Re: [linux-audio-user] Recommendatio wanted

New Message Reply About this list Date view Thread view Subject view Author view Other groups

Subject: Re: [linux-audio-user] Recommendatio wanted
From: Joerg Anders (j.anders_AT_informatik.tu-chemnitz.de)
Date: Mon Dec 01 2003 - 13:31:11 EET


On Mon, 1 Dec 2003, Chris Cannam wrote:

> On Monday 01 Dec 2003 10:45 am, Joerg Anders wrote:
> > Do you really get different results
> > with FluidR3 soundfont and FluidSynth ?
>
> Well, one problem is that I don't think you've ever actually said
> which patch you're using, except that it's "strings".
... Ok, my mistake!
>
> String Ensemble 1 (49): pretty much the same as you. I'm guessing
> this is what you're using.

Yes, that's what I'm talking about!

>
> String Ensemble 2 (50): a nice smooth string wash, no attack.

Yes, but these are slow strings. They always come too late.
>
> Synth Strings 1 (51): a smooth sound with some pretty bizarre tuning.
>
> Synth Strings 2 (52): similar but more so (reminds me of that
> out-of-tune synth in Joy Division's The Eternal).

The synth strings are unusable in most cases. Only patch 49
behaves and sounds like real strings.

> So you see, I've generally assumed that most of these fonts were just
> designed to have the string ensemble 1 sound a bit more vigorous for
> faster passages and the string ensemble 2 be the friendly wash. That
> is actually quite a useful distinction. It hadn't occurred to me
> that other synths would play it differently.

The string ensemble 2 always comes too late. If you want an accompainmant
you cannot use string ensemble 2.

>
> Given that this isn't exhibited with all string patches, it really
> might be worth taking a bit more time to check that the hardware
> synth is actually the correct one...

... TiMidity gives also smooth strings. The same happens with AWE32/64.
On M$-Windows are also non-attacking strings. Believe me: The
soundfont creator certainly had no attacking strings in mind!

> Just because you like it better
> doesn't necessarily mean it is (although it seems probable). I

It is certainly a FluidSynth error.

> certainly agree with you that the difference is rather strange.
>

My question is: What is the conclusion ? Is there any FluidSynth
programmer on this list ?

If not, that means: There is no soft synth which creates a
good bigband sound. The hardware sound is the only option.
But the Audigy2 driver produces gaps:

 http://rnvs.informatik.tu-chemnitz.de/DROP/gap.html

And apparently nobody will fix this. Thus, I'll throw away my
Audigy2 soundcard and I'll follow the recommendation of

On Wed, 26 Nov 2003, Hartmut Z Noack wrote:

> If SB-Live does not work sufficiently, why dont trying to use Terratecs
> EWX24/96? The card does not have Hardware SF-Support but you can load
> SFs into RAM (if you have enough...) and it works perfectly well for me.
>
> Conclusion: try Delta 66 or some similar Pro-Gear or try it with a
> ICE-chipbased card like terratecs muse, whitch are verywell supported by
> alsa and can play SF from RAM.
>

The whole dilemma comes from: The AWE64 is an ISA card.
Otherwise, I'd further use AWE64 ... :-((

-- 
J.Anders, Chemnitz, GERMANY (ja_AT_informatik.tu-chemnitz.de)


New Message Reply About this list Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b28 : Mon Dec 01 2003 - 13:28:56 EET