Ayre QB-9 Twenty upgrade

[CG]

11 hours ago, AB835 said:

CG

I have had issues with my ISO REGEN  connected to my QX-5 20 as you described earlier, original USB REGEN functions normally. I installed a simple on/off switch in the power line providing power to the ISO REGEN. Normally power to the ISO REGEN is always on. If there is a problem I cycle the power off and back on to the ISO REGEN and it then works fine till the next time. What is so weird is that I have not had any issues with either REGEN when connected to my QB-9 20!

 

Thanks to you, and the others, for the suggestions.

 

I know the isolation device used in the ISO Regen can be flaky.  The chip designer has even admitted as much.  With mine, if I let the QB-9 DSD play for more than five days, the USB signal would be lost.  A power cycle of the ISO Regen fixed that every time for me, too.

 

I've eliminated the cabling as an issue, through substitution.  So, it's down to the power supply and the ISO Regen itself.  The internal regulator of the ISO Regen has never seen use here, since the QB-9 DSD had its own VBus supply.  So, that's a possibility.

 

Hopefully, I'll have time this weekend to investigate further.  My hope is that the new USB system in the Twenty with its better isolation will eliminate the desire for external isolation and reclocking.  The server I'm using has a pretty complete isolation system from common mode currents that I built myself (and actually measured...) and there's no cables other than the isolated power and USB connection to the server.  That may be fine in itself.  Or, not.  It certainly eliminates the common mode circuit added by the power connection to the ISO Regen.  In any event, I'm sure it will get straightened out.  

 

If the difference between cold and new and ~170 hours of use extends over the next 200-300 hours in the same way, I think it'll be really great.  The minor sonic imperfections I'm hearing were about 90% mitigated over those 180 hours.  Maybe the additional burn-in will get that remaining 10%.  We'll see.  Already, the Twenty has shown itself to be well worth the upgrade price.

[FelipeRolim]

On 12/13/2019 at 3:08 PM, Ryan Berry said:

It's interesting you should mention the driver.  We actually have a brand new one this year that's completely updated and necessary for raw/native DSD playback on Windows.  You can find it on our page at https://www.ayre.com/support/, select Firmware, and then select the USB Driver tab.  Let me know if you have any trouble with it.

 

I tried driver version 4.67.0, but it didn't work. In principle, I experimented with foobar2000 and JRiver, but neither could play sound. I tried using Kernel Streaming and ASIO modes, but none of them worked. foobar2000 gave the message "Unrecoverable playback error: Could not create ASIO buffers". JRiver gave the message "Format not supported". When I reinstall version 1.26S, everything works again. Too bad. I'd like a more updated driver, because this old one (1.26S), still prepared for Windows 7, causes me many problems installing on Windows Server 2019 Datacenter Core Mode.

 

Does this happen because my QB-9 is 192/24, not the "Twenty" version?

 

PS: version 4.67.0 I tried to install on a normal Windows 10. On the dedicated server I use JPLAY FEMTO.

 

PS 2: I would like buffering and latency options to remain available, just as it was in version 1.26S. I do not know if this is possible, but are key parameters to make the most of the DAC.

[FelipeRolim]

A friend installed the new version of the driver and was able to get his QB-9 DSD to work. I don't know if the problem is only in version 192/24 or if it is on my computer, but I'll reiterate the experiments calmly and post again here.

[Ryan Berry]

On 12/17/2019 at 5:04 AM, GAR said:

Quoting from this article: https://audiophilestyle.com/ca/reviews/my-quest-for-a-new-dac-part-2-ayre-qx-8-r737/

 

Everything I like about the Codex, the QX-8 does similarly, but better. I've described the Codex as having a lovely midrange, a satisfyingly solid mid-bass, and a smooth, refined treble. Dynamics-wise, it can be a bit polite-sounding, which I attribute to its necessarily small PSU, to fit its small case.

 

Ryan, do you agree with this statement, in particular, the PSU? Does the same thing apply to the QB-9 Twenty? The size and quality of the PSU seems to be SO important to all audio gear.

 

The power supply has a pretty huge impact, though it's more than simply how big the transformers are.  In fact, oversizing the transformer usually leads to the circuit sounding worse from our experiments here.  For a DAC, hitting high amperage ratings isn't what's going to make the product sound better, but the circuit that follows the transformer is what has a huge impact.  When you're designing a DAC, it's pretty simple to have an properly sized transformer in place, even with fully linear power supplies like we use.  As such, I wouldn't expect many audio manufacturers to struggle ensuring the transformer is large enough to give it enough headroom to sound its best.  Beyond that, it's making sure that the transformer you're using is a high quality transformer becomes key. 

 

The Codex was effectively an extension of the Pono player that we designed the audio circuitry for without using a large battery or making it fit inside of a pocket.  Since the unit wasn't as limited in size as the Pono we were able to give it an extra output circuit to play back to a system as a DAC or be used as a stand alone headphone amplifier.  We really jammed as much as we could get away with inside the chassis, even giving it two separate transformers to allow the digital and analog portions of the circuit to have their own linear power supplies.  The transformers weren't undersized as we could have really just made the chassis larger if we needed it to be.  In the end, we almost made it "too good" as it was a definite bump over sound quality compared to the QB-9 DSD.  We weren't willing to continue to sell a unit for twice the price of the Codex when we knew it sounded worse as that's not how we do choose to do business, so we ultimately killed the line at that time.  I absolutely love the Codex and use one exclusively with my computer at home for gaming and movies and I am convinced it's the best thing out there at its price.  Heck, it was good enough to cannibalize one of our more expensive lines when it was released.

That said, the QX-8 is really a different design in many ways, from the regulator design through the audio design.  Components selected in the QX-8 are more expensive because we could afford them to be and the circuit design overall represents the next evolution in what we've learned in what makes audio sound its best.  This can make it difficult to nail down exactly what makes one product of ours sound different than another, because we're not in the habit of simply reusing the same circuit over and over and slapping it in a different chassis just so we can charge more.  Every time we release a new product we've learned a little more and we want to share what we've learned as soon as possible.  Since everything from audio design to chassis design is done in-house at Ayre, this lets us constantly tweak things...we usually have come up with 4 new things by the time the unit we just designed hits production!  

 

I hope that helps explain it...but simply put: it's a bit more complicated than that.

[Ryan Berry]

On 12/17/2019 at 5:51 AM, CG said:

 

Thanks to you, and the others, for the suggestions.

 

I know the isolation device used in the ISO Regen can be flaky.  The chip designer has even admitted as much.  With mine, if I let the QB-9 DSD play for more than five days, the USB signal would be lost.  A power cycle of the ISO Regen fixed that every time for me, too.

 

I've eliminated the cabling as an issue, through substitution.  So, it's down to the power supply and the ISO Regen itself.  The internal regulator of the ISO Regen has never seen use here, since the QB-9 DSD had its own VBus supply.  So, that's a possibility.

 

Hopefully, I'll have time this weekend to investigate further.  My hope is that the new USB system in the Twenty with its better isolation will eliminate the desire for external isolation and reclocking.  The server I'm using has a pretty complete isolation system from common mode currents that I built myself (and actually measured...) and there's no cables other than the isolated power and USB connection to the server.  That may be fine in itself.  Or, not.  It certainly eliminates the common mode circuit added by the power connection to the ISO Regen.  In any event, I'm sure it will get straightened out.  

 

If the difference between cold and new and ~170 hours of use extends over the next 200-300 hours in the same way, I think it'll be really great.  The minor sonic imperfections I'm hearing were about 90% mitigated over those 180 hours.  Maybe the additional burn-in will get that remaining 10%.  We'll see.  Already, the Twenty has shown itself to be well worth the upgrade price.

 

Hey CG,

 

Thanks for the report on the burn-in.  Since Charley isn't around any longer, I'll be the first one to say that burn-in has a huge impact on our products.  I'd expect that you'll continue to hear improvements for the next few months if our other products are any indicator (and they should be!)  The first few hundred hours are definitely the most impactful, but they really take some time to settle in.  After that, it will stabilize and should sound it's best for years to come.

[Ryan Berry]

5 hours ago, FelipeRolim said:

A friend installed the new version of the driver and was able to get his QB-9 DSD to work. I don't know if the problem is only in version 192/24 or if it is on my computer, but I'll reiterate the experiments calmly and post again here.

Hi Felipe,

 

I'm sorry for the trouble you're having.  Similar to your friend, the unit I personally tested this on was the QB-9 DSD.  I would assume this would work with both as the actual USB portion of circuit is the same between the two units with some minor firmware changes (which is why the old drive worked with both), but I hadn't tested the 192 version explicitly myself.  I'll speak with R&D to see if they have done that experiment and try and recreate it here just in case there's some strange incompatibility we're missing.  Unfortunately, this MAY be after the Christmas break with holiday schedules hitting us, but we'll do our best.

[CG]

8 hours ago, Ryan Berry said:

 

Hey CG,

 

Thanks for the report on the burn-in.  Since Charley isn't around any longer, I'll be the first one to say that burn-in has a huge impact on our products.  I'd expect that you'll continue to hear improvements for the next few months if our other products are any indicator (and they should be!)  The first few hundred hours are definitely the most impactful, but they really take some time to settle in.  After that, it will stabilize and should sound it's best for years to come.

 

Charley and I discussed this on end to try to figure out what might cause that.  Lots of ideas, but no time or energy to really get to root cause.  Besides, even if you do figure out the why, does it really matter in the end?  It just is and letting the gear run takes care of it.  There's other challenges to address in the audio world.  It's the owner's patience that is the big hurdle...

 

Just as a point of reference for everybody else, my own observation in our own system is that the QB-9 Twenty tends to sound kinda "thin" and "dry", especially in the midrange, right out of the (cold) box.  This has perhaps 2/3 gone away after a week or so.  I'll see where it is after the second week of 24/7 operation over the weekend.  YMMV and all that stuff.

[beetlemania]

11 hours ago, CG said:

 

Just as a point of reference for everybody else, my own observation in our own system is that the QB-9 Twenty tends to sound kinda "thin" and "dry", especially in the midrange, right out of the (cold) box.  This has perhaps 2/3 gone away after a week or so.

 

I plugged mine in within minutes of delivery off the UPS truck in winter conditions. There was a vague character that something was not quite right but your description doesn't reflect my experience. Regardless, that character was gone within a day or two and I otherwise heard a notable increase in overall clarity and an open, unbounded soundstage even after I first got it running. I'm now approaching the 250 hour mark and plan to play my reference tracks this weekend.

[CG]

Here's the ~350 hour update.  Probably my last words on audio...

 

That dryness or thin characteristic I mentioned before is almost entirely gone.  Who says burn-in isn't a thing?

 

It turns out that the problem with the UpTone ISO Regen was that the USB adaptor widget they sell wasn't fully engaging the USB connection in the QB-9.  A modest, ahh, circumcision of the heat shrink on the adaptor fixed that up.  The ISO Regen now does make a small improvement to the sound.  The aural picture is just a bit sharper and clearer.  Worth it?  I guess we'll see after another week of burn-in.  The Regen gets rather warm when used with the QB-9 Twenty, even with a 7.5 VDC supply.  If I keep the Regen in the system, I'll probably build a new supply that provides around 6 VDC to the Regen.  The regulator UpTone uses for the VBus supply in the Regen has a very low dropout voltage, so going from 7.5 VDC down to around 6 VDC is well within spec and would reduce the power dissipation on that regulator by more than a factor of two.  The added complexity and the additional current loops when using a device like the ISO Regen is just one of those trade-offs.  The other question is whether the Regen is providing improvement by cleaning up the VBus or whether it's due to the USB isolation and reclocking.  Or, both.  So, there may be other solutions that do the job better.  Time will tell.

 

Finally, Kimmo released a new version of Daphile today.  One of the features is that it now supports Native DSD playback with the QB-9 Twenty.  I was able to play a DSD256 track without a problem.  The QB-9 display even said so.  I was also able to play a 384K PCM track, too.  I compared one of the free demo tracks from 2L in both PCM and DSD256 format.  Personally, I liked PCM better.  But, they were close.  Again, YMMV, etc.

[desbiss]

1 hour ago, CG said:

Here's the ~350 hour update.  Probably my last words on audio...

 

That dryness or thin characteristic I mentioned before is almost entirely gone.  Who says burn-in isn't a thing?

 

It turns out that the problem with the UpTone ISO Regen was that the USB adaptor widget they sell wasn't fully engaging the USB connection in the QB-9.  A modest, ahh, circumcision of the heat shrink on the adaptor fixed that up.  The ISO Regen now does make a small improvement to the sound.  The aural picture is just a bit sharper and clearer.  Worth it?  I guess we'll see after another week of burn-in.  The Regen gets rather warm when used with the QB-9 Twenty, even with a 7.5 VDC supply.  If I keep the Regen in the system, I'll probably build a new supply that provides around 6 VDC to the Regen.  The regulator UpTone uses for the VBus supply in the Regen has a very low dropout voltage, so going from 7.5 VDC down to around 6 VDC is well within spec and would reduce the power dissipation on that regulator by more than a factor of two.  The added complexity and the additional current loops when using a device like the ISO Regen is just one of those trade-offs.  The other question is whether the Regen is providing improvement by cleaning up the VBus or whether it's due to the USB isolation and reclocking.  Or, both.  So, there may be other solutions that do the job better.  Time will tell.

 

Finally, Kimmo released a new version of Daphile today.  One of the features is that it now supports Native DSD playback with the QB-9 Twenty.  I was able to play a DSD256 track without a problem.  The QB-9 display even said so.  I was also able to play a 384K PCM track, too.  I compared one of the free demo tracks from 2L in both PCM and DSD256 format.  Personally, I liked PCM better.  But, they were close.  Again, YMMV, etc.

Hi CG,

 

I’ve always found 500 hrs to be the point where everything Seems to be truly settled in, and at home. Certainly with my Ayre gear, but with almost everything component-wise in my experience. 
 

It’ll get more coherent in the last stretch to 500.... AND that’s where the presence and emotion seem to come in to their fullest.

Question - you’re playing in Native DSD 256? Jealous... 🙂

[CG]

13 minutes ago, desbiss said:

Question - you’re playing in Native DSD 256? Jealous... 🙂

 

Yeah.  Native.  The display on the QB-9 even says "256".  

 

Honestly - it's not that big a deal.  I heard zero angels singing.  I'm certain the sound quality is more of a function of how the original recording was formatted and all that jazz.

 

Then again, I may be the wrong guy to talk about this.  Presence and emotion are secondary to me compared to transparency and lack of crud.  Presumably, when you attain good performance in those areas, your brain should be able to take care of the rest.  (My wife is a real-life musician.  Or was, anyway.  Was a musician, that is - she's still real-life.  She prefers transparency so she can hear what the musicians are actually playing.  For emotion, she points out that pretty much all the musicians she knows can read the sheet music for a piece and derive the emotion from that.  That, says she, is why musicians usually don't play in the audiophile sandbox.  It's already in their heads.  The emotion processing is done there.)

 

[desbiss]

13 hours ago, CG said:

Presence and emotion are secondary to me compared to transparency and lack of crud.  Presumably, when you attain good performance in those areas, your brain should be able to take care of the rest.

Thanks for confirming Native 256 CG! I upsample everything to DSD using Roon but am not yet able to play Native DSD, only up to DSD 128 via DoP. 
 

I find a balance is achieved in the final 500 hour stretch. The QB-9 certainly hits high in terms of transparency and lack of crud before then. It just seems that the final stretch is where the icing on the cake becomes most apparent, with the presence and emotion solidifying. 
 

Thanks for your reply!

[beetlemania]

Approaching the 400 hour mark. I think the sound has stabilized although I haven’t been careful about playing the same songs at varying intervals.

To my ears, in my system and room, the main improvements are 1) improved clarity and resolution and 2) an expanded, almost unbounded, soundstage. The overall sound is refined, pretty much top-shelf albeit it’s been a while since I’ve heard competing contenders. I always considered the DSD version to be excellent but the Twenty is a notable step up. Easier than ever to relax into the music. 

The increased clarity makes it easier to hear individual voices and instruments during complex passages. The added resolution adds a bit more texture all around; performers and instruments are more “fleshy”. This is most obvious in the midrange but I also hear improvement in the bass and, to a lesser extent, the treble. Ayre claims quieter backgrounds and that sounds accurate to me. Well recorded music is more dimensional with an expansive soundstage. I don’t hear any aspect of performance that is a step backwards from the DSD version. 

This is the fourth time I’ve upgraded an Ayre product! Been really happy every time. Thanks Ryan, Ariel, and crew 😊

[Sonny Oh]

Can any QB9 version (non-DSD old) be upgradeable to Twenty? Also at the same cost? How long the waiting list is? 

[beetlemania]

2 hours ago, Sonny Oh said:

Can any QB9 version (non-DSD old) be upgradeable to Twenty? Also at the same cost? How long the waiting list is? 

From the Ayre website, this upgrade is applicable to all versions. I think they change everything except the chassis, display and power supply (although I think the DSD version had a power supply modification from the original version?). 

[zek4u]

I get mine Thursday and super excited. It’s up against an Aqua La Voce S3. 

[GrantsPCA]

Are you in a position to provide an update on how the Ayre QB-9 Twenty compares with your Aqua?

[The Computer Audiophile]

Let the fun begin. 
 

QB-9 DSD vs. QB-9 DSD Twenty

 

[barrows]

5 minutes ago, The Computer Audiophile said:

Let the fun begin. 
 

QB-9 DSD vs. QB-9 DSD Twenty

 

Cool!  i am looking forward to hearing your thoughts on the comparison.  Also a little bit on how the Ayres compare (just generally) to typically, more expensive DACs.

[Mercman]

Looking forward to your results!

[zek4u]

I let the QB-9 Twenty break in for 200-300 hours and it really did get more organic, not as much as the La Voce but started to sound extremely nice on the AX-5 Twenty. Would say punchier, more dynamic and more resolution without being bright when compared to the La Voce. The La Voce still has this magic in the midrange and timbre is impeccable. Think with the right equipment, each is a winner. 
 

Anyone here pair a QB-9 with Naim and have good results?

 

I need to figure out how to change my signature on this site as the system has changed completely. 

[The Computer Audiophile]

I'm using optical Rendus to feed each QB-9, then into a Constellation Audio preamp, then into the direct input of the Constellation monoblocks. This input bypasses a gain stage because it isn't needed with this preamp. The whole setup sounds really good. 

 

I haven't had a QB-9 in my system for many years. I'm very impressed. 

[zek4u]

That’s my next step, compare an opticalRendu to my ultraRendu. Maybe also an audiophile grade switch for my NAS and media server. Soon I will also be trying the QB-9 Twenty on a Supernait 2 as it seems great on a Nait 5i. Would like to also bring a Chord Qutest and Dave into the mix.  

[bsinc]

Have played my QB 9 Twenty all day, everyday now for almost 90 days. I don’t think 2-300 hrs gets  it to where it is illustrative of the full capability of the unit.  The owners manual for the original specified up to 500 hours.  After two months I felt it was pretty much where it was going to be.... but I was wrong. After another month it continued to show subtle, incremental improvement. Not huge performance leaps but more refinement on the manifested qualities and a continuation of refinement in depth and high frequencies. As we get near the end of the fourth month it hasn’t changed much at all so I’ll call it “done”. It doesn’t surprise me with some new attribute anymore, but it delights me everyday with some sessions lasting 8 hrs. It takes longer than you might think to run it in.

[FelipeRolim]

I had scheduled the upgrade for February, but a number of things happened and I'll have to postpone it a little longer. I don't want to delay, as I'm very optimistic about the improvements. I currently have version 192/24, and tell you: try the Synergistic Research Orange fuse. If you are surprised by the QB-9, wait until you hear it with this fuse...

 

Please, keep the topic updated with your impressions and also with comparisons with other DACs....