Archived from groups: microsoft.public.win95.general.discussion (More info?)
The PC has a com1 (external, unused) port. An internal modem uses com3. If
I install a second internal modem, which com port does it want to be
assigned to ? This is all on win95.
Archived from groups: microsoft.public.win95.general.discussion (More info?)
What sort of internal modem? If it includes its own serial port then it
will have to be configured (probably by jumpers on the board) to an I/O port
and interrupt, so you will completely control the COM port number that it
gets. In your case this is presumably COM2.
If it is a software modem then it will use a virtual COM port - a port
created in software by the modem driver that fools the system into thinking
that a physical COM port is installed. In this case, the number that gets
assigned could be anything at all, and is often not controllable. The
documentation supplied with the modem should explain the process, but in
many cases you can't do anything other than let the installation procedure
make these decisions for you.
--
Jeff Richards
MS MVP (Windows - Shell/User)
"Bill" <billnh2001nospam@yahoo.com> wrote in message
news:eaC5LrDJFHA.3788@tk2msftngp13.phx.gbl...
> The PC has a com1 (external, unused) port. An internal modem uses com3.
> If
> I install a second internal modem, which com port does it want to be
> assigned to ? This is all on win95.
>
>
Archived from groups: microsoft.public.win95.general.discussion (More info?)
"Bill" <billnh2001nospam@yahoo.com> wrote in message
news:eaC5LrDJFHA.3788@tk2msftngp13.phx.gbl...
> The PC has a com1 (external, unused) port. An internal modem uses com3.
If
> I install a second internal modem, which com port does it want to be
> assigned to ? This is all on win95.
>
>
2 or 4 to prevent an IRQ conflict with the existing modem.
Archived from groups: microsoft.public.win95.general.discussion (More info?)
"Kay Archer" <kayhyphenarcher@mail.com> wrote:
>
>"Bill" <billnh2001nospam@yahoo.com> wrote in message
>news:eaC5LrDJFHA.3788@tk2msftngp13.phx.gbl...
>> The PC has a com1 (external, unused) port. An internal modem uses com3.
>If
>> I install a second internal modem, which com port does it want to be
>> assigned to ? This is all on win95.
>2 or 4 to prevent an IRQ conflict with the existing modem.
Right. One of the problems with COM ports is that COM1 and COM3 use
the same IRQ line (I forget which one it is), and COM2 and COM4 share
another. So, although there are four COM ports designed, you really
can only have two active at once.
Another reason to be glad the world has moved to USB.
--
Tim Slattery
MS MVP(DTS)
Slattery_T@bls.gov
Archived from groups: microsoft.public.win95.general.discussion (More info?)
"Tim Slattery" <Slattery_T@bls.gov> wrote in message
news:h90u21d2arrku3tv434al8u1jb04mr5j1o@4ax.com...
> "Kay Archer" <kayhyphenarcher@mail.com> wrote:
>
>>
>>"Bill" <billnh2001nospam@yahoo.com> wrote in message
>>news:eaC5LrDJFHA.3788@tk2msftngp13.phx.gbl...
>>> The PC has a com1 (external, unused) port. An internal modem uses com3.
>>If
>>> I install a second internal modem, which com port does it want to be
>>> assigned to ? This is all on win95.
>
>>2 or 4 to prevent an IRQ conflict with the existing modem.
>
> Right. One of the problems with COM ports is that COM1 and COM3 use
> the same IRQ line (I forget which one it is), and COM2 and COM4 share
> another. So, although there are four COM ports designed, you really
> can only have two active at once.
>
> Another reason to be glad the world has moved to USB.
>
> --
> Tim Slattery
> MS MVP(DTS)
> Slattery_T@bls.gov
IIRC I used to disable serial port 2 in the bios , then the internal modem
would take over the function of com 2 and use IRQ 3 (com3 is IRQ 4)
Archived from groups: microsoft.public.win95.general.discussion (More info?)
Tim Slattery wrote:
> Right. One of the problems with COM ports is that COM1 and COM3 use
> the same IRQ line (I forget which one it is), and COM2 and COM4 share
> another. So, although there are four COM ports designed, you really
> can only have two active at once.
>
> Another reason to be glad the world has moved to USB.
I could never figure out why, with all the leaps and bounds in computer
technology, that IRQ's were for the longest time limited to 15? To
boot, as you mention the first group of 5 or 7 IRQ's are used by
different devices and some are cascades or redirect to other IRQ's! Not
to mention sound cards from hell that can be picky when it comes to IRQ.
Windows IRQ sharing took care of some of the problems but still I
can't understand why the motherboard manufacturers didn't engineer 50
IRQ's years ago! I went here and took a little 2 minute test, but I
won't tell you my score, too embarrassing! http://studynotes.net/QIRQ.htm
John
PS my score was 21%! It's been a long time since I fiddled with IRQ's.
Archived from groups: microsoft.public.win95.general.discussion (More info?)
"John John" <audetweld@nbnet.nb.ca> wrote in message
news:eZedhyLJFHA.572@tk2msftngp13.phx.gbl...
> Tim Slattery wrote:
>
> > Right. One of the problems with COM ports is that COM1 and COM3 use
> > the same IRQ line (I forget which one it is), and COM2 and COM4 share
> > another. So, although there are four COM ports designed, you really
> > can only have two active at once.
> >
> > Another reason to be glad the world has moved to USB.
>
> I could never figure out why, with all the leaps and bounds in computer
> technology, that IRQ's were for the longest time limited to 15?
16, zero through fifteen (g).
To
> boot, as you mention the first group of 5 or 7 IRQ's are used by
> different devices and some are cascades or redirect to other IRQ's! Not
> to mention sound cards from hell that can be picky when it comes to IRQ.
> Windows IRQ sharing took care of some of the problems but still I
> can't understand why the motherboard manufacturers didn't engineer 50
> IRQ's years ago! I went here and took a little 2 minute test, but I
> won't tell you my score, too embarrassing! http://studynotes.net/QIRQ.htm
>
> John
>
> PS my score was 21%! It's been a long time since I fiddled with IRQ's.
>
Archived from groups: microsoft.public.win95.general.discussion (More info?)
Kay Archer wrote:
> 16, zero through fifteen (g).
Oh just great, now my test score goes down from 21% to 14%!
John
Archived from groups: microsoft.public.win95.general.discussion (More info?)
"Kay Archer" <kayhyphenarcher@mail.com> wrote:
>16, zero through fifteen (g).
But I believe number 2 was used to connect the first IRQ bank to the
second, wasn't it? So you could only actually use 15.
--
Tim Slattery
MS MVP(DTS)
Slattery_T@bls.gov
Archived from groups: microsoft.public.win95.general.discussion (More info?)
Tim Slattery wrote:
> But I believe number 2 was used to connect the first IRQ bank to the
> second, wasn't it?
Cascade to IRQ 9. How many IRQ's in todays computers, 26 or 29?
John
Archived from groups: microsoft.public.win95.general.discussion (More info?)
John John <audetweld@nbnet.nb.ca> wrote:
>Tim Slattery wrote:
>
>> But I believe number 2 was used to connect the first IRQ bank to the
>> second, wasn't it?
>
>Cascade to IRQ 9. How many IRQ's in todays computers, 26 or 29?
I have no clue. With all kinds of peripherals using USB you very
rarely have to worry about IRQs anymore.
--
Tim Slattery
MS MVP(DTS)
Slattery_T@bls.gov
Archived from groups: microsoft.public.win95.general.discussion (More info?)
The architecture was set by IBM with the selection of the Intel 8259 PIC in
the PC and the addition of a second 8259 device (daisy chained from IRQ2) in
the AT. So it's still 16 total.
--
Jeff Richards
MS MVP (Windows - Shell/User)
"John John" <audetweld@nbnet.nb.ca> wrote in message
news:en5IhuXJFHA.1476@TK2MSFTNGP09.phx.gbl...
> snip <
> How many IRQ's in todays computers, 26 or 29?
>
> John
>
Archived from groups: microsoft.public.win95.general.discussion (More info?)
"Tim Slattery" <Slattery_T@bls.gov> wrote in message
news
ek03195sg9ugpo2rotm5iqdj5g15t3ipo@4ax.com...
> "Kay Archer" <kayhyphenarcher@mail.com> wrote:
>
> >16, zero through fifteen (g).
>
> But I believe number 2 was used to connect the first IRQ bank to the
> second, wasn't it? So you could only actually use 15.
>
Use? I'd say less than 15. Did you ever put anything on IRQ0? Depends on
how you define useable... (Still with a (g)).
Signature:
Never attribute to malice anything that can be ascribed to simple stupidity.
Archived from groups: microsoft.public.win95.general.discussion (More info?)
In that case, as Tim said, thank God for USB. But that's of little use
to Windows 95 users so someone has to know the abc's (or 123's) of IRQ's.
John
Jeff Richards wrote:
> The architecture was set by IBM with the selection of the Intel 8259 PIC in
> the PC and the addition of a second 8259 device (daisy chained from IRQ2) in
> the AT. So it's still 16 total.
Archived from groups: microsoft.public.win95.general.discussion (More info?)
It's used by the System Timer.
John
Kay Archer wrote:
> "Tim Slattery" <Slattery_T@bls.gov> wrote in message
> news
ek03195sg9ugpo2rotm5iqdj5g15t3ipo@4ax.com...
>
>>"Kay Archer" <kayhyphenarcher@mail.com> wrote:
>>
>>
>>>16, zero through fifteen (g).
>>
>>But I believe number 2 was used to connect the first IRQ bank to the
>>second, wasn't it? So you could only actually use 15.
>>
>
> Use? I'd say less than 15. Did you ever put anything on IRQ0? Depends on
> how you define useable... (Still with a (g)).
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> Signature:
> Never attribute to malice anything that can be ascribed to simple stupidity.
>
>
Archived from groups: microsoft.public.win95.general.discussion (More info?)
I always hated having to disable something to free up an IRQ for use
....(only about 5 that you could really fiddle with)
XP shares them better ...but not perfect by any means
"John John" <audetweld@nbnet.nb.ca> wrote in message
news
GpAJcjJFHA.2136@TK2MSFTNGP14.phx.gbl...
> In that case, as Tim said, thank God for USB. But that's of little use to
> Windows 95 users so someone has to know the abc's (or 123's) of IRQ's.
>
> John
>
> Jeff Richards wrote:
>> The architecture was set by IBM with the selection of the Intel 8259 PIC
>> in the PC and the addition of a second 8259 device (daisy chained from
>> IRQ2) in the AT. So it's still 16 total.
>
Archived from groups: microsoft.public.win95.general.discussion (More info?)
It depends on the hardware more than the operating system. The original PC
ISA bus was constructed in such a way that it was not possible for more than
one device to properly share the IRQ. USB uses one IRQ for the controller,
and decodes device IDs for itself. The technique that USB uses is not
actually new - it's the same technique that had been used for many years if
many (eg, 16) serial ports had to be used with an ISA bus. The serial port
interface card used one interrupt for its interface to ISA, but decoded port
numbers for itself to provide individual access to each serial port through
a single interrupt number. PCI works similarly, providing it's own internal
interrupt structure to enable multiple devices to function through each CPU
interrupt.
--
Jeff Richards
MS MVP (Windows - Shell/User)
"John John" <audetweld@nbnet.nb.ca> wrote in message
news
GpAJcjJFHA.2136@TK2MSFTNGP14.phx.gbl...
> In that case, as Tim said, thank God for USB. But that's of little use to
> Windows 95 users so someone has to know the abc's (or 123's) of IRQ's.
Archived from groups: microsoft.public.win95.general.discussion (More info?)
"John John" <audetweld@nbnet.nb.ca> wrote in message
news:%23i5p2cjJFHA.2136@TK2MSFTNGP14.phx.gbl...
> It's used by the System Timer.
>
> John
>
From memory:
0 system timer (I might have said clock without John's help)
1 keyboard
2 redirect to IRQ9
3 com 2/4
4 com 1/3
5 typical uses: NIC or soundcard (some of which could not be changed,
argh!)
6
7 lpt1
8
9 redirect from IRQ2
10 p/s 2 mouse port
11
12
13
14 scsi controllers (typical)
15 hdd controllers
I used to be able to do this on a chalkboard while giving an hourlong
lecture on the subject...
I may have 14 and 15 reversed.
10 out of 16 only gives me about 60%.
Archived from groups: microsoft.public.win95.general.discussion (More info?)
According to the test on page http://studynotes.net/QIRQ.htm you scored
57%. You did indeed have 14 and 15 reversed. Had you given those
answers your score would have been 71%, but as you already know the
Academy cannot allow students to change their answers after tests have
been turned in so we must inscribe a test score of 57% in your student
records. At this time we would like to inform you that Mr. John John
was caught last night trying to hack the Academy student records to
falsify his test score. Mr. John has been suspended.
Sincerely;
Dewey Dweeb
Academy Dean
Kay Archer wrote:
> "John John" <audetweld@nbnet.nb.ca> wrote in message
> news:%23i5p2cjJFHA.2136@TK2MSFTNGP14.phx.gbl...
>
>>It's used by the System Timer.
>>
>>John
>>
>
> From memory:
>
> 0 system timer (I might have said clock without John's help)
> 1 keyboard
> 2 redirect to IRQ9
> 3 com 2/4
> 4 com 1/3
> 5 typical uses: NIC or soundcard (some of which could not be changed,
> argh!)
> 6
> 7 lpt1
> 8
> 9 redirect from IRQ2
> 10 p/s 2 mouse port
> 11
> 12
> 13
> 14 scsi controllers (typical)
> 15 hdd controllers
>
>
>
> I used to be able to do this on a chalkboard while giving an hourlong
> lecture on the subject...
> I may have 14 and 15 reversed.
> 10 out of 16 only gives me about 60%.
>
>
Archived from groups: microsoft.public.win95.general.discussion (More info?)
Thanks Jeff, you have explained something that I just couldn't and
didn't bother to figure out for many years, IRQ sharing. If I may
indulge and pick at your brains a bit more with this device IRQ sharing,
this is the IRQ list for my Windows 2000 home pc. IRQ2 cascades to IRQ9
and ...hmmm... how can all these devices on IRQ9 work without
conflicting with each other or rather, how can they all work simultaneously?
******************** IRQ SUMMARY ********************
IRQ Usage Summary:
(ISA) 0 System timer
(ISA) 4 Communications Port (COM1)
(ISA) 6 Standard floppy disk controller
(ISA) 8 System CMOS/real time clock
(ISA) 9 Microsoft ACPI-Compliant System
(PCI) 9 NVIDIA GeForce2 GTS/GeForce2 Pro
(PCI) 9 Win2003 Promise Ultra133 TX2 (tm) IDE Controller
(PCI) 9 Creative SB Live! Basic (WDM)
(PCI) 9 U.S. Robotics 56KVoice PCI
(PCI) 9 Intel(R) 82801 BAIBAM USB Universal Host Controller -2442
(PCI) 9 Intel(R) 82801 BAIBAM USB Universal Host Controller -2444
(PCI) 10 Intel(R) 82801 BAIBAM 5MBus Controller -2443
(ISA) 12 Microsoft PS/2 Port Mouse (IntelliPoint)
(ISA) 13 Numeric data processor
(ISA) 14 Primary IDE Channel
(ISA) 15 Secondary IDE Channel
I'm sure glad IRQ sharing exists because I would have had a bitch of a
time trying to get all that stuff to work the old fashion way!
John
Jeff Richards wrote:
> It depends on the hardware more than the operating system. The original PC
> ISA bus was constructed in such a way that it was not possible for more than
> one device to properly share the IRQ. USB uses one IRQ for the controller,
> and decodes device IDs for itself. The technique that USB uses is not
> actually new - it's the same technique that had been used for many years if
> many (eg, 16) serial ports had to be used with an ISA bus. The serial port
> interface card used one interrupt for its interface to ISA, but decoded port
> numbers for itself to provide individual access to each serial port through
> a single interrupt number. PCI works similarly, providing it's own internal
> interrupt structure to enable multiple devices to function through each CPU
> interrupt.
Archived from groups: microsoft.public.win95.general.discussion (More info?)
When IBM added the second 8259 in the AT they had to cope with the fact that
the CPU only supported one hardware interrupt line, so they had to make the
two 8259s look like one large one. They did this by connecting the
interrupt output of the second device to IRQ2 of the first device, so the
interrupt line of the CPU is only driven by the first 8259. So ANY interrupt
from the second device looks like interrupt 2, and the routine at this
interrupt has to decode the status of the second 8259 to find what the
interrupt really was.
Of course, this would have made IRQ2 unavailable, so this interrupt line was
wired into IRQ9 (interrupt line 1 of device 2). The service routine at
interrupt 9 calls interrupt 2 (like all interrupts through the second device
do). So IRQ2 is still available (via the deciding at the service routine)
but IRQ9 is not available because that interrupt line isn't connected to
anything.
However, this arrangement, designed for PC add-on devices that might have
used IRQ2, seems to have been dropped somewhere along the way, and interrupt
line 9 is now wired through to IRQ9, and IRQ2 is marked unavailable. It
doesn't actually matter which way around things are, as long as the
interrupt service routines understand the physical arrangement of the
machine, and plug-in cards don't attempt to use the unavailable interrupt.
It is correct that ISA interrupts cannot be shared, but this is due to the
ISA bus architecture. Sharing interrupts depends on two things - the signal
levels and protocols that allow more than one device to drive the interrupt
line, and a mechanism for the software to work out which device actually
generated the interrupt. Although the ACPI interrupt is listed as ISA, it's
actually raised by the motherboard chipset, so presumably there are features
in the chipset design that allow it to be shared.
--
Jeff Richards
MS MVP (Windows - Shell/User)
"John John" <audetweld@nbnet.nb.ca> wrote in message
news
t4GlTwJFHA.3916@TK2MSFTNGP14.phx.gbl...
> Thanks Jeff, you have explained something that I just couldn't and didn't
> bother to figure out for many years, IRQ sharing. If I may indulge and
> pick at your brains a bit more with this device IRQ sharing, this is the
> IRQ list for my Windows 2000 home pc. IRQ2 cascades to IRQ9 and
> ...hmmm... how can all these devices on IRQ9 work without conflicting with
> each other or rather, how can they all work simultaneously?
>
> ******************** IRQ SUMMARY ********************
> IRQ Usage Summary:
> (ISA) 0 System timer
> (ISA) 4 Communications Port (COM1)
> (ISA) 6 Standard floppy disk controller
> (ISA) 8 System CMOS/real time clock
> (ISA) 9 Microsoft ACPI-Compliant System
> (PCI) 9 NVIDIA GeForce2 GTS/GeForce2 Pro
> (PCI) 9 Win2003 Promise Ultra133 TX2 (tm) IDE Controller
> (PCI) 9 Creative SB Live! Basic (WDM)
> (PCI) 9 U.S. Robotics 56KVoice PCI
> (PCI) 9 Intel(R) 82801 BAIBAM USB Universal Host Controller -2442
> (PCI) 9 Intel(R) 82801 BAIBAM USB Universal Host Controller -2444
> (PCI) 10 Intel(R) 82801 BAIBAM 5MBus Controller -2443
> (ISA) 12 Microsoft PS/2 Port Mouse (IntelliPoint)
> (ISA) 13 Numeric data processor
> (ISA) 14 Primary IDE Channel
> (ISA) 15 Secondary IDE Channel
>
> I'm sure glad IRQ sharing exists because I would have had a bitch of a
> time trying to get all that stuff to work the old fashion way!
Archived from groups: microsoft.public.win95.general.discussion (More info?)
Thanks Jeff.
John
Jeff Richards wrote:
> When IBM added the second 8259 in the AT they had to cope with the fact that
> the CPU only supported one hardware interrupt line, so they had to make the
> two 8259s look like one large one. They did this by connecting the
> interrupt output of the second device to IRQ2 of the first device, so the
> interrupt line of the CPU is only driven by the first 8259. So ANY interrupt
> from the second device looks like interrupt 2, and the routine at this
> interrupt has to decode the status of the second 8259 to find what the
> interrupt really was.
>
> Of course, this would have made IRQ2 unavailable, so this interrupt line was
> wired into IRQ9 (interrupt line 1 of device 2). The service routine at
> interrupt 9 calls interrupt 2 (like all interrupts through the second device
> do). So IRQ2 is still available (via the deciding at the service routine)
> but IRQ9 is not available because that interrupt line isn't connected to
> anything.
>
> However, this arrangement, designed for PC add-on devices that might have
> used IRQ2, seems to have been dropped somewhere along the way, and interrupt
> line 9 is now wired through to IRQ9, and IRQ2 is marked unavailable. It
> doesn't actually matter which way around things are, as long as the
> interrupt service routines understand the physical arrangement of the
> machine, and plug-in cards don't attempt to use the unavailable interrupt.
>
> It is correct that ISA interrupts cannot be shared, but this is due to the
> ISA bus architecture. Sharing interrupts depends on two things - the signal
> levels and protocols that allow more than one device to drive the interrupt
> line, and a mechanism for the software to work out which device actually
> generated the interrupt. Although the ACPI interrupt is listed as ISA, it's
> actually raised by the motherboard chipset, so presumably there are features
> in the chipset design that allow it to be shared.
There are 817 identified and unidentified users. To see the list of identified users, Click here.
You are about to answer a thread that has been inactive for more than 6 months.
If you still wish to proceed, please ensure that your posting is original and does not duplicate or overlap any prior responses to this thread.
