John Siau

There are a couple of important reasons why the bass response of the Benchmark AHB2 is different than that of the other amplifiers. First, the bass response of the AHB2 extends down to 0.1 Hz. The reason for designing in this extended response is that it keeps the phase response flat to below 20 Hz. This flat phase response means that the bass will be delivered in the proper time relationship to the rest of the music. In contrast, most audio devices deliver the bass late. This is especially true of power amplifiers. Late bass will often sound louder because the bass decays too late and this obscures reverberant details in the music. Detail is lost when bass is late. The bass will sound louder, but the kick drum will have less impact. In contrast, the AHB2 accurately reproduces the bass with the proper timing and this tends to reveal details in the music that are often obscured when the bass is delivered several milliseconds too late. Second, the AHB2 does not produce any audible distortion when reproducing low-frequency tones. Most amplifiers add audible harmonic distortion, especially when encountering difficult phase angles in the speaker load. In contrast, the AHB2 is not adversely affected by the load angle and it stays clean under severe loads. If an amplifier produces audible 2nd harmonic distortion this will give the impression that the bass is deeper and louder. The low-power 1-watt THD performance can have an impact on these effects. The feed-forward error correction in the AHB2 keeps it exceptionally clean at all power levels, including the first few watts. Whenever bass is emphasized through one of these two mechanisms, details will be obscured. The AHB2 is designed to reveal details captured in the recording. It is not designed to modify the sound in any way. The other amplifiers in this group are also designed to be transparent, but there seem to be some audible differences which may be due to the effects I have described above. These effects do not show up in a simple full-power THD+N measurement, or a simple frequency response measurement. They do show up when you dig deeper into the measurements.

We would love to sell you a second AHB2, but the honest answer is that you do not need it if you are never lighting up the clip lights. Unlike virtually all other power amplifiers, the THD produced by the AHB2 does not increase under load. The 8-ohm and 4-ohm THD vs. Power curves are identical. Even when loaded with 2 ohms, the THD is virtually unchanged. The red clip lights are driven by a circuit that measures the THD. If it exceeds 1% the lights will turn on. The AHB2 delivers an astonishing 0.00011% THD, 1 kHz at full rated power into any rated load. The transition from 0.00011% to 1% is abrupt, but it occurs above the rated output power. This means that the AHB2 stays clean, even when you drive it almost to the rails. Ordinary amplifiers cannot do this. With most amplifiers, THD gradually increases with output level and with increasing current load. The reason the AHB2 stays clean is the patented feed-forward error correction system. The AHB2 is a very unique design and many of the usual rules of power amplifiers do not apply. I would love to sell you a second AHB2, but you do not need it!

This is not an insignificant difference. Every added bit reduces the wide-band noise by 6.02 dB. The quantization noise produced by a 5-bit modulator is 24 dB lower than that produced by a 1-bit modulator. This is why virtually all PCM A/D and D/A converters have abandoned 1-bit modulators.

This is correct. The SNR can improve by up to 3 dB when the number of converters is doubled. In the DAC2 we use four ESS Sabre channels running in parallel to achieve a 6 dB improvement over the performance of a single Sabre channel. Internally each Sabre channel drives a parallel array of 1-bit D/A converters.

This is a common misunderstanding about PCM. All of the band-limited information between samples is preserved and can be reconstructed. This technique is used in upsampling systems. Any content that falls below the Nyquist frequency (1/2 of the sample rate) can be recovered at any point in time between samples. I discuss this in detail in one of my application notes: High-Resolution Audio - Sample Rate - Benchmark Media Systems, Inc. This application note presents some non-audio sampling systems as examples that show how information is accurately captured between samples. The application note should help dispel this common myth about sampling.

Multi-generation loopback tests are actually very useful. They can highlight defects that may not be as easy to detect after a single generation. Once the listener's ears are trained to the defects, it is often easier to detect them after fewer generations. The loopback tests can also make it easier to determine what objective measurements are most useful for detecting the defects that have been identified. We have created 16th generation loopback tests using our Benchmark A/D and D/A converters. These tests help us determine the transparency of our PCM A/D and D/A loop. We are way overdue for some DSD loopback tests. Based on the mathematics, I don't expect that DSD loopback tests will match the transparency of PCM. The problem with multi-generation DSD is the cascading of 1-bit noise shaping processes (an undesireable necessity in a 1-bit system).

I agree that these labels are producing some wonderful recordings but I would contend that it is due to the care and skill that they put into the production and not due to the fact that the delivery method is DSD.

Don't fall for this! The square waves show the time-domain response before the ultrasonic noise is removed. An FFT of these waveforms would show high levels of ultrasonic noise which must be removed to prevent damage to tweeters. After filtering out the ultrasonic noise the square waves are no longer square. The DSD marketing people love to show these kinds of misleading time-domain responses that show the waveforms before the mandatory low-pass filters at the end of the DSD chain. Notice how rare it is to see frequency-domain plots in the DSD marketing literature.

Modern PCM sigma-delta converters produce much lower error signals than1-bit sigma-delta DSD converters. The errors in the DSD system are due to the1-bit quantization that occurs in 1-bit sigma-delta DSD converters. Multi-bit PCMsigma-delta converters can be fully dithered and do not suffer from this un-dithered truncation. Plus, every added bit reduces the noise signal by 6 dB. A 4-bit sigma-delta converter is 24 dB quieter than a 1-bit sigma-delta DSD converter. Right from the start, 1-bit DSD signals have much higher losses than multi-bit PCM signals. Conversion from 1-bit DSD to multi-bit PCM is a lossless process inside the audio band. The only thing that is removed is the out-of-band noise above the Nyquist limit of the PCM system. Nothing else is lost. Don't believe the DSD marketing hype. In contrast, conversion from multi-bit PCM to 1-bit DSD is always a lossy process.The loss is due to the 1-bit truncation. This truncation introduces a very large ultrasonic error signal that makes the ultrasonic region unusable for audio. But remember the ultrasonic region of DSD is always unusable for audio because of the high noise levels. This ultrasonic noise produced by 1-bit DSD systems must always be removed before reaching power amplifiers and tweeters. When the noise is removed, the ultrasonic audio content is also removed. Processing a 1-bit signal to create a 1-bit signal is also always alossy process. A volume control is one of the simplest processes in a multi-bit PCM system, but it creates a large error signal when applied in a 1-bit DSD system.The same is true for any other 1-bit to 1-bit DSP process. The lossy part of these DSP processes is the quantization back to 1-bit. Cascaded 1-bit truncation processes can rapidly degrade the audio quality. For this reason, DSD is almost always processed as multi-bit PCM. Any DSP process applied to a 1-bit DSD signal produces a multi-bit PCM signal. No loss of information occurs until this multi-bit signal is quantized back to a 1-bit signal. Why incur the loss by going back to a 1-bit signal after the processing inherently produces a multi-bit PCM signal? All practical DSD systems require some sort of DSP processing (gain control, mixing, filtering, etc.) and all of these processes produce multi-bit PCM results. Taking these lossless multi-bit results and adding loss by truncating them back to a 1-bit DSD signal makes absolutely no sense. DSD complicates the signal processing and adds unnecessary losses in several places along the signal path. DSD does not simplify the signal path. There is absolutely no truth to the marketing hype that claims that 1-bit DSD is a simpler system than multi-bit PCM. The exact opposite is true. 1-bit DSD is a lossy system. John Siau, VP Benchmark Media Systems, Inc.

Modern PCM sigma-delta converters produce much lower error signals than1-bit sigma-delta DSD converters. The errors in the DSD system are due to the1-bit quantization that occurs in 1-bit sigma-delta DSD converters. Multi-bit PCMsigma-delta converters can be fully dithered and do not suffer from this un-dithered truncation. Plus, every added bit reduces the noise signal by 6 dB. A 4-bit sigma-delta converter is 24 dB quieter than a 1-bit sigma-delta DSD converter. Right from the start, 1-bit DSD signals have much higher losses than multi-bit PCM signals. Conversion from 1-bit DSD to multi-bit PCM is a lossless process inside the audio band. The only thing that is removed is the out-of-band noise above the Nyquist limit of the PCM system. Nothing else is lost. Don't believe the DSD marketing hype. In contrast, conversion from multi-bit PCM to 1-bit DSD is always a lossy process.The loss is due to the 1-bit truncation. This truncation introduces a very large ultrasonic error signal that makes the ultrasonic region unusable for audio. But remember the ultrasonic region of DSD is always unusable for audio because of the high noise levels. This ultrasonic noise produced by 1-bit DSD systems must always be removed before reaching power amplifiers and tweeters. When the noise is removed, the ultrasonic audio content is also removed. Processing a 1-bit signal to create a 1-bit signal is also always alossy process. A volume control is one of the simplest processes in a multi-bit PCM system, but it creates a large error signal when applied in a 1-bit DSD system.The same is true for any other 1-bit to 1-bit DSP process. The lossy part of these DSP processes is the quantization back to 1-bit. Cascaded 1-bit truncation processes can rapidly degrade the audio quality. For this reason, DSD is almost always processed as multi-bit PCM. Any DSP process applied to a 1-bit DSD signal produces a multi-bit PCM signal. No loss of information occurs until this multi-bit signal is quantized back to a 1-bit signal. Why incur the loss by going back to a 1-bit signal after the processing inherently produces a multi-bit PCM signal? All practical DSD systems require some sort of DSP processing (gain control, mixing, filtering, etc.) and all of these processes produce multi-bit PCM results. Taking these lossless multi-bit results and adding loss by truncating them back to a 1-bit DSD signal makes absolutely no sense. DSD complicates the signal processing and adds unnecessary losses in several places along the signal path. DSD does not simplify the signal path. There is absolutely no truth to the marketing hype that claims that 1-bit DSD is a simpler system than multi-bit PCM. The exact opposite is true. 1-bit DSD is a lossy system. John Siau, VP Benchmark Media Systems, Inc.

THX has been demonstrating the AHB2 with ATC SCM-40 speakers.

I cannot answer this. I would contact Wireworld. They are very good about standing behind their products. I believe they have the ability to rework the existing connectors.

Wireworld was very responsive to the compatibility problem with their cables. They quickly reworked all cables in stock and changed their tooling. Some of the old-style cables may be in circulation. The problem was caused by the two bumps on the side of the B-Type USB plug. Normally these bumps have a 45 degree bevel to aid in the insertion and removal. On the original Wireworld plugs these bumps were formed at a 90 degree angle. B-type USB receptacles have fingers that engage with the bumps on the side of the plug. These fingers can jamb if the bumps do not have the proper bevel. High-retention USB connectors (identified by orange inserts) are most prone to damage from the old-style Wireworld cables. Old-style cables will damage orange-colored high-retention USB connectors. Wireworld assures us that very few of the old-style cables were sold.

You are correct, the scale factors are the same. I thought I was looking at a comparison between 24-bit PCM and DSD64. This is DSD256 which has very good noise performance in-band. The difference is the out-of-band noise which is not shown. DSD 256 can achieve very low noise in-band (as shown). However, the out-of-band noise must be removed by down-stream processing.