- Messages sorted by: [ date ][ thread ][ subject ][ author ]
- Next message: David Fox: "Re: Suggestions requested"
- Previous message: Susan Fulton: "Re: Suggestions requested"
- In reply to: Joseph S. Wisniewski: "Re: PCI Sound Card Advantage?"
that the AWE64-Gold (NOT the AWE64) had less noise by a factor of 10 or
so. Is this not the case?
-Chip
On Thu, 12 Feb 1998, Joseph S. Wisniewski wrote:
> Every additional function you make the card perform adds to it's noise
> (or makes removing the noise even more complicated). Take a look at an
> AWE64 or an AWE32 and you will see that they have everything an old SB16
> has (Codec chip, FM synth chip, interface chip, plus the additional
> wavetable synthesis chip and bank of DRAM). Adding more stuff means you
> do three things:
>
> 1) you add more components, which generates more electrical noise. A
> wavetable synthesis chip (or daughterboard) and it's associated RAM
> generates much more noise than everything else on the soundcard put
> together.
>
> 2) you make the board more crowded and more expensive. The soundcard
> designer might be tempted to remove (or downsize) a component or two to
> save space and cost. And, if the components that get removed or reduced
> are the ones that make the wave input a bit quieter, then the voice
> users loose, while the game users gain.
>
> 3) you make the layout much worse. Now, there is less space for ground
> planes, clean signal routing, and isolated areas of board for the analog
> circuitry.
>
> So, at the same time you add a new "noisemaker" to the soundcard, you
> remove components that protect you from the noise, and change the layout
> so it's harder to get away from the noise.
>
> In short, an expensive wavetable card is not a solution to speech
> recognition problems, because adding that expensive wavetable music
> synthesizer actually cheapens and degrades the part of the card we need
> for voice.
>
> The solution is either to spring for a card that is designed to do one
> thing really well (Fiji or Card-D) instead of a card that is designed to
> do many things but not as well (AWE 64).
>
> Or, ou can look for cheap and simple sound cards, and hope that the stuff
> that makes them cheap and simple is the removal of noisemaking wavetable
> chips and such, and not the removal of necessary noise reduction
> components. But this is a totally hit-and-miss proposition. Creative labs
> changes their board layouts and compnonet selections constantly, so you
> never know if this weeks SB-16 is quieter or noisier than last weeks.
> No-name soundcards are even worse for unexpected changes and radical
> quality differences from card to card.
>
> On Wed, 11 Feb 1998, Judy Jackson wrote:
>
> > I have found that the sound quality of the SoundBlaster AWE64 cards can be
> > fairly noisy. On two systems I've configured, I've had to replace these
> > supposedly high end cards with old style SoundBlaster 16's in order to get
> > good recognition with the Dragon. Nothing was trying to play any MIDI at
> > the time. I am not suggesting that the bus speed is the problem; instead
> > I'm suggesting that the quality of the 64's is not as good as the older cards.
>
> Joseph S. Wisniewski | Views expressed are my own, and don't reflect
> Ford Motor Company | those of the Ford Motor Co. or affiliates.
> Project Sapphire | LeMans, Daytona, Bonneville, and Sebring are
> jwisniew@@ford.com | just races, won by people driving Ford cars!
>
>
>
-------------------------------------------------------------------------------
Chip Sumner
sumner@@cco.caltech.edu
-------------------------------------------------------------------------------
- Next message: David Fox: "Re: Suggestions requested"
- Previous message: Susan Fulton: "Re: Suggestions requested"
- In reply to: Joseph S. Wisniewski: "Re: PCI Sound Card Advantage?"
 |  |