Production Expert

View Original

Analogue To Digital Converter Test - We Compare Four; UA, Korg, RADAR & Metric Halo

Two of our professional contributors, Steve DeMott and William Wittman, decided to put analogue to digital converters, in fact A/D and D/A convertors, to the test. They wanted to know if they could hear the difference and which one they preferred of the four units. This article is the follow on from a podcast where they discuss the test and the results in detail.

Steve said, “What happened is there was a discussion elsewhere about converters. It actually was because of a discussion about DSD versus PCM and there was an ongoing discussion. I chimed in saying that I tried DSD a long time ago, late nineties and was completely unimpressed. So several people said the Korg MR-2000S DSD unit that will change your mind about DSD. So I got to try one in the studio for a month.”

Cambridge Audio has an excellent explanation about DSD Audio, here is an extract;

“DSD stands for Direct Stream Digital. It’s a high-resolution format that produces a high-resolution signal in a different way to that employed by the PCM system that can be transmitted as WAV, FLAC, ALAC or AIFF. If we were to look at a ’24 bit 96kHz’ file- a commonly used high resolution sample rate, this contains a stream of information 24 bits in size. This stream of information is then sampled 96,000 times a second- giving us the 96kHz part- to produce a signal that is then converted to an analogue signal by a DAC. The system has been used since the dawn of CD (which uses a 16/44.1kHz signal) and it also represents the basis of how our ATF upsampling process works.

DSD takes a different approach to the creation of a high resolution audio signal. Instead of using many bits of information in the signal, DSD uses a single bit. However, instead of sampling the information several thousand times a second, this single bit samples 2.8 million times a second to generate the audio signal. The result is still a high-res signal it’s just generated in a completely different way.“ You can read more here

William Wittman picks up the reason for the test. “Because there were a few choices of converters on site, we thought while we're at it, we might as well try a few and not just one converter versus DSD.”

Some of the units available are no longer made, but they are good benchmarks in the test. The units chosen were.

The Methodology

The original recordings were analogue, so no prior conversion took place to create the recording. The recordings consisted of guitar, bass, drums and vocal all straight to tape. The multitrack recorder is the Otari MX5050 MKIII 8-Track 1/2” and the two track master recorder/playback is the Otari MTR-12 1/2”. The mixer used is the APB-DynaSonics' ProDesk-8. Tape was ATR Magnetics Master Tape.

The stereo playback for the session was then sent out to all four convertors at the same time using matrix outputs. All audio to the converters was also lined up using a test tone to match input gain. Recordings were coming in at around -18db to give the convertors plenty of headroom.

Conversion for the UA, Radar and Metric Halo was 24/96. The DSD uses 1 bit conversion.

The entire test was a double blind test with the listener unable to know which unit was being played back at any time. Unlike some convertor comparisons this was a comparison of both A/D and D/A conversion, an important point.

Otari MX5050 MKIII 8-Track 1/2” and the two track master recorder/playback is the Otari MTR-12 1/2”

The Results

William Wittman: The clear loser was the Universal Audio. It just sounded flat, but not in a positive way. Other adjectives for it would be; flat, flattened, lifeless, smaller, narrower, and image was shorter front to back. Just in every way was it was plain and in uncomfortable kind of way. Of course you could go back and listen to just the tape and it was clear that it was losing something relative to the tape.

Steve DeMott: I also felt the low mids were a little crowded, or narrowed. Everything seemed to collapse in, on itself in those low mids and it made it really unpleasant to listen to.

Steve: Next for me. And I think also for William was the Radar. The stereo field collapsed. It really wasn't as wide. It wasn't as deep, it also felt very flat.

William: Yes. I would not have shot myself if I had to use the Radar and way back, I did lots of things with Radar and it sounds pretty good. But in comparison, it was when we had the opportunity to compare it to some other things. In this instance, it was noticeably smaller.

William: Number two, I think for both of us, is going to be the Korg DSD. It was much closer. This is why they don't usually announce number two first because then you know who number one is. So yes, the Metric Halo was the clear winner, the Metric Halo was by far the closest to listening just off tape versus converted.

It had both the size and an immediate feeling of wow, the stereo was what it's supposed to be just completely full range. The Korg DSD was by far the closest, but somehow lacked a little feeling of excitement. I want to say this is where we're into that very subjective territory, but there was something just more alive feeling about the Metric Halo versus the DSD.

And again, not that I would have killed myself with either of those two, but I did have still have a clear preference for the Metric Halo, which I will admit surprised me. Steve was probably less surprised as he has invested in them.

APB-DynaSonics' ProDesk-8

Summing Up

This test was in no way scientific, in terms of using measuring equipment. However, in true Experts style, the approach was far more practical and real world, taking the approach of if it sounds good then it is good, it’s hard to argue with that, even if the data were to say otherwise.

What the test does demonstrate is that there is a clear audible difference between A/D-D/A converters and a significant enough difference for the listener to notice. This varied between tone, stereo image and clarity.

Often discussions about the quality of converters, especially in audio interfaces, is dismissed by some when raised. If this experiment shows anything of value, it is that it’s worth auditioning converters before investing.

See this content in the original post

The Listeners

See this content in the original post