So, do we know what the status is on using events in C.1.1+? Turns out=20=20
I need to use an event right now and don't want to tear my hair out=20=20
trying to debug the un-debuggable. Did someone determine that behavior=20=20
was always anomalous or only under certain conditions? I need to use=20=20
oSerialPort and want to use an event based on the arrival of a=20=20
character.

--

Brian Lloyd Granite Bay Montessori
brian AT gbmontessori DOT com 9330 Sierra College Blvd.
+1.916.367.2131 (voice) Roseville, CA 95661, USA
http://www.gbmontessori.com

I fly because it releases my mind from the tyranny of petty things . . .
=97 Antoine de Saint-Exup=E9ry

PGP key ID: 12095C52A32A1B6C
PGP key fingerprint: 3B1D BA11 4913 3254 B6E0 CC09 1209 5C52 A32A 1B6C

=20

=20


(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

Brian,

Events are broken in the compiler for C1.1+. The firmware is fine,
but the compiler needs rev'd. The developer is working on it now.

DLC

Brian Lloyd wrote:
> So, do we know what the status is on using events in C.1.1+? Turns out
> I need to use an event right now and don't want to tear my hair out
> trying to debug the un-debuggable. Did someone determine that behavior
> was always anomalous or only under certain conditions? I need to use
> oSerialPort and want to use an event based on the arrival of a
> character.
>
> --
>
> Brian Lloyd Granite Bay Montessori
> brian AT gbmontessori DOT com 9330 Sierra College Blvd.
> +1.916.367.2131 (voice) Roseville, CA 95661, USA
> http://www.gbmontessori.com
>
> I fly because it releases my mind from the tyranny of petty things . . .
> — Antoine de Saint-Exupéry
>
> PGP key ID: 12095C52A32A1B6C
> PGP key fingerprint: 3B1D BA11 4913 3254 B6E0 CC09 1209 5C52 A32A 1B6C
>
>
>



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

On Mar 4, 2008, at 9:58 PM, ooPIC Tech Support wrote:

> Brian,
>
> Events are broken in the compiler for C1.1+. The firmware is fine,
> but the compiler needs rev'd. The developer is working on it now.

I know that it is broken but is it enough not-broken to allow some
event programming to work if I hold my tongue just right? Or is it so
broken that no amount of effort will make events work? I am planning
to introduce events to two of my students today and I am hoping that
they can get something simple to work even if I have to slap on some
band-aids.

Brian Lloyd
Granite Bay Montessori School 9330 Sierra College Bl
brian AT gbmontessori DOT com Roseville, CA 95661
+1.916.367.2131 (voice) +1.791.912.8170 (fax)

PGP key ID: 12095C52A32A1B6C
PGP key fingerprint: 3B1D BA11 4913 3254 B6E0 CC09 1209 5C52 A32A 1B6C



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

--- In o...@yahoogroups.com, Brian Lloyd wrote:
> On Mar 4, 2008, at 9:58 PM, ooPIC Tech Support wrote:
>
> > Brian,
> >
> > Events are broken in the compiler for C1.1+. The firmware is
fine,
> > but the compiler needs rev'd. The developer is working on it now.
>
> I know that it is broken but is it enough not-broken to allow some
> event programming to work if I hold my tongue just right? Or is it
so
> broken that no amount of effort will make events work? I am
planning
> to introduce events to two of my students today and I am hoping
that
> they can get something simple to work even if I have to slap on
some
> band-aids.
>
> Brian Lloyd
> Granite Bay Montessori School 9330 Sierra College Bl
> brian AT gbmontessori DOT com Roseville, CA 95661
> +1.916.367.2131 (voice) +1.791.912.8170 (fax)
>
> PGP key ID: 12095C52A32A1B6C
> PGP key fingerprint: 3B1D BA11 4913 3254 B6E0 CC09 1209 5C52 A32A
1B6C
>
>From a recent thread, events are so broken you can't even link
CPU.Hz1 to an LED.

HOWEVER, I have this code running on a B.2.2+ chip and it WAS
compiled with V6:

' Code for V6 of compiler

Dim BLINK As New oEvent
Dim WIRE As New oWire
Dim LED As New oDIO1

Sub Main()
LED.IOLine = 7
LED.Direction = cvOutput
WIRE.Input.Link(ooPIC.Hz1)
WIRE.Output.Link(BLINK.Operate)
WIRE.Operate = cvTrue
End Sub

Sub BLINK_Code()
LED = cvOn
ooPIC.Delay = 250
LED = cvOff
ooPIC.Delay = 250
End Sub

I wouldn't necessarily buy into the problem being totally V6
related. I might believe V6 & C.1.1 combined are an issue. Or, more
likely, I wouldn't believe any of it unless I tested it myself.

Is there any way you can get your hands on B.2.2+ chips? Depending
on you time requirements, perhaps
http://www.junun.org/MarkIII/Info.jsp?item=29 Tim has always shipped
my stuff pretty quick.

If you get B.2.2+ chips, there is no reason NOT to back up to V5. It
certainly lacks the 'gee whiz' objects and features of V6 and it has
warts but, for the most part, it works and its' deficiencies are well
understood.

Richard



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

--- In o...@yahoogroups.com, "rtstofer" wrote:
> Get some B.2.2+ chips, use V5 and ship the thing.
>
> Richard
>
Richard,

I've been debating this in my mind. I'd hate to re-write the code, get
the older chips and then turn around and migrate back as soon as the
patch is out, but since there is not even an ETA for the patch, I might
have to bite the bullet. Unfortunately, I am writting software for
another company who is providing the hardware, so I don't have that much
control anyway.

Dogulas



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

> I too, am waiting for the patch, because I can't deliver on my project
> until this is resolved. Its going to start costing me money. :-(
>
> Good luck,
>
> Dogulas
>

Get some B.2.2+ chips, use V5 and ship the thing.

Richard



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

Brian,

Not really - There is a memory slot after the event definition that
the compiler is not filling in. You could do a real hack using oRAM to
punch values into that slot, but it is not simple and it would suck up a
lot of object RAM to do it. If you are interested I can find the
details of the hack, it hasn't been tested yet though.

DLC

Brian Lloyd wrote:
> On Mar 4, 2008, at 9:58 PM, ooPIC Tech Support wrote:
>
>
>> Brian,
>>
>> Events are broken in the compiler for C1.1+. The firmware is fine,
>> but the compiler needs rev'd. The developer is working on it now.
>>
>
> I know that it is broken but is it enough not-broken to allow some
> event programming to work if I hold my tongue just right? Or is it so
> broken that no amount of effort will make events work? I am planning
> to introduce events to two of my students today and I am hoping that
> they can get something simple to work even if I have to slap on some
> band-aids.
>
> Brian Lloyd
> Granite Bay Montessori School 9330 Sierra College Bl
> brian AT gbmontessori DOT com Roseville, CA 95661
> +1.916.367.2131 (voice) +1.791.912.8170 (fax)
>
> PGP key ID: 12095C52A32A1B6C
> PGP key fingerprint: 3B1D BA11 4913 3254 B6E0 CC09 1209 5C52 A32A 1B6C
>
>



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

Does this impact the ocounter2h object? I've verified the square wave
going in, but am getting consistent counts (regardless of the input
frequency).

--kn

On Mar 4, 2008, at 11:58 PM, ooPIC Tech Support wrote:

> Brian,
>
> Events are broken in the compiler for C1.1+. The firmware is fine,
> but the compiler needs rev'd. The developer is working on it now.
>
> DLC
>
> Brian Lloyd wrote:
>> So, do we know what the status is on using events in C.1.1+? Turns
>> out
>> I need to use an event right now and don't want to tear my hair out
>> trying to debug the un-debuggable. Did someone determine that
>> behavior
>> was always anomalous or only under certain conditions? I need to use
>> oSerialPort and want to use an event based on the arrival of a
>> character.

--
Kevin Nickels



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

If the problem can be worked around by patching a byte or two in the
.oex file, a simple patching program (DOS or Win32 .exe) could be
written in a few minutes. I'd do it right now if the necessary
information was made available. It would let those with deadlines
get on with their work.

At 07:12 PM 3/5/2008, ooPIC Tech Support wrote:
>Brian,
>
> Not really - There is a memory slot after the event definition that
>the compiler is not filling in. You could do a real hack using oRAM to
>punch values into that slot, but it is not simple and it would suck up a
>lot of object RAM to do it. If you are interested I can find the
>details of the hack, it hasn't been tested yet though.
>
>DLC



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

On Mar 5, 2008, at 4:12 PM, ooPIC Tech Support wrote:

> Brian,
>
> Not really - There is a memory slot after the event definition that
> the compiler is not filling in. You could do a real hack using oRAM
> to
> punch values into that slot, but it is not simple and it would suck
> up a
> lot of object RAM to do it. If you are interested I can find the
> details of the hack, it hasn't been tested yet though.

No, I will find a B.2.2+ chip to use. I stupidly upgraded mine but I
know where I can lay hands on one. I can even do what needs to be done
in a commutator loop. Using events was just more elegant.

Brian Lloyd
Granite Bay Montessori School 9330 Sierra College Bl
brian AT gbmontessori DOT com Roseville, CA 95661
+1.916.367.2131 (voice) +1.791.912.8170 (fax)

PGP key ID: 12095C52A32A1B6C
PGP key fingerprint: 3B1D BA11 4913 3254 B6E0 CC09 1209 5C52 A32A 1B6C



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

On Mar 5, 2008, at 10:07 PM, Andrew Porrett wrote:

> If the problem can be worked around by patching a byte or two in the
> .oex file, a simple patching program (DOS or Win32 .exe) could be
> written in a few minutes. I'd do it right now if the necessary
> information was made available. It would let those with deadlines
> get on with their work.

I changed the code to use a commutator loop. It is way inefficient as=20=20
compared to using an event but it works. I can probably wait for the=20=20
compiler fix.

--

Brian Lloyd Granite Bay Montessori
brian AT gbmontessori DOT com 9330 Sierra College Blvd.
+1.916.367.2131 (voice) Roseville, CA 95661, USA
http://www.gbmontessori.com

I fly because it releases my mind from the tyranny of petty things . . .
=97 Antoine de Saint-Exup=E9ry

PGP key ID: 12095C52A32A1B6C
PGP key fingerprint: 3B1D BA11 4913 3254 B6E0 CC09 1209 5C52 A32A 1B6C

=20

=20


(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

--- In o...@yahoogroups.com, Brian Lloyd wrote:
>
> I know that it is broken but is it enough not-broken to allow some
> event programming to work if I hold my tongue just right? Or is it so
> broken that no amount of effort will make events work? I am planning
> to introduce events to two of my students today and I am hoping that
> they can get something simple to work even if I have to slap on some
> band-aids.
>
Brian,

I am in the same boat; needing an event for a project. Maybe not with
as tight a time constraint. Someone else suggested that by moving the
oEvent object and the event handler around in memory you might get it to
start working again, but in my case no movement had any effect. In the
C.1.1+ compiler, the effect that I am seeing is that when the Operate
property is set the code branches to Never Never Land. No code in the
event handler executes and it never returns to the interrupted code.
Everything just stops. It seems to me to be completely consistant with
no work-arounds.

I too, am waiting for the patch, because I can't deliver on my project
until this is resolved. Its going to start costing me money. :-(

Good luck,

Dogulas



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

--- In o...@yahoogroups.com, Andrew Porrett wrote:
>
> If the problem can be worked around by patching a byte or two in the
> .oex file, a simple patching program (DOS or Win32 .exe) could be
> written in a few minutes. I'd do it right now if the necessary
> information was made available. It would let those with deadlines
> get on with their work.
>

And apparently it only affects C.1.1 because V6 does produce proper
event code for B.2.2+. Or, at least it seems to work...

The real difference must be in the C.1.1 firmware for events and the
compiler output wasn't changed to accomodate the new firmware.

A workable solution would be to have the chips reprogrammed with
B.2.2+ firmware. There would be a mail turn-around issue. Perhaps
the chips could be ordered direct from SavageInnovations.

With Priority Mail, the delivery could be quite fast. I order from
Digikey in Minnesota on Monday and receive the parts in California on
Wednesday.

There is never anything good that comes from being on the bleeding
edge. If C.1.1 works today, it will still work in a year or so. Same
with V6 vs V5. There's no hurry to upgrade; better the devil you know...

Richard



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

Dennis,
I would be very interested in this hack. I have one C.1.1+, two
C.1.1's, one B.2.2+ and one B.2.2. I do not want to go into
developement of two different programs to be able to use C.1.1
chips. I have plenty of available object space to use an oRAM
object. The speed of looping through code to check for changes will
be too much time for my purposes as I am dealing with a good amount
of communication. Thanks in advance,
Bill
--- In o...@yahoogroups.com, ooPIC Tech Support
wrote:
>
> Brian,
>
> Not really - There is a memory slot after the event definition
that
> the compiler is not filling in. You could do a real hack using
oRAM to
> punch values into that slot, but it is not simple and it would suck
up a
> lot of object RAM to do it. If you are interested I can find the
> details of the hack, it hasn't been tested yet though.
>
> DLC
>
> Brian Lloyd wrote:
> > On Mar 4, 2008, at 9:58 PM, ooPIC Tech Support wrote:
> >
> >
> >> Brian,
> >>
> >> Events are broken in the compiler for C1.1+. The firmware is
fine,
> >> but the compiler needs rev'd. The developer is working on it
now.
> >>
> >
> > I know that it is broken but is it enough not-broken to allow
some
> > event programming to work if I hold my tongue just right? Or is
it so
> > broken that no amount of effort will make events work? I am
planning
> > to introduce events to two of my students today and I am hoping
that
> > they can get something simple to work even if I have to slap on
some
> > band-aids.
> >
> > Brian Lloyd
> > Granite Bay Montessori School 9330 Sierra College Bl
> > brian AT gbmontessori DOT com Roseville, CA 95661
> > +1.916.367.2131 (voice) +1.791.912.8170 (fax)
> >
> > PGP key ID: 12095C52A32A1B6C
> > PGP key fingerprint: 3B1D BA11 4913 3254 B6E0 CC09 1209 5C52
A32A 1B6C
> >
> >
> >
> >
> >
> >
> >
> >



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )