I vote for a hardware upgrade for the DirectStream in 2020

I vote for a hardware upgrade for max. USD 1500 for the DirectStream with:

  • the Spartan 7 FGGA (instead of the version 6)
  • A Jensen line output transformer
  • Maybe 2 clocks instead of one?
  • An upgrade of the existing Crystek clock with an even lower phase noise
  • More dedicated PSU’s
  • A better isolation of the Bridge II/III to other components

What do you guys think?


How about letting PS audio get the transport and server out and save this development for a future time. It would be nice to have a better functioning code for the DMP. Multiple clocks are not a good thing anyway.

I think it ain’t gonna happen. Never mind all the other things going on that sixpack mentions - the TSS is next, DAC-wise. This is sort of like asking for the “TSS Junior” before that even comes out.

1 Like

I can comment on a few of these in the abstract.

Probably the Artix 7, it has a better mix of resources for this application. The TSS is using an Artix 7 so a smaller one would be trivial.

There are few Jenson line output transformers. Any one of which would eat most of the budget for a $1,500 upgrade.

I don’t understand two clocks. Even having a way to select between them would add jitter. The purpose of upsampling to 20 × the DSD64 frequency is to make upsampling from 352.8k and 196k simple. Upsampling 196k to some other high rate wouldn’t be any better.

I’m always on the lookout for lower phase noise VCXO’s.

Definitely more local regulators and better separation of the global PSU’s.

Definitely better Bridge isolation…

Some other wishlist items:

Tho I don’t know how to pay for it, a headphone output might be nice (and expensive.)

Better isolation of inputs from each other (to lower ground loop opportunities) would be good.

More configurable output grounding should be doable.

An optional external PS might be interesting.

Support for Quad rate, perhaps for octal rate (that might be a little harder)

A lower analog noise floor (the digital noise floor is more of a software issue.)

Perhaps optical isolation between the digital and analog sections?


Hi Ted: many thanks for your reply.

I understand the Jensen trafo could be a $$$ problem…

But maybe a DS hardware upgrade would cannibilize the TSS DAC sales…

I sure hope PSA would consider a hardware upgrade in 2020/2021.

And I am willing to spend another $1000 for a Jensen transformer…

Me too. (Speaking as a consumer, not a producer.)


I (by the way) do not vote for a tube ouptut stage. I once had a Jadis integrated amp and thereafter I decided never to buy tube gear ever again…


Would also encourage an incremental ‘mid-cycle-refresh’ for the DS. This device is definitely on my list but I do feel some hesitation about the by now 5-6 year old hardware. A mild update with some trickle-down tech from the TSS would make this ‘new’ again in 2020 without requiring a huge new development project. Increased FPGA processing power provides more room for future software improvements as well. Will keep an eye out next year.

Is better bridge isolation a matter of bridge or DAC design? How much of it will come with a bridge III?

I don’t know a lot about the newer/upcoming streaming products/solutions, but insofar as they are external to the DAC that’s good. Being at the mercy of Converse Digital isn’t fun and I sure don’t want a SOC (system on a chip) inside the DAC. I’m pretty sure any new DAC will have better streaming processor isolation than the Bridge 1 or 2.

Do you say Bridge III will be an external solution?

No, as I said “I don’t know a lot about the newer/upcoming streaming products/solutions”, i.e. I don’t know if Bridge III would be the solution, one of multiple solutions or not relevant to any new DAC (unlikely, but as I said I don’t know.)

Back in the day, we used Vectron VCXO’s for our Application Evaluation boards to show off our MUX/DEMUX products in the 10Gb/s world in 2001 ! 40Gb/s Chips were already in prototype production.


I’m sure you’ve looked at these Ted.

What’s so special about it?

Indeed, my original prototype used a custom Vectron part built to my specs (darned expensive.) The list of standard products from Vectron don’t compete with the Crystek parts I’m using. When the DS project was starting I went to Crystek to see if they had anything like what I needed - their web site didn’t but their engineering team was working on one that was almost perfect for the DS. They grabbed a few from engineering for the DS prototypes. They are still (to the best of my knowledge) the best fit for my use.

1 Like

Networking, streaming solution shouldn’t be inside a high-end DAC, there are great solutions out there, and aparently PS Audio working on one… Not only you get less noise, interference, but you also gain better reliability, and much improved interface, software, apps, potential in the future.

Ah ok, I overlooked your „insofar“, sorry. But then what you say is that an internal solution is not preferable in terms of isolation, right? In previous Bridge III discussions I presumed the new lens and isolation design of it would isolate similar to an external solution…but it’s all speculation I understand.

Yes, for me it’s just interesting insofar, as I wouldn’t go for a proprietary solution as the Octave SW then if it’s external anyway.

1 Like

I don’t know if the Bridge III would be an internal or external device, I’m not paying attention to it since I just don’t care about streaming solutions: I want the best sounding DACs, and that implies than internal solutions for streaming are a problem.