Why vinyl?

Les, my friend, no offense taken at all. Yeah, we're all just talking here and trying to learn. I understand what you saying and feel the spirit.

I suspect from you comments that you've never really done any serious recording. The 20dB of ambient noise that you mention uses up part of the dynamic range and it's often really higher.

Say you have a fine consumer reel-to-reel recorder with 115 to 120dB of dynamic range at its highest speed. Your goal is to have as little tape noise as possible, so you'll need to record at the highest level possible without exceeding the recorder's dynamic range. So, you have the group play their loudest passage and set the recording level so it's just below or just touches 0dB (115dB in this case). You're recording trumpets and trombones that easily have dynamic range of 40 to 50dB: therefore, your recorder will be set so that the quietest levels are recorded at 65dB (115dB -50dB) which is very soft, even in a quiet room.

With tape, those quiet levels will have too much noise. Of course if you use the very best professional recorder a 30ips (burning very expensive tape) you'll gain 10 more dB of dynamic range. Anyway, back to our consumer machine, you've got a problem, the dynamic range IS too huge. What you do is move your mic back or use a compressor.

With the Korg, I've an extra 15 to 20 dB of dynamic range, which is a huge advantage. For instance, at the Rocky Mountain Trumpet Fest, which features a 64-trumpet ensemble at one point, I can set my mics at the front of the stage and have someone play a few loud notes from close range and then back of 15 or 20dB for my average recording level. Since the mics are 20-feet or so from the performers and the noise level of the Korg is so low that I don't have to sit at the recorder and "ride the gain" to avoid overload. I can actually play in the ensemble, record everything and edit later.

Good pop recording have 20-30 dB of dynamic range in them. Get a Radio Shack SPL meter and hold while you listen to Nora Jones or Jane Monheit. They'll start many songs at around 72dB at the beginning then average around 83dB and then peak for just a few seconds at 92 or so dB. On a good system with a good recording, that's very pleasant, not fatigueing at all.

Now listen to a something really dynamic, like Hugh Mesekala's Simela on 45rpm LP. It goes from about 65dB to just under 100dB at my listening position. It's really exciting. They hold the peak for more than a second or two, making the peak VERY dramatic by holding it for 15 seconds or so. That's loud, but still just under 100 dB. Some on this forum might actually let that peak get up to 110 dB, particularly if they set the level based on the very quiet beginning.

So, you're still saying, "that's not 130dB" and you're right. The extra 20 to 30dB allows the recording engineer ease of recording. If Stimela were ever going to get any air play, it'd need to be compressed. Listen to Motown from the '60s and '70s or EW&F from the '70s. The trumpets are actually blowing their brains out, but they sound like toys on those recordings, due to very serious compression, used to fit the trumpet within the context of the rest of the music. Drums are also seriously compressed on most pop recordings.

So, to summarize, the 130dB of dynamic range gives the recordist the luxury of being able to set and forget and still get a high quality. Brass and percussion can be difficult to put on any recording other than those meant to be played on the very best equipment; therefore, post recording mixing will often change the level on the final master.

Dave

Les, in response to my comment, you are effectively saying that for the same digitally recorded event, most people (you included) will prefer the LP playback to the digital playback due to the nicer distortions added during the playback. I agree, but I still contend that the digital recording and playback is more faithful to the original event.
Dave, ever try recording 72 brass + 25 percussion of drum & bugle corps? I perfectly understand your need for 130+ db dynamic range in recording!! Do the microphones even have that range?
I feel that the preent state of digital, recording and playback has more potential to do justice to the recorded event, but unfortunately, not many recordings are done with quality in mind, just loud sounds.

Bob P.

03-15-08: Inpepinnovations asked:
"Dave, ever try recording 72 brass + 25 percussion of drum & bugle corps? I perfectly understand your need for 130+ db dynamic range in recording!! Do the microphones even have that range?
I feel that the preent state of digital, recording and playback has more potential to do justice to the recorded event, but unfortunately, not many recordings are done with quality in mind, just loud sounds."

No, I haven't recorded a DC, but I've heard it up close, when I helped provide the side for one of the Texas Region regionals. Florida State now marches 430+ in their marching band and I've been 20-feet from the front line. AMAZING. Next November I hope to record over 100 trumpets at the Rocky Mountain Trumpet Fest. I love the BIG sounds and nothing beats being right in the middle of it.

Yes, my AKG condensor mics have 130dB and that's it. It works for just about anything recorded at any distance. I've got a 20dB pad in case I want to record trumpet with the mic right in the bell or a drum close up.

I'm hearing a lot of good all-digital recordings, by the likes of Diana Krall, Jane Monheit, Rebecca Pidgeon, Dave Gruisin, many Concord artists and others. Playback in 2-channel SACD or DVD-A is stunning. I even bought a Radiohead album in vinyl that was DDA and found it amazing and only 16-bucks.

So, I think there's hope for all formats. Digital playback is progrssing so fast (32-bit will be readily available at responsible prices within a few months) that the digital-to-vinyl gap is disappearing quickly.

I'm very happy that'll I'll be able to buy a true hi-rez harddrive, music server by the end of this year that'll do justice to my D2D recordings that I've been collecting since the 1970s. BTW, I will NOT be sending $20k Linn's way. (What a crock).

Dave

I don't believe digital is inherently more accurate. It has its strengths but is has its flaws. It takes an analogue signal and chops it up into little digital packets where to play back a microprocessor tries to reassemble the packets to recreate an analogue wave form.

I've been trying to think of a visual analogy so if I may.

16 bit digital is like looking through a screen door. You see the outside world image but the screen effectively blocks the image into little packets invariably obscuring the cleanliness of what the real world image should look like. Higher rez digital just has a less apparent screen door effect. Your mind sees the outside world image and can understand it but the mesh is acting like an A to D converter effectively breaks the image into its little block pattern. I hope this makes sense. Analogue recording is like looking out a window without screen but through a pane of glass that may not be perfectly crystal clear and may even have a film of dust and dirt on it. You see the outside world image. Your brain can understand it but its imperfections in the window glass and maybe the film of dust obscures it from perfection. Both formats suffer inaccuracies but maybe different ones. IMO we humans are more willing and accept and maybe even enjoy the inaccuracies of the typical analogue playback over inaccuracies over the digital ones.

Where digital can be quieter and can have expansive dynamic range etc it also in the A to D and D to A process suffers and loses things humans not only can hear but can sense.

Fact is whether one records on analogue or digital from the master on down to the final consumer product we must accept errors and inaccuracies in the final sound.

There are certain luxuries for a engineer to record in digital but so to there are luxuries to record in analogue. Its about what compromises you are willing to accept.

Just one point to compare in this simple regard:

Push a signal above digital 0db and you run out of bits and instant massive clipped distortion. Go too far into the red on an analogue recording and you get a general but progressively higher distortion.

Pick your poison I guess.

Les, you're talking about OLD digital technology.

Using DSD @ 5.6MHz the "packets" are as dense as the oxide particles on tape and dozens of times denser than the 16-bit technology you're using as a reference. I agree with you about old 16-bit with harse filters and other shortcomings, but the new digital is leaping forward by an order of magnitude every few months. Your arguement no longer holds water.

A DSD, i-bit, 5.6MHz sample does NOT lose data in replication. It's truly archival. Using professional-level programs you can copy the file and it will be "bit-perfect" in ALL generations. This issues you address have all been addressed and corrected.

With 130dB of headroom I hardly need to worry about overload; however, if I am worried I can add attenuation or a filter. The 130dB is only the starting point, but it gives the recordist tons of options to never reach clipping.

Let me think a little. I've read some good summaries of today's state of the art. You seem interested enough that if I can find you some links you'll start getting up to date. Have you heard good DVD-A? It's not as good as DSD, which has twice the resolution, but it'll give you some idea of what's possible in the digital domain.

Please note, I still have a very good TT, tube, phono-preamp and a big library of LPs. I believe that digital is just now catching up with analog, but, based on what I hear, it's the gap has disappeared. Unfortunately, you can't buy much at the highest possible resolutions.

Dave