Griffinator, et al who participated in the "Harsh CD" Thread...

[Mallette]

First, it was a great thread. I learned a lot and that is always good. I'd like to explore further one main issue.

1. What evidence is there to prove that ultrasonic harmonics have no impact on the audible spectrum and the response of the human brain to same?

2. As a teenager and young man, my hearing extended at least to, if not beyond 22khz. I could hear silent dog whistles, and those 140db "silent" ultrasonc motion detectors in banks would cause me to reel away in pain if I walked under them. I have reason to believe my 5 year old daughters hearing is similarly "gifted." What do you think a person will hear from a CD if their hearing goes out to, say, 23khz? Must be pretty brutal.

3. Why don't 78's sound as good transferred to CD as they do played back for real? I have about 400, including some very excellent wide range recordings from the 30's, 40's and 50's. Minimum sample rate that produces a recording that doesn't make me want to run for the disc is 24/88.2. Further, I've heard transfers of jazz greats and other material from 78 to CD where extreme efforts (so they said) were made to clean them up and provide the highest possible fidelity. Not one sounds good to me at all. I've never heard a successful 78 to CD transfer.

4. What about this http://www.stereophile.com/fullarchives.cgi?203 offered during the last thread? This and similar experiments have produced similar results ever since the infamous Dr. Diamond was booed down in the late '70's for suggesting that his music therapy patients didn't seem to get the same benefits from CD as from analog sources.

5. Finally, just what is it that people of all ages, persuasions, and experience are suddenly hearing when they experience LP again or for the first time? In my time on this thread, I've seen downright miracle conversions of people who stated up front there was no way they could be convinced and all this talk about analog was horse hockey.

In the last thread, there was an awful lot of "science says this and science says that." While I have a deep respect for and faith in science, I believe the input from my eyes and ears first. Should a mother ship suddenly appear over my house, I am not going to simply say "Must be swamp gas since science has conclusively shown that time or translightspeed travel is impossible."

Now, for any new participants, please don't flame me for being either anti-digital or anti-CD. I am not. I've released 10 CD's and build high resolution digital recorders of my own design. My experience causes me to believe there are good points and bad points to all the media we have available at the moment, and they should be used and judged accordingly.

Dave

[Deang]

So Dave, is this a matter of inches or miles?

[Marvel]

Dave,

Great thoughts here. Can you give us a rundown of your hardware? I know you like to record at 24/192, or at least 24/96. What soundcard or A/D do you use, and is your software totally homebrew or do you use off the shelf software, i.e., Cool Edit Pro will do 24/192? BTW, I just learned last month that Adobe bought Syntrillium's assets, and has done some kind of remake of CEP, released as Adobe Audition.

I admit that I think the ultrasonics make a difference, but I am also one who isn't driven crazy just listening on cheap computer speakers (some of the time). I probably listen at levels around 80db, and the days are long past when I would listen to a couple of hours a day. The days are too short and I'm doing too many other things.

Marvel

[Mallette]

Jeez, Dean. Probably depends on the particular set of ears, if I follow you. Some ears are immediately appalled at Rat Shack compared to Cardas. Others could care less, but prefer Stradivari to Amati.

BTW, as to your mention of DVD-A as 24/192 in the other thread, you are half right. The spec has provision for two channels of 24/192, but not more.

Marvel: As to my current recorder, it uses an ESI 192X card providing 8 channels of full duplex at up to 24/192. I am using Antonio Fabioli's N-Track software to record. While it has a few issues, Antonio responds directly and that makes up for it when you are going where few have ventured before. My mic preamp is an HHB Radius 10 vacuum tube design, and I use various ribbon mics exclusively. These range from 1936 vintage RCA BK3A to modern Russian Oktavas. I believe you have to get to classic Neumanns far out of my price range to find condensors that bring more to the table than the electronics involved take away. Those are out of my price range, and I find a lot to love in ribbons...they work a lot like ears.

I thought it would be easy. It's not. That much data flowing creates all sorts of issues, which may be software (OS), software (audio), hardware (bios) hardware (soundcard), hardware (drive and drivecard), hardware (regular old audio issues), or any combination of the above.

Something I discovered but still have figured out why, though I know a real computer geek or two that can probably tell me. When Sound Forge gets to 1:07:00 at 24/88.2, it just stops and says it can't do any more. You have to break the tracks down smaller even to save. N-Track, recording the same material looped out at 24/92, makes a new file every 28:00, thought it does not stop. I suspect both of these are software issues where the programmers never assumed that much data as well as the old OS limits. Since I am running XP and Win2K, there are no file size limits. OK, I am rambling, but those are just a few of the unespected issues one runs into at the edge.

It was a lot easier with the old DAT deck. But not as fulfilling or nearly as much fun.

While I always have fun with any direction a thread takes, I'd really like to hear some direct input on the stated premises.

Best regards,

Dave

[Deang]

Hmmm, you didn't take the bait. Are the Dave Mallett of the Group "Inches and Miles"?

[leok]

Dave,

Why would a second generation of anything sound better or even as good than the original? Assuming that the original was recorded well onto the medium (78s in this case), sending the signal through several additional gain stages alone should mess it up (cd recorder input / volume control and cd output .. probably both op-amps) plus a possible additional pass through a pre-amp (no necessarily, I don't know your setup). Add encoding and decoding to that. I would expect an additional layer of haze.

leok

[Marvel]

Leo,

There was a test some folks did a few years ago with one of the Panasonic DAT recorders, where they took a recording, copied to another deck of the same model. Copied back again ... They then did tests comparing the original with the fitfth generation copy. Most people could not tell the difference (I'm not saying there wasn't a difference, they just couldn't tell). The interesting thing was they were using the analog ins and out to make the transfers. One would have thought there would have been a big difference in the end.

I agree that a copy, going through any electronics should be degraded somewhat.

Dave,

Thanks for the rundown on your hardware/software. Did I tell you of a friend of mine who works at the Edison Historical site in New Jersey? They have recorded on Edison's original equipment, with current talent.They are then transfering to CD. Would be nice to transfer to 24/192 for archival sake, or until the medium is there for common playback. I could get their number if you would like.

N-Track is really good, but I have a hard time with the interface. It is hard to beat for the price though. I just noticed this weekend that Sonar only goes to 96k.

Marvel

[leok]

Marvel,

I understand your point, but in the context of this forum, I think it may not apply. People here can and do notice such things .. that is exactly why they are here. People on this forum are not "most people."

leok

[Griffinator]

I'll second Leok's assessment of the 78-to-CD results. You've got 4 stages of playback to get the vinyl to disc, then ADC. Follow that with DAC at playback and it's painfully obvious why the CD doesn't sound as good as the 78 you captured it from.

As far as ultra-sonics, the vast majority of humans cannot perceive sound above 20Khz. The only stock exception to this is young children (such as your 5 year old) who typically can hear up to 25Khz. Obviously there are rare cases where people retain this HF sensitivity, but it's not typical.

Repeated double-blind testing using realistic audio events (cymbals, car keys, etc) have revealed that the vast majority of humans do not perceive ultrasonics, nor do ultrasonics impact the way we perceive audio.

I would also like to know what you're using in your recording system. If it would make you feel more comfortable, I'll lay out my system, and we can compare notes.

- Mackie 1604VLZ-Pro mixer

- Alesis AI3 24/48 8 channel A-D converter

- Terratec EWS-88D digital interface card

- Athlon 1.77Ghz, 256MB RAM, 2x40GB RAID array

- Sonar 2.2XL, Waves Gold plugin package

- M-Audio Midiman Super DAC 24/96

This is chained into my 2-channel system as laid out in my sig for monitoring, along with an extra pair of Event 20/20 monitors and a pair of Infinity RS-2 bookshelf speakers. I use Monster interconnects throughout the whole system because I've found them to be dramatically superior to the standard Horizon, Hosa, et al cabling I used to employ.

Most of my mastering clients are DIY'ers who work in DAW software like Sonar, so I have them forward me bundle files from their software packages and I work with those.

My system certainly has its flaws, but it's a simple matter of extracting the best results possible with the equipment you have, and addressing issues wherever you can.

[Mallette]

Dean: Not familiar with I & M.

All: Let me state again.

LP or 78 transfer to 24/192: Identical

LP or 78 to 24/88.2 or above: Acceptable

LP or 78 to anything below: Maybe OK, but never in the league with the original.

Those are my personal results and experience. As to double blind audio tests whether the unwashed masses or golden ears, we've hashed and rehashed on the list and few, at least in this forum, have faith in them. I lost mine many years ago when I read of the early 20th "real vs. recorded" double-blind test in which no one could tell the real violin from the recorded one. The human brain just doesn't make for a reliable link in a scientific procedure.

I started to list here the things I DO rely on, but it is such a mass of various little things that is only useful to me to viewed as one and in context. Just no way to really describe it in a debate-worth mode.

Griff: Looks like a great system. It sounds as if you do a lot of electric/MIDI type music. I'd certainly agree that CD is a perfectly suited medium for that use. I cannot tell the difference (all things equal) between an all-electric band on CD or from analog.

In my case, I have gone to considerable effort to insure I never have to go through a mixer, mainly because it is unneccessary to my work. Even though I may use a spot mic on a piano or voice occasionally (rarely, I much prefer two mics only per stereo channel), I keep it separate and do the mix in software if necessary.

Marvel: I agree with you about the N-Track interface, and I dearly wish there was a multi-track version of Sound Forge, which I love. However, N-Track is field reliable, and handles 24/192 in multi-track well. I use Sound Forge for 2 track.

I have Cool Edit, but am not that fond of it. One thing I don't understand about multi-track software is why I have to jump through hoops to simply delete more than two tracks at a time. While no programmer, I just can't see any reason why I shouldn't be able to select across 4 tracks and hit the "Delete" button.

Dave

[Mallette]

So I'm thinking about this today and realizing that, to some degree, I am making assumptions. While there is no question in my mind that the presence or lack thereof of ultrasonic information influences percieved accuracy, I have been assuming that somehow these impact the audible directly.

While it may exist, I know of no evidence to support that particular mechanism. It could be something else entirely. There is much evidence for the existence of many things for which no mechanisim is understood. We can't even explain a simple thing like gravity. The current state of science is vastly overrated, though that is the case at every point in human history.

Dave

[DrWho]

our ears aren't the only thing doing the hearing either...for example, you can feel the bass of a kick drum on your chest, which further adds to the realism of the music. i see no reason why you wouldn't be able to feel higher frequencies and all that. i know my neck tingles when i hear high pitches that are loud enough.

http://www.anstendig.org

i know they talk about this concept somewhere on that site and how a certain level of relaxation is needed. ever notice how music sounds duller when you're exhausted from doing chores and sit down and listen during a break? i have no doubt that someone really really relaxed could "hear" as high as 30khz.

a friend of mine in the military was talking about using speakers playing inaudible high frequencies to induce fear in the enemy. dunno how much crap he was given me, but i thought it was interesting enough to mention.

[LynnOlson]

Heres my two cents worth on the HF bandwidth discussion. I'm 53, so no spring chicken here, but haven't been exposed to any severe acoustical truama I can recall. I know I can hear 15.75kHz just fine, because I hear a loud annoying whistle from most NTSC television sets.

However ... and this is where it gets a little strange ... I own a 36" Proscan that's a multisync TV (not very common, most HDTV's resample at 480 progressive). With this TV, switching from conventional NTSC to progressive-scan is pretty obvious; just like a computer monitor, the picture goes black, you hear a loud CLICK, and the pictures re-syncs at a visibly different resolution. It's basically a really big computer monitor that can run at 480 interlaced, 480 progressive (60 Hz VGA), or 800 x 600 at 60 Hz, re-syncing each time (I've tried all the resolutions with a laptop, they're really there and displayed with no visible convergence problems).

Anyway, to make a long story short, I watch DVDs and some cable programming in progressive-scan mode (using an iScan Pro), and the rest in good old NTSC, which I know has a horizontal rate close to 15.75kHz. Like many people, I hear the 15.75kHz whistle. What more unusual is that I *still* perceive a whistle sensation at progressive-scan rates ... where the horizontal refresh is 31.5kHz! I shouldn't be able to hear this, according to this or that academic study, but there it is. Maybe I'll have to cart the MLSSA system in the TV room and measure what the TV set is actually emitting ... although I wouldn't expect much 15.75kHz subharmonic, otherwise it would be visible in the picture as an interference pattern.

[artto]

Dave.......what do you think of this? The ultrasonic frequencies are having an intermodulation effect in conjunction with the frequencies below it. Sort of (for lack of better words) like frequency intermodulation distortion. But from the top down.

[Griffinator]

Dave - there's one critical question you're still not answering.

What are you using for A/D conversion? Is it an internal soundcard? Is it an outboard box? Is it some hybrid version (a la Aardvark's Q10)?

Until you answer that question, I can't get any insight into your process or what effect your process is having on the audio.

You've clearly stated that you're recording into your computer. Have you done waveform analysis on the material you've recorded? Have you verified that ultrasonics are actually present in your source material? Or are you going on faith that since your soundcard has 24/96 converters, it's recording at 24/96?

I suspect I already know why you're hearing an improvement, but I'm not going to give you my answer until I see your recording gear list.

Oh - and BTW - I don't do any MIDI work. My rig is set up to record whatever I have to, whether that be drums, acoustic or electric guitar, bass, violin, pedal steel, organ, piano, or a host of other instruments.

The only purpose for that mixer? 1) the preamps. 2) multichannel routing for detailed drum recordings.

[Marvel]

Griff,

He already answered with what he uses -- the newest WaMi Rack -- 192x.

http://www.esi-pro.com/

And uses N-Track for his software.

As I recall, mostly uses Royer ribbon mics.

Dave,

I don't know why you can't delete more than two tracks at once. You certainly can in Sonar (limited to 96k), and I thought you could in CEP as well, but I only had the demo on my PC right after 2.0 came out.

For what it's worth, my system is comprised on an ADAT XT (20 bit), Alesis Studio 32 mixer, Tascam DA 30 MKII DAT, M-Audio Audiophile 2496 Sound Card, and other assorted goodies. I haven't decided on the software yet, as this is also an extension of my video editing setup, which is a dual Athlon 1.4Ghz running Win2k and Adobe Premier 6.5. I mostly use some AKS C1000S mics, but still use some RE16 mics on occasion.

My 1/4" four track TEAC 3340S doesn't get much use these days, but earlier tapes I've done are excellent on this rather small format.

Marvel

[Mallette]

Thanks, Marvel. I use Oktava, Beyer, and vintage RCA ribbon mics. I'll check out Sonar for 24/96 edit. I passed it by when I found it incapable of recording at 24/192 and did not look further into it's edit capabilities. Since this will be issued in 24/96 (assuming it sees the light of day), Sonar might be just the ticket.

Griffinator: As to the question about my equipment, I don't see how it's relevant to the questions at hand. I'll provide any information you'd like, however, if it makes you happy. No secrets amongst friends.

I think it a bit strange you'd ask if I "were going on faith" about 24/96. I've mentioned a number of tests, including from above, this thread,

"All: Let me state again.

LP or 78 transfer to 24/192: Identical

LP or 78 to 24/88.2 or above: Acceptable

LP or 78 to anything below: Maybe OK, but never in the league with the original or the higher sample rates."

I don't want to sound cranky, but I DO have some idea both what I am recording and playing back.

Best regards,

Dave

[Marvel]

Dave,

Since interfaces can make or break the tools you use, I might be worthwhile to also check out Vegas and/or Cubase. While most people only think of Vegas as a video editing app, there are lots of studios now using it for just doing audio. Again, it will only do 24/96. If you get Vegas +DVD, you get a full AC-3 encoder with it as well as SFs DVD Architect.

Cubase SX/SL is another of my favorites, but only does 24/96 as well. My favorite one though is Samplitude (now at vers. 7.1).

http://www.samplitude.com/de/sam.htm

More tracks than you will ever need, and a very good sounding audio engine/summing buss. It WILL do 24/192. This software is nice, but the pricetag goes right along with it, at close to $1k. You can download the demo. It just won't save projects, etc. Way more elegant than Sonar.

Marvel

I think that subsonic and ultrasonic frequencies do influence how music sounds to us, but I don't know that you can quantify it in a real/definable way. The monitor/tv issue is a real problem for a lot of people. I'm not sure how high my hearing goes these days, but probably not as high as I think it goes.

[Mallette]

Problem with Vegas + DVD is that the encoder only does up to 48, even though the Vega editor handles 24/96. Go figure.

I'll take another look at Samplitude. I thought it either didn't handle more than two tracks or didn't do 24/192. Can't remember which.

The only DVD encoder I've found besides Sadie ($10k) is Minnetonka DiscWelder Steel at 475.00. It will do 24/96, but you can't even use your own graphics or slides. If you want that, you've got to go to Chrome at 2500.00. Add another 2500.00 if you need Meridian Lossless Processing.

I really need to sell something before converting the kids college funds for such things...

I've not even been able to locate a SERVICE outside Hollywood that can deal with 24/192. Nobody in Dallas/FW even handles 24/96. Bleeding edge!!!!

Dave

[Griffinator]

----------------

On 8/4/2003 8:37:21 PM Mallett wrote:

Griffinator: As to the question about my equipment, I don't see how it's relevant to the questions at hand. I'll provide any information you'd like, however, if it makes you happy. No secrets amongst friends.

I think it a bit strange you'd ask if I "were going on faith" about 24/96. I've mentioned a number of tests, including from above, this thread,

I don't want to sound cranky, but I DO have some idea both what I am recording and playing back.

Best regards,

Dave

----------------

OK, about your equipment. The reason I kept asking about it? It's real simple. When the ADC in your recording rig is located inside your computer (as the WamiRack is) it is subjected to a barrage of EMI from the CPU and other extraneous internal components. This has a huge impact on the resulting capture. I submit to you that the WamiRack at 192Ks/S is certainly going to be much better not because it captures 96Khz frequencies, but rather because at that samplerate it can self-correct the jitter errors, where at 88.2 and certainly at 44.1 it cannot.

About the "going on faith" statement. What I'm asking you to do is take the recordings you have made and put them through WaveLab, Cool Edit, or Soundforge and do a spectrum analysis. Prove to yourself and to me that you have actually captured ultrasonic information. Until you do that, you are "going on faith" that ultrasonic information actually exists in your recordings.

Throughout this whole discussion, I never said you don't hear what you hear - I merely argued why you're hearing it.