szhu@tasha.poly.edu (A1 shu bin zhu (cs) ) (02/27/91)
Is there a way to set the serial port above 19200 baud on the ST? The MFP chip can handle upto 56.2K baud, there must be a way. Anyone..
carter@cat27.cs.wisc.edu (Gregory Carter) (02/27/91)
The MFP chip only has a clock rate of 4Mhz. I wonder if it is is possible to go as fast as 56K??? Any confirmations on this? I wonder if the possibility is assuming minimum reliability at that speed? ??? Gregory
hase@netmbx.UUCP (Hartmut Semken) (03/03/91)
szhu@tasha.poly.edu (A1 shu bin zhu (cs) ) writes: >Is there a way to set the serial port above 19200 baud on the ST? >The MFP chip can handle upto 56.2K baud, there must be a way. There is no way to do it in software: One of the timers in the MFP is used to generate the send and receive clock for the serial interface. Using the standard clock frequency and a pre-divider by four (this is necessary for async. operation) it can generate clocks up to 19200 Baud. The only way is to cut the send clock and receive clock input pins of the MFP an feed it with a new clock (from extra hardware). I planned to try it, but never found the time to do it... Try it out. You can still reverse the little hack and up with nothing more but some experience :) hase -- Hartmut Semken, Lupsteiner Weg 67, 1000 Berlin 37 hase@netmbx.UUCP Hi! (Zaphod Beeblebrox)
hyc@math.lsa.umich.edu (Howard Chu) (03/03/91)
In article <2030@netmbx.UUCP> hase@netmbx.UUCP (Hartmut Semken) writes: >szhu@tasha.poly.edu (A1 shu bin zhu (cs) ) writes: >>Is there a way to set the serial port above 19200 baud on the ST? >>The MFP chip can handle upto 56.2K baud, there must be a way. The Motorola Spec says 64kbps max asynch, 1Mbps synch. >There is no way to do it in software: I hate to disagree with you, but you're wrong about this... >One of the timers in the MFP is used to generate the send and receive >clock for the serial interface. Using the standard clock frequency and a >pre-divider by four (this is necessary for async. operation) it can >generate clocks up to 19200 Baud. Actually the predivide is by 1 or by 16, with div 16 for async and div 1 for synchronous modes. However, even tho div 1 is meant for synchronous comm., the MFP will operate asynchronously with this setting. Unfortunately, the crystal frequency is so bizarre that 38400 is the highest "standard" bit rate you can coax out of it. (2.4576MHz.) You can set the counter for 76800, but I've not tried it, I'm sure it would fail miserably. I got a lot of noise running at 38400 between my ST and a NeXT box. Of course, I haven't tried between two STs yet. Someone else commented to me (sorry don't remember, this was a while back) that there's no way you can get a consistently good signal using this trick, because the MFP is struggling so hard to detect the pulses at these speeds. So, if you're really serious about it, you're probably better off doing the hardware mod. -- -- Howard Chu @ University of Michigan Flame all you want - we'll take more.
Shervin.Shahrebani.Of.250/744@p0.f744.n250.z1.fidonet.org (Shervin Shahrebani Of 250/744) (03/06/91)
What is the MFP chip? I do have a desk acc from Germany that claims to let you go way beyond 19200 baud. S.S.
hase@netmbx.UUCP (Hartmut Semken) (03/09/91)
hyc@math.lsa.umich.edu (Howard Chu) writes: >Actually the predivide is by 1 or by 16, with div 16 for async and div 1 for >synchronous modes. However, even tho div 1 is meant for synchronous comm., >the MFP will operate asynchronously with this setting. Unfortunately, the >crystal frequency is so bizarre that 38400 is the highest "standard" bit >rate you can coax out of it. (2.4576MHz.) You can set the counter for 76800, >but I've not tried it, I'm sure it would fail miserably. I got a lot of noise >running at 38400 between my ST and a NeXT box. Of course, I haven't tried >between two STs yet. I tried it once between my St and my 386 box with Telix. Tried again between my ST and my SE/30 at 38400 Baud. Did not work. Too much noise. The hardware mod should work fine, cause the clock divided by 16 is more stable... # Out of time hase -- Hartmut Semken, Lupsteiner Weg 67, 1000 Berlin 37 hase@netmbx.UUCP Hi! (Zaphod Beeblebrox)