When do you think DACs became "transparent"?

[Chrispy]

With the reliability of whathifi in mind (or hifichoice I suppose, not familiar with that) what's the question?

 

[AnalogSteph]

I am curious how much hearable improvements have truly happened since Red Book.

That's kind of a bad question. Red Book is just a format, and one with enough bandwidth to cover the humanly audible frequency range and with arguably more dynamic range than any listener even wants (and if that's not enough anyway, some fancy shaped dither gets it to a perceived 18ish bits). A CD is only ever as good as the recording, processing and mastering behind it. It can sound as good as you want if the playback chain is up to it. Of course 16/44 is only the end format these days, virtually nobody is recording straight to 44.1k any more (with a lot of ADCs it is inadvisable to use any less than 48k, and with some you want to go even higher, although ones with "44.1-proof" filters do exist).

You can take a 24/96 master, downsample to 16/44 (I recommend using the Foobar2000 converter with the SoX resampler DSP, quality "Best", plus dither), and upsample back to 24/96 and get something that is 100% indistinguishable in an ABX comparison. Well, unless some bozo thoroughly brickwalled it, in which case a small level adjustment to eliminate the overs would be required first. (Brickwalled hi-res sounds stupid and it is, but has absolutely happened before.) With all that dynamic range below, some people still insist on bumping their head on the ceiling it seems...

 

[Waxx]

The digital part was clean from start, but it's in the analog output parts that there was a difference at least untill the late 90's in consumer electronics in my pricebracket (budget) my experience. Maybe there were more pricey dacs that were invisible, but not that i'm aware of. All early cd players were kind of coloured some better (Denon, Marantz, Philips) than others, but they were coloured.

Even the infamous Apogee AD8000 and DA8000 studio convertors from the late 90's were coloured, but sounded good coloured and therefor still is very popular in music studio's. Even if you had very clean convertors also already (Prism was already arround and famous). Apogee got famous in the late 80's with building anti-allias filters for older gear (improving sinad and linearity of the old Sony recorders and others) before they made their own dacs.

And even today there are still dac's that are coloured. But the days of heavy pricetags for clean dac's (like the Prism) are definitly over. This site proofs that you can even have them for 50€/$/... And that is the biggest revolution i think. And with the right dac (that does not have to cost a lot) ad/da conversions became a non-issue.

 

[Bach]

When a DAC was shown with an acrlyic case, during an audioshow.

 

[VientoB]

Asynchronous Sample Rate Converter. Typically used for upsampling with the 96 and 192 kHz capable DACs that became common around the time. Unfortunately the things are known for hard-clipping overs and generating some nasty distortion doing so.



You can probably see now why that's a problem...

Interesting that the TEAC with the ASRC was released in 1993.

 

[VientoB]

When a DAC was shown with an acrlyic case, during an audioshow.

Geshelli Labs?

 

[Old_School_Brad]

1982-1986? Philips developed the TDA1541, a highly popular chip with specifications that made it virtually transparent in performance.

 

[VientoB]

1982-1986? Philips developed the TDA1541, a highly popular chip with specifications that made it virtually transparent in performance.

Were the output stages up to the task back then?

 

[sergeauckland]

1

Were the output stages up to the task back then?

Yes, very much so. Just look at the performance of the original CD100/CD63 or the Sony 101. The overall performance as measured at the analogue output was so far below audibility as to be impossible to improve on audibly. That modern DACs do a lot better on their measurements is irrelevant. As I mentioned above, the Sony PCM F1 was shown to be fully transparent and that was even before CD was released.

S.

 

[Snarfie]

Skynet became self aware Saturday 5th October 2024.

So Keith what are you/we going to do about it

 

[DSJR]

I got into hifi in about 1996, reading UK magazines like What HiFi? and HiFi Choice. They would always assure us that there were clear differences in sound between CD players. What HiFi? magazine loved the Marantz CD63 KI Signature for two or three years. I think it was priced at £500 initially and they reckoned it beat everything up to £1000 Were they fooling themselves back then or were there audible differences between DACs and CD players back in the 90s?

I thought the KI CD63 was a 'HiFi-impressive' sounding scrapper of a thing, sounding more scrappy the better the system it was put with (Rotel did similar with one of their otherwise nice sounding machines - 965BX?). Interestingly, the original CD63 wasn't like that at all...

The better specced CD players from the mid 80s onwards were pretty good I think, but subjectivism for whatever reason (to sound impressive to amateur WHF reviewers, or to slug the tones for a more vintage 'analogue' tone) had taken over and we were all the worse for it. We'd look to 'enthusiast' brands (and the KI Signature models for sales which to me seemed worse than the vanilla issue models) for a nicer tone and ignore what Sony, Denon and many others were doing at the higher end of their ranges with what I've lately discovered were superbly truthful 'sounding' machines and dacs, with performance about as good as 'red book' would allow.

 

[sergeauckland]

I got into hifi in about 1996, reading UK magazines like What HiFi? and HiFi Choice. They would always assure us that there were clear differences in sound between CD players. What HiFi? magazine loved the Marantz CD63 KI Signature for two or three years. I think it was priced at £500 initially and they reckoned it beat everything up to £1000 Were they fooling themselves back then or were there audible differences between DACs and CD players back in the 90s?

What HiFi and HiFi Choice required there to be differences they could hang articles on. It required differences to maintain readership, as saying that the new CD XYZ sounded the same as its predecessor wouldn't sell advertising. Never mind that the measurements didn't back them up, their Golden Ears could hear the difference.

The professional journals such as Studio Sound were much more based on reality. If I remember correctly it was Studio Sound or possibly Broadcast Engineering that published the tests showing that the Sony PCM F1 was transparent in blind listening tests.

Incidentally, one separate test that was only very sparingly reported was one Ivor Tiefenbrun, who claimed that his Linn Sondek was hugely better than digital, that digital destroyed the sound, yet he failed to hear the difference when the PCM F1 was switched in and out of circuit when playing one of his LPs...

 

[AnalogSteph]

Even the infamous Apogee AD8000 and DA8000 studio convertors from the late 90's were coloured, but sounded good coloured and therefor still is very popular in music studio's.

From this review, I suspect the (optional) soft-limiter may have been to blame for this quirk:

This circuit on the AD-8000 is supposed to work like dbx's Type IV: to avoid overloading the AD converter, a limiter provides non-linear limiting in the area -4 to 0 dBFS; the nearer the level comes to zero, the stronger. However, either Apogee's circuitry is defective or the attack and release times are to[o] short, since as soon as Soft Limit comes into action, there is around 3 per cent distortion. Of course, a measurement of such a peak limiter always looks worse than it sounds. Whilst with ordinary pop material, the Soft Limit circuit can produce an intensification and volume increase without pumping effects, it leads to a raw sound with critical material.

Otherwise frequency response was found to be basically ruler-flat 10 Hz - 20.4 kHz (+ we know that the AK5391 ADC has a ±0.001 dB ripple spec and -110 dB ultimate), dynamic range to be between 108 and 113 dB(A) and THD+N to be as low as -105 dB. In other words, it ought to be capable of perfectly transparent performance.

(BTW, the article is wrong on one thing, the ADI-8 DS uses the AK5393 as you'd expect in a 24/96 capable unit, rather than the AK5392 as stated. The latter is single speed only.)