PS Audio Obsidian "Ted Smith Signature" "TSS" DAC.

[CKKeung]

PS Audio announced a new flagship DAC at the Axpona Show 2019 :
The Obsidian "Ted Smith Signature" "TSS" DAC.

Some features are :

• Separate digital and analog sections.
• New FPGA has 250% greater computing capacity.
• High isolation power supplies.
• Galvanic isolation via fiber optic cable.

It seems that there is a new wave of using "optical isolation" on CAS/digital products.

Its tentative msrp is USD20-25k.

From Audio Bacon website.

 

[rockitman]

Interesting. That's a big price jump from the Direct Stream DAC....it will be interesting to see how they compare against each other at 5 times the cost for the new DAC.

 

[Macattack]

I find it interesting there hasn’t been a word about this ‘mega’ DAC from Paul. He has been talking about his new speakers for a year and hasn’t been shy about speaking other new products early.

 

[CKKeung]

A few photos from audiostream.com
https://www.audiostream.com/content/axpona-part-3-chicago-ps-audio

 

[Rob181]

PS Audio announced a new flagship DAC at the Axpona Show 2019 :
The Obsidian "Ted Smith Signature" "TSS" DAC.

Some features are :

• Separate digital and analog sections.
• New FPGA has 250% greater computing capacity.
• High isolation power supplies.
• Galvanic isolation via fiber optic cable.

It seems that there is a new wave of using "optical isolation" on CAS/digital products.

Its tentative msrp is USD20-25k.

From Audio Bacon website.

That was a great example of a genius speaking with the average person - Ted Smith is one VERY smart guy & the interviewer really had NFI what Ted was talking about. Digital & analogue in separate boxes - makes perfect sense to me - NOTE - I have as much DAC knowledge as the interviewer. Cook great bacon & eggs though

 

[Lee]

I spoke with Ted at the show (the prototype looked fantastic by the way). It's definitely an all out design. He also mentioned the updates for us DirectStream users will continue to come out.

I hope I can afford the TSS when it comes out!

 

[barrows]

While I agree with what Ted Smith is doing here, and I understand that some audiophiles would rather pay only for this than figure it out on their own, there is another way to achieve a similare approach as what the Obsidian does. The basic approach behind the Obsidian is that the high noise part of the component is in the first box (including digital receiver circuitry, and all DSP/oversampling/and modulator(s). And then the second box contains the actual D-A conversion (presumably via a discrete LPF circuit) and the analog output stage. A more powerful FPGA (than the DirectStream) in the first box, means the unit can use more sophisticated modulators, and the optical connections scheme (if well implemented) means the processing noise from he first box will not be a problem for the analog output.

Anyone who has played around with oversampling to DSD, 2x, 4x, or 8x, with HQPlayer can attest to how more processing power can equal better sound with the right DAC. Hell, Jussi's latest modulators are so sophisticated that they require an I9-9700K at a minimum to achieve DSD 256 output (and no one has been able to bull;d a machine capable of DSD 512!).

By using HQPlayer running on a powerful computer in another room doing all the oversampling, and connecting via optical Ethernet to an optically based renderer and a simple discrete DSD DAC (like Holo audio, for example), one could take advantage of orders of magnitude higher processing power of the computer to do the heavy processing, while still having an isolated, low noise, simple conversion stage in the audio system. In either case, one has the flexibility of using new oversampling/modulators when they become available, but by using a real computer for the heavy processing the sophistication of the oversampling/modulators is much less limited in terms of processing power available.

Chord's upscales use a similar approach as well, but with a PCM emphasis: and the eschew optical isolation, although they are designed to provide electrical isolation from the powerful upscaling box to the less powerful (processor) in the DAC box.

 

[CKKeung]

Yes, my diy CAS friends found the HQPlayer's new dsd-converaion algorithm excellent sonically but they could build desktops capable of dsd256 conversion only.
DSD512 may need water-cooling monster desktops but then there will be a hell of other problems to solve in the sonic aspect.

 

[Carlos269]

With a powerful enough PC you can run almost all the filters while transcoding to DSD512. It is only when using the new EC modulators that you are limited to DSD256 because of the PC processing-power/cooling limitations.

 

[barrows]

Yes indeed Carlos, I was referring to the EC modulators in my post, with them only DSD 256 is currently possible: which is fine with me, as most report that DSD 256 via the EC modulators sounds better han anything else. This will be DAC dependent of course. I prefer to build a DAC optimized for DSD 256 myself, as this rate is now relatively easy to achieve with many approaches.

 

[Carlos269]

My experience concurs with what others have reported, that the new EC modulators (EC7 in particular) at DSD256 sounds better than my previous reference DSD512 settings. The new EC modulators render additional focus and detail while remaining organic and analogue sounding as opposed to being hyper detailed and analytical.

 

[barrows]

Yeah, I was only able to play with the EC modulators for a short period (insufficient computer here) but what I did hear was extremely natural sounding, while still being very highly detailed. Seemed to me the best of all worlds, natural sounding, organic, and highly detailed. Like the best of digital and analog combined.

 

[CKKeung]

Yes, the diy audiophile computers of my audiophile friends can play up to EC7 DSD256 only.
The sound is excellent.