----- Original Message ----- From: Ellis Breen <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Friday, September 29, 2000 3:06 AM Subject: [linux-audio-dev] SAOL/Quasimodo > >wait - how can a firewire device access host RAM other than through > >the PCI bus ? is there some alternate route to RAM ? > Sorry, someone gave me some misleading figures on Firewire speed... turns > out it's only about 40% the speed of PCI66 (1Gbs) thus far... although they > are > attempting to match that. Still, we need faster transfer busses for > auxiliary > hardware - AGP manages it... > > Someone suggested Creamw@re use the AGP4x slot and leave the > PCI slots to the graphics cards. It seems a shame to waste all this upcoming > RAM bandwidth (4Gbs for the new Athlon motherboards) on GPUs/x86s... > maybe a 16-way Crusoe/PPC/Thumb10 (running PPC and ARMLinux > respectively of course ;-) ) setup would be worthy though... > > > >Are potential optimisations lost in the translation to C by SFront/SFX? > > >Could one spec up a superior intermediate API for SAOL to machine code > .. > > There's some wins possible in that direction (for a micro-optimization > > example, many table operations could go better with ASM-optimized rounding > .. > > > yikes. please don't use SAOL. saol was an excellent reformulation of > > > Csound, but as a general purpose synthesis/fx language, i consider it > > > to be weak and to have a number of specific defects. > . > > Let's take Paul's judgement of SAOL vs Quasimodo as a sound language > > as fact for a moment -- I have my own opinions, but its irrelevant to > .. > > It may seem like a minor quibble, but its really important to > > understand that Quasimodo is *NOT* a sound language. Quasimodo > > dynamically loads the parsers+compilers for the sound languages used > > by the modules it runs, and does not contain *any* built-in language > > > > But on the other hand, there are lots of other systems where the > > > installed base of MP3 as a standard overwhelms efficiency gains you'd > > > get by moving to a technically superior format. > > > This is true, without a doubt. What's sad for me here is that MP3 is > > not a programming language, and when MP4-SA is used in the vast > > majority of the systems where it will be deployed, it will also not > > really be a programming language in the sense that no human will sit > > down and express algorithms with it. > > Paul, I checked out Quasimodo and it looks very interesting. I actually > had Lexx/Flex and Yacc/Bison in mind when I was considering the proposed > intermediate API for SAOL compilation. It strikes me that one could define > a superior Lexx grammar and Yacc implementation for SAOL/CSound than > that of the C language... Is it not true that optimisation at an abstract > level > dictates implementation at a machine level? > > So maybe Quasimodo would be a suitable umbrella for both my doctyped XML > idea and my > intermediate API? I have no desire to re-invent the wheel (unless "Wheel MK > 2" is significantly better). If you (or I) could create a sequencer module > for Quasimodo, then we could encapsulate everything neatly. I loathe nothing > more than unnecessarily replicated effort so the more one can control and > setup > everything in the same place, the better. > > > >
