Search results

The latest version supports mono downmixing. I think it is useful for areas with a single speaker, and I wish more receiver devices (Bluetooth etc.) would support it.

The speaker's self-noise is more than that.

Genelecs at their highest (default) sensitivity produce 100 dB SPL at -6 dBu (0.388 V). This is 20 dB attenuation from 4 V. So if you want something like 50 dB SPL, it makes 70 dB attenuation.

Have you checked Topping's published specs?

Archimago measured DSD implementations: https://archimago.blogspot.com/2021/10/measurements-review-topping-d90se-dac.html (Conclusions, paragraph 5) It seems ESS chips produce an elevated noise floor with DSD256 compared to AKM.

IEM output Noise level (A-weighted, 16/44.1) <8.4uVrms So in this case the noise level is limited by the 16-bit signal. Does the device's own volume knob still attenuate at higher resolution even if the input has 16 bits?

Supporting balanced and unbalanced connections seamlessly from the same output requires a cross-coupled circuit. Many balanced audio devices use it, in particular professional devices, but many don't. It requires an additional op-amp, which is more expensive and increases distortion.

Roon Ready not ready.

Oratory1990 publishes a list on Reddit. It has EQs for various headphones tuned for ADI-2 DAC.

I suppose True Peak Level is the Peak Level (as opposed to RMS) and takes into account inter-sample overs, so the True Peak Level of a digital sound wave can be more than 0 dbFS.

S/PDIF supports only DoP64, and Wiim's internal DAC doesn't probably support DSD.

E1DA's Cosmos series (currently ADC and pre-amp with a notch filter) is intended for affordable measurements.

I wonder if AKM's traditionally reliable S/PDIF receiver is also integrated.

Quality is a work in progress.

No, delays need more memory than included.

DAC is self-powered, so it has its own power supply and doesn't consume bus power.

If the voltmeter doesn't detect the pulses, it means there is a frequency of pulses that is too large. That is the bandwidth limit of the meter. DC switches cannot switch current immediately, either. Inductance and capacitance mean that the current and the voltage take some time to settle, so...

The FFT looks like the effect of some lossy encoding.

E30 II tries to maximize compatibility. It has proprietary signal processing for S/PDIF. Jitter performance is respectively worse. I don't know why they didn't add the same to other products.

I suspect 24/192 is Wiim's unofficial extension to Chromecast. Applications following Google's documentation would not try to stream more than 24/96, but Wiim doesn't refuse to play 24/192 if it is sent.