What are the DX outputs for, and what happened to the pulse array power amplifiers?

What the DX outputs are

In November 2015 he said Dave has 705/768 kHz DX digital outputs "to go with" his digital power amps, which are non-switching, and that he had just finished testing the DX outputs Head-Fi, Nov 2015. In May 2016 he described the DX output as transmitting 705 or 768 kHz, 24 bit, at full level, with the volume control setting sent in the user data so that the DX power amp does the volume control "at the optimum point" Head-Fi, May 2016. In February 2017 he wrote that DX outputs are always dual BNC 705/768 kHz SPDIF "with one important difference - the volume control data is in the user data of the SPDIF, together with a preamble in the user data identifying the data as DX" Head-Fi, Feb 2017. He added that DX is a regular SPDIF in AES dual data format with special user data carrying the volume data, so the audio itself is not attenuated Head-Fi, Feb 2017. Each channel needs one cable; he put two identical outputs on Dave "in case one wanted to bi-amp" Head-Fi, Feb 2017.

In January 2018 he wrote that the only difference between Dave's DX output and a regular 705/768 kHz signal is the embedded volume information, so "when Dave is used with a DX power amp then Dave becomes a DX pre-amp, as it controls the volume"; Qutest accepts the DX data but ignores the volume part, as it has no volume control Head-Fi, Jan 2018; in August 2018 he added that Qutest "does not have a DX input" but will accept dual data 768 kHz and ignore the volume information Head-Fi, Aug 2018. The same day, on Qutest's thread, he wrote "Only Dave has DX outputs" Head-Fi, Jan 2018. In July 2019 he wrote that both TT2 and Dave have DX outputs from their volume controls, and that switching the DX option on in the M scaler lets it control the volume for future DX-compatible products Head-Fi, Jul 2019. At CanJam London in July 2018 he said the M scaler has "a DX mode for DX power amplifiers", devices then under development c w 24:49.

On the M scaler thread in February 2019 he wrote "DX OP is for future products only" Head-Fi, Feb 2019. In May 2022 he wrote that the DX protocol with the volume data "is intended for future DAC power amps that won't have a volume control built in" Head-Fi, May 2022. In the same period he said the DX outputs are 705/768 kHz outputs from Dave's WTA1 filter, which he gave as 164,000 taps Head-Fi, May 2022.

In February 2023, answering an audience question in Bucharest, he said the DX outputs "are for future DACs that are powerful enough to drive big inefficient loudspeakers directly", that he wants DACs that can do 100, 200 or 400 watts, and that the outputs are "To go to other DACs that don't have volume controls" (the figures should be checked against the recording) Dusty Audio 1:01:31Dusty Audio 1:02:25. In the same session he said you can connect a second Dave to a Dave's DX outputs and it would sound better because the second Dave is not doing any WTA processing, and that this is not something Chord talked about because he did not want to encourage people to sell two Daves Dusty Audio 1:02:44.

He framed the roles in several ways. In May 2021: "DACs as they are with headphone outputs, and DX amps (without headphones) for high power loudspeaker use, being fed from M scalers or DX OP DACs. It's analogous to analogue pre-amps and power amps" Head-Fi, May 2021. In March 2022: "M scalers to feed DACs for headphones, and DX amps for loudspeakers" Head-Fi, Mar 2022. In October 2024: "I think of M scalers as digital pre-amps, feeding headphone DACs or loudspeaker only DX amps" Head-Fi, Oct 2024.

What the power amps were to be

2014 to 2015: Hugo-style sound with far more power. In August 2014 he said the next big thing for him was DACs that can drive loudspeakers directly, and that he had been working on digital power amps for ten years article, Aug 2014. In October 2014 he replied "its not Class D" and called it a Hugo with much more power Head-Fi, Oct 2014. Also in October 2014 he said he had a more powerful version of Hugo under development, "(20W into 8)", with the prototype being built Head-Fi, Oct 2014. He added that it was risky because it contained untested innovations, that he would not release it until it sounded as good as Hugo driving efficient horn loudspeakers, and that "we don't have target dates for it - it will be ready when its ready!" Head-Fi, Oct 2014. In November 2015 he said power would run from 20W to at least 200W, and that the first DX power amp would be available "next year" at 20W RMS with a 70W mono-block option Head-Fi, Nov 2015.

2016 to 2017: prototype on the desk, the name, and the topology. In February 2016: "The power amp prototype has been stuck on my desk waiting for me to test it for a year now", with the amp and Davina as his priorities that year Head-Fi, Feb 2016. He said the topology is new and scalable, so he can create any power, and that 20W (70W mono-block) was just the first model, repeated in October 2016 as "merely the first amp in a series of differing power" Head-Fi, Feb 2016Head-Fi, Oct 2016. In April 2016 the amp was his second priority behind Davina, with the code word Hugo amp, "unlikely to be actually called that" Head-Fi, Apr 2016. In June 2016 he said Dave's output stage and second-order noise shaper came from the digital power amp, "You can think of it as a more powerful version of Dave", and that the amp is the same except for the power supply Head-Fi, Jun 2016. In April and June 2016 he said he wanted a better name than "digital power amp" because it sounds like Class D: "I want to get away from digital power amp word" Head-Fi, Apr 2016Head-Fi, Jun 2016. In January 2017 he said the DX outputs go to "my DX input power amps", digital-input power amps that decode Dave's DX outputs, with four passive components in the direct signal path and one global feedback loop Head-Fi, Jan 2017. On Class D, in February 2017: "No way is it Class D!", giving his reason as the timing errors of any switching output, and "no it's linear Class A, with no output reconstruction filters" Head-Fi, Feb 2017. In May 2017 he named the technology Power Pulse Array and said it was mostly intended for speaker drive from Dave's DX outputs, with one or two headphones that would appreciate the extra power Head-Fi, May 2017Head-Fi, May 2017. In July 2017 prototypes were being built, and: "It won't get released I am 100% happy with it" Head-Fi, Jul 2017.

2018: a series of DX amps, and the output stage goes into TT2. In January 2018 he wrote that power pulse array would give hundreds of watts "with no loss at all in transparency and refinement - at least that is the intent" Head-Fi, Jan 2018. The next day he answered confusion about the DX amps: for a conventional linear power amp use Qutest; for 1W into 8 use Hugo 2; for 2W use Dave; for greater powers where transparency should be preserved, a DX amp, which "can be a dedicated power amp using a DX signal from Dave" or an integrated-style unit with on-board volume Head-Fi, Jan 2018. In February 2018 he said all DX amps would have mono-block mode as a capability, and in July 2018 that "The intent is if the DX amp sees only one DX data, then it will auto configure to monoblock mode" Head-Fi, Feb 2018Head-Fi, Jul 2018. In February 2018 he wrote that power pulse array "shares the efficiency of Class D - but the output stage can be full Class A, and the signal path is completely linear" Head-Fi, Feb 2018. In May 2018 he said "the TT2 OP stage comes from the power pulse array design" Head-Fi, May 2018, and in an interview that month described power pulse array as a DAC capable of "200 watts capable of a kilowatt or whatever power you require" Darko Audio 2:10; he also said that the DX amp "had a much more complex OP stage - but this turned out to be not suitable", so it was redesigned and the new stage went into TT2 Head-Fi, May 2018. In July 2018 he said the DX amps are intended for power amps "although of course HP drive will be available" Head-Fi, Jul 2018, and that "the DX amps will offer a big step forward in transparency, judging from my listening tests so far" Head-Fi, Jul 2018. At CanJam London he described them as DACs with the same pulse array topology but capable of "four or 500 watts" (check the recording) c w 25:32. On headphones he said in August 2018 that the first DX amp will have headphone outputs and that Dave would be more transparent than it through headphones Head-Fi, Aug 2018Head-Fi, Aug 2018; in November 2018: "Never say never but no, headphones won't need a power pulse array DX amp - but some models will happen to have HP sockets" Head-Fi, Nov 2018. His later wording differs: in May 2021 he described DX amps "(without headphones)" and said the DX amp "will have DX inputs only" Head-Fi, May 2021Head-Fi, May 2021.

2020 to 2021: delays and prototype problems. In January 2020: "No news on the amps, been hyper busy on other projects" Head-Fi, Jan 2020. In a podcast published June 2020 (recording date not stated), he described a long-running project with the efficiency benefits of Class D but a linear output stage, said he was testing it, and "we hope to have products available in the next year or so" Part-Time Audiophile 26:10Part-Time Audiophile 26:53. In July 2020 he gave two problems with the prototypes, one practical and one concerning signal-dependent switching activity coupled through the ground plane, said he had solved the second by eliminating it, and said that a Dave DX amp "will be greater than 1 year, and for sure less than 5" away Head-Fi, Jul 2020. In March 2021: "There were some problems with the prototypes", naming getting the second-order analog noise shaper to work at high output power and worry about signal-correlated ground-plane noise from the power rails, and "I am not interested in doing it until I achieve parity with the performance of the DACs" Head-Fi, Mar 2021. In May 2021 he said he set the goal that DX amps must have innately similar performance to Dave, was not happy with the first attempt, and "I guess it will take a few years for production" Head-Fi, May 2021. He described the DX amp as "a DAC - with the same basic architecture as all the other DACs but with lots of power" with DX inputs only Head-Fi, May 2021. In July 2021: "No timeline for DX amps; but I guess after the ADC" Head-Fi, Jul 2021. In September 2021 he reported finishing a prototype PCB testing a new concept, "progress is being made, if somewhat glacial" Head-Fi, Sep 2021.

2022 to 2025: stalled, with the idea kept alive. In March 2022 he wrote "The pressure to do DX amps is off me somewhat as the TT2 is adequate for driving my loudspeakers" Head-Fi, Mar 2022. The next day: "the DX amp project did get stalled", with two technical issues from the first prototype, one now solved (stability of the second-order noise shaper at high voltages); if the next prototype works he can go back to Chord with a development proposal, and "my personal view would be to do something at the Dave level" Head-Fi, Mar 2022. Later that month, on replacing passive crossovers, he wrote that it would be done with a loudspeaker company: "Once the ADC and DX amps are done..." Head-Fi, Mar 2022. In February 2023 he said the DX program is "a future program" and "It's kind of been stuck" (the sentence is cut off) Dusty Audio 1:01:31. In October 2024: "The large power DX DAC concept is certainly not dead in my mind", and in the same post said he had been thinking about a design detail, how to prevent turn-on thumps without using an output relay Head-Fi, Oct 2024.

What he has not said

This archive has no statement from him giving a product name, price, final power rating or release date for a DX amp. It has no statement that a DX amp was released, canceled or handed to Chord for production. His power figures differ by date and context (a 20W first model with a 70W mono-block option and a range of "20W to at least 200W" in 2015, "many hundreds of Watts" in January 2018, "a kilowatt" in a May 2018 interview, 100 to 400 watts in a 2023 talk) and are stated aims, not specifications of a finished product Head-Fi, Nov 2015Head-Fi, Jan 2018Darko Audio 2:10Dusty Audio 1:01:31. The last dated statement found that names the DX amp concept is from October 2024 Head-Fi, Oct 2024. In a podcast recorded in October 2025 and published in March 2026 he described a project starting "next year", that is 2026, with loudspeaker companies on active drivers, "one DAC, one amplifier feeding each individual drive unit", but in that passage he did not call it DX or connect it to the DX amps ecoustics 1:28:33ecoustics 1:29:27ecoustics 1:29:55.

Quotes

“So the DX amps are just DAC's, exactly the same as existing pulse array DAC's - but with very high power OP stages”2018-01-11 Head-Fi, Jan 2018

“DACs as they are with headphone outputs, and DX amps (without headphones) for high power loudspeaker use”2021-05-12 Head-Fi, May 2021

“So the DX amp project did get stalled.”2022-03-04 Head-Fi, Mar 2022

“The large power DX DAC concept is certainly not dead in my mind”2024-10-06 Head-Fi, Oct 2024

Dated positions

Caveats and open points

• Almost all evidence is Head-Fi forum posts; many replies answer questions that are not shown, so the exact product or context is inferred from the thread.
• This archive has no statement giving a product name, price, final power rating, release date, cancellation or handover to Chord for a DX amp; the latest dated statement found is October 2024.
• Head-Fi, Jul 2022 (2022-07-19) says 'The amp is still in the background' without naming the amp, and Head-Fi, Mar 2025 (2025-03-13) does not say what 'something' refers to; neither is used. The active-loudspeaker project from the ecoustics podcast (recorded 2025-10-28 per the publisher's page, published March 2026) is reported but not treated as the DX amp, since he did not link the two.
• Head-Fi, May 2018 and Head-Fi, May 2018 (May 2018) carry dates estimated from their post IDs.

Topics: Dual BNC, dual data and DX links, Power amplifier and speaker-drive design, Future products and project status