Michael Fremer defends Hi-Res digital while chewing out Gizmodo

You need the extra precision because you are running transformations on the data which given the way digital data is represented are lossy. You don't need it for playback. This is ABC stuff.

 

186 billion per nano-second. That'll get Fremer's jollies off, if not his Ya-Yas out.

 

;^) Right, but I'm interested to know what this particular audio professional has to say.

 

Me too!!!

 

For what? Recording/production or playback on a consumer level? I personally think bit depth is more important than sample rate, but I currently consider 24/96 the sweet spot for most music. 24/192 might be preferable for some classical or jazz recordings, but I think that 24/96 is better handled by most recording (a few exceptions of course) ADCs and consumer DACs. 192 sample rate brings with it a host of new problems that are still being addressed.

Most digital projects that I work on now are 24/96. Sometimes higher for processing and mastering, but I prefer to deliver final mixes at 24/96.

One of the best sources for information on hi-res digital is this white paper by Dan Lavry. Much more credible info here than the usual half-baked crap flatearthers disseminate on the internet in their boneheaded quest to discredit high resolution digital. Mr. Lavry kinda knows his stuff. I've used his converters and they're among the best in the world.

http://lavryengineering.com/pdfs/lavry-sampling-theory.pdf

 

Real engineers, people who actually build the machines, don't use language like that in technical discussions.

 

Some engineers are definitely more serious than others

 

Recording.

Yes, I'd come across that paper (as well as a more recent one, The Optimal Sample Rate for Quality Audio) via this other article, written by another pro: The Science of Sample Rates (When Higher Is Better — And When It Isn’t).

I was mainly interested to hear where you stood on the 192 kHz rate. By the way, I think it would be useful if those advocating higher rates on the forum stated where they drew the line, and why.

 

I enjoyed Fremer's article, and honestly it doesn't bother me. Some of you delicate flowers need to HTFU!

 

I didn't stop him, did I?

 

In 2004 the AES had a seminar, "From Hear to Eternity - Sampling, Conversion and the Limits of Hearing" which concluded that a sampling rate of 60k would likely be enough to cover anything that could be audible in the human hearing spectrum. That's why 88.2 or 96 has always been fine with me. 192 is a bit of an unnecessary storage hog, IMO, but as drives get larger and cheaper I suppose that could potentially change.

 

Yeah, but there's more than just a storage and price issue here; there's a sound quality issue, e.g., distortions, according to Dan Lavry (and others).

 

I can understand Mr. Fremer's hostile reaction: for 30 years, people who said that they thought LPs sounded better than CDs have been called Luddites and much worse. Personally, I also find it pretty offensive when someone with little knowledge of a subject, tries to lecture me on that subject. Absent definitive knowledge and experience, one should tread lightly. IMHO.

 

It puts the record on the turntable!

 

not that it matters, but I was listening to Wayne Fontana and the Mindbenders playing Girl Can't Help It ('65 og mono on vinyl) whilst posting this.

 

You do realize you just acknowledged what Robin L stated and in so doing rendered your comments pointless, right? You just defended Sony/Philips' right to make a business decision; a business decision that had nothing to do with "best sound quality at any cost". So, you're arguing, cyclically, against yourself. It's a bit comical. Nevertheless, I defend your right to listen to whatever recordings you want to at whatever quality you want. But, if I was a Chevy Vega owner I would not walk up to an F-1 driver and tell him I have the faster car. Know what I mean?

 

I didn't "defend" anything. I mocked the juvenile idea that it's somehow shocking or wrong for businesses like Sony and Phillips to make business decisions about the products they sell. I can pretty much be certain that unless you are as rich as Croesus and built it yourself, your own preferred system was also ultimately subject to "business decisions" that compromised its performance. In the real world, engineering is all about striking the right compromises. Or did you actually believe your audio vendor when they told you different?

 

Has "The CD Story" already been linked here? If not,

http://web.archive.org/web/20141104160226/http://www.exp-math.uni-essen.de/~immink/pdf/cdstory.htm

Pretty clear that the 16/44.1 Redbook standard has little or nothing to do with optimal sound quality and everything to do with the computer hardware available in 1979! By the development of the DVD (digital versatile disc) in 1995, it was decided (by engineers!) to make audio up to 24/96 available, even though at that time there was hardly any equipment capable of performing at that level. Why did they do this? They saw what happened to CD when it was locked into 16/44.1 and hoped to allow for continuing improvements in hi-res digital audio.

 

Engineers use higher sample rates when recording/mixing because In the DAW age 96khz means a lower latency ( delay ) when recording tracks through plug ins. It also means lower aliasing in some plug ins when using higher sample rates.

 

Thanks for posting. Interesting articles, at least for a novice lie me

 

Also if you argue higher sample rates don't effect what we hear be because we can't hear above 20khz,you really are showing your ignorance. What happens above 20khz most definately effects what we hear in the audible range.
You can equate it in result but obviously not functionality, to resonances in a tone arm on a turntable. You can get resonances at say 10hz and 30hz which you may think is ok as they are both out of the audible spectrum. However,they can combine to produce artifacts that fall within 20hz to 20khz.

 

Youch! Twist that knife! Funny, though. And even if it is true, I don't care! It's a matter of a small thing but mine own.

 

Okay, man, whatever...

(SIDE NOTE: What Veni Vidi Vici doesn't know is that I have actually DIY'd/rebuilt half the stuff in my "preferred system" just so I could avoid some of those "business decisions" that compromised performance. And, I don't have an audio vendor. Yet again by admitting "In the real world, engineering is all about striking the right compromises," he continues to discredit his previous statements. , don't tell him.)

 

That's easy ... Headroom. When mixing many sources together, or even recording a source live, its easy to run out of headroom. Tracking examples ... the singer gets closer to the mic, drummer plays harder, guitarist turns up his amp. Instead of a take being ruined because these moments blew past the original level set, they get captured due to the greater headroom available and adjusted later.. When mixing, its easy to overload the stereo bus as more instruments are added (if you are not paying attention). Greater headroom makes this less of an issue. Higher sample rates also give the in-box effects more bits to crunch on.

Usually higher bit depth is more important that sample rate. I typically record at 48K/24bit.

 

I'm an engineer - with a sense of humor. All engineering is a compromise, even the DIY kind you did on your system. Time, cost, capability. Pick two.

The compromise made with 16/44.1 was to pick a sample rate that was adequate to capture the entire audible spectrum as perceivable by the immense majority of human beings while having a readily available storage system to work with it. This considerably reduced the development expense and time to market of CD-DA and promoted its adoption.