Shelford Interviews: The Development of Early Faders

Rupert Neve discusses some of the early trials and tribulations of incorporating faders into the early Neve consoles. For more information on the Shelford Series please visit: rupertneve.com/shelford-series

[@iamthat787]

This man changed my music and audio world. Thank you Mr. Neve, you are a legend. See you in the next life!

[@joepalooka2145]

I hope Mr. Neve and his wife are doing well during this current COVID-19 crisis. This man is a true legend who should have been knighted by the Queen a long time ago.

[@waterwoodguitars6871]

These videos with Mr. Neve are absolutely amazing and an important piece of engineering history. Glad these interviews will always exist for others to enjoy and learn from. Absolutely beautiful

[@benperry490]

so cool and at his age it's amazing how well he recalls all these stories in detail.I am glad you are talking with Mr. Neve and documenting these stories. I wish i had done the same with my grandparents before they passed on Rupert won;t be with us forever and getting these stories now is fantastic. If I could afford his products i'd definitely purchase a few of his classic mic pre amps and his great new mixing console

[@tovsteh]

This was very relaxing and interesting to listen to. Thank you.

[@Microtonal_Cats]

"Who do you think you are, Mr. Neve..." Yup. Exactly. This cat is such a legend. And it's a blast to listen to him talk.

[@miguelk7891]

Un genio que amo el buen audio , su serie 8000 de sus consolas 8016 8032 , 8078 , sus EQ 1066,1073,1084 y 1081 fueron y son sin duda lo mejor que se hizo en transistores Podría decir que Saúl Walker diseñador de API hizo un gran trabajo en sus primeros Pre y EQ 550 A Felicitaciones a estos 2 próceres del Diseño del Audio Miguel Krochik

[@mixedbymikael]

RIP to a real one!

[@JaminatoRRR]

Love you Mr Neve !!! Someday will own a neve console !!! :) greetings from India !

[@solootto]

Same here bro 🙏🏾

[@prodjincmusicproductions6587]

Beautiful!

[@chrisroberts2266]

In fact it's still an issue with the fader wiper.solder breaking and them falling off! These interviews are amazing capturing snippets from Rupert's mind. I just wish a TV sound engineer had recorded the voice and put the microphone in the right place.

[@thaexception3406]

Insightful

[@Jamesrevelations]

2nd question is about harmonics, if a fundamental frequency of a signal is captured and filtered out its harmonics that might go beyond 20k, will recorded material without the harmonic content in it, generate those harmonics again when played back..

[@OttoOreh]

No. (But the listening room will produce some new ones)

[@Frank55]

Did Mr. Neve maybe mean EAB faders? I used these in the past and they indeed had ball bearings.

[@Jamesrevelations]

I have a genuine question for,Mr. Neve or any one who can answer. When a device converts analog audio signal to digital, does it ignore the data below and above the 20 to 20khz range ? Second if no then what component or spec on a audio converter should we look for the details of the conversion...

[@OttoOreh]

Well an audio CD can represent frequencies up to 22.05 kHz. With all those diffrent types of mp3 and likewise there are now the same or more. But the biggest lose (in terms of audio quality) is the compression of the file in size. If you looking for a natural sound or more of a old sound you shouldn't be to worried because the frequency response back then wasn't to good any way. Try fix the acoustics of the studio-room instead, rather than go for expensive tape machines (those are great but will not give you that much in return)

[@axelv21]

+James Manasseh You should look for the resolution and the SNR. Make sure the sample frequency is above 44.1Khz so you can record all audible frequencies.

[@mattsyson3980]

The whole topic of analogue to digital conversion is very complicated. It is always a compromise between cost and performance of which some 'standards' were established in the early 1970's. Audio has almost always been hanging on the coat tails of other industry and going it alone is exceedingly expensive.Some companies (notably AMS) were early in the use of 8 then 12 bit a/d conversion where 8/12 bits could be converted in an (expensive) IC (chip) but as the quality was not too good they used a large quantity of logic chips to 'add' an extra 4/8 bits, where 16 bit depth was at least reasonable. I think it was Sony (among others) who manufactured convertors with more bits and could operate at higher frequencies to bring 'quality recording to a state where 'worldwide' standards became important. A the conventional wisdom at the time was that 20Hz to 20KHz would be 'adequate' for 'high quality audio' and with the birth of compact disk the bit 'depth' was specified to 24 bits (I think) even though then, as now you struggle to achieve full 24 bit resolution. there are other constraints (physics) that actually prevent full use of 24 bits and I think a little over 22 bits are realised when you consider a bandwidth of 20 - 20khz. The analogue stages surrounding the A/D and D/A place a restriction that cannot be overcome unless you limit the bandwidth or use 'fancy processes' to counteract physical conditions. A CD with it's 44.1KHz sample rate can capture up to 20KHz audio but there must be a very 'sharp' filter to actively remove audio between 20 and 22KHz to avoid anti aliasing 'distortion'. This sharp filtering itself created it's own problems for which Mr Rupert Neve pointed out and he collaborated with some Japanese researchers in Tokyo to demonstrate the 'advantages' of having music that could be recorded and reproduced possibly up to 50kHz.. In the early days computers would struggle to store the huge amount of data that 44.1kHz, 16 or 24 bit data produce as computer hardware was only beginning to move to 16 bit operation. A 16 bit A/D or d/A 'chip could easily cost a hundred Dollars in the early days whereas now you can buy combined A/D d/A chips that can handle 24 bit stereo at 192KHz sample rate that are about 5 dollars apiece that have much of the logic circuitry to interface to something useful, built in. The Dynamic range (difference between 'impossible for a human ear to hear, up to a sound level that would cause major bodily damage) can be captured by a 24 bit convertor. Unless your listening room is a totally silent anechoic chamber where all yopu can hear is the blood flowing in your head, and a speaker system capable of bursting your eardrums and eyeballs the theoretical capability of the A/D/A can match it. Bit depths greater than 24 are useful and necessary for mixing systems where multiple signal sources will be added together and manipulated to do things like equalisation (tone controls) but ultimately the 'mixed' digital audio has to be scaled back down to fit real world amplifiers and speakers.

[@silverXnoise]

@@mattsyson3980 That’s a pretty comprehensive (and accurate) overview. Mix down CD standard is 16-bit/44.1KHz, and DVD video is 16-bit/48KHz. Digital recording standards have fairly well settled into 24-bit/192KHz, which in my opinion is the reasonable thinking person’s (and engineer’s) ideal. 24-bit depth provides 16,777,216 discreet levels of sampling with which to represent the signal’s analog voltage, generally a range of about 3V peak-to-peak, so imperceptibly smooth sine waves and most any complex arbitrary waveform can be accurately modeled up to around 96KHz. There are some audiophile products that boast 384KHz sampling with the ostensible claim that the higher slew rates it provides (ie-the sharper edges for sudden amplitude changes, like the crack of a snare drum) make it superior even within normal hearing ranges, but I don’t think there’s any speaker in the real world that can respond fast enough to make good use of such high frequencies and slew rates. The physical motion of a speaker’s driver and its impedance begin to limit frequency response well before hitting the 96KHz mark in my experience. But we’re also still living with the limits of chip designs and costs at the higher frequencies, so it’s not like we have the luxury (yet) of just over-speccing our equipment just because we can. For one thing, while 24-bit AD/DA has become inconsequential for audio frequencies, if you look at chipsets for things like RF or test equipment, getting anything above 12-bits (4096 levels) at frequencies beyond 100s of KHz is still quite rare and expensive, not the least of which because of the additional memory and support components necessary to handle the volume of data such sampling requires. But also because of some physical limitations within the circuit designs that grow increasingly problematic to deal with the higher one goes with sample rates. It’s a scenario where the diminishing returns begin to grow exponentially much past 192KHz, so most products that advertise such high specifications are largely doing so for the express purpose of marketing them to people without the technical knowledge to understand what it all means, and as we all know: when in doubt, bigger numbers equals better equipment to lord over your friends with. Some purists may claim that no amount of bit depth or sampling rates can ever really reproduce an analog signal, and although they’re mathematically incorrect, I will admit that the qualitative features that are found at the edges-things like headroom and distortion-are indeed a good enough reason to continue to design, build, service, maintain, and restore things like Neve or API consoles and signal processors. I believe both analog and digital signal workflows have their respective places within a professional audio production process, but it’s critical that today’s engineers understand their features and limits, and know how and when to leverage them appropriately. To any aspiring audio professionals still reading this, get an associate’s degree or better in electronics engineering, it will make everything else about the job easier, cheaper, and better.

[@mattsyson3980]

@@silverXnoise Indeed there are many constraints, down to basic physics that make recording/playback of material beyond 24 bit 192k rather moot because even that range (24bit ) represents a range of voltage /air pressure levels beyond the capabilities of humans, so if the lower end is surrounded by random movement of air the upper limit is so great that hearing damage and even bodily organ failure would result. If course creating a microphone and speaker system that can actually transfer the levels is a different problem altogether. Of course the requirements of 'recording' a single microphone/source and being able to reproduce it are diferent to the situation where the signal is 'processed' be it some form of Equalisation, compression, expansion etc where the TIME taken to carry out these activities represents a different problem. Indeed mixing (summing) adds yet another layer of complexity.

[@mattsyson3980]

Minor correction Quadrant (or any stud faders) do not 'create' clicks UNLESS there is a leakage of DC from the preceding or following amplifiers. If taken to extremes you might perceive a 'disturbance' if the amplification following the fader has sufficient gain that you can hear the DIFFERENCE in level of thermal noise over the 1 or 2 dB step between studs. Depending on the frequency and audio content being manipulated by the fader a 'step change' can be interpreted by the ear as a 'click' however this is not a technical fault but a function of human hearing. If you listen to a sinewave your ear/brain expects it to continue. If you abruptly alter the level (intensity) of that sinewave your brain interprets it as a 'click' or a sinewave of a different frequency. the BBC (among others) evaluated these effects and proposed that a step change of 1dB or less is inaudible in most circumstances. The use of multiple studs and wiper arrangement allows a 'silent' fade to be achieved with the majority of programme material. This is all frequency and level dependent.