search for: dm642

I'm trying to make a design decision between a TI 6416 or DM642 (fixed point) and 6713 (floating point) platform. The application is a 32 channel speech encoder. (CBR only, 8khz, 8kbps) To get a feel for the computational load, I am running 1 second (50 frames) of voice through the encoder. My profile of the 6416 indicates I'm at 27.4M cycles/channel. I...

<speex-dev at xiph.org> Hi all, I'm a begginer with DSP and i need your help and suggestions. I'm trying to use the Speex and a DSP DM642 to implement a solution with voice! I'm using the 1.2 beta 3 distribution and the TI's Code Composer Studio v3.1 simulator. I coosed speex_C64_test.pjt and modified the speex_C64_test.cmd to only use the DM642 external RAM memory. I'm not sure about the input file, the speex file and th...

...s I'm at 27.4M cycles/channel. I need to get > below 720Mhz/32 channels = 22.5M cycles per channel. I did a little work on > inner_prod() and normalize16() and I'm confident I can get 32 channels by > optimizing 5 or 6 functions. I expect these numbers to translate over the > DM642. have you tried defining PRECISION16? That should reduce the computation cost. > A lower cost option would be to use a floating point 6713. I thought that a > 300Mhz floating point would come out even or ahead in an encoding > comparison. Instead of the 300M/32=9.3M cycles per channel...

Hello! I'v downloaded speex lib 1.1.11.1. I am trying to port speex lib to Blackfin processor. I am using VisualDSP++ 4.0. If I am compiling source codes with using floating point everything ok. When I am compiling with FIXED_POINT defined everything's ok and code works about two times faster. But when I am defining BFIN_ASM I am getting several compiling errors in Blackfin assembler

...s I'm at 27.4M cycles/channel. I need to get > below 720Mhz/32 channels = 22.5M cycles per channel. I did a little work on > inner_prod() and normalize16() and I'm confident I can get 32 channels by > optimizing 5 or 6 functions. I expect these numbers to translate over the > DM642. have you tried defining PRECISION16? That should reduce the computation cost. > A lower cost option would be to use a floating point 6713. I thought that a > 300Mhz floating point would come out even or ahead in an encoding > comparison. Instead of the 300M/32=9.3M cycles per channel...

Hi, My company just recently bought a TI DaVinci development board and I'm trying to port Speex to run on it. The TI DaVinci chip has an ARM926 processor and a TMS320C64+ DSP in it. I was able to get Speex somewhat working on a TMS320VC5510 about a year ago but I was wondering is there anything special I need to do to get Speex working on the TMS320C64+ DSP in the DaVinci chip. I

...s I'm at 27.4M cycles/channel. I need to get > below 720Mhz/32 channels = 22.5M cycles per channel. I did a little work on > inner_prod() and normalize16() and I'm confident I can get 32 channels by > optimizing 5 or 6 functions. I expect these numbers to translate over the > DM642. have you tried defining PRECISION16? That should reduce the computation cost. > A lower cost option would be to use a floating point 6713. I thought that a > 300Mhz floating point would come out even or ahead in an encoding > comparison. Instead of the 300M/32=9.3M cycles per channel...

...s I'm at 27.4M cycles/channel. I need to get > below 720Mhz/32 channels = 22.5M cycles per channel. I did a little work on > inner_prod() and normalize16() and I'm confident I can get 32 channels by > optimizing 5 or 6 functions. I expect these numbers to translate over the > DM642. have you tried defining PRECISION16? That should reduce the computation cost. > A lower cost option would be to use a floating point 6713. I thought that a > 300Mhz floating point would come out even or ahead in an encoding > comparison. Instead of the 300M/32=9.3M cycles per channel...

...----- Original Message ---- From: Jerry Trantow <jtrantow@ieee.org> To: ndno72-speex@yahoo.com; speex-dev@xiph.org Sent: Tuesday, December 5, 2006 1:55:13 PM Subject: RE: [Speex-dev] Speex support for TI DaVinci processor Do you need to encode and/or decode? I did some work optimizing the DM642 speex 1.1.11.1 encoder earlier this year. I was able to reduce the narrow band MIPS from approx 30 to 16 using some of the #pragmas and optimizing the functions (Following the BlackFin example.) The project I was working on has been put on hold so I haven?t looked at the more recent changes. If...

...erry Trantow <jtrantow@ieee.org> > To: ndno72-speex@yahoo.com; speex-dev@xiph.org Sent: Tuesday, > December 5, 2006 1:55:13 PM Subject: RE: [Speex-dev] Speex support > for TI DaVinci processor > > > Do you need to encode and/or decode? I did some work optimizing the > DM642 speex 1.1.11.1 encoder earlier this year. I was able to reduce > the narrow band MIPS from approx 30 to 16 using some of the #pragmas > and optimizing the functions (Following the BlackFin example.) The > project I was working on has been put on hold so I haven?t looked at > the more...

Do you need to encode and/or decode? I did some work optimizing the DM642 speex 1.1.11.1 encoder earlier this year. I was able to reduce the narrow band MIPS from approx 30 to 16 using some of the #pragmas and optimizing the functions (Following the BlackFin example.) The project I was working on has been put on hold so I haven't looked at the more recent changes....

...@ieee.org> >> To: ndno72-speex@yahoo.com; speex-dev@xiph.org Sent: Tuesday, >> December 5, 2006 1:55:13 PM Subject: RE: [Speex-dev] Speex support >> for TI DaVinci processor >> >> >> Do you need to encode and/or decode? I did some work optimizing the >> DM642 speex 1.1.11.1 encoder earlier this year. I was able to reduce >> the narrow band MIPS from approx 30 to 16 using some of the #pragmas >> and optimizing the functions (Following the BlackFin example.) The >> project I was working on has been put on hold so I haven?t looked at &...