In a way, this is not a review of one Weiss device, but two. You see, the Weiss Audio DAC50x platform is made up of two almost identical products; this one, the half-sized DAC501 streaming DAC, and the full sized, but half the height DAC502 streaming DAC. The difference between them is the latter adds a rear-mounted four-pin headphone XLR socket. Everything else, from the interface to the connections right down to the colour of the front panel, is functionally identical. We went for the more conventional audio-sized DAC501 box, complete with a standard 6.3mm stereo headphone jack on the front (which turns out to be a bit of a star).
Daniel Weiss’ concept behind both is to create a new category in audio. “With the DAC50x we are creating a new paradigm for what used to be a black box device”, he says. “A typical D/A Converter is a “set and forget” device. Not so with the DAC50x. It adds a number of interesting signal processing features and sports a variety of digital inputs.” To this end, the DAC501 and 502 both include an extensive range of novel DSP settings.
On the hardware side of both DACs, there are a total of five inputs: AES/EBU or S/PDIF via XLR, Toslink, and RCA sockets, UPnP/DLNA via Ethernet, and USB. There is also a USB A connector used ‘for various applications’ (although these are not described anywhere, I’m suspecting international espionage or juggling are probably not on the list of potential applications). Additionally, both DACs have single-ended line out on a pair of RCA connectors, balanced line-out on a pair of XLR connectors, and a headphone output on a 1/4” stereo headphone jack.
Where most DAC-first audio products consider the headphone amplifier as an afterthought, the Weiss uses discrete output stages for both line and headphone outputs. The output levels of both can be set to adapt for the headphones or amplifier in the chain, using a coarse four-step adjustment. These levels can be set independently for line- and headphone outputs, and there are no sound-degrading servo mechanisms used anywhere in the Weiss DAC50x models.
Common to Weiss Engineering models of past and present, Daniel Weiss is reluctant to discuss the make and model of the DAC chips used inside his devices, and the DAC501 is no exception, preferring instead to describe the digital conversion as using two 32 bit D/A Converter chips, with two D/A conversion channels used for each audio channel. This chipset is governed by an internal high precision, low jitter clock generator. Uniquely, the sampling frequency of that generator is fixed at 195kHz, not 192kHz or 176kHz. The input signals are converted to the 195kHz sampling frequency because Weiss claims this gives optimal signal quality, reducing any jitter related effects in the process. All standard PCM sampling frequencies from 44.1kHz up to 384kHz plus DSD x64 and x128 are supported.
Weiss opted for a linear power supply in the DAC501, with separate regulators for left and right channels and two separate toroidal mains transformers and automatic selection of mains voltage from country to country, measuring the voltage before power up, so your fine DAC is not blown apart by mistake. Even the power switch is different from most, as it activates a semiconductor relay which only switches on or off at the zero crossing of the mains voltage. This means glitch free power switching.
This on its own would make the Weiss DAC501 interesting in its own right, but where it really moves into a league of its own is when you look to the software and DSP engines. The DAC501 has a digital signal processing chip built in, which supports a range of clever algorithms, including the following:
Room Equaliser – to suppress room modes for a decent bass reproduction.
Creative Equaliser – a tone control with low boost/cut, high boost/cut and mid boost/cut.
De-Essing – the automatic removal of overly bright sibilance from human voices. The sibilance effect can be more or less pronounced depending on your speakers or room acoustics.
Constant Volume – adjusts the audio volume (loudness) to a constant value across all tracks played. Useful for ‘party mode’ when the volume control should stay untouched.
Vinyl Emulation – Designed to give a vinyl-like performance from digital. This is seemingly developed in tandem with the DMM-CD emulation procedure offered by the Stockfisch label. More on this later
Crosstalk Cancelling (XTC) – for the playback of dummy head recordings or live recordings via speakers for an incredible live sensation. Dummy head recordings usually are listened to via headphones because they only work properly if the left channel goes to the left ear only and the right channel to the right ear only. With speakers this is difficult to achieve as the left channel goes to the left and the right ear. But with some clever signal processing of the speaker channels it is possible to suppress the crosstalk: i.e., the audio going from the left speaker to the right ear and vice versa. If that works properly then the recording sounds as if one would be in the space where the recording has taken place. All the reverberation and 3D representation of the sound sources is there. Naturally this is for speaker based playback only.
Out Of Head Localisation algorithm – an attempt to get the music ‘out of your head’ when listening via headphones. The goal is to achieve a similar listening sensation as one gets when listening via speakers. Naturally, this only applies or headphone based playback.
In addition, Weiss is at the consideration phase for implementing Roon, MQA, and Apple’s Airplay. Future formats such as these can be accommodated for via software updates; in fact, midway through the review the vinyl emulation algorithm was effectively unlocked through software update.
Both DSP and updates are driven through the DAC501’s web interface, which is configured via a web browser. This controls volume and balance controls, input selection, output type, DSP configuration, and snapshots of your DSP configuration. There is also an infra-red remote, which controls the input selected for conversion, the output type, level, muting, absolute polarity, the choice of DSP preset, and power handling. Then there is also the rotary encoder knob and colour front panel LCD touchscreen, that can also navigate these functions, albeit in a slightly more modal way than the web interface.
Put simply – phew!
With so much on offer, you might be thinking the DAC501 is difficult to use, and you’d be part right. Like any advanced digital device, much of the complexity is front-loaded, meaning you spend a lot of time configuring the device at first, and then day-to-day navigation and driving is relatively straightforward. That is mostly the case here, but the sheer diversity of DSP options open to the listener does make the learning curve steep and long lasting. If you are the kind of person who gets lost trying to operate a toaster, this is not the product for you. Instead, if you are the kind of person who loves to tinker, to play with their music just for the hell of it, who wants to know how it can sound different, and is prepared to experiment without prejudice, the Weiss DAC50x models are a sustaining and continued interest.
Yes, you can used the Weiss DAC501 as a DAC on its own, and it’s a good one in its own right, but it’s using a tiny fraction of the DAC501’s performance potential. In fact, I’d go so far as to say if you were looking for just a DAC, there are ones that deliver the performance of the Weiss DAC501 for a fraction of its price and not find that an aggressively negative statement. The desktop version of the Chord Hugo, for example, has a musical performance that is on a par with the Weiss DAC501, and if all you want is a device that converts digital data into analogue music, go down that route.
Instead, the Weiss DAC501 takes that basic performance and runs with it. Not all of the DSP options are fully unlocked at this time, but those that are prove surprisingly effective, and that inspires confidence that the rest will work when they are finally released to the public. Take lateralisation, for example; some are perfectly comfortable with the sensation as if there is a tiny orchestra or band living in the space between your ears, but a system that offers a modicum of reduction of these lateralisation effects and makes it seem like you are listening in a room with some musicians is gratefully received. To date, this has been of mixed success, in that the best remove musicians from inside my head to an echoic, bathroom-like environment wrapped just around my head. Different… but nothing like reality.
Given Weiss actually makes those instruments seem to live outside the space between my ears even without DSP, iI am expecting very good things when this lateralisation system is unlocked. The Weiss presently shows this best with either really large-scale music (Mahler’s Eighth Symphony, Solti [Decca]) or paradoxically really simple music (‘Because He Was a Bonny Lad’ from Here’s the Tender Coming by The Unthanks [Rabble Rouser]), where the stability of the generated image is best portrayed. Medium-sized bands can still get slightly ‘in-head’ in their image solidity.
Vinyl emulation is fascinating, and I’m still in two minds as to what it does, and whether or not I like it. It seems to be generating a small amount of low-level crosstalk (mixing a small amount of left channel output in the right channel and vice versa) and possibly some harmonic enrichment, and even a slight high frequency roll-off, but in the process it seems to create that kind of natural-sounding deep soundstage and a sort of effortless ‘listenability’ that typifies good vinyl. One of my current regular tracks at the moment is ‘The Ghetto’ from Everything is Everything by Donny Hathaway [Atco]. This started life as an LP back in 1970, and while it’s a good transfer, you are always wishing it was the LP playing (my vinyl copy is so scratched, it appears to have been used to clean ice skates at one time). Weiss’ emulation is good here. Not perfect – there’s no noticeable processing effects, but it’s not as vinyl as I’d like, but it’s a good stab at making obviously digital systems less digital. These were my two go-to DSP settings.
There’s one other setting that I found myself using a lot; if I’m honest, more than the vinyl emulation described above. The Creative Equaliser, which sounds like a late 1980s TV show. In fact, used carefully, a judicious bit of tone shaping can help out a lot of recordings. Not all of us own a well-manicured, carefully managed record collection. Some of us have great but badly recorded music in our collections. The equaliser is what a good tone control always should have been: a first-do-no-harm modifier of the tonal balance of a signal, making those bright and compressed recordings of the last 20 or so years less bright. Which means I can listen to Oasis records at last. I’m not sure that’s a good thing, but the equaliser is a very good thing.
Not all the DSP systems are fully beneficial, but in many cases they are going to work better or worse in different systems. This is kind of giving them the benefit of the doubt, but often that’s the way it is with DSP systems. I feel this only scratches the surface of what this excellent device is capable of. The hardware platform is excellent, with extremely fine, very detailed, and extraordinarily even-handed sound quality (as befits a company with 30 years of pro-audio engineering under its belt), and the headphone amplifier in particular is extremely detailed and capable of driving even difficult headphones. But the performance of the Weiss DAC502 transcends mere sound quality. It’s truly a digital device for the future!
- Type: Streaming DAC/headphone amp
- Digital inputs: XLR connector, S/PDIF RCA connector, TOSLINK connector (optical), USB type B connector, RJ45 Ethernet connector
- Analogue outputs: 2x RCA single-ended, 2x XLR balanced, 1x 1/4” headphone jack socket
- D/A Converter chip: Over-sampling multi-bit sigma-delta converter. Two converters per audio channel
- Sampling frequencies supported: 44.1 kHz, 48 kHz, 88.2 kHz, 96 kHz, 176.4 kHz, 192 kHz, 352.8 kHz, 384 kHz, DSD64, DSD128
- Linearity: At 0 dBFS to -120 dBFS input level: less than ±0.4 dB deviation from ideal
- Crosstalk: Better than 110 dB, 20 Hz-20 kHz
- Total Harmonic Distortion plus Noise (THD+N):116 dBr (0.00016 %) at -3 dBFS input level
125 dBr (0.000056 %) at -40 dBFS input level
125 dBr (0.000056 %) at -70 dBFS input level
- Headphone Output options: 5.2 Vrms +16.53 dBu,
2.7 Vrms +10.84 dBu
0.9 Vrms +1.30 dBu
0.2 Vrms -11.77 dBu
- Headphone output (THD+N):-115 dBr (0.00016 %) at -3 dBFS input level
-122 dBr (0.0000795 %) at -40 dBFS input level
-122 dBr (0.0000795 %) at -70 dBFS input level
- Dimensions (W×H×D): 18.8 × 6.6 × 30cm
- Weight: 6kg
- Price: £7,200
Manufactured by: Weiss Engineering Ltd
Distributed by: Padood
Tel: +44(0)1223 653199
Read Next From ReviewSee all
Gold Note DS-10 EVO
Gold Note's small but perfectly formed small boxes of high-performance electronics have been impressing us since we tried its excellent PH-10 phono stage. Now, it seems the DS-1 EVO streamer/DAC/preamplifier is just as splendid, according to Chris Thomas.
- Chris Thomas
- Nov 2022