Conference hydra::amiga_v1

I had a similar problem some months ago.  It cured itself after a power off/on.
It appears the clock was somehow left in a hung state.  If you haven't tried a
power on/off give it a shot.  Otherwise that battery (nicad? lithium?) probably
lost it's jam.

...Ed

Ed, I had the same trouble on my 500.

Get to a CLI and issue a TIME and DATE command to set the clock right, 
then do a SETCLOCK OPT SAVE.

This fixed it for me...

John

    Re: .0
    
    If the problem persists after setting the clock with setclock, try
    a "setclock reset" with the Workbench 1.32 setclock version, before
    you heat up that ole soldering iron. 
    
    Regards,
    
    <CB> 

    
    
    This has been discussed before, but this is a known bug with W.B 1.3
    setclock program.
    
    The first fix was a small routine from Germany(?) called CallBack2Life.
    It resets the clock to "0"
    
    And the second fix came from C= in the distribution of WB 1.3.2
    (a new setclock program)
    
    For more info, consult previous notes.
    (I had the same problem you are having.)
    
    If you want the 1.3.2 in zoo, let me know I will upload it for you.
    
    Vince.

    
    I'm thinking that it must be a bad battery, since...
    
    1.  I have the 1.3.2 update(s)
    2.  I've done the 'reset' and 'save' options a few times.
    
    Thanks for the help, but I think it's time to heat up the old soldering
    iron.
    
    Ed.

    If the way to fix the clock is to power down, since it is
    battery-backed up, you can't power down the clock without pulling out
    the a501 card for a few minutes.  I think this disconnects a loopback
    from the battery tothe circuit, but I'm guessing, since that could
    improve shelf life of the battery but what do I know?
    So try powreing down the computer, pulling out the a501, count to
    1000, put back the a501, power up, set the date (there is no time
    command in the standard release OS 
    as given above) and setclock option save, then setclock option
    load, date.
    Tom

    re: .0
    
    Had exactly the same problem myself on my A501.  Don't know if that is 
    what you have Ed, but I couldn't clear the problem by doing a setclock	
    opt save.  Pulled the A501 out and checked the battery voltage.  It
    looked OK so I put it back in and it's run fine ever since.
    
    Ken

    My clock gains 7-12 minutes a day.  The battery backed clock keeps time
    just fine.  If I do not do a SetClock load daily my system time stamps
    would be rather hosed.  Are there alternate time keeping routines that
    are more accurate than what is used now?  
    
    mark
    

    re: .6 And if you don't have an A501 in your 2000 as Ed probably
    doesn't   8^) !
    
    

    re: .8
    You should be able to calibrate your clock to keep better time.  I
    remember some earlier notes from others who did.  Look back through
    some of the older topics (around 2 years ago?).  Somebody even had a
    program to help check the calibration.

Re: .10

>    You should be able to calibrate your clock to keep better time...
>    Somebody even had a program to help check the calibration.

I'm the author of that program.  However, it was to calibrate the
battery backed up clock, and .8 reports problems with the line
frequency clock.  So, it'll be no help, I'm afraid.


Re: .8

The non-battery backed up clock is mostly software.  On the the CIA
(complex interface adapters?) chips is the hard ware part of it.
(The Amiga has two CIA chips which control the serial port, the
parallel port, and various system timers.)  The CIA chip counts
the AC current cycles, and the operating system periodically
takes an interrupt, resets the counter, and adds the elapsed time
to the system's time.  This clock does lose time because of lost
cycles during the interrupt servicing, but it is less than a few seconds
a day, I believe.

If you have a 2000, there is a jumper on the motherboard that can cause
the CIA chip to count video vertical sank pulses instead of the line
frequency.  I've heard that one of the symptoms of a misplaced jumper
here is that the line frequency clock runs slightly fast.

The back of the Amiga 2000 owner's manual gives describes the jumpers.

    Thanks for the info.  I've broken out my system programmers guide and
    rolled my shirt-sleeves back.  I will let you know what I find and do.
    
    mark
    
    p.s. I am using an A500.

    
    I've go a similar problem to .8, except that my clock LOSES 7-12
    minutes per day. (At leats I think it's different. Mark says that he
    gains 7-12 mins, but the title says he's losing 7-12 mins). As in .8,
    the battery backed on the A501 keeps perfect time.
    
    I've had my A500 for 6-8 months now, but only noticed it when I started
    running JR-Comm (maybe it's always been wrong and I only noticed
    because JR-Comm displays the system time for me).
    
    Anyway, any answers/info to .8 would interest me too.
    
    Cheers,
    
    Steve Bee.
    

    I've noticed problems with my clock - nothing consistent but it seems
    to be related to downloading - the more I download the further off the
    clock is when I notice the problem. I also use JR-Comm but it certainly
    isn't consistent. I have a 2000 w/ 3Meg and Quantam harddisk.
    
    John

    
    I've been keeping a closer eye on my clock since I read this note, and
    I don't think downloading with JR-Comm has anything to do with it. I've
    been running JR-Comm for 2 days now (continuously), but have NOT done
    any downloads. It is now 9:56 am (UK time), but my clock says that it's
    only 9:52.
    
    	Over the weekend, I'll spend 2 days not running JR-Comm, and report
    the results here.
    
    Steve.

    I get consistant time losses with all comms packages. sort of of the
    order of 1 hour lost in 2 or 3. Get really frustrating.
    
    regards
    kevin