The Neuman U87 is a very popular microphone in the audio world in general (TV, radio, post-production and even music), especially for voices[1].

Going a little further, the mixing console you can see on the picture is a Lawo Sapphire[2], which NPR seems to have installed in 2012. According to this gearslutz post[3], a 12 faders + central section was starting at 20,000€ back in 2010. You can see on the picture that NPR has a 24 faders version.

The audio monitor on top on the console is an RTW[4]. It's kind of hard to tell which model it really is, but the TM9 for example starts at $4,200[5].

[1]https://www.gearslutz.com/board/low-end-theory/618100-mic-vo...

[2]https://www.lawo.com/products/radio-consoles/sapphire.html

[3]https://www.gearslutz.com/board/music-computers/864209-what-...

[4]https://www.rtw.com/en/products/audio-monitors.html

[5]http://www.rspeaudio.com/RTW-TouchMonitor-TM9-p/rtw-tm9.htm

So let's say they have one mic per five employees. Those employees cost (I'm guessing) at least $300 000 per year, and most of what they produce of value for NPR goes through that mic. Doesn't look very expensive now?

And I haven't factored in cost/rent for the studio, music licensing, and a whole lot of other OPEX. Since there are quite a few local radio stations around, I'd say the CAPEX is not a big deal.

Talent for NPR might pull $300k (I have doubts) but if you're including everybody else when you say "employees" you're shooting astronomically high.

Radio pays notoriously bad. Local TV is just as bad, and they can't even afford photographers any more so reporters do their own shooting; I was laid off as a photographer when that change was made at a former haunt and newscast quality plummeted. (It's popular and spun positively now: Google "backpack journalism.")

Particularly local, you have to really love the craft to be in it. National/international is better, but to an extent. On-air talent soaks up the salary budget at pretty much every shop, universally, and to use a local example like KTVU or KRON I wouldn't be surprised if on-air talent was making $500k or less. You can do better at Google. The producer running the newscast is probably struggling to make rent. The camera operator certainly is, if they haven't yet been replaced with automatic, motorized cues driven by an overworked TD, who is just as worried about their own rent.

Ad-based business models suck, particularly when owners hate spending money. All media owners do. We got punished for the millions that had to be invested for the HDTV transition with 15% pay cuts at one station in the middle of nowhere.

NPR's last CEO, Gary Knell, made about $750K in his last year in the job.

They have a number of highly paid mid-level people, such a diversity officer (Keith Woods) who makes $220K.

More notable hosts make over the stated $300K, including Steve Inskeep ($400k) and Michele Norris ($350K –– wage gap!).

Highest honors, at least according to the 2013 990 linked here[1], go to Renée Montagne, who pulled in $412,581.)

These are not fully loaded costs. You probably want to add 10% for D.C.-based employees for employee overhead incl. labor lawyers, retirement contributions, payroll tax, &c.

Not bad to be a corporation that doesn't pay taxes, takes federal taxpayer money, takes state taxpayer money, takes corporate gifts, solicits foundation grants, and asks for individual donations.

[1] http://990s.foundationcenter.org/990_pdf_archive/520/5209076...

Their non-profit status doesn't have much to do with their staff compensation. Broadcasting at this level takes a lot of talent. I realize this every time someone sits in for a good broadcaster. Sometimes the substitutes are nearly unbearable. Good broadcasters don't work for cheap.

They wouldn't be successful at soliciting donations if they had bad broadcasters. Most podcasts have a tiny budget, which is part of the reason most podcasts are terrible.

I largely agree with your comment that talent doesn't come cheap. Software developers, venture investors, bank CEOs, &c.

In the case of NPR staff, I question whether anyone else is bidding on their services.

Michele Norris previously worked for ABC, the Chicago Tribune, the Washington Post, and the LA Times.

I worked in radio in a fairly large market. On-air talent made good (not great, compared to tech salaries) money. The best talent did lots of events & wrote books to push income, always moving to a bigger market if possible. Upper management (as in, national corporate management) made bank. Everyone else made garbage.

As someone upthread inferred, I was suggesting 300k or more for five people in total.

On a differen note, it is a dollar a day for 10 years; not so much, is it? For the added value it gives.

Am I misunderstanding, or do you seriously think NPR on-air talent is making in excess of $300,000/yr?

Employees cost a business much more than their salary. There are benefits, taxes, insurance, administrative overhead, etc. There's also the cost of replacing them.

Morning Edition and All Things Considered are hugely popular. Broadcast personalities on a national stage usefully get compensated pretty well.

According to NPR's IRS form 990 the hosts of Morning Edition and a couple other big shows do indeed make over $350k.

I think the GP intended this to be the figure for five employees combined, not per person.

Exactly.

The ones you can name offhand (Steve Inskeep etc.) generally are.

http://www.andymboyle.com/2013/07/21/an-updated-look-at-ira-...

I'd be stunned if the console it's connected to were any less than $20k. More likely in the $50-100k range.

Audio is a very expensive game to play well.

It's not that hard to get very good audio these days. Truly incredible audio will always be expensive. But the mid-range MI gear is arguably good enough to use.

When you're buying U87s by the dozen, mid-range MI gear won't cut it.

I suppose - but I'll probably never know if that's for actual performance reasons or for object fetishization reasons.

Object fetishization is not to be dismissed - having dozens of U87 and Manley stuff might be a competitive advantage.

An audio signal is only as clean as the worst element in the chain.

But these days, past a certain price point, you don't really get much cleaner. It's not like it was when you found studios with a ( relatively nice say, Peavey AMR ) console and a Fostex 1" 16 track, to where a nicer console and better tape machine were audibly better.

I have a CD or two done with a D&R console and a Mitsu DASH, and I'd say this is comparable. Room and human factors dominate.

You can get 8 channels of A/D-D/A with Lightpipe for 8 more for $500 bucks and while I can't say definitively it's as good as it gets, it loops back to < -90dB for noise & distortion. It's good enough to where the "worst element" is usually me :)

Mics (and to a lesser degree high-end preamps) are not so easily replicated as digital gear.

If you want the sound of a Neumann U87, or U47, or Shure SM7, or 57, then nothing - nothing - is going to scratch that itch like the right mic for the job, if you really know what it is you're listening for.

Preamps and signal processors too... I once thought as you did: that digital DAWs have so many advantages over hardware that it just never made sense to put $5000 into a particular piece of hardware when the same money would buy a pile of digital plugins.

Until I had all the digital gear possible, and I came face to face with GIGO: garbage in, garbage out. Audio starts life as analog and that's the beginning of everything that follows in the signal chain. Past that there are certain non-linear things that happen in analog that are particularly euphonic and hard to reproduce digitally.

For example I've got pretty much every plugin compressor ever made, but nothing can take the place of my UA/RE175 compressor, because it's delightfully nonlinear and chaotic and valve-y.

And even if you could perfectly model it, what you can't do is put it where you need it: in the signal chain before A/D conversion, so you're packing as much information into the track as possible. The same goes for mics. Mics distort the sound more than anything in the straight signal chain. Every mic has a fingerprint - once that fingerprint is in the sound, it's there. Same goes for the fingerprint of the musicians instrument / amp. And the room. Etc.

And thus my maxim: the earlier in the signal chain you can get the sound you want, the better.

"Fix it in the mix" is the devil. "Fix" is something you do to a cat.

Considering things like gold-plated speaker cables and crystal sound purifiers, audio is also a very expensive game to play stupidly ;-)

Studio audio technicians are not so easily fooled. I'm sure the field has its fair share of peculiar magic, but it's not like home audio.

People who spend thousands on exotic speakers in the name of "transparency" in their home setup could do so much better in that respect with a pair of active studio monitors. But they might not like the results as much.

Keep in mind though, the frequency response curve of studio monitors is more intended for pointing out flaws in a mix than enjoying music. Some of the engineers I've talked to pointed out that NS-10 monitors (once very poor selling consumer speakers, they've been considered legendary in studios for 25/+ years) are particularly harsh, and they don't enjoy listening to things on them when they don't have to.

https://en.wikipedia.org/wiki/Yamaha_NS-10

Things like Beats have frequency response curves that're basically a polar opposite from the NS-10s. Even less cringe-worthy headphones, like Sennheiser HD-280s have sort of a "smiley face" (dip from ~900-2800 hertz, while boosting very low and very high frequencies) EQ curve. Not to discredit them - they are good headphones, but in doing my own projects, I honestly feel like I've done better work on terrible sounding, $5 headphones.

"If it sounds good on NS10's it'll sound good on anything"

I have fucked up a few mixes by using $5 headphones. But yes properly monitoring gear will point out flaws rather than make something sound good. When I get to the end of a mix (using Sony MDR7506's for example) and I think "man this sounds SO CLOSE but it's still a little shitty" then I'll go and check it on iPhone headphones or something and it'll sound fantastic.

>When I get to the end of a mix (using Sony MDR7506's for example)

This. All that talk about "great detailed sound" that is "as close to the original as possible" of some $1000+ headphones and yet most music is done on something like $100 MDR-7506. Moreover, 7506s are very comfortable and sturdy, I can't comprehend why anyone in their mind would shell out hundreds or even thousands bucks for their "everyday" headphones (granted, noise-cancelling or sport-y ones are a bit different thing).

> most music is done on something like $100 MDR-7506

Musicians wear headphones in studios when recording, but audio engineers pretty much never use headphones for mixing except as microscopes to listen in terrific detail for something in particular.

Mix engineers mix on speakers almost exclusively.

Considering how many people listen on headphones as opposed to past decades, would it not be a valid choice to do more headphone mixing?

It's true that you can "mix for the listening format."

This is just my personal experience and opinion (and that of pretty much every audio engineer I've worked with): headphones lie. Ears are designed to perceive sounds in 3-space, not to have individual channels shot down the ear canal bypassing the outer ear, with a sealed transducer directly loading the eardrum. As an audio engineer, you're looking for the truth, and this isn't the way for your ears to best perceive it. Moreover headphones are generally fatiguing compared to a good set of studio monitors in a tuned control room.

So the most representative mix - regardless of listening format - is going to come from speakers. I think most audio engineers are going to agree on this point.

Some audio engineers "mix for the format" - I suppose if you want a mix that is optimized for headphones, then yes you're going to at least check the mix with headphones, but I still doubt many engineers would do the mix on headphones.

Seasoned engineers know that listening fads come and go - the way to make a mix that will stand the test of time is to make a mix that sounds good on anything, and "good studio monitors in a good room" is going to get you there best.

What makes these devices (e.g. the microphones, monitor speakers, the mixing console...) so hideously expensive isn't even the sound quality (even if I wouldn't expect anything short of the technically achievable performance in mic-preams, dynamic range/noise, ...).

This equipment will have to run 24/7, be constantly worked on, be easy to service, will likely have redundant power-supplies, maybe redundant DSPs/Routing/I/O. And if anything breaks, you'd expect to get a 100% drop-in replacement (in terms of frequency response, gain, ...) delivered over-night, or by a courier/technician.

Regarding the mixing consoles: For live-sound (the only area I know a little about), only recently a wealth of very affordable mixing-consoles have flooded the market, but anything slightly larger still sets you back ~€10k for an entry system (Midas Pro, Yamaha CL, Soundcraft VI, Allen-Heath GLD). And there, I suspect the number of units shipped is one or two orders of magnitude larger than what Lawo sells for fixed installation in studios.

Generally, the equipment will work flawlessly, even when hauled around every weekend and not handled very gently.

A home stereo has to take maybe 2 channels through a hilariously blunt EQ stage (bass/treble knobs) and then up to suitable wattage for a pair of small speakers, which only need to sound right in an area of a couple square feet.

I won't get into it all unless you want me to, but a system for live performance or broadcast has orders of magnitude more work to do, functionality which is demanded by a tiny market relative to consumer electronics, with far higher expectations on reliability, flexibility, etc.

There are certainly people (Pyle, Behringer, etc) who claim to sell cheap solutions, but they are substantially more likely to bite you in the ass in concrete, embarrassing ways if you try to use them in performance environments.

> I won't get into it all unless you want me to...

My experience is 100% studio and 0% live, and though I'm not ygra, if you're willing to type, I'd love to read.

You don't find real audio engineers using stuff like that. Gear has its place or it leaves the studio.

That comment actually intended to say »Even when you have no clue about audio (i.e. are doing this for your home, not a studio), you can leave a lot of money (in this case unjustified). Apparently that wasn't clear enough :-)

That's not as bad as it sounds because the console, preamps, compressors, etc... all cost a ton of money too.

Even the headphones themselves are probably at least a grand and they've got way more of those laying around. $3600 is a lot, but audio equipment gets REAL expensive pretty quickly.

Edit: Fair enough, I'm entirely incorrect as far as the headphone price point.

Sony's MDR7506 is, or at least used to be a few years ago, the reliable radio workhorse. When you spend all day with your headphones, especially in the field, super expensive cans most people like for music are a liability. I love the 7506s: flat, cheap, and comfortable all day. <$100. NOT the consumer version (-V6 or something). Real, purebred, MDR7506. NegativeK is right, too; the $100 AKG range is also popular in working radio, as well as some Sennheiser HD280s.

On the production side of things where quality super matters, you'll see the more expensive headphones for sure. But every single person in the last newsroom I saw rocked the 7506, and though I've moved on to more expensive headphones for programming and my music collection, I still miss my 7506s on occasion.

I have owned multiple pairs, and still use one for my main work. The bass and treble roll-off isn't great for mixing and mastering, but it's a truly great value for the money for most monitoring uses in performance. It's better on the production vs. consumption end - where people like all their colored sound.

It's like the NS-10's of the headphone world. Well-understood and consistent, while offering a great average, flat experience that translates well over the air.

Yep! Headphone nerd Marco Arment hates them for some reason, but you see tons of actual professional musicians using the MDR-7506 in the recording studio. Not just radio/broadcast people.

I don't think there's anything that comes close for ~$100 in terms of sound quality and comfort; you'd have to spend at least $300 for a real upgrade. And even then, there's an element of subjectivity.

They shoot for flat when they make them and they're indeed flat enough for sleeves-up radio work, but I have heard folks say (somewhat correctly) that they're a bit bright up high and don't have enough gas down low. I can understand Marco's opinion in the context of being an audio nerd, because they're flat, but potentially not flat enough for some kinds of listening.

But yeah, couldn't agree more. Price point.

And they last ten years, then you change the pads for $10 and they go another ten years...

I think they are good enough for the application they are designed for and somewhat of a bargain at the price point.

Having listened to most and owning/owned for both studio and personal listening use some of the headphones that Marco reviews I can’t help but wonder if there is some bias going on with the music he reviews with or whether he actually does have very skewed hearing. Having been used to fine equipment and rooms whilst working in mastering suites, recording studios and I have picked up a preference for probably a certain type of sound but when auditioning and getting used to audio equipment I often test with a wide variety of sources that includes some music I don’t personally like but know quite well (such as country and western) as a reference to understand capabilities across the board.

Whatever ones thoughts it highlights the importance of auditioning oneself and getting to know something well (goes for speakers, rooms and headphones). It’s possible to do produce a good result using MDR-7506 or NS10s if you are comfortable with them and have the skill to do so.

What is a can?

Sorry, "cans" is radio slang for headphones, from the can-and-string metaphor. I'll make that clearer.

Inside one of my Focusrite audio interfaces, the internal headphone-amp, and wiring to the headphone connector, is labeled "cans" ;-).

Actually, I've seen AKG studio monitor headphones quite a bit -- and they're right around $100. Whether NPR uses them is a different thing, though.

Apropos of not much, I like the venerable Beyer DT250 for everyday use -- they're similar to the DT150 studio cans but with a velour kind of finish, that makes them comfier at home but not so suitable for passing around in the studio. They're also in the 100-200 dollars/quid range.

Preamps and compressors will be integrated in the "console" (even though typically the user-interface will be separate from I/O and DSP, the latter being housed in a 19" rack).

This is radio, not some boutique studio with a ton of outboard gear.

preamps will cost a good couple thousand but compressors, etc... have now moved into digital computers. An expensive set of plug ins will cost $2,000 but you could get by with $200 easily as a professional vocal shop.

I use a AKG C-414 ($900 +) with the bass roll and you are pretty much there.

All mics colour the sound in different ways. For some reason the U87 adds a colouration that's perceived as expensive, larger than life, and authoritative.

Other mics - including cheaper Neumanns - don't have quite the same effect.

It's not about capturing the sound perfectly accurately, but about creating an impression with sound. That's why some pro audio gear is so insanely expensive.

It's very hard to create the same impression digitally because this kind of subtle analog distortion is complex, non-linear, and not very well understood.

There is a reason why AKG 414 are probably the second most popular mic for studios that have huge budgets. 414 sound is also considered an awesome mic.

I've never gotten a better drum overhead sound than with a spaced pair of 414's (I think TL II, but it's been awhile). Love that mic.

That said, I don't think the 414 color is the same as the U87 color. My impression has been that the U87 is much more flattering to vocals, but I've admittedly never had the opportunity to use one.