Does a Multi-channel "output only" PC card exist?

  • Thread starter Thread starter sonusman
  • Start date Start date
I guess I would say that I don't know why for sure. I don't have any realtime effects that can be applied destructively so I can't do the tests you are doing.

One thing I will say for certain is that this has nothing to do with "errors" in the computer. It's just not possible as I explained above.

It might be possible that the effects your talking about have two internal functions. 1) Real Time 2) Computer Time. Perhaps the realtime version isn't as accurate (e.g. due to performance expectations).

Also, if you apply an effect destructively to a wave file, it's working directly with the file so ANY OTHER adjustments made with software, e.g. levels, panning, other effects, will not be involved.

Say you have a track and you bumped the volume up on it by 5db via the software mixer. Then let's say it's stereo and you panned it 75% right via the software mixer. Finally, let's say you added some realtime reverb to the left channel. Now, let's say that you decide to do a destructive compression on the wave. The resulting wave will be compressed PRIOR to the volume change, the panning, and the reverb. If you had applied the compression in realtime, then the compression would take place AFTER the volume change, panning, and reverb. The two will sound VERY VERY different.

Think of things like this:

A wave is basically a collection of samples. Let's think of an 8bit wave for simplicity. Sample rate is unimportant.

The data in our wave might look like this (made up numbers):

0110 0100 'sample 1
0110 0010 'sample 2
0110 0011 'sample 3
etc

Now when we play the wave, those samples, or bytes, are taken from the disk, placed into memory, and eventually shot off to the soundcard which gets an EXACT copy. Nothing is EVER modified on accident. This isn't an analog process where you can pickup radio interference or something.

Let's talk "realtime effects". In this case the data is modified just before going to the soundcard (let's assume no DSP acceleration). We typically use the term "stream" when talking about realtime audio. At 8/44, a samples in the stream are modified by a "plugin" at a rate of 1 every 44,100th of a second. That is, as it's being played.

Let's say for the sake of argument that our effect divides the sample by two. Why? Because the math is easy :)

0110 0100 'original sample
0011 0010 'sample/2 in realtime

Now when we talk about destructive editing, we're just taking the samples from the disk and modifying them via the effect algorithm as fast as possible. There's no timing involved. The SAME number crunching takes place however.

0110 0100 'original sample
0011 0010 'sample/2 in computer time

The sample is still divided by two regardless of how long we took to do it.

I know that my little bits don't mean much, I'm just trying to get you to think in computer terms...what's really going on or at least what's really possible. Bits are bits are bits. Time is not a factor in how a processor functions.

Anyway, I do agree with you 100% that mixing with a mouse sucks. BUT, there are plenty of external mixers that will control your software mixer in realtime. Now that's fucking cool. Right now processors are kinda lacky when it comes to realtime processing, but in 2-5 years I bet you won't be able to tell the difference between using your old digital mixer and your software mixer controlled via some external mixing console.

You gotta remember that the computer is the greatest device of all time. It's litterally a machine that can be made to perform an infinite number of tasks. Really incredible. Also remember that we're still in the diaper stage.

I wouldn't recommend that you, as a professional with buttloads of gear trade everything for an iMac running Pro Tools. That'd be ridiculous. BUT, I bet that in 5 years you'll be exploring a similar option.

Also, for those of us who are new and/or hobbyists, this computer option saves big bucks and offers SO much more versatility over a mixer and an ADAT. And as us "newbies" start buying this shit up, you should see great advances in computer-related audio technology. I'm excited about it.

Slackmaster 2000
 
Ooh I should add one more thing.

Some plugins are not designed for high bit depths. That is, if you use a 16bit plugin on a 24bit file...it should still work, but it's going to sound like shit.

There is a program out there...I think it can be found on AnalogX...that will basically tell you how good your plugins are (e.g. what bit depth are they working at).

This might provide you with some clues.

Slackmaster 2000
 
Well, all the Steinberg stuff is held in fairly high regard. I am assuming that the Mastering Edition Suite of plug's is 24 bit capable as it is advertised that way, and Steinberg makes some pretty impressive stuff for the PC.

Anyway.

I guess I will need to hunt down this issue about real time processing not sounding like applied processing. It is not too terribly big of a deal on my system. It is only a tad bit annoying at this point, but, the more I use it, the more I wish that I didn't have a setup that reveals so damn much about the audio... :) My days with my DAT recorder and stand alone CDR burner and cheapy Yamaha 16 bit soundcard made all this computer stuff so much easier because I couldn't hear the bad effects. It would seem that 24 bit has kind of opened up a whole can of worms for me. Damnit anyway. I should have never started reading Bob Katz articles... :(

Those killer interfaces to digital mixing and processing already exist. Go to any of the leading recording/mixing consoles websites (I am of course talking about brands like NEVE, Amek, SSL, Soundcraft, etc...not Yamaha, Mackie, Tascam, and Panasonic) and look at their digital consoles... :D There is your interface, and they are very expensive. Prohibitively so.

I am having a hard time believing that PC's are going to catch up to their quality anytime soon Slack. There just has to be a reason why good quality digital mixers cost over $100k. If ProTools and a snappy tactile interface could do the same job, how come these high end digital consoles are prefered? Is it a case of studio engineers that just can't get over having something that looks like a console? Is it a case that they just like to spend way too much money on something? Is it a case of they don't know that a PC can deliver the same quality? I think not. These consoles are used in studios with engineers that know more about this kind of crap than any 5 of us combined. These consoles supply superior processing power that I find it hard to believe that PC's are going to offer for a long time to come. Come on Slack, applying your same logic, a 486 with 8 MB of RAM should work just fine as a application and file server right? Sure, as long as you don't require it to do too much. So, maybe a PC can handle one or two tracks of audio at a time with comparable quality if it utilized most of it's resources to the task. But when we are talking about 24, 32, 48 tracks at once? Something has to give. If a PC was even close, those $100k+ consoles would never sell at all. Most studios would just wait for the PC to come of age rather than spend that kind of money... :)

Anyway, I am not trying to shoot down your theory per se. Just trying to get some answers. You of course know a hell of a lot more about software in's and out's that I do, so maybe I am missing something here. I just know that in audio, more times than not, higher price equals higher quality. Seldomly is there that much of a mark up on Pro Audio equipment. I have had this verified at a sound company that I work for that is part of a music store. I have been shown by my manager the price the store pays for audio equipment compared to what they sell it for. Seldomly does any of it sell for more than a 15% mark up over cost. So, the retailers are not getting rich on this high priced stuff. In fact, usually the better quality gear sells for as little as a 5% mark up. Often, because it is so damn expensive, it sells at cost to just get it out of the store to make room for the cheaper stuff that everyone seems to think sound as good or better... :)

Ah crap, I just hit the "f*ck it" stage on this again. :D

When I get to the bottom of this whole computer as a mixer thing, I will let you all know what I find out. I am sure that it will make sense why the expensive digital processing stuff is expensive, and why computers are good for word processing and posting lengthy posts on BBS's on the internet... :)

Ed
 
You misunderstood what I was trying to say.

Yes external mixers for multitracking software exist, but your software mixer does not react instantly to your commands...that is, computer multitracking is not quite realtime yet...it's more "wait a sec" time.

This is completely tolerable when you're using a mouse to mix because you expect delayed reactions on a computer....but it would be very annoying with an external mixer since your brain will have switched over into analog mode :)

And don't ever say that a computer won't be able to do "such and such" any time soon. Notice I said 2-5 years. Do you remember what your computer looked like 5 years ago? Perhaps it was that 486 you described. How about 10 years ago? That 386sx with 1MB RAM, 256K Video Card, 40MB Hard drive, 5 1/4" Floppy Drive, 2400 baud modem, and 14" EGA monitor was a real screamer running Windows 3.1 or DOS 5 wasn't it :) (my first computer actually)

Software is cheaper to develope than hardware also. So yes, if a software version of a digital reverb box was developed, it would (or could be) be cheaper than its hardware counterpart. Why? a) Your development team is probably going to be 1-3 people on a project like this. b) People expect flaws so QC isn't too rough and you can fix the product after it has been released. c) CD's cost less than $1 to produce. Online "download" sales are even cheaper since your company already has a web server. d) Software cannot break down therefore there is no real warranty to worry about. e) You're competing with the hardware vendor. The coolest thing? People will always think that they need to upgrade relatively often, regardless of how well their current system is working, therefore you'll have repeat customers for what is basically the same damn product.

And I'll tell you why that expensive hardware is so expensive. 1) As with analog devices, good sounding digital devices require a lot of hard work...trial and error ...nasty physics...etc. People expect perfection when it comes to hardware, right from the box. 2) Like diamonds and Harley Davidsons, prices are inflated to give a sense of "worth" or "value". High prices are common in the music industry because they're expected, and there is a relatively small consumer base for the fancy stuff. (the same happens in the software industry)

And finally I'll take a crack as to why we don't have GREAT software yet. The software is NEW! You have all these old school companies that have been producing hardware and shit for years and years and years. These are the companies that really know their stuff. Are they going to one day wake up and say, "Hey, let's put a whole shitload of effort into software. An untested market that hasn't proven itself!" Hell no. So you have these startup audio software companies. They probably aren't quite as smart as their hardware competitors, and they only have a handle on a small segment of the market. Really, who the hell is going to pay $1000 for some unfamiliar software plugins when they could buy some time-tested shiney box for $2000! I don't think I would even do it. So the software companies probably aren't putting out the greatest shit yet...regardless of who gives them awards.

The thing is that I can forsee no reason for the computer to disappear any time soon. Nore do I see any reason why there will no longer be a need for research into audio solutions for the computer. Five years man, you're going to see some shit.

Slackmaster 2000
 
On this topic: I don't think a PC can replace that expensive gear Sonusman is talking about. If I was paying 100$ an hour a studio to get my stuff recording, I wouldn't expect them to use a PC. But listening to the tapes I made when I was using a 4-tracks, comparing them to what I can do with the PC, for low-buck, entry-level homestudio, it's hard to beat a PC. But I admit it's really time consuming to make a good mix, you spend a lot more time fiddling with fx controls. But upgradability is incredible. I was able to upgrade to 24-bit without changing anything else in my gear. Imagine you bought a 150000$ digital console a couple of years ago when 16 bit was considered enough and then you discover just how better 24bit is...
 
Yes, but there is a limit as to what the human ear can pickup and therefore there is a point we can reach that is good enough forever as far as *current* digital technology is concerned.

There is also a no need to do things like use up 100 tracks with 50 realtime effects each. Unless music changes substantially, 48-64 tracks will probably be the limit for most people...and there will be a time when a computer will chuckle at such a silly task. The same is true with sound processors and effects. You only need so many at one time.

Just think, 5 years ago you couldn't really do ANY multitrack recording. You could do some editing, but you weren't working with a lot of waves at once on your 850MB 2MB/sec PIO hard drive. And that Pentium 75 wasn't exactly running 50 realtime effects.

Remember, if you have a piece of purely digital gear, and you think it absolutely rocks....a computer can mimic it exactly.

Slackmaster 2000

[This message has been edited by Slackmaster2K (edited 04-14-2000).]
 
You must log in or register to reply here.
Back
Top