dw@rocksanne.UUCP (Don Wegeng) (01/15/87)
Awhile back someone mentioned that a clock driver had been written to support the 6300 realtime clock under PC-DOS 3.1. I'd like to ahold of this beast, if it is available. Any details that anyone can supply will be most appric- iated! Thanks, /Don -- "I smell catastrophe upon the wind..." "That's just my breath." --Bloom County Don Wegeng Wegeng.Henr@Xerox.COM seismo!rochester!rocksanne!dw
aptr@ur-tut.UUCP (01/16/87)
In article <254@rocksanne.UUCP> dw@rocksanne.UUCP (Don Wegeng) writes: >Awhile back someone mentioned that a clock driver had been written to support >the 6300 realtime clock under PC-DOS 3.1. I'd like to ahold of this beast, >if it is available. Any details that anyone can supply will be most appric- >iated! If you are using AT&T Dos 3.1 you do not need any special drivers. It just works. AT&T DOS 3.1 is available for about $50 (I think) for any AT&T supporting store. -- The Wumpus UUCP: {seismo,allegra,decvax}!rochester!ur-tut!aptr BITNET: aptrccss@uorvm Disclaimer: "Who? When? Me? It was the Booze!" - M. Binkley
king@uvm-gen.UUCP (John King) (01/17/87)
> > Awhile back someone mentioned that a clock driver had been written to support > the 6300 realtime clock under PC-DOS 3.1. I'd like to ahold of this beast, > if it is available. Any details that anyone can supply will be most appric- > iated! > Could you make that two requests, but change mine to PC DOS 3.2 (if it makes any difference). Also, does anyone out there have any information on what AT&T DOS does with other things (interrupts, etc), that normal PC DOS dosen't need to bother with? Thanks a lot! -- John R. King USENET: decvax!dartvax!uvm-gen!king The University of Vermont MODEM: (802) 862-2186 The SLC TBBS (SYSOP)
henkp@nikhefk.UUCP (Henk Peek) (01/20/87)
=Could you make that two requests, but change mine to PC DOS 3.2 (if it makes =any difference). Also, does anyone out there have any information on what =AT&T DOS does with other things (interrupts, etc), that normal PC DOS dosen't =need to bother with? =John R. King USENET: decvax!dartvax!uvm-gen!king =The University of Vermont MODEM: (802) 862-2186 The SLC TBBS (SYSOP) I have a realtime clock support program. A friend of me has written it. He is not on the net. Who was the first guy which wanted it?? I will send it by mail. So far as I known there are no special interupts... There are only 2 special bios calls to read and set the realtime clock. (See the bios listings in the hardware manual). henk peek, henkp@nikhefk.UUCP ..mcvax!nikhefk!henkp.UUCP
jeffm@mmintl.UUCP (02/03/87)
*
I too have noticed the 6300 clock problems with PCDOS 3.1 but only recently,
although I have been running it for almost 2 years. I am on an ethernet
system here at work. When I brought my machine home the other week I took
3Com's driver, enet.sys, out of config.sys. Lo and behold, my clock stopped
working automatically. When I brought the machine back to work and added
enet.sys back in, the clock started working again. As long as enet.sys is
in there it works. I'll be leaving it in there even after I retire this
machine! :-) BTW, I have ROM version 1.21, but the same thing must have
been true with version 1.1 since the clock always worked fine then with PCDOS
3.1. I also use an Adaptec 2070A controller w/ a Seagate ST238, but this
occurred when I had a WD1002S controlling a CMI 1/2 ht 20 Meg. I don't
think the controller has any effect but I thought I'd mention it. BTW BTW,
the Adaptec/Seagate combo is dynamite.
Strange but true........
Jeff
*************************************************
* Jeff Miller *
* Multimate International, an Ashton-Tate Co. *
* 52 Oakland Avenue *
* East Hartford, CT 06108-9911 *
* (203) 522-2116 x257 *
* UUCP: *
* ...!seismo!utah-cs!utah-gr!pwa-b!mmintl!jeffm *
*************************************************
*wtm@neoucom.UUCP (02/10/87)
A: I run AT&T DOS 3.1 without an ethenet card and also without enet.sys. I have experienced little trouble with the system clock. Congrats to AT&T for not having the date rollover bug that so many other implementations of DOS have. On many flavors of MS-DOS and (all flavors??) of PC-DOS, if the first BIOS call after midnight is a disk operation, the flag that indicated date rollover is lost, and the date is a day behind at the occurance of the next timer tick. Either I have been lucky or this is not a problem on AT&T MS-DOS. B: Possible clock problems. 1. Sappy batteries cause the clock to loose time. This happens when the battery itself poops out, or if the machine is used infrequently. 2. It has been my experience that about 3/4 of the 6300s that we have have a tendency to loose time at about 1 min/day. You could try to adjust the trimmer cap. next to the clock chip. With a low-capacitacne scope lead, a frequency counter should read 32 768 . 000 Hz. 3. Resident Diagnostics version 1.43 cures some problems related to getting goofy times, if the time rolls over while the clock chip is being read. 4. Some overly IBM-specific programs are confused by the 6300's way of keeping the system time-- Norton Utilities, for instance. 5. This shouldn't matter, but could leaving the ethernet board in your computer but taking enet.sys out of your config.sys file leave the hardware on the ethernet controller in a state that keeps the clock from working correctly? That is, IF you left the board in. C: One of our 6300s is weird. It stops keeping time when the a.c. power is off; almost as if the clock oscillator is not running. The time when the system was shut off is frozen while the a.c. is turned off. I haven`t had time to put my oscilloscope to it yet to see what is going on. --Bill Bill Mayhew Division of Basic Medical Sciences Northeastern Ohio Universities' College of Medicine Rootstown, OH 44272 USA phone: 216-325-2511 (wtm@neoucom.UUCP ...!cbatt!neoucom!wtm)
wtm@neoucom.UUCP (02/10/87)
I incorrectly stated that there is a trimmer cap on the the 6300 motherbard for adjusting the clock accuracy. You can add your own trimmer, but you shouldn`t do it unless you know what you are doing. Since the resonant frequecy tends to be less than desired, you can add a small capacitory in series with the crystal to raise its resonant frequency. A 3-10 pF capacitor would be a good choice. Radio Shack sells 6-50 pF capacitors 2 for $1.59, part number 272-1340 that might give you an acceptible adjustment range. The clock chip is located at position D-9 on the motherboad, it is numbered MM5328, if memory serves me right (too lazy to take the cover off right now). The crystal is at approximately C-10. It is the wee little cylinder with two wires coming out one end that is taped to the motherboard. Some crystals are metal, and some are coated with white plastic. They may be numbered "32768". You can unsolder one of the leads of the capcitor (IT IS VERY DELICATE!) and install the capacitor in series. A screw driver and a high accuracy frequecny counter are required for adjustment. Unless you have access to an electonic shop, you might as well settle for the chronically slow time of day clock. Secure the capacitor to the motherboard with a drop of silicone rubber "bathtub" compound to keep it from breaking loose. PLEASE don't fiddle with the machine unless you know what you are doing and are willing to take the responsibility for anything that you happen to goof up. Watch out for shorting out the NiCd battery that powers the clock if you decide to monkey around. You can disable the battery by removing the jumper at approximately D-10.
jeffm@mmintl.UUCP (02/22/87)
*
My comments about the need for 3com's enet.sys driver being necessary for
my 6300's clock to work properly are specifically directed at PC-DOS 3.1
on my machine. I don't know about and MS-DOS version. I use PC-DOS due
to the reasonable cluster size. The clock works automatically and per-
fectly. The time seems to be extremely accurate even over long periods.
My ROMs are V1.21. I've run the same PC-DOS on it since the summer of 85,
at which time I had ROMs V1.1 and the clock worked fine then also.
Sometimes the clocks don't work when you get the machine because the
batteries have never been charged. Often this can be remedied by leaving
the machine on for 48 hours or so. When that doesn't do it the battery
is just deader than a door nail. I fixed one of those once by removing
the jumpers connecting the battery to the mboard and zapping it with about
3.3 volts for a few hours. Then I left the machine on for 2 days and it
worked. Sometimes you can really shock a nicad into working again by
hitting it with reverse voltage momentarily. You've got to know what
you're doing with that one but it can be done. Unfortunately, when all
fails with the battery there's nothing you can do but void your warranty
and replace the battery yourself (not recommended) unless you care to
wait a long time for factory repair. Why the battery must be soldered
to the mboard is beyond me. Nothings perfect, not even the 6300 (it is
close). P.S. All 6300s should have V30s for best performance.
Jeff
*************************************************
* Jeff Miller *
* Multimate International, an Ashton-Tate Co. *
* 52 Oakland Avenue *
* East Hartford, CT 06108-9911 *
* (203) 522-2116 x257 *
* UUCP: *
* ...!seismo!utah-cs!utah-gr!pwa-b!mmintl!jeffm *
*************************************************
*