I have known Bob Stuart since 1972, when we met in London at an AES talk. I had a problem with him when he took my advice to use a complementary differential input stage and then tried to patent it, without telling me. Still, he started out as a very good analog designer, but I really don't think that his analog designs ever evolved outside of hi mid fi, from what I have been able to hear, including his demo at the Newport audio show this year. I had really hoped to be able to evaluate his and Dr Peter Craven's MQA project, but I found his electronics/speakers too opaque to determine anything. I think that Bob has gone on to other things, like advanced digital, and left analog audio behind. This is why, I suspect, that Bob might be more interested in recent years in digital problems, rather than analog problems.
This is very true (as in the abstraction I mention below, when writing code for a digital processor people don't often think of various transistors being in/near saturation or cutoff), and as Stephen Jay Gould might have said, these are non-overlapping magisteria.
On the other hand...
"Digital" is only an abstraction of discrete voltage levels, and "digital circuitry" is just analog circuitry with active devices operated in or near saturation, or in cutoff, or switching between these two states. In this way, many "subtle problems in the digital design" can be seen and approached as analog.
cmi.fairlightus.com
CMI Series I
RM4136
CMI Series II
LF347
CMI Series III
ADC module: 5534 (TL072 for DC servo)
Audio Module (DAC/LPF/VCF/VCA): LF347, 5532
Audio buffers, mixing: LF347, LF353
Headphone amp: LM380
🙂
George
>Edit
From another source
http://www.synfo.nl/servicemanuals/Fairlight/CMI-IIx_SERVICE_MANUAL.pdf
CMI Series IIx
ADC module: LM301 (MC3140 for DC servo)
Audio Module (DAC/LPF/VCF/VCA): MC34004
Audio buffers, mixing: LM301, TL084
Headphone-Speaker amp: discrete
You know what is wrong with opamps if you are wearing socks and step into one laying upside down on the floor. Like I did just the other day. They have 8 fangs that bite deep. No discrete component has ever done that to me.
I reckon, but have never experienced it, that an upturned TO3 could potentially inflict a couple of rather nasty puncture wounds...
Ha Jay I see our posts crossed in the ether.
Ha Jay I see our posts crossed in the ether.
Save
The problems you get in the design of digital systems is not usually with the hardware, but a failure to think through the number systems, and catch / deal with number overflow, saturation, sign errors etc. In that sense they are similar to analogue issues, like failure to consider a device in the middle of the design saturating - not just focusing on the input and output devices.
Alternatively, you can view these subtle problems at the border of analog as "BER" and treat them in the digital domain (CRC or some other error correction scheme).
Wait until you step on a tube.
I had two amazingly great analog designers at the dinner table last night, both of whom also have deep knowledge about digital. I wish I had seen John's post so I could have read it to them.
- Status
- Not open for further replies.