View RSS Feed

Miska

90's DAC shootout

Rate this Entry
I took a pair of my old early 90's DACs from the storage and fed bunch of measurements through both. Source for both was MuFi V-Link192 at 44.1k output. As result of doing this at midnight I forgot to change the output to 16-bit, so you may see some truncation quantization effects too... Both of the DACs claim 18-bit resolution.

Participants:

1) Crystal Semiconductor CS4328 64x oversampling 1-bit DAC chip with 8x interpolation and fifth-order modulator. IOW, so it's pretty much a "DSD DAC". S/PDIF receiver is Yamaha YM3623B.

As numbers:
THD: 0.003%
IMD: 0.004%

1 kHz tone:
blogs/miska/attachments/6232-90s-digital-analogue-converter-shootout-cs4328-1k.png

Digital silence:
blogs/miska/attachments/6233-90s-digital-analogue-converter-shootout-cs4328-silence.png

19 + 20 kHz IMD test tone:
blogs/miska/attachments/6234-90s-digital-analogue-converter-shootout-cs4328-imd.png

Frequency response:
blogs/miska/attachments/6235-90s-digital-analogue-converter-shootout-cs4328-freqresp.png

0 - 20 kHz sweep, wide band:
blogs/miska/attachments/6236-90s-digital-analogue-converter-shootout-cs4328-sweep-wide.png

16-bit J-test:
blogs/miska/attachments/6237-90s-digital-analogue-converter-shootout-cs4328-jtest16.png

Impulse response (note vertical scale):
blogs/miska/attachments/6238-90s-digital-analogue-converter-shootout-cs4328-dirac.png




2) Burr-Brown DF1700P 8x oversampling digital filter configured for 18-bit output and PCM1700P 18-bit R2R ladder DAC. S/PDIF receiver is Crystal Semiconductor CS8412.

As numbers:
THD: 0.007%
IMD: 0.024%

1 kHz tone:
blogs/miska/attachments/6239-90s-digital-analogue-converter-shootout-pcm1700-1k.png

Digital silence:
blogs/miska/attachments/6240-90s-digital-analogue-converter-shootout-pcm1700-silence.png

19 + 20 kHz IMD test tone:
blogs/miska/attachments/6241-90s-digital-analogue-converter-shootout-pcm1700-imd.png

Frequency response:
blogs/miska/attachments/6242-90s-digital-analogue-converter-shootout-pcm1700-freqresp.png

0 - 20 kHz sweep, wide band:
blogs/miska/attachments/6243-90s-digital-analogue-converter-shootout-pcm1700-sweep-wide.png

16-bit J-test:
blogs/miska/attachments/6244-90s-digital-analogue-converter-shootout-pcm1700-jtest16.png

Impulse response (note vertical scale):
blogs/miska/attachments/6245-90s-digital-analogue-converter-shootout-pcm1700-dirac.png




Conclusions:
- 2.8 MHz sampling rate of CS4328 is quite apparent from the wideband sweep spectrum.
- DF1700 + PCM1700 has higher distortion.
- Jitter figures are similar and anyway depend on the S/PDIF receiver chips.
- Impulse responses are very similar, by eye almost the same.
- DF1700 + PCM1700 is more silent when playing silence.
- Amount of noise-shaping noise from the 1-bit 2.8 MHz CS4328 is minimal.
Categories
Uncategorized

Comments

  1. Miska's Avatar
    I'll redo some of the relevant measurements with 16-bit output dithering at some point to clean up the spectrums. (doesn't apply to impulse response or J-test measurements, those are from 16-bit source, bit-perfect)