Bryston Outboard DAC

James Tanner · « **on:** 1 Dec 2005, 01:19 pm »

Hi All,

We have had a lot of interest in an outboard (PCM) DAC from Bryston so we are going to go ahead. The DAC will have 2 models - one that can be used with our MPS-2 power supply and one that will contain a built in power supply.

Any input is welcomed.

james

kfr01 · « **Reply #1 on:** 1 Dec 2005, 02:16 pm »

Priced at least in the same ballpark as the Benchmark dac-1, if possible. Most folks that want an outboard DAC just want an outboard DAC. Not a headphone amp. Not a preamp. I, for one, hate to see features I don't want or need add to the cost.

Balanced and single-ended outs would be nice.

If you do decide to go with a volume control, make it remote optional and one of high quality.

If you decide to stray from a minimalist approach, how about a USB input? This would distinguish Bryston's from just about everybody elses right now.

Some quick thoughts...

Phil A · « **Reply #2 on:** 1 Dec 2005, 02:27 pm »

I'd 2nd the USB input idea. I think media servers and PCs for HT and music are becoming more mainstream. The playback of music from a hard drive is something that can be greatly complimented by a DAC with friendly inputs

nicolasb · « **Reply #3 on:** 1 Dec 2005, 04:50 pm »

Quote

We have had a lot of interest in an outboard (PCM) DAC from Bryston so we are going to go ahead.

Yippee!!!

What shall we try and bully James into agreeing to next?

Yeah, the Benchmark DAC1 should definitely be your... well... benchmark for this. You need to be aiming either for that level of performance at a lower price, or better performance for not too much more.

I have to say, though, I think you really cannot get away from the need for a proper dejitter circuit in a device like this: not just asynchronous upsampling, as seen in the BP26DA, but the sort of full buffer-and-reclock that you see in the Benchmark DAC1 and the (comparatively overpriced) Chord DAC64.

I think it'd be neat to have a headphone stage as an optional extra, but I suspect that would require a design comparable in complexity to a BP26DA to achieve, which rather defeats the purpose of the device....

Jon L · « **Reply #4 on:** 1 Dec 2005, 06:09 pm »

Please consider the new AKM AK4396 DAC chip over the usual suspects.

If you are going to use upsampling, please make it defeatable. I would personally prefer that Bryston save the cost spent on upsampling circuit and use higher quality parts elsewhere.

perose · « **Reply #5 on:** 1 Dec 2005, 11:34 pm »

Balanced and single-ended outputs plus USB input woud be great! Also, don't forget the standard AES/EBU, coax and optical inputs.

I agree with nicolasb's comments but, if you intend this for the pro market you might need the headphone output. As a home user I wouldn't be broken hearted if it wasn't included.

Interesting times!!

thomaspf · « **Reply #6 on:** 2 Dec 2005, 06:17 am »

I thought the Benchmark DAC1 just uses an AD1896 asynchronous sample rate converter with an AD1853 and has no other reclocking circuit. That should not be much different than the current Bryston design. The area that the Bryston DACs seems to be behind the DAC1, Univeral Audio 2192, Lavry Blue, or Apogee Mini-DAC is the analog stage. Frequency response, DNR, and distortion are nothing special which is sort of a surprise since you would expect Bryston to be strong in the analog domain.

For a more sophisticated synchronous dejittering circuit you might have to look at Weiss or Lavry. The new Lavry DA10 is probably the DAC to beat in this class. The DA10 also has a pretty advanced digitally controlled analog volume control althoug they are missing a remote for it.

Another option is of course of course to put a high quality clock into the DAC and provide a master clock output like the UA2192. That will probably provide the best jitter performance of all these solutions. I would love to see a USB port as well but this again brings up the clock question. A simplistic USB implementation with an isochronous USB audio mode will probably not be a good idea. However, if you put a master clock in the DAC and go for asynchronous USB audio you can get the best jitter performance with a single cable connection. This would be a really cool!

This is going to be good

. I was waiting for the DA10 but maybe if James tells us a bit more about this new DAC I might spend my money differently.

Cheers

Thomas

nicolasb · « **Reply #7 on:** 2 Dec 2005, 02:27 pm »

Quote from: thomaspf

I thought the Benchmark DAC1 just uses an AD1896 asynchronous sample rate converter with an AD1853 and has no other reclocking circuit. That should not be much different than the current Bryston design.

Well, it does it by upsampling, but the timing of the upsampled signal is derived entirely from an internal clock rather than from the timing of the input signal.

Quote

The DAC-1, DAC-104 and the ADC-104 employ Benchmark’s new UltraLock™ technology to eliminate all jitter-induced performance problems. UltraLock™ technology totally isolates the conversion clock from the digital audio interface clocks. Jitter on a DAC digital audio input, or an ADC reference input can never have any effect on the conversion clock of an UltraLock™ converter. In an UltraLock™ converter, the conversion clock is never phase-locked to a reference clock. Instead the converter oversampling-ratio is varied with extremely high precision to achieve phase-lock to the reference clock. Jitter cannot effect the audio conversion, and test bench performance is repeatable in any installation!

So, yeah, I s'pose. But the device nonetheless seems to be regarded as sounding as good as devices that cost two or three times as much, so they're obviously doing something right.

Maybe the Chord DAC64 would be a better example, then.

That's definitely a "buffering" system: the output can be anything up to 4 seconds behind the input.

Watson · « **Reply #8 on:** 2 Dec 2005, 02:47 pm »

Quote from: nicolasb

But the device nonetheless seems to be regarded as sounding as good as devices that cost two or three times as much, so they're obviously doing something right.

They just have better marketing. Their UltraLock system uses the AD1896 with a local clock, just like everyone else who uses the AD1896 for jitter reduction.

thomaspf · « **Reply #9 on:** 2 Dec 2005, 05:12 pm »

Yes, there is absolutely nothing special about the circuit in the DAC1.

The big downside to these async sample rate converters is that you effectively listen to different samples being converted every time you play a track, so there is an element of randomness in the playback chain.

What the DAC1 has achieved is a pretty clean analog stage from standard components. Their headphone output is also very strong and can easily drive an HD600 without an amp.

Cheers

Thomas

kfr01 · « **Reply #10 on:** 2 Dec 2005, 05:47 pm »

Quote from: Watson

They just have better marketing.

And a better price. The DAC-1 is popular, yes, because of marketing, but also because it provides a very high percentage of the performance of devices costing 2-3x as much.

The DAC-1 is a capable device. It makes little sense to all but the most well-off consumers to spend two or three thousand dollars to obtain an arguably inaudible increase in performance over the $950 Dac-1.

Levi · « **Reply #11 on:** 2 Dec 2005, 09:43 pm »

I am assuming upsampling will not be supported.

Did anyone mentioned tubes. A DAC w/6922 or 6H30 output tubes should sound pretty good.

James Tanner · « **Reply #12 on:** 2 Dec 2005, 10:20 pm »

Have a look at these two Bryston NewsLetters. They pretty well lay out the thoughts and designs in our current DAC's.

http://www.bryston.ca/newsletters/83_files/vol8is3.html

http://www.bryston.ca/newsletters/81_files/vol8is1.html

james

Watson · « **Reply #13 on:** 2 Dec 2005, 10:36 pm »

Thanks for that. (It's pretty amazing that the first article is written by one of your customers -- it reads like advertising copy written in consultation with the engineers!) The CS8620 + CS43122 is a solid combo.

James Tanner · « **Reply #14 on:** 2 Dec 2005, 10:40 pm »

Well it was totally unsolicited but the customer obviously has expert knowlege in the workings of all things digital.
I know I learned a few things from it.

james

Watson · « **Reply #15 on:** 2 Dec 2005, 10:46 pm »

Does the BP16 use the same DAC architecture as the BP25/26?

number6.mi · « **Reply #16 on:** 2 Dec 2005, 10:55 pm »

I don't totally see the point of developing this before having equal quality digital handling features built into the SP2. It was a major dissapointment to see the SP2/SP1.7 upgrade won't have any re-clocking functionality. If I have an SP2, I don't want to have another box, much less pay for it, to get better sound from digital signals. I am an SP1.7 owners and think it is a wonderful product in all respects but this. I was eagerly waiting for the SP2/SP1.7 upgrade thinking it would address this (auto room correction was a second want). Once I learned it wouldn't, instead of waiting to upgrade my SP1.7, I went out and bought a Meridian G68 (which also has auto room correction). Maybe these features won't matter as much to other potential Bryston customers...

James Tanner · « **Reply #17 on:** 2 Dec 2005, 10:58 pm »

Quote from: Watson

Does the BP16 use the same DAC architecture as the BP25/26?

Yes the DAC in the B60, B100, BP6, BP16 and BP26 are the same.

james

James Tanner · « **Reply #18 on:** 2 Dec 2005, 11:16 pm »

Quote from: number6.mi

I don't totally see the point of developing this before having equal quality digital handling features built into the SP2. It was a major dissapointment to see the SP2/SP1.7 upgrade won't have any re-clocking functionality. If I have an SP2, I don't want to have another box, much less pay for it, to get better sound from digital signals. I am an SP1.7 owners and think it is a wonderful product in all respects but this. I was eagerly waiting for the SP2/SP1.7 upgrade thinking it would address this (auto ...

The thing is that the SP2 uses a dedicated surround processing MDS digital board and the priorities are different than straight 2 channel PCM digital processing.

Also the thing that differentiates the SP2 from the competition is the Class A Discrete output stages - the separate ground planes for Digital and Analog signals, the separate power supplies for Digital and Analog signals, Linear power supplies instead of Switching supplies etc. The difference in sound quality is not sutle.

I know I sound like a broken record but the SP2 represents an " surround audio" first approach and I realize that we may have issues competing on a 'feature first' bases.

james

thomaspf · « **Reply #19 on:** 3 Dec 2005, 02:34 am »

I sympathize with number6.mi on this issue. My hope when I bought the SP1 a long time ago was, that I could use one system for both surround and stereo sound.

While I have been patiently upgrading and waiting for a few fundamental fixes I am now resigned that I will relegate the SP1.7 and soon I hope SP2 to movies only and that I need a higher quality stereo DAC.

That is a bit unfortunate and I frankly don't quite understand why Bryston could not add a dejitter stage to the SP2.

Cheers

Thomas