suggestion for PC card

I am seeking the sound equivalent of a graphics
accelerator card (nvidia,ATI) to the extent that
hardware generators and effects can be leveraged
by Reaktor and Reason (the programs of most
concern to me) in the same way that graphics software
leverages the hardware graphics pipeline. The problem
with Reaktor on my system is that it eats way too
much CPU time. I don't want it using the CPU but instead
accessing the hardware on the sound card.

Right now, I have an Audigy2, and I am not aware of
whether this card has hardware effects. Can anyone
recommend cards under $300 that will "accelerate"
Reaktor and Reason?

Thanks for your suggestions.

 

you need a faster CPU. Audio algorithms cannot be processed on a different board in the same way graphics can.

Maybe you need to look at Nord G2E ?

Or check out a second laptop to do nothing except Reaktor, and keep your other audio apps on another machine linked by OSC, MLan, or somne of the other fast sync systems.

 

I appreciate your suggestion on the Nord and the
laptop idea...


> you need a faster CPU. Audio algorithms cannot be
> processed on a different board in the same way graphics can.

Consider the following architecture for a board: memory
for synths, where each synth is a network of functions,
and each function is either a generator or effects
processor. The card would come with a huge collection of
basic generators and effects. When I build a softsynth
on my computer, it gets sent to the card, which then issues
sound. So Reaktor and Reason would help the user to build
the softsynth, which is then "rendered" on the audio card.

In theory, such a card should not be difficult to construct,
and the CPU would indeed not be necessary to achieve
effective sound synthesis.

 

the card would need to process whatever you are sending to it. Otherwise you are talking about a 2nd hard drive with your ensembles/vst on it.

 

Such dedicated cards do exist, but they are not, by definition, for 'native audio'. The korg oasys is an example of such a thing.

 

There are several hardware accellerators but as said above they only run special software applications, but they are all of high quality.

Google these:

UAD from Mackie
ProTools
Creamware

 

It would be great if the softsynth vendors,
Microsoft, and Apple would get together and
create some software API standards so that
the cards need not be special-purpose, tied
only to one piece of software (as was done
for graphics with OpenGL and DirectX9). This
way, all of the softsynths could use the
hardware by making the appropriate API calls.
I guess it is a matter of time and market size..

 

The strategy behind Native Instruments is to utilise the computer CPU avoiding dedicated hardware. The reasoning is that the CPUs are becomming more powerful every year making the dedicated hardware platforms redundant. Also the multi-CPU cards will get you more power.

So I think time and market size works in this direction, making it harder for the dedicated hardware plattforms to justify there existens by the year...

 

NI Strategy

You said:

The strategy behind Native Instruments is to utilise the computer CPU avoiding dedicated hardware. The reasoning is that the CPUs are becomming more powerful every year making the dedicated hardware platforms redundant. Also the multi-CPU cards will get you more power.
................


The CPU has traditionally fared poorly when compared with
the GPU, which has far exceeded the CPU's capability at least
with regard to graphics processing. Over the past 5 years,
GPU speed has run circles around the CPU, which has evolved
at a relative snail pace. Thus, for graphics, the
industry is moving in the opposite direction to NI's
reasoning. As a consumer, I hate to have to toss my CPU
every two years, when buying a new card would be relatively
inexpensive.

Now, the question becomes whether audio will follow in the
footsteps of graphics? Clearly, the two types of processing
have differences. I noticed a paper from INRIA on using
GPUs for audio processing...

 

Re: NI Strategy

Intresting... I would also like to have a dedicated music processor
just as for graphics. But a GPU everybody have a need for, music (MPU?) very few out of all PC/Mac users would have use for.... so I don't see it happening :-(

 

audio processing

Yes, games drive the GPU card companies to
compete with one another. The thing is: games
use sound in incredibly simplistic ways. What
about a game that synthesizes sounds and music
using softsynths, based on where the character
is located, what is happening to the character,
the mood, ambience.. This is the next step in
gaming, at least from where I sit. Playing
pre-recorded samples and soundtracks seems
antiquated.

 

A link to this website just went up on KVR, about a yet to be released application that allows real time audio data to be processed on a GPU. The posting on KVR expressed some doubt about whether or not this is for real.

Here's the link:
http://www.bionicfx.com/

 

Cobalt and others interested in this subject:

I just saw the same post at KVR's website and came here to paste it. Interesting that there are doubts as to its veracity. Figures. Hope the doubts are only fueled by the novelty of the technology. After all, there are still people who think to this day that the US Apollo moon expeditions were filmed in a secret Hollywood studio. (Uh-oh. I hope I haven't started a new thread within a thread.)