Thanks for impressions and the nifty spreadsheet comparative review of several high-profile DACs.

What pointed me towards the Google machine (and this site) was John Darko giving the Mytek a 5+ rating, the highest he's given any DAC. Huh? I've listened to the Mytek at some length on my system. Yes, DSD128 sounds pretty impressive, but it sounds impressive on just about any DAC ... and DSD128 converts very well to 176.4/24 PCM if you want to stay in the R2R camp. The Mytek's way with PCM, which is 99.9% of my music library (thanks to SACD's being nearly impossible to rip), is not as impressive. The Resonessence Invicta does a better job with PCM sources, although still not quite as good as my old-school Monarchy PCM-1704 DAC.

I was under the cheerful impression that my Monarchy had been thoroughly overtaken by modern 192/24 & DSD DACs. It only has a S/PDIF input, only goes up to 96/24, and has late-Nineties tech for the input receiver, oversampler, PCM-1704 converter, passive I/V conversion, and a 6DJ8 SRPP analog stage. Nothing cutting-edge about it.

So when a good audio-friend of mine threatened/promised to bring his Berkeley BADA DAC after the RMAF show a couple of years ago, I was expecting the worst. The BADA is three times the price of the Monarchy, designed by the very talented HDCD team, and represents the latest thinking in delta-sigma converters and high-speed analog sections. I really didn't want to spend $5000 on a DAC just then ... and the Berkeley USB -> S/PDIF converter was another $2000. So, $7000 to get a ticket on the modern DAC train.

But ... I was very surprised when I heard the two in a direct comparison. On my system, which I admit has a rather idiosyncratic power amplifier that doesn't sound like most other amps, the Berkeley was dynamically compressed and quite flat tonally, with a smaller and more closed-in soundstage. The expensive USB -> S/PDIF converter, on every selection played, was not as good as S/PDIF content from my SuperClocked Denon DV2900 transport. The Monarchy had far better dynamics and more vivid tone color, and the ambient impression filled the room, instead of being restricted to a space a little wider than the speakers.

Vacuum-tube euphonics? Well, maybe, but euphonic coloration doesn't usually add musical information, it just smooths out the rough edges. I heard more music, not less ... much more. The difference was larger than I expected, comparable to good-quality 320k AAC versus uncompressed Red Book. With 88.2/24 and 96/24 content, the differences between the two DACs narrowed considerably, but retained the same character .... the tonal colors, dynamics, and sense of pacing was noticeably flattened on the BADA. A very surprising result, not what I expected at all.

On my system at least, I had results completely the opposite of what I'd read in Computer Audiophile. That reviewer's speakers and amplifiers are completely different than mine, so perhaps that was the reason. I'm coming to a rough hypothesis that delta-sigma sounds better on Class AB transistor amplifiers than it does on Class A DHT triodes, which bring tonal vividness and dynamic shadings to the listener's attention. If a DAC is deficient in those areas, it's less noticeable on Class AB transistor amplifiers, which tend to have a flatter presentation.

But plenty of audiophiles prefer 200-watt amplifiers ... it's the majority choice, after all, and 20-watt DHT's are very much a speciality item for the fanatics who like that sound. All-transformer-coupled PP DHT's are even more rare, and almost never seen at hifi shows.

Thanks again, purrin, for your impressions of the AGD Master 7 fed with I2S data from the OffRamp 5 converter. Although the OR5 is pretty expensive for a mere format converter, I've had enough startling experiences with different transports sounding very di fferent that the OR5 might just be worth the moola. I'm guessing you sprang for the clock and power supply upgrades as well?

I spoke to Steve at THE SHOW, and did mention a better PS would make the turboclocked units sound better. I really don't want to spend around $1k for a specialty power supply, so I'm exploring the use of an AMB Sigma11 power supply with +15V to replace the wall wart.

P.S.

I'm not sure what the Dynamo is going to cost, but I wouldn't be surprised if its that high, the old Monolith battery was something like $1200. Chris at BPT does linear supplies on custom order for closer to $400-500, it might be worth asking him to do one for the OR5. He also does 12V batteries as well, which is another route. Similar pricing IIRC.

Thanks for the tidbit on Ultramatch Pro. I've heard others say great things about it as a re-clocker. I may pick one up now. Thinking about modding it with custom clocks, better PSU, etc.

Tried AGD M7 tonight in NOS mode. Very interesting. I like it a lot - to the extent of me thinking about putting in a toggle switch tied to the jumper. NOS mode is a departure from the M7's standard presentation. Tonally dense with a jet black background, but at the expense of hyper resolution. NOS is basically BYPASS mode. Bypasses the Alterra DSP. Had to use coax with NOS/Bypass mode since it does not gel well with i2s. WIth i2s, the center image gets shifted to the left and is unstable. I'm assuming one the bit or master clocks if off by a tick, or the LRclock got mixed up.

Also tried with PLL off. The Altera chip evidently has a PLL on/off option. PLL on is better. PLL off = shallow depth, less focus, less separation and layering.