Serial Port Baud Rates faster than 115200 on PC
Makhan
Is it possible to obtain Baud Rates faster than 115200 on PCs? What
should I look for in the hardware that would give me a clue as to how
fast I can get?
I have a Dell with dual processor 3GHz, so I guess the system is
pretty new. I have tried opening the com1 port on hyper terminal with
240300 baud but it gives an error opening port message.
Any ideas or pointers will be very helpful.
Thanks in advance
Makhan
Randy
> Hello,
>
> Is it possible to obtain Baud Rates faster than 115200 on PCs? What
> should I look for in the hardware that would give me a clue as to how
> fast I can get?
>
> I have a Dell with dual processor 3GHz, so I guess the system is
> pretty new. I have tried opening the com1 port on hyper terminal with
> 240300 baud but it gives an error opening port message.
As I understand it, the maximum data rate of the 16550a UART is 115,200 BPS.
However, a 16C650 can be driven to 230,400 BPS. (Your number of 240300 seems to
have exchanged the 3 and the 4 from what it should be).
I think the only alternative to better hardware (like USB) would be something
like channel bonding several RS-232s. I have no idea whether any drivers exist
that can do this. Of course, you'd need them at both ends of the link.
A useful URL: http://www.beyondlogic.org/serial/serial.htm
Randy
--
Randy Crawford http://www.ruf.rice.edu/~rand rand AT rice DOT edu
Florian Liebig
Makhan schrieb:
> Hello,
>
> Is it possible to obtain Baud Rates faster than 115200 on PCs? What
> should I look for in the hardware that would give me a clue as to how
> fast I can get?
two days ago, i had to do with this problem, and i would like tell you
what i got out.
since the FIFO is used for simulating beaudot rates in a normal com port,
it has been designed to have a CLOCK-IN electronics circuit that gives
that FIFO a input rate of about 1.1MHZ (quartz/pll).
when programming in/output on standard COMx-port, the FIFO have the need
to use a scanrate that is 16 times higher than the Baudrate in order to
ensure error detection and so on.. So, in this case, 115200 is the only
maximum rate you can use in order to have no errors while transferring.
when needing higher io-rates, you have to do direct USB-1.2/2.0 register
programming, because they obtain higher baudrates at a quite useful
error-correction.
USB2.0 has been even more worked out.
You can get good information for USB-port programming since it has become
standard in ~1995&6.
Florian.
Philip Homburg
Sometimes.
>What
>should I look for in the hardware that would give me a clue as to how
>fast I can get?
Some implementations of the serial port allow you to switch to higher speeds in
non-standard ways.
I think that
http://lkml.org/lkml/2001/11/9/32
should give you the basic idea of what is available.
'mizu...@st.rim.or.jp' would be a good name to put into google to find
most of the stuff related to high speed serial ports.
I used to connect to my Zyxel TA-128 ISDN adapter at 460800 bps (these days
I have ADSL).
--
This Monk had first gone wrong when it was [...] cross-connected to a video
recorder that was watching eleven TV channels simultaneously, [...] The video
recorder only had to watch them, of course. It didn't have to believe them all
as well. This is why instruction manuals are so important -- Douglas Adams
Markus Humm
>
>
> two days ago, i had to do with this problem, and i would like tell you
> what i got out.
> since the FIFO is used for simulating beaudot rates in a normal com port,
> it has been designed to have a CLOCK-IN electronics circuit that gives
> that FIFO a input rate of about 1.1MHZ (quartz/pll).
> when programming in/output on standard COMx-port, the FIFO have the need
> to use a scanrate that is 16 times higher than the Baudrate in order to
> ensure error detection and so on.. So, in this case, 115200 is the only
> maximum rate you can use in order to have no errors while transferring.
Why? There exist special advanced UARTs which allow double or four times
mode, but afaik then the minimum possible rate is also double or fourth
than normal => these chips are not really standard, but some newer (1998
or so) PCs do really have these on board. Google around and you'll see
that. Windows even support them afaik, when present and properly detected.
Anbother suggestion for higher rates would be data compression, e.g. run
length or LZW. But since the original poster doesn't tell what he will
do with this serial port, we're clueless on what to advice...
Greetings
Markus
Florian Liebig
Markus Humm schrieb:
> [snip]
> >Why? There exist special advanced UARTs which allow double or four times
> mode, but afaik then the minimum possible rate is also double or fourth
> than normal => these chips are not really standard, but some newer (1998
> or so) PCs do really have these on board. Google around and you'll see
> that. Windows even support them afaik, when present and properly detected.
>
I never heard of such chips. And I never read something about this. Have you a
little guideline?
"googling around" is not my favourite.
Even the FIFO advanced UART doesn't provide standards on overclocking serial
input / output data streams but I have only documents defining such standards
up to 1999.
>
> Anbother suggestion for higher rates would be data compression, e.g. run
> length or LZW. But since the original poster doesn't tell what he will
> do with this serial port, we're clueless on what to advice...
as I read out from your letter, În this case; both transmitter & receiver -
should have hardware compression or how do you think it could be established ??
Florian.
Florian Liebig
Markus Humm schrieb:
> [snip]
> >Why? There exist special advanced UARTs which allow double or four times
> mode, but afaik then the minimum possible rate is also double or fourth
> than normal => these chips are not really standard, but some newer (1998
> or so) PCs do really have these on board. Google around and you'll see
> that. Windows even support them afaik, when present and properly detected.
>
I never heard of such chips. And I never read something about this. Have
you a
little guideline?
"googling around" is not my favourite.
Even the FIFO advanced UART doesn't provide standards on overclocking
serial
input / output data streams but I have only documents defining such
standards
up to 1999.
>
> Anbother suggestion for higher rates would be data compression, e.g. run
> length or LZW. But since the original poster doesn't tell what he will
> do with this serial port, we're clueless on what to advice...
as I read out from your letter, În this case; both transmitter &