Shmuel

> >... at least I think that's what you are claiming. This is
> >interesting since using such devices - if anyone still has them - or,
> >I would expect, later generations of 3270 displays which had a fancy
> >name I can't remember - was it something weird like InfoTerminal or
> >InfoSystem? - it would be possible to exercise Edward Jaffe's
> >"enhancement" using these devices.
>
> No, because the sizes were fixed during the BIND process.

I think here you are saying "No" because the dimensions can only be "set"
using the BIND and cannot be "configured" using the device "configuration"
possibilities. (See the earlier post concerning the use of the words "set"
and "configure".)

> You'd need
> to enhance the data stream beyond what those devices supported.

You lost me here.

>  I
> believe that they term you are looking for is InfoWindow and, yes,
> they could be configured to support a BIND with primary other than
> 24x80.

Thanks, At least I got the "Info" part right. Probably I have a blank spot
where the word "window" belongs - understandable, don't you think?

> Note, however, the reference to 14-bit
> addressing; you can't run an application that uses 12-bit addressing
> if the screen size exceeds 4KiB.

That's a good point about the 12-bit addressing. I expect this is in
reference to having default presentation space dimensions where the number
of rows is set to more that 24 but the application knows only to use 24 -
and uses 12-bit addressing. There is another requirement - most unlikely not
to be met but possibly not - which is that the application mustn't rely on
"wrapping" at the 1919 address. Come to think of it I think I relied on this
once in an USS (at least I don't have to explain which USS that is to you)
message 10 since I wanted to use the whole of the remainder of row 24 for
input data. I relied on an attribute byte with the "protected" bit at
address 0 (row 1, column 1) to be the delimiter for the field.

> >On the other hand I can't make sense of "No, I was talking about the
> >dimensions on the 3275 and 3277, which didn't have EWA or Read
> >Partition Query."
>
> You made a general statement about EWA on the 3270, and it was only
> valid for the 3278 and later.

Ok, this is my fault, clearly I have to be ultraprecise in my referencing or
the faintest chance to misunderstand will be taken. The sentence upon which
I was commenting was "In particular, the 3180 and the 3192 work just fine
with a primary size of 43x80.", to which I replied initially "I believe you
are talking about dimensions you set using the X'7F' code in the penultimate
byte of the PSERVIC operand.", to which you replied "No, I was talking about
the dimensions on the 3275 and 3277, which didn't have EWA or Read Partition
Query." at which point I was lost ...

> >[1] My presentation notes say that the 3179, specifically the 3179G,
> >does not support the use of the "unspecified screen size" BIND.
>
> The 3179 and the 3179G are very different beasts. My recollection is
> that the 3179 was intended strictly as a 3278-2 equivalent and
> supported only 24x80 for both primary and alternate. The 3179G was an
> APA graphics terminal.

Thanks for the clarification concerning the 3179. I think I only ever saw
the G-flavour. It's odd that it had the lack of support for the 03 option.

Chris Mason

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

Reply via email to