Truncated match.
PICList
Thread
'SPI vs I2C'
1997\04\04@103756
by
Antonio Almeida
What's the difference between using an SPI or I2c EEPROM with the
PICs (ignoring the fact that one needs 3 and the other needs 2 wires)?
Thanks.
1997\04\04@162018
by
Sarunas Cepulis
Antonio Almeida wrote:
>
> What's the difference between using an SPI or I2c EEPROM with the
> PICs (ignoring the fact that one needs 3 and the other needs 2 wires)?
>
> Thanks.
Hi,
SPI works on higher speed (up to 1Mhz),I2C - 100Khz.
SPI simplest exchange protocol (simple software).
Saras.
1997\04\04@163058
by
Sarunas Cepulis
Antonio Almeida wrote:
>
> What's the difference between using an SPI or I2c EEPROM with the
> PICs (ignoring the fact that one needs 3 and the other needs 2 wires)?
>
> Thanks.
Hi,
SPI works on highest speed (up to 1Mhz),I2c - 100Khz.
SPI simplest exchange protocol (simple software).
I2C can handle many devices on same wires (with diferent slave
addresses),SPI needs individual CS lines for eatch device.
Saras.
1997\04\04@212937
by
Robert Lunn
> What's the difference between using an SPI or I2c EEPROM with the
> PICs (ignoring the fact that one needs 3 and the other needs 2 wires)?
SPI is normally rated up to 2MHz.
IIC is normally rated up to 100kHz (though most manufacturers
also supply 'high speed' devices rated up to 400kHz).
Addressing of multiple devices is also different. SPI addresses
each device using a discrete 'chip select' line, so you need as
many CS's as you have devices. IIC, on the other hand, embeds
the addressing in the data stream so _no_ additional lines are
needed.
Given that the high data rates are rarely needed, my own pref-
erence is for IIC.
___Bob
1997\04\05@092919
by
Luiz Marques
IMHO, for low port counts I2C is preferable but for hi-speed aquisition (eg.
(eg. A/D) or low code space and overhead SPI is preferable.
In SPI EEPROMs you can read the whole device just inssuing clks to SPI,
quick and simple.
Luiz Marques
1997\04\06@021812
by
Harold Hallikainen
And.. it appears no one makes SPI RAMs, just EEPROMs, right? I
searched all week and finally went with the Dallas RAMPort (thanks to the
list!) for more memory.
Harold
1997\04\06@175453
by
Martin Lund
Harris do 128 and 256 byte SPI RAM - CDP68HC68R1e and CDP68HCR2E respectively.
-----Original Message-----
From: Harold Hallikainen [SMTP:spam_OUTharoldhallikainenTakeThisOuT
JUNO.COM]
Sent: 06 April 1997 08:16
To: .....PICLISTKILLspam
@spam@MITVMA.MIT.EDU
Subject: Re: SPI vs I2C
And.. it appears no one makes SPI RAMs, just EEPROMs, right? I
searched all week and finally went with the Dallas RAMPort (thanks to the
list!) for more memory.
Harold
Attachment converted: wonderland:WINMAIL.DAT 1 (????/----) (0000D942)
1997\04\06@185210
by
Sarunas Cepulis
Martin Lund wrote:
>
> Harris do 128 and 256 byte SPI RAM - CDP68HC68R1e and CDP68HCR2E respectively.
>
> {Original Message removed}
1997\04\07@103737
by
Andy Kunz
> And.. it appears no one makes SPI RAMs, just EEPROMs, right? I
>searched all week and finally went with the Dallas RAMPort (thanks to the
>list!) for more memory.
I have a guy coming in from RAMTRON. It's RAM speed, EEPROM
non-volatility, and pin compatible with various EEPROMs.
http://www.ramtron.com
Andy
==================================================================
Andy Kunz - Montana Design - 409 S 6th St - Phillipsburg, NJ 08865
Hardware & Software for Industry & R/C Hobbies
"Go fast, turn right, and keep the wet side down!"
==================================================================
1997\04\07@103751
by
Andy Kunz
>Try PHILIPS
>http://www.semiconductors.philips.com/
>
> Saras.
Phillips is a very poor choice for USA products. They have availability
problems. Worse the Motorola, if that's possible.
Andy
==================================================================
Andy Kunz - Montana Design - 409 S 6th St - Phillipsburg, NJ 08865
Hardware & Software for Industry & R/C Hobbies
"Go fast, turn right, and keep the wet side down!"
==================================================================
1997\04\07@141505
by
Luiz Marques
>> And.. it appears no one makes SPI RAMs, just EEPROMs, right? I
>> searched all week and finally went with the Dallas RAMPort (thanks to
>> the list!) for more memory.
> Harold
> Harris do 128 and 256 byte SPI RAM - CDP68HC68R1e and CDP68HCR2E >
respectively.
Harold, I'm designing a product with a MCU with SPI and up to 512K bytes
of non-volatile RAM. The RAM is from Dallas.
I send the address by SPI to two S/R and then to address lines of RAM.
Additional addressing lines (3) is provided direcly by MCU.
The data path use a 8 bit port (shared with LCD module) to read and write
data to RAM but you can do this also by SPI.
My two cents (if worth to someone else)
Luiz Marques
1997\04\07@225826
by
)
|
I've seen this URL posted numerous time to PICList. When I go there all
I get is an advertisement for an ISP. Anyone else having this problem?
Frank Richterkessing
Experimental Methods Engineer
GE Appliances
FRANK.RICHTERKESSING
KILLspamAPPL.GE.COM
{Quote hidden}> ----------
> From: Andy Kunz[SMTP:
.....montanaKILLspam
.....FAST.NET]
> Sent: Monday, April 07, 1997 10:21AM
> To:
EraseMEPICLISTspam_OUT
TakeThisOuTMITVMA.MIT.EDU
> Subject: Re: SPI vs I2C
>
> > And.. it appears no one makes SPI RAMs, just EEPROMs, right?
> I
> >searched all week and finally went with the Dallas RAMPort (thanks to
> the
> >list!) for more memory.
>
> I have a guy coming in from RAMTRON. It's RAM speed, EEPROM
> non-volatility, and pin compatible with various EEPROMs.
>
>
http://www.ramtron.com
>
> Andy
>
> ==================================================================
> Andy Kunz - Montana Design - 409 S 6th St - Phillipsburg, NJ 08865
> Hardware & Software for Industry & R/C Hobbies
> "Go fast, turn right, and keep the wet side down!"
> ==================================================================
>
1997\04\07@235858
by
Robert Lunn
>I've seen this URL posted numerous time to PICList. When I go there all
>I get is an advertisement for an ISP. Anyone else having this problem?
>>
>> http://www.ramtron.com
Well, yeah, I got the same message...
But poke about a bit and the site tells you what to do (the
essence of computing is experimentation, ie: poke about a bit).
Try http://www.ramtron.com/ramtron
By the way, you need a frames capable browser!
___Bob
1997\04\11@223331
by
Harold Hallikainen
On Mon, 7 Apr 1997 15:12:51 -0700 Luiz Marques <omarques
spam_OUTUFBA.BR> writes:
>Harold, I'm designing a product with a MCU with SPI and up to 512K
>bytes
>of non-volatile RAM. The RAM is from Dallas.
>I send the address by SPI to two S/R and then to address lines of RAM.
>Additional addressing lines (3) is provided direcly by MCU.
>The data path use a 8 bit port (shared with LCD module) to read and
>write
>data to RAM but you can do this also by SPI.
This looks like a good approach for the amount of RAM you need.
Since I only need 1K to 2K bytes or RAM, it appears a lower chip count
approach is to use the Dallas RamPort (2 Kbytes). I'm surprised there
aren't more larger SPI RAMs out there. I'd love to have a few Kbytes in
an 8 pin package!
Harold
1997\04\11@231246
by
John Payson
|
This looks like a good approach for the amount of RAM you need.
Since I only need 1K to 2K bytes or RAM, it appears a lower chip count
approach is to use the Dallas RamPort (2 Kbytes). I'm surprised there
aren't more larger SPI RAMs out there. I'd love to have a few Kbytes in
an 8 pin package!
I would like to see someone (anyone at Microchip listening?)
produce a RAM chip with a synchronous serial input, the clock
wire of which was designed to run off a small micro's XTAL
signal directly (i.e. with an adjustible clock divider so the
CPU could simply input or output a bit every, say, 12 oscil-
lator clocks. Such a part would not be terribly hard to de-
sign, but it would save an I/O pin (as compared with I2C-style)
or two (compared with SPI) and it would be faster than either
of those other protocols as well.
Better still would be if Microchip could produce a PIC with a
feature to input/output data on such a device at a rate of one
bit per raw clock [each byte being sent as probably a 12-bit
datagram]. The hardware would not be overly complex, but it
would be able to transfer data almost as fast as a "wide" bus
but at a fraction of the cost in pins and real estate.
What does anyone think of this idea?
More... (looser matching)
- Last day of these posts
- In 1997
, 1998 only
- Today
- New search...