Slashdot Mirror

Of the top of my head:
CME
TC-electronic
M-Audio
and at the more average sound quality level
Behringer

Here's a very good paper on the subject from TC Electronic's tech library:

http://www.tcelectronic.com/media/lund_2004_distor tion_tmt20.pdf

Although it's a couple of years old it's still very valid.

See my sibling post to yours. ProTools is not 'inherently' about manipulating audio on a note-by-note basis, in the manner described by the GP -- although it is favoured within the industry because the interface allows you to work with your audio recordings in that kind of way quite quickly and easily, from the arrange page.

Conceptually, it's probably best described a DAW, though I'll grant you "multi-track recorder" is a pretty close approximation for all intents and purposes ;-)

These days, imho, ProTools is by no means far out in front of the competition, as it arguably was in the 90s. Rivals like Steinberg's Nuendo give you the same highly slick chop-up-your-audio-to-sample-accuracy-from-the-arr ange-page capabilities. Also, if you're doing any sort of even semi-professional audio, you're going to need a (semi-)professional audio interface (24 bit, 96khz minimum, probably multiple I/O) - this is one big chunk of what the ProTools hardware does, the other big chunk being DSP to run your plugins without straining the native CPU. Again, ProTools is no longer unique in this regard - TC's PowerCore is probably the best known similar product. Finally, as you say, there are plugins available for ProTools for things like autotuning, but again - this is very far from unique to ProTools these days. You can get Antares Autotune in TDM (the ProTools format) but also as native VST or DX (which basically means any serious music software under the sun can utilise it). There are a few plugins that are only sold as TDM, not in VST/DX/AU/RTAS, but there arent that many anymore, and you can bet there are roughly equivalent alternatives in the native (or powercore) formats.

Perhaps I'm over using the term but I am using convolution as one would use comvoluted, to make something more complecated. The EMU DSPs do really process sound. They can do EQs, reverbs, delays, etc. Basically any FIR function that can be executed in sufficient time (they execute a fixed number of instructions per sample), I'm not sure if they can do IIR or not, but regardless. So when a game requests a sound be played from a certian location, in a certian environment, the card does the calculation for how that wave should be changed, and what speaker(s) it should be played from. Go to the kX project homepage if you want more technical information on the topic.

As for VST plugins yes, I am fully aware of how they work. YOU seem to be the one that doesn't understand. VST is just an interface, an API, for plugins to talk to hosts. There is no reason one can't have the other side of that talk to hardware instead of having it be a software alogrithm. Some companies do just that. Universal Audio, for example, has a card the UAD-1. It's basically just a DSP on a PCI card. It supports VST, DX, RTAS, and AU, but the work isn't done by the CPU for those plugins, it's done on the card's DSP (http://www.uaudio.com/products/digital/ultrapak/F AQ.html). The PowerCore PCI MkII is the same basic thing, different company (http://www.tcelectronic.com/default.asp?id=1233).

For that matter Cubase 3 does something like this. You can set it up to use basically any external hardware processors as VST plugins. It controls the automation of the external device, and handles sending and recieving audio (http://www.steinberg.de/ProductPage_sb51ba.html?P roduct_ID=2442&Langue_ID=4).

It's not all that common, since most people would rather spend the money on something else and just bounce tracks when needed, however it's not at all a technical problem. You write a plugin as normal, but rather than implementing and processing in software, you have it do it on your DSP.

That's a good response and helps to clarify the differences. I'm interested in accurate quality conversions, and I don't play games much any more. There are a couple professional DSPs, UAD-1 and PowerCore, which are add-on DSP cards which you can use to process effects plugins that are specific to their hardware. And actually, the Emu 1212 has a DSP with effects, but I knew someone who bought one and returned it. Said he didn't like the quality of the DACs and the effects were sub-par.

UAD-1
http://www.uaudio.com/products/digital/ultrapak/in dex.html

PowerCore
http://www.tcelectronic.com/PowerCoreConcept

So it sounds like there could be some potential behind this new Creative card if Creative wanted to allow that to happen. What you said makes a lot of sense with the target towards gaming and what-not. My favorite card right now is Lynx Audio. Tight clock and great sounding DACs, but another great product that may work for some people, is Benchmark DAC-1. It takes a digital input signal, and is a little box that converts the signal to analog, and it is really good at it. Costs about $1000 but it may be the best converter for anything in that ballpark. So it has audiophile and pro-audio applications. It will be interesting to see how this Creative card goes once it's been out there considering there is at least potential both in the converters and the DSP even though it's geared towards the gamer/media consumer type. Anyways, just daydreaming here I suppose.

Computers already do real time FX processing, along with realtime everythingelse. You can head over to kvr-vst.com and browse through hundreds of them right now (for windows, anyways). The real problem with FX is quality. There are 800 software reverbs out there, and most of them pale in comparison to a 200$ hardware unit (convolution reverbs sound great, but eat all your cpu). I'm far more interested in things like TC's Powercore system. You can drop 300$ on this thing (the pci version, anyways) and get a suite of pro-quality FX that Hans-freaking-Zimmer uses. Bonus for not eating up the processor, because those 8 Absynth and Virsyn Cube instances are already bringing the machine to its knees.

What kind of latency does this pose?

There are currently lesser expensive audio DSP cards on the market (UAD 1 by Universal Audio/Kind of Loud, and the TC Powercore, and nowadays they don't cost much more than a GPU. However on both of those cards the latency is pretty harsh. Many audio system will compensate for the latency in some instances, although some can't/don't compensate for bussed effects, which is unfortunate as reverb is the greatest reason to use a card like this, and it is a bus effect typically, and the extra delay incurred acts to set a huge, usually inappropriate predelay.

Of course there will always be those willing to work around the potential latency issues, however that defeats the purpose that they state on their site (no more freezing/bouncing/yelling at the machine).

This is exactly why Protools TDM systems are still in vogue for higher end studios and producers. The TDM hardware does just about everything as offloaded DSP, therefore the latency is extremely low, fixed, and documented. You can look up (command-click on the track volume display actually) to find out the amount of latency on a track in samples, and if there is a need to compensate than you can figure it out. Although typically one doesn't need to compensate for only 20 samples of latency as that is less than you might find in a analog studio using digital effects.

Well they already make DSP cards for audio processing. Simply do a google(TM) search for "DSP card" and you will get several vendors.

I can't imagine it would take a whole lot to hack them for just their processing power outside of audio applications.

I have a Mexican pop album, which will remain nameless (except to say said album became platinum in Mexico), which has some very noticable processing of this nature. I am extremely familiar with the piece of gear they used and its sound (which is unmistakable on this album); it costs considerably less than $1000. I also have a another similiar device which costs twice as much, but is noticably more transparent.

Fixing singers that can't hit notes is actually one of the cheaper parts of a recording. It is the microphones and the room which cost the most when making a good recording.

- Sam