[Coco] Max baud rate for 68681 in MM/1?

[Dave Philipsen]

That implies a maximum data rate of 230,400 bps but I'm not so sure the 
1488/89 chips would go that fast.  They were the de facto chips of 
choice for RS232 back then but, as you say, most people never attempted 
anything much faster than 38,400.  Maybe someone can recall the 
practical limit on those drivers/receivers.

There's a strong possibility that you could replace the crystal with a 
7.3728 MHz crystal, replace the Signetics with Exar, and split out the 
TTL signals and connect them to a more modern RS232 transceiver and 
you'd be in business at least up to 460,800 bps.

Dave


On 1/31/2016 11:46 PM, Joel Ewy wrote:
> On 01/31/2016 11:33 PM, Dave Philipsen wrote:
>> In my opinion, the biggest factors in determining whether the MM/1 
>> could do this are:  1) What kind of clocks feed the 68C681 chip(s) in 
>> the MM/1.  2) What kind of driver/transceiver chips are downstream of 
>> the 68C681.
>>
> The crystal connected to it is 3.6864MHz, and the level converters are 
> 1488/1489.
>
> JCE
>
>> Perhaps if the proper clock is available to the 68C681 the RxD and 
>> TxD signals could be intercepted before the driver chips and you 
>> could coerce a pretty fast data rate out of it.  The RS232 interface 
>> is inherently weak when it comes to running at higher speeds.  You 
>> are much better off using RS422/485 or just straight TTL data if the 
>> distance is short.
>>
>>
>> Dave
>>
>>
>> On 1/31/2016 10:26 PM, Joel Ewy wrote:
>>> On 01/31/2016 08:44 PM, Kip Koon wrote:
>>>> Hi Joel,
>>>> This is a link for the MC68681 datasheet.
>>>> < 
>>>> http://pdf1.alldatasheet.com/datasheet-pdf/view/4164/MOTOROLA/MC68681.html>
>>>> The maximum speed is 1Mbps.  That should have been a lower case 'b'.
>>>
>>> That's kinda what I was thinking.  The question remains as to 
>>> whether the MM/1 can run the 68681 at a higher baud rate than 
>>> 34,800, and if so, is it likely to lose characters.  Part of why I'm 
>>> asking the question is that I'm wondering about the feasibility of 
>>> doing Drivewire over the 68681 serial ports. On the CoCo, 115Kbps is 
>>> achieved through great trickery and masking interrupts for however 
>>> long it's necessary.  This seems to work acceptably well in 
>>> NitrOS-9.  The ordinary floppy driver halted the CPU for a while, 
>>> waiting for a sector.  The major side-effect of this was that the 
>>> software clock would lose time, hence the desirability of no-halt 
>>> FDCs, like Disto's.  Drivewire largely makes up for this problem by 
>>> giving the CoCo access to the DW server's clock.  The MM/1 has a 
>>> DS1287 on the I/O board (with the 68681) which probably doesn't have 
>>> a very good battery anymore after 20 years.  I don't know how often 
>>> the clock module reads from the RTC -- is it only at boot time?
>>>
>>> My other thought is that there are some fairly inexpensive 
>>> serial<->ethernet adapters out there with hardware TCP/IP. But it 
>>> would sure be nice to have just a little more speed than 34,800bps.
>>>
>>> JCE
>>>
>>>> Kip Koon
>>>> computerdoc at sc.rr.com
>>>> http://www.cocopedia.com/wiki/index.php/Kip_Koon
>>>>
>>>>
>>>>> -----Original Message-----
>>>>> From: Coco [mailto:coco-bounces at maltedmedia.com] On Behalf Of Joel 
>>>>> Ewy
>>>>> Sent: Sunday, January 31, 2016 4:34 PM
>>>>> To: coco at maltedmedia.com
>>>>> Subject: [Coco] Max baud rate for 68681 in MM/1?
>>>>>
>>>>> I read ambiguous information about the 68681 on the MM/1's I/O board.
>>>>> Does anybody have a definitive word?  Page 2-5 of the MM/1 
>>>>> technical manual claims that it can be run at "1 megabytes per 
>>>>> second"
>>>>> (seriously 8M bits/sec?), but the table on page 9-2 and the text 
>>>>> on 9-2 state that the maximum baud rate is 34,800.  Is this a 
>>>>> pragmatic
>>>>> maximum based on the CPU's ability to service the DUART's on-chip 
>>>>> buffers without interrupt-masking trickery, is it the real hard
>>>>> maximum for the 68681, or is it a hard maximum imposed by the 
>>>>> MM/1's hardware?  Anybody have any insight?
>>>>>
>>>>> JCE
>>>>>
>>>>>
>>>>> -- 
>>>>> Coco mailing list
>>>>> Coco at maltedmedia.com
>>>>> https://pairlist5.pair.net/mailman/listinfo/coco
>>>>
>>>
>>>
>>
>>
>> ---
>> This email has been checked for viruses by Avast antivirus software.
>> https://www.avast.com/antivirus
>>
>>
>
>


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus