…the electronics tech in me says that after a certain point of component quality, there is no increase in sound quality.
I think this is generally a fair statement, but there are variables including what the device does and where it sits in the signal chain, as well as the complexity of the chain…certain negative impacts tied to component quality can be cumulative. But my opinion and experience is the circuit design has more bearing on the sound than the quality of the components…within certain boundaries. You can’t make a garden-variety signal path sound boutique by upgrading opamps and capacitors and resistors and transistors. You’ll improve the sound (the system slew, clarity, etc.), but the basic character will remain, and this is all dependent on the upgrades being prudent…tested…like making sure new opamps are happy in the circuit. And I agree it is a process of diminishing returns. And in many cases upgrading components can take away things that made a device desirable. A couple Tascam-related examples: when I was refurbishing my RS-20B spring reverb box, I recapped it, and used factory values because Teac did a good job with all that (which is rarely otherwise), analyzed the power supply filtering and determined it was fine (which is usually the case), and considered opamp changes. IIRC it’s all 4558. But I left it alone because I like how it sounds. It, by nature, is not a high-slew device…rather soft and washy…why try and make changes to change the clarity? I left it alone. In the case of my prototype Tascam mixing console, I’ve made a bunch of changes. But in some cases, like with opamps, I didn’t like how it sounded with more contemporary “faster” opamps. That sounded brittle to me. But in some stages I have installed different period opamps, just based on my experiences with differences between say a TL072 and a 5532. But the 5532 doesn’t always work for an 072. So testing is required. Take the EQ section for instance…there are 6 dual opamps just on the EQ board. And in stock form the EQ is not really enjoyable to use. I knew I wouldn’t be able to change the nature of how the EQ works and what problematic artifacts it imparts depending on how extreme the settings are set, but I wondered what different opamps might do. With DC offset measurements and square wave response testing I determined the 5532 only works as a replacement for 3 of the 6 TL072 parts originally installed. For the other 3 I tried OPA2134. I liked the changes. I believe overall the circuit block has better system slew rate, and better clarity…maybe quieter. But subjectively it’s a little less muddy and sterile, a little more musical. The same annoying things happen when the controls are not used conservatively (because it’s the same circuit architecture), but it sounds “better” to me and I like the changes so I kept them. The other thing I’m doing across the entire console is replacing the factory 5% tolerance carbon film resistors with 1% metal film, but only in the feedback loops of opamps that are in critical signal pathways…why? The thermal self-noise of a resistor is exponentially taxed in a feedback loop compared to, say, an input or output or loading resistor…not worth it in those cases. And maybe none of this is worth it as far as noise goes, but the prototype mixer has, relatively, a LOT of opamp stages. And I believe, like in a case where I’ve got all 12 input modules open and I’m mixing and am using the EQ on all those, having the better spec metal film parts in allllll those feedback loops likely reduces the noise floor. Does that noise floor differential matter once the program is playing…no…don’t think so. But I’m doing it anyway. Now contrast the examples above and the points about component quality and diminishing returns, limitations imparted by the circuit design in and of itself, with the, for example, Studer console, where you’ll find multiple circuit blocks, especially the critical input, summing and output stages, that are relatively very unique…boutique-ish…designed for quality performance and to deliver high headroom and drive with low voltage power rails…not at all garden variety…more expensive…AND expensive high-performance opamps and all 1% tolerance metal film resistors throughout…so it is the combination of high-performance circuit design AND high-quality components, which bring better returns because of the high-zoot circuitry.
…gold contacts will not tarnish…but they cannot make the sound better…
Totally agree. My drooling about all the of the gold plating on all the card-edge connectors and switch contacts isn’t about better sound, but rather reliability, longevity and stability. And I guess sound quality in terms of crackles and skritchies…like standard-grade switches and connections with tin plating or brass in any 20, 30, 40 or more year old analog console are going to make for headaches…so common…intermittent signal continuity, crackles, noise…ugh. Exercising and contact cleaner help, usually, but the need to re-switch a console with gold-plated contacts is SO much lesser compared to tin plated or brass contacts…and my Studer has zero issues with any of those symptoms. My Ampex MM-1000 has a lot of gold plated stuff…it’s almost 60 years old. These are the easiest connections to get working reliably or they already do. So it’s just a measure of quality IMO when that’s what you find under the hood, and there’s a LOT less fiddling to get things working.
Sweetbeat's* Studer is a wonder of its age but I suspect it is the opposite of Neve in that it has a very clean, transparent signal path?
I’ve never been up close to a Neve console…been able to hear signal raw coming out of it…or better yet to A/B compared to other makes. So I can’t compare. What I do know is Rupert Neve was genius and I’ve done some study of his designs, and very subjectively I am a child of the 70s and many of my favorite music across a range of styles came out of Sound City and for me those productions steer what I consider to be good sounding music, and I believe that console that was there has no small part in that. I think the Studer is definitely less of a “color console” compared to that Neve console for sure. In fact the Studer 928 really gets a lot of up-turned noses from Studer aficionados…really any of the later generation 900 series consoles (928, 970, 980, 990)…anything after Harman acquired Studer. They don’t consider it a true Studer device. Which I understand. And there is some mistaken lore that the 928 is just a reskinned Soundcraft B800…that it’s not Studer at all. None of which is true…entirely. The 928 is the only all analog console of that last generation of analog consoles (the others are digitally controlled analog consoles and, unfortunately, are often plagued with reliability issues related to the digital side), was designed in collaboration with Soundcraft drawing on their global position in the broadcast industry, but was built in Switzerland at the Studer factory by people that predated the Harman acquisition, and is packed with physical and electronic design features that predate the acquisition…clear Studer signature all over it. The Soundcraft B800 however is a reskinned watered-down Studer 928 that lacks the fancy mic amp, hybrid transformer outputs, stereo width filters, the Studer compressors and limiters, the high quality meter bridge modules and all the options, etc. I digress…so the 928 is comparatively not a “color console”, but the project was fixated on developing a robust and reliable high-performance, well-featured analog console for professional broadcast and production environments, that delivered “Studer good sound.” This is highlighted in the 928 brochure. And what I experience as “Studer good sound” is impressive clarity across the frequency spectrum without any mud or harshness, from the lowest lows to highest highs, and an exceptionally low noise floor. I wouldn’t call it “invisible” or totally “transparent”…like I think there’s are consoles out there that are really lauded as being completely transparent and there are some for whom that is the ultimate pursuit that defines a good signal chain. Me personally that experience is boring. I prefer different devices to bring *some* element of character to the signal chain, and the 928 definitely has a sound, and it’s characterized by the lack of mud in the lows or low mids, lack of harshness in the upper mids and highs, a really wide sound stage, like between the clarity and width of the stereo field it’s just very open, and with the lack of harshness the sound through it is devoid of undesirable edges if that makes sense. You can turn things up louder and it doesn’t hurt the ears, you just get more immersed in the sound. It’s just really usable that way. I have to believe the input and output transformers have something to do with the musical nature of the sound. I just love hearing stuff through it. It’s not boring. It’s exciting sounding to me. And then I’ve mentioned before what happens when you drive the summing or output stages, the mid-range fine fur that brings some “bite” and presence to edges of sounds and helps certain things to cut in the mix. So, no…not Neve mojo or musical color to that degree, but open, musical clarity that lacks harshness and can really be pushed, with the bonus of some very musical non-linearities I think from the transformers in the circuits.
…I could well be wrong and this super high end gear does somehow contain "magic"…But my 60 years of messing with components tells me…There is no magic, caps are caps and resistors resistors SO LONG as you use the right part in the right place and understand how it works.
And I think this is another way of saying the circuit design itself has more to do with the sonic performance than the components themselves, and in my experience I agree with this.
