Work at SourceForge, help us to make it a better place! We have an immediate need for a Support Technician in our San Francisco or Denver office.

Close

#2 ES1878 ISA sound chip is supposed to be supported but doesn't work well

v1.0 (example)
open
nobody
None
5
2013-12-12
2013-12-12
dos386
No
BUG: ES1878 ISA sound chip is supposed to be supported but doesn't work well.

"README.TXT" wrote :

  • ESS 688,1688,1868,1869,1878,1879,1887,1888 chips (?)
  • SB Pro or compatible (DSP 1.5, 2.x, 3.x)(a lot of cards)

"SNDDARDS.TXT" wrote :

ESS 688,868,869,878,887,etc. | 4k-48k | 16 | 2 | BLASTER=A220 I5 D1 H5
-----------------------------+----------+------+----+----------------------
SB 1.0,2.0,2.01,Pro | 4k-22k | 8 | 1,2| BLASTER=A220 I5 D1 T4

Slightly inconsistent and buggy ?

Windaube reports (2 ES1878 devices !!!) :

> ES1878 Control Interface (WDM)
> * port $0250 to $0257

> ES1878 Plug and Play AudioDrive (WDM)
> * port $0220 to $022F
> * port $0388 to $038B
> * port $0300 to $0301
> * IRQ 5
> * DMA 5
> * DMA 1

"BLASTER.EXE -v" reports :

No legacy devices found.
Found SB compatible DSP. checking resources...
Waiting for IRQs...ok.
Testing DMA 1...timeout.
Testing DMA 0...timeout.
Testing DMA 3...timeout.
SoundBlaster Pro or compatible on port address 220H found. DSP version 3.1
Sorry, no Sound Blaster compatible sound chip found.

I am NOT smart from this report :-(

As an excellent side effect, "BLASTER.EXE" trashes the system so I can't
launch any other program (including "BLASTER.EXE" itself) anymore. No way
to fix apart from cutting the power cable. :-( Note that MPXPLAY doesn't
suffer from this BUG, it doesn't trash anything (apart from my ears ...).

"MPXPLAY -sct" :

> Autodetecting/testing available outputs/soundcards, please wait...
> ESS : soundcard found (type=688B) : SET BLASTER=A220 I5 D5 H255
> SBP : SB-pro card found : SET BLASTER=A220 I0 D5 T4
> Autodetecting finished... Exiting...

- ESS found just at begin, few seconds later heavy "doubleclick" from
  speakers, then delay / hanging cca 60 s, then another single "click",
  then delay / hanging cca 30 s, finally reporting "SBP" and exit

"MPXPLAY -scs SBP" :

  • MPXPLAY start takes 6 s, afer 3 s heavy "doubleclick"
  • same heavy "doubleclick" when playing starts
  • playing runs at double speed (!!!), with reducing to "50%" it's almost OK
  • well not fully OK, there is noise, especially remarkable with audio
    files recorded with low volume

    "MPXPLAY -scs ESS" :

    • MPXPLAY start takes 3 s
    • playing doesn't work, horrible noise with maximum volume,
      volume control doesn't work either
    • well it somewhat works if I set the volume to 1 (ONE !!!), then it plays,
      sound is distorted (less with audio files recorded with low volume)
      and double speed too, "50%" helps

Speculations :

There seems to be a sample format problem: LE-vs-BE or
8bit-vs-16-bit or mono-vs-stereo.

For both "-scs SBP" and "-scs ESS" playing runs at double speed.

For "-scs SBP" the chip gets only the higher 8 bits of the 16 bit
samples and this causes the weak quantization noise ?

For "-scs ESS" the chip gets only the lower 8 bits of the 16 bit
samples and this causes the heavy volume control insensitive noise,
amd makes it partially working with volume set to 1, when higher
8 bits are ZERO for most samples sent to the chip.

Discussion

  • dos386
    dos386
    2013-12-12

    "MPXPLAY -scs SBP" :

    • MPXPLAY start takes 6 s, afer 3 s heavy "doubleclick"
    • same heavy "doubleclick" when playing starts
    • playing runs at double speed (!!!), with reducing to "50%" it's almost OK
    • well not fully OK, there is noise, especially remarkable with audio
      files recorded with low volume

    ~ ~ ~ ~ ~ ~ ~ ~

    "MPXPLAY -scs ESS" :

    • MPXPLAY start takes 3 s
    • playing doesn't work, horrible noise with maximum volume,
      volume control doesn't work either
    • well it somewhat works if I set the volume to 1 (ONE !!!), then it plays,
      sound is distorted (less with audio files recorded with low volume)
      and double speed too, "50%" helps