Dear all,
question I?m thinking about, has clock for the 81C55 to run sychronized with
CPU clock. If
it is independent would it be an option to use a separate 2,4576MHz oszilator
to drive PIO
@ its original speed.
Any idea if this can work?
Kind regards
Georg Käter
On Thu, Nov 26, 2015 at 8:33 AM, Stephen Adolph <[email protected]> wrote:
I think doubling the speed of the processor might be too much however. I know
that is convenient - to make a /2 circuit. It is an experiment.
I found the original OKI datasheet for MSM80C85ARS indicates that it can
support a clock period of 320 nsec minimum.
There are faster processors that are rated down to 200 nsec and 125 nsec.
On Thu, Nov 26, 2015 at 8:11 AM, Stephen Adolph <[email protected]> wrote:
I'd want to take a look and understand how the downstream parts would react to
the faster I/O and memory timing. The RAM and ROM systems are basically
asynchronous - they look for edges and respond accordingly. So long as they
are fast enough it should be ok.
The PIO chip - unclear there, but I bet it would work
On Thu, Nov 26, 2015 at 3:53 AM, Georg Käter
<[email protected]> wrote:
Dear all,
thank you for your comments/ideas. I?ll do further investigations in that
matter. Maybe M100
can run CPU @ 9,8304MHz outer clock, related systemclock @pin 37 has then just
to be
divided by 2 through a 74LS293 or equal to feed the PIO and further components
with the
"old" base clock.
What?s your opinion about this?
Kind regards
Georg Käter
On Nov 25, 2015 11:55 AM, "Alex ..." <[email protected]> wrote:
The cassette routines will probably be broken by this too, though you might be
able to "overclock" the tape recorder (or MP3 file) to match. :)
On Nov 25, 2015 9:43 AM, "Stephen Adolph" <[email protected]> wrote:
Hi Georg,
Very interesting! I think there could be impact on sounds as well.
Tundra semiconductor made a processor capable of 10MHz also. I have not yet
played with that.
It could be possible to make an adapter that includes a /4 circuit to feed
correct clock to the UART, and also the 81C55.
This is an interesting idea, worth exploring.
Steve
On Wed, Nov 25, 2015 at 9:33 AM, Georg Käter
<[email protected]> wrote:
Hello together,
I did some try out on overclocking the 80C85 in my M100. I was quite
successful using an oszilator
with 7,3728MHz which gives a CPU clock 50% higher than the original one. My
M100 run with this
w/o any issues in running BASIC so far, gain in performance is visible and
amazing. But there is one thing
you might have an solution or even idea how to solve for. TELCOM and any
program communicating via
serial port are not working as PIO 81C55 gets the clock from CPU which is now
50% above the original
value.
Any comment from you is welcome
Kind regards
Georg Käter
---------- Forwarded message ----------
From: "Alex ..." <[email protected]>
To: Model 100 Discussion <[email protected]>
Cc:
Date: Wed, 25 Nov 2015 11:55:39 -0500
Subject: Re: [M100] M10X/T200: Overclocking CPU
The cassette routines will probably be broken by this too, though you might be
able to "overclock" the tape recorder (or MP3 file) to match. :)
On Nov 25, 2015 9:43 AM, "Stephen Adolph" <[email protected]> wrote:
Hi Georg,
Very interesting! I think there could be impact on sounds as well.
Tundra semiconductor made a processor capable of 10MHz also. I have not yet
played with that.
It could be possible to make an adapter that includes a /4 circuit to feed
correct clock to the UART, and also the 81C55.
This is an interesting idea, worth exploring.
Steve
On Wed, Nov 25, 2015 at 9:33 AM, Georg Käter
<[email protected]> wrote:
Hello together,
I did some try out on overclocking the 80C85 in my M100. I was quite
successful using an oszilator
with 7,3728MHz which gives a CPU clock 50% higher than the original one. My
M100 run with this
w/o any issues in running BASIC so far, gain in performance is visible and
amazing. But there is one thing
you might have an solution or even idea how to solve for. TELCOM and any
program communicating via
serial port are not working as PIO 81C55 gets the clock from CPU which is now
50% above the original
value.
Any comment from you is welcome
Kind regards
Georg Käter
--- Begin Message ---
I think doubling the speed of the processor might be too much however. I
know that is convenient - to make a /2 circuit. It is an experiment.
I found the original OKI datasheet for MSM80C85ARS indicates that it can
support a clock period of 320 nsec minimum.
There are faster processors that are rated down to 200 nsec and 125 nsec.
On Thu, Nov 26, 2015 at 8:11 AM, Stephen Adolph <[email protected]>
wrote:
> I'd want to take a look and understand how the downstream parts would
> react to the faster I/O and memory timing. The RAM and ROM systems are
> basically asynchronous - they look for edges and respond accordingly. So
> long as they are fast enough it should be ok.
> The PIO chip - unclear there, but I bet it would work
>
>
> On Thu, Nov 26, 2015 at 3:53 AM, Georg Käter <
> [email protected]> wrote:
>
>> Dear all,
>>
>> thank you for your comments/ideas. I´ll do further investigations in that
>> matter. Maybe M100
>> can run CPU @ 9,8304MHz outer clock, related systemclock @pin 37 has then
>> just to be
>> divided by 2 through a 74LS293 or equal to feed the PIO and further
>> components with the
>> "old" base clock.
>> What´s your opinion about this?
>>
>> Kind regards
>> Georg Käter
>>
>>
>> On Nov 25, 2015 11:55 AM, "Alex ..." <[email protected]> wrote:
>>
>> The cassette routines will probably be broken by this too, though you
>> might be able to "overclock" the tape recorder (or MP3 file) to match. :)
>> On Nov 25, 2015 9:43 AM, "Stephen Adolph" <[email protected]> wrote:
>> Hi Georg,
>> Very interesting! I think there could be impact on sounds as well.
>> Tundra semiconductor made a processor capable of 10MHz also. I have not
>> yet played with that.
>> It could be possible to make an adapter that includes a /4 circuit to
>> feed correct clock to the UART, and also the 81C55.
>> This is an interesting idea, worth exploring.
>> Steve
>>
>> On Wed, Nov 25, 2015 at 9:33 AM, Georg Käter <
>> [email protected]> wrote:
>> Hello together,
>>
>> I did some try out on overclocking the 80C85 in my M100. I was quite
>> successful using an oszilator
>> with 7,3728MHz which gives a CPU clock 50% higher than the original one.
>> My M100 run with this
>> w/o any issues in running BASIC so far, gain in performance is visible
>> and amazing. But there is one thing
>> you might have an solution or even idea how to solve for. TELCOM and any
>> program communicating via
>> serial port are not working as PIO 81C55 gets the clock from CPU which is
>> now 50% above the original
>> value.
>> Any comment from you is welcome
>>
>> Kind regards
>> Georg Käter
>>
>>
>> ---------- Forwarded message ----------
>> From: "Alex ..." <[email protected]>
>> To: Model 100 Discussion <[email protected]>
>> Cc:
>> Date: Wed, 25 Nov 2015 11:55:39 -0500
>> Subject: Re: [M100] M10X/T200: Overclocking CPU
>>
>> The cassette routines will probably be broken by this too, though you
>> might be able to "overclock" the tape recorder (or MP3 file) to match. :)
>> On Nov 25, 2015 9:43 AM, "Stephen Adolph" <[email protected]> wrote:
>>
>>> Hi Georg,
>>> Very interesting! I think there could be impact on sounds as well.
>>> Tundra semiconductor made a processor capable of 10MHz also. I have not
>>> yet played with that.
>>> It could be possible to make an adapter that includes a /4 circuit to
>>> feed correct clock to the UART, and also the 81C55.
>>> This is an interesting idea, worth exploring.
>>> Steve
>>>
>>> On Wed, Nov 25, 2015 at 9:33 AM, Georg Käter <
>>> [email protected]> wrote:
>>>
>>>> Hello together,
>>>>
>>>> I did some try out on overclocking the 80C85 in my M100. I was quite
>>>> successful using an oszilator
>>>> with 7,3728MHz which gives a CPU clock 50% higher than the original
>>>> one. My M100 run with this
>>>> w/o any issues in running BASIC so far, gain in performance is visible
>>>> and amazing. But there is one thing
>>>> you might have an solution or even idea how to solve for. TELCOM and
>>>> any program communicating via
>>>> serial port are not working as PIO 81C55 gets the clock from CPU which
>>>> is now 50% above the original
>>>> value.
>>>> Any comment from you is welcome
>>>>
>>>> Kind regards
>>>> Georg Käter
>>>>
>>>
>>>
>>
>
--- End Message ---
