3 inputs: Toslink, RCA Coax, Balanced AES/EBU
2 outputs: RCA Coax, Balanced AES/EBU
Signal boosting x10
Sampling frequency conversion: Input from 32K to 96K and user selectable output 44.1K or 192K
Bit extension: from 16 bits to 24 bits
Dual choked and filtered power supplies
Dim: 6.625 W x 7.5 D x 2.25 H
Weight: 3 lb. Net, 4 lb. Shipping
Power consumption: 6w
100-120VAC (500mA FB fuse) / 200-240VAC (250mA FB fuse)
50-60Hz voltage set by factory
GW Labs, San Francisco
While AUDIOPHILE AUDITION has been one of the first audio publications to begin extensively reviewing software in the two new high-res. formats, it must be admitted that not everyone is going to be an early adaptor, as happened rather sooner with DVD video. There are many reasons - the battle of two conflicting formats probably being utmost. Then there is the higher cost of both new formats, and until recently the lack of moderately-priced players. No really good universal player is out yet. Lastly, many of us have huge collections of CDs by now, numbering in the thousands. Are we going to ignore all of those - as some of us did LPs in the early days of the 44.1 CD? Or are we going to want to get the best possible resolution out of our libraries without investing in brand new software - often of the same albums we already had bought in several earlier versions - each said to be an enhancement over the previous one?
One possible solution for some audio buffs has been to invest mucho $$ in the dCs Purcell/Elgar upsampling DAC combo. There is no denying that what it does in upsampling 44.1 digital data to 96K or 192K can bring many standard CDs close to the high resolution levels of many DVD-A or SACDs. An often amazing enhancement of depth, soundstaging and ambience retrieval can be heard - depending on the 44.1 source material. There is a bit more air in the high end and generally a sonic veil lifted. I hope to have a review of these products for you sometime soon. The next step down to a fraction of that cost (but still $3500) is the Camelot upsampling DVD player, The Round Table.
But most of us have to make do with something requiring a bit smaller investment. The Bel Canto DAC2 has been highly recommended to me and I hope to also cover it in a future review. But at hand I have the brand new DSP upsampler-processor from San Franciscos GW Labs. It is a small box that is not a complete DAC but a digital-to-digital upsampler that converts PCM signals with sampling frequencies anywhere from 32K to 96K into either 44.1K or 192K (the early units only went to 96K). The GW Labs must be used between the digital out of a transport (using either coaxial or Toslink cable) and the digital input of a separate DAC. It will work with any DAC but for the best results it should be a DAC capable of 96K playback - such as originally designed for the 96K audio DVDs issued by Chesky, Classic and a couple other labels (now a dead issue since the two new high-res formats).
What It Does
The DSP enhances the digital signal from the transport and then outputs this cleaner signal to the DAC. Multiple filters suppress noise generated within the unit and prevent mains-borne AC noise from entering the circuitry. The DSP boasts over-engineered power supplies. Separate transformer windings feed two low-noise rectifiers, followed by choke filtered capacitor banks - each with 20,000 microfarad filtering capacitance. The unit is designed to be powered on continuously and only unplugged when away for prolonged periods.
Three are three LEDs and two switches on the front of the DSP. First from left to right is the Lock light, which comes on when a valid digital input is received. It is next to the Input selector, which switches between optical and line inputs. Either the coax or balanced line input can be used, but not both at once. On the right side of the DSP is the switch to select either 44.1K output or 192K. An LED to the left lights at 44.1 and another to the right at 192. When used with DACs that lack 96K processing, this switch should be set to 44.1 for best results.
The two players I had on hand at the moment with digital outputs were the Pioneer DVD-AX10 so-called universal player and the Sony 9000ES DVD and stereo SACD/CD player just modified by Dan Wright Sound with a complete set of Bybee filters and many other upgrades. I started with the Pioneers digital coax out feeding my MSB LinkDAC III with their P100 outboard power supply, the Nelson Upgrade and 96K upsampling. After running in the DSP for about a week I did some comparisons. My first source was the Philip Glass soundtrack for Powaqqatsi (Nonesuch) I felt this would be great material for A/B testing due to the minimal nature of the music which wouldnt require the hassle of setting up programming for an endlessly repeating section of the CD - the constant repeating cut from the end of the section to starting over being psychoacoustically distracting as well. Other CD sources included the gold CD from Opus 3 Records titled Testrecords 1, 2 & 3 - my favorite test tracks are nos. 2 & 3 - the first the Stockholm Guitar Quartet in Telemann and the second the Tomas Ornberg Blue Five playing Buddy Bolden Blues. I also used The Antiphonal Music of Gabrieli, a Masterworks Heritage Series reissue (Sony Classical).
I should preface this by saying I have never been able to hear any improvement or difference between the 44.1 processing and the 96K upsampling of the MSB DAC. I used it with various CD and DVD players and with a variety of CD sources. I even inserted two Monarchy SuperDrives (jitter filters) between the transports and the DAC to no avail, though they did greatly enhance the sound at both sampling settings. And at the start I was faced with the same situation using the GW Labs DSP. Then I inserted just one of the SuperDrives in the digital circuit. I began to hear an extremely subtle enhancement at the 192K setting. Next I added the second SuperDrive and now the enhancement was more pronounced. Lastly I fed the analog output from the MSB DAC to my Taddeo Digital Antidote (see my past review). This combination of six different components finally produced some hearable improvement in the 192K mode vs. The 44.1K. On the Glass soundtrack there was more depth and air, the vocal chorus seemed more like genuine voices, and the soundstage widened slightly. On my favorite guitar quartet track the initial transients of all four guitars had more snap and the almost holographic impression of the four instruments seated in front of you was enhanced over the 44.1K setting. The brass instruments on the Gabrieli collection had a richer tone and seemed somewhat more separated from one another in space.
I tried a number of different CDs old and new. I had thought that perhaps the upsampling would be more successful on much older CDs that suffered from problems of digititus that have been cleared up in most releases of the last few years. I chose one early one with especially annoying string tone. The result was very little if any noticeable improvement at the 192K setting. In other words, if it was bad before, it was still bad after upsampling.
When I got my Sony 9000ES back from Dan Wright I was anxious to try it as the transport with the GW Labs DSP. Here the results took me back to the start again. I heard almost no improvement, although the sonics seemed even better at the 44.1K setting than I recalled from using the Pioneer as the transport. Considering the 52 lbs. weight of the Pioneer, the tight quarters of my rack setup and my recent carpal tunnel surgery, I couldnt easily switch back and forth to do more comparisons. I did hear that at least one other reviewer ran into problems using Sony transports with any upsamplers.
I didnt feel that with the Pioneer deck the DSP enhanced 44.1 CDs to anywhere near the level of a good SACD or DVD-A, though with close listening there were some enhancements on many CDs. I found that in my system adding the Taddeo Digital Antidote or even just one of the SuperDrives made a greater improvement than switching the DSP to 192K. Also spinning a CD on the Bedini Ultra Clarifier. Removing the DSP from the chain completely didnt appreciably worsen the sonics either. However, I left the two Superdrives and the Taddeo in the chain when I removed the DSP. Now that the DSP is on its way back to the manufacturer I realize I failed to try it alone with the DAC and without the other three black boxes. Its enhancement features aside from the upsampling might be similar to what the Superdrives achieve - although the Taddeo works on phase in the analog domain - quite a different approach.
Now dont laugh, but I was thinking that the only other player I had around that comes close to high end quality and is also just a plain CD player instead of DVD was my nine-year-old Radio Shack Optimus CD portable. One of its features is a digital out jack and some audio buffs are still using it as their main CD transport with a suitable DAC. (As I recall someone even made a special base and power supply upgrade for it that cost much more than the player.) I found the proper mini adaptor, hooked up a digital cable and auditioned the same three above CDs using it as the transport. The results were almost exactly identical to those with the $6000 Pioneer DVD player.
Others have had greater praise for the upsampling capabilities of this unit. While it appears that if the particular transport doesnt make that much difference, the DAC that it is mated with could have a major influence on the results of the upsampling. I would advise trying out the DSP within your own system before deciding on purchase. Of course thats always a good idea with most any component if it can be so arranged.
- John Sunier
To Next Equipment Review
To Monthly Index of Equipment Reviews
Return to April 2002 Home Page