FYI: SOY-chips for DSP

[puretube]

http://www.dtelectronics.com/focus_oki.htm

personally I like the phrase at the end of the first chapter:

..ass transistors are isolated...

but seriously: scroll down a little...

[puretube]

http://elm-chan.org/works/vp/report.html

[puretube]

nice example of what could have been done with that "n.c."  A9 pin
:lol:

http://www.stc.itri.org.tw/CCL/docs/IC_Catalog/datasheet/es56090.pdf

[Peter Snowberg]

Cool links!

Yes, I think it is very important to make sure your ass transistors are isolated. A separate ground plane is advisable in most instances.

nice example of what could have been done with that "n.c." A9 pin

Are you referring to the /MOTOR output?  :lol:

They must have been thinking of digital distortions when they wrote the following text about the CODEC...

....Designed for hands-free communications systems, it is equipped with grain and mute controls for data transmission and reception....

The MS81V10160 and MS81V06160 look VERY cool!

The ML2500 looks like a very neat chip, but I don't like the idea of cell failure after 10,000 writes.  :?


Thanks for the URLs!

[puretube]

the ISD chips go 100.000 rec/plays: enuff for answerphones...

[puretube]

http://www.willas-array.com/prod/products/directory/pdf/toshiba/AU_0001.PDF

http://www.asahi-kasei.co.jp/akm/en/product/ak7750/ek7750.pdf

http://www.asahi-kasei.co.jp/akm/en/product/ak7712a/ek7712a.pdf

[Peter Snowberg]

That AK7712A looks like a VERY cool chip!

Code Select Expand

3) External Memory Access (SRAM·Pseudo-SRAM·DRAM)
·Objective memory: 256k(32k ´ 8-bit),1M(128k ´ 8-bit) ´ 1 / SRAM
256k(32k ´ 8-bit),1M(128k ´ 8-bit) ´ 1 / Pseudo-SRAM
256k(64k ´ 4-bit),1M(256k ´ 4-bit) ´ 2 or ´ 1 / DRAM
(Half volume of 1M DRAM is used as 512k memory.)
·Treating bit length: 16-bit (24-bit is available, but double time is needed for access.)
·The number of times to access:
SRAM, 256k Pseudo-SRAM ; 76 at 384fs
: DRAM, 1M Pseudo-SRAM ; 64 at 384fs (32 at one DRAM)
: SRAM, 256k Pseudo-SRAM ; 51 at 256fs
: DRAM, 1M Pseudo-SRAM ; 42 at 256fs (21 at one DRAM)
·Memory access time:
less than 100nsec
·Maximum address length:
65535 sampling times (at 1M SRAM)
2.048sec at 32kHz, 1.486sec at 44.1kHz, 1.365sec at 48kHz

SRAM, PSRAM, & DRAM.... very nice flexability when you consider that A/D and D/A is also built in.

Bank switching could easily be used to extend the SRAM or PSRAM. Hehehehe.... It's 1981 all over again! :mrgreen:

[StephenGiles]

Ah.......youv'e lost me there!
Stephen

[Peter Snowberg]

Stephen, if you have questions.... the forum provides answers. Just start asking!

I think we're going to see more and more of this type of info around here so now is a great time to start answering questions about this stuff.

[puretube]

from the website, it says the 7712 will soon be discontinued for the future,
but being run in "mass production" at the moment....   gotta run!

:lol: soon come: NOS DSP chips ...  :lol:

[puretube]

http://documentation.renesas.com/eng/products/assp/rjj01f0001_audio.pdf

(7MB !)
:wink:

[puretube]

just to put it into some drawer:
http://focus.ti.com/lit/an/slaa123/slaa123.pdf

[zachary vex]

the ISD chips go 100.000 rec/plays: enuff for answerphones...

yes, but what happens at rec number 100,000?  one cell goes bad and starts making a "tick" sound as that bad cell gets played?  i have replaced one 1420 chip so far (lo-fi loop junky) because of a "tick."

[puretube]

who`ll be the first to use a "Crossbar Latch" for distortion???

http://mms.itweb.co.za/sections/computing/2005/0502020915.asp?S=Reuters&A=REU&O=FRGN

of course, the "CBL-Dist" should come on a printed paper/foil PCB,
and have a thinfilm "OLED" display,
and, most of all: work in its linear region...

[puretube]

ADAV400

[SeanCostello]

A new SigmaDSP chip was announced at AES. Not as much delay memory, but it also has built in stereo ADC and quad DAC, plus auxillary ADC to hook 4 knobs up.

The nice things about the SigmaDSPs are that the development kits are very nice, come with some great graphic software for dragging and dropping DSP blocks, and the price is right. Also, the chips themselves seem humanly solderable.

Sean Costello (who works for Analog Devices, but not in the SigmaDSP line - I work on high priced software for super fast DSPs that are in no way humanly solderable - well, a few crazy mofos can solder them, but I sure couldn't)

[A.S.P.]

signal-wizard 2 filter-dsp

[A.S.P.]

http://products.zarlink.com/product_profiles/MT8816.htm

[TELEFUNKON]

interesting new device for replacing opto-couplers:
digital isolator:
http://www.analog.com/en/content/0%2C2886%2C767%255F827%255F95767%2C00.html

[puretube]

P-RAM
perfect?