circuits resulting in greater internal impedance mismatches/reflections; shielding & level of internal reflectivity (return loss) of cable used (what cable did you use?); length of time OCK-1 has been left on - how long have you had it?

So only obvious controllable variables, things to try that jump out to me are: more time (eg days/week); sine rather than square output (former better for me); putting DAC in NOS/DSD Direct mode (also better); maybe a different cable but totally understand if you don’t want to go there - good money after bad etc. That’s it off the top of my head, will advise if I think of anything else.

Again, darn!

Jake

No worries Jake. I knew when I purchased the clock, it was a 50/50 (and I had seen other reports around the traps praising it). Most of those reports though were on PC driven systems (unlike mine) - there was a dearth of reports referencing the clock with a LAN set up straight to the gustard DAC.

I will give it a few more days and check out the Sine Wave output.

Where are you at with HQplayer? I am back on the Gustard's OS on the 'mid' filter setting. I was finding HQplayer a little fatiguing on extended listens (played with a wide variety of filters).

 

[Jake2]

No worries Jake. I knew when I purchased the clock, it was a 50/50 (and I had seen other reports around the traps praising it). Most of those reports though were on PC driven systems (unlike mine) - there was a dearth of reports referencing the clock with a LAN set up straight to the gustard DAC.

I will give it a few more days and check out the Sine Wave output.

Where are you at with HQplayer? I am back on the Gustard's OS on the 'mid' filter setting. I was finding HQplayer a little fatiguing on extended listens (played with a wide variety of filters).

Cheers man, and all the best with that, whatever you try or conclude.

Re HQP I've been out of town for 5 days so no change from whatever I last posted - preference for HQP PCM768 + PCM NOS on. Certainly prefer it to other HQP setting as the most dramatic/weighty sound. But appreciate some others find oversampling not as natural (certainly I agree some DSD isn't). I haven't recently compared it to HQP native signal bitrate (i.e. HQP NOS) + R26 NOS on or off or just Roon native and/or upsampled. I should do. So many variables, so many possibilities. What I can confidently say is that for less perfect material that isn't and can't easily be upsampled externally (e.g. s/pdif from TV) I find PCM NOS Off (i.e. R26 is oversampling ) is a lot more forgiving and smooth/natural than NOS on.

Speaking of TV, on my return yesterday evening my broken router ethernet (wifi still ok) meant I couldn't listen to any Mac HQP music, just watched a bit of TV. Tried the first episode of the somewhat panned Rings of Power, which has a massive production budget, so visuals and sound is top shelf. In fact the sound in some scenes, for e.g. in the closing credits with some seriously big bass lines, drums and clashing swords was so unusually dynamic, grand in scale and and engaging for a TV s/pdif sourced signal I couldn't resist. I AB'd (with just the Sopras, subs were disconnected) whether I could hear the clock on or off with that SPDIF source, albeit first reclocked by the iFi Ipurifier s/pdif 2. The Ifi made a big improvement vis straight into the R26 when I first tried it. Marks out of ten, if straight into R26, no Ifi and internal clock was say 3/10 (thin and edgy), then the ifi would've taken it to say 6-7 and adding the OCK-1 maybe took it to 8. So much better bang for buck of the Ifi for the that use case. By comparison I could hear the external clock on/off difference - weightier bass, longer and more nuanced reverberation of swords clashing, along with a hint of actual soundstage depth (unheard of for a TV source IME), but it was more subtle. Subtle enough to be a fair risk of missing it in a blind test, but big enough to make the overall sound noticeably more engaging, if harder to immediately pinpoint the particular differences, if that makes sense. Whereas from memory using the OCK1 with R26 fed HQP via ethernet has a definitely more dramatic effect in my system - say +3 notches. Numbers just for illustrative purposes. Just my 2c of experiences once again - take with a pinch of salt if you will. But the result for me is my TV sound with R26 + Ifi + OCK1 in play with just two speakers - not even the subs connected - is now good enough that my motivation to hook up my mothballed Denon X3700H receiver and surround speakers is really low. Add them to the new years' garage sale list methinks.

 

[camrector]

DSD over LAN sounds amazing!

 

[Alenotta]

I downloaded the Roon trial a while back so I could test the direct LAN. It sounded really nice, but I actually sort of preferred the flavor the AudioQuest Carbon usb c cable gave things. I think the LAN was likely more accurate, but the advantage seemed more minimal than other changes I've made. The usb 2 side of the Gustard is slow, but it is going into a Thunderbolt 4 port so I'd like to think there is some speed benefit. Also, I stream all my music so dealing with Tidal directly has been a better experience. I too am hoping for that Tidal Connect...

I haven't noticed anything that the October firmware update fixed. One of my main gripes is still there - the popping sound when skipping around in Tidal on MQA tracks when exclusive mode is on with PCM NOS. Luckily PCM NOS doesn't seem to make a big difference to my ears.

 

[Forgisound]

I downloaded the Roon trial a while back so I could test the direct LAN. It sounded really nice, but I actually sort of preferred the flavor the AudioQuest Carbon usb c cable gave things. I think the LAN was likely more accurate, but the advantage seemed more minimal than other changes I've made. The usb 2 side of the Gustard is slow, but it is going into a Thunderbolt 4 port so I'd like to think there is some speed benefit. Also, I stream all my music so dealing with Tidal directly has been a better experience. I too am hoping for that Tidal Connect...

I haven't noticed anything that the October firmware update fixed. One of my main gripes is still there - the popping sound when skipping around in Tidal on MQA tracks when exclusive mode is on with PCM NOS. Luckily PCM NOS doesn't seem to make a big difference to my ears.

AQ Carbon is a very good USB cable. It would be fair if you have a similar quality LAN cable.

 

[manishex]

The Gustard is excellent, I prefer it on the fixed output which is 5v, however I do wonder how much this DAC compares to others is also due to the voltage difference. The pegasus is 3.55v and the holo may 5.8v. I'm somewhere around -10db to get my amp to play at the same volume as it was on the pegasus.

 

[565321]

Level matching is paramount to comparisions, just adjust the level.

 

[manishex]

Level matching is paramount to comparisions, just adjust the level.

Right, but the Gustard doesn't sound as good as it can when I do so, is that fair? It helps take some load off the amp which is another factor in the equation (or maybe the amp prefers higher?). I find lower voltage output pulls back the treble a bit, so was alluding to the warmer/weightier sound of the musician/denafrips to their lower voltages? And if people did want something in between perhaps lower the output on the dac a few db.

 

[Odezra]

Right, but the Gustard doesn't sound as good as it can when I do so, is that fair? It helps take some load off the amp which is another factor in the equation (or maybe the amp prefers higher?). I find lower voltage output pulls back the treble a bit, so was alluding to the warmer/weightier sound of the musician/denafrips to their lower voltages? And if people did want something in between perhaps lower the output on the dac a few db.

The DAC outputs 2.5v on RCA - have you tried that on your burson instead of xlr and found any differences?

 

[manishex]

The DAC outputs 2.5v on RCA - have you tried that on your burson instead of xlr and found any differences?

That's to be expected since single-ended is half the balanced output. I haven't tried it yet, although I have no issues with the sound, it's more of a hypothesis for those who miss some aspects of other lower voltage dacs or something to consider during DAC comparisons.

 

[fullbass]

I find lower voltage output pulls back the treble a bit, so was alluding to the warmer/weightier sound of the musician/denafrips to their lower voltages?

for me, the musician does not sound better than the gustard at any volume level. and a gustard is more weightier and bassy for me than pegasus which has a very primitive bass in comparison