RE: A82: statram?
[Prev][Next][Index][Thread]
RE: A82: statram?
Well, it looks like STATVARS make themselves their own structure dynamically
in memory just as you run a regression.
I have crash loaded along with edit-82 and several other programs
{0,1,2,3,4}->L1
{5,6,7,8,9}->L2
my memory was at 11796 bytes at this point.
QuartReg
Now, I wind up with 11680 bytes. What the? Where did 116 bytes of my mem
disappear? If statram was static, then it wouldn't suddenly deplete 116
bytes.
So to see what's up I looked up the vat in edit-82.
Looking up VAT ram (around FE6E) seems to show that there is a list variable
(type 01) called 01 2E after the two variables 5D 00 (L1) and 5D 01 (L2),
and also a Yvar (type 03) variable called 62 01.
The variable 01 2E (stat list) has 00 06 entries, thus is 6 * 9 + 6 (vat) +
2 (size) = 62 bytes
It can be renamed as a list (such as L3, 5D 02) and you can view the list
just fine.
The variable 62 01 (RegEQ) is a Yvar string that is 00 2E long, so it is 46
+ 6 (vat) + 2 (size) = 54 bytes
It seems that "a" is actually simply stat(1), "b" is stat(2) etc. "n" is
stat(6) when you run QuartReg.
RegEQ is simply Rcl reg (which is that 62 01 variable).
This showed that STATRAM doesn't quite exist in the TI-82 as it does in the
TI-83.
The area of RAM from $868E which was believed to be STATRAM seems to be
window vars in the following order: (and it says there are 00 9A of them...
26 of them?? maybe it only looks at the lowest 7 bits of that)
Xmin, Xmax, Xscl, Ymin, Ymax, Yscl,
thetamin, thetamax, thetastep, Tmin, Tmax, Tstep,
nMin, nMax, UnStart, VnStart, nStart, ?, ?, XFact, YFact, ?, ?, ?, ?
This seems to be it, as the next entries seem to be pointer data. So that
was a total of 26 variables, which totals to 234 bytes not to mention it
would be annoying to some users if it were modified all the time.
Now, traveling up to $85F3 we seem to find another structure of data, but
this also seems to be another list of Z window vars. It is the exact same
thing as the list above, except that there seem to be only 25 entries, 225
bytes. This may also be annoying to some users if modified.
so far looking into more free mem, i haven't found any.
how about if the shell just supports a few pages at page boundaries? maybe
a k or two.
when i finish my current projects i'll run back to crash2000 and all will
be well for us.
-crashman
> -----Original Message-----
> From: owner-assembly-82@lists.ticalc.org
> [mailto:owner-assembly-82@lists.ticalc.org]On Behalf Of Eric Piel
> Sent: Friday, May 19, 2000 10:18 AM
> To: assembly-82@lists.ticalc.org
> Subject: Re: A82: statram?
>
>
>
> Well, first sorry to be a little bit slow to reply but I have exams :-)
>
> You are right stat, zoom and table vars seem to be manage the
> same way on all
> calcs. But THAT IS NOT what I'm talking about: The stat vars are
> the vars that
> can be found in VARS/Statistics and those ARE managed in a different way.
>
> On the 82 they are like any other vars (you even can read them
> when the flag of
> statvarsvalid is reset!) whereas on the 83(+) they are in a
> predefined area that
> can't move.
> So NO statram area in the 82
>
> Eric
>
> Note: All my observations have been made in VTI but I don't think
> it can change
> anything (just quicker)
>
>
>
> On Sun, 14 May 2000, Dines Justesen wrote:
>
> >
> > >From what I have seen in the disassembled ROM the stat vars
> are handled in
> > the same way as on the TI83. Investigations of backup files
> also seems to
> > show that the stat vars are not in the VAT.
> >
> > After finding the stuff in the ROM I set all stat, zoom and
> table vars to
> > known values and looked at the RAm contents of the calc
> afterwards. These
> > results matched the stuff found in the ROM and that is what the info in
> > 82ram is based on.
> >
> > Dines
> > -----------------------------------------------------------------------
> > Dines Justesen
> > "There are two major products that come out of Berkeley: LSD and UNIX.
> > We don't believe this to be a coincidence." -- Jeremy S. Anderson
> > Email: dines@aub.dk
> > WWW: www.student.dtu.dk/~c958362
> > -----------------------------------------------------------------------
> >
> > ----- Original Message -----
> > From: "Eric Piel" <pieleric@etu.utc.fr>
> > To: <assembly-82@lists.ticalc.org>
> > Sent: Friday, May 12, 2000 8:04 PM
> > Subject: Re: A82: statram?
> >
> >
> > >
> > > I'm sorry after more investigations I think I was wrong:
> > > In the TI82 the management of the statvars is differrent from
> the TI83.
> > The vars
> > > aren't somewhere precise and so a refernced in the VAT and
> don't take any
> > place
> > > when not used. So there isn't 531 free bytes in the 82.
> > >
> > > The data located at $868E are the window data and can't be
> (at least, I
> > don't
> > > think) overwritten. The error comes from the fact TI call them the
> > 'statvars',
> > > that's why I did this mistake.
> > >
> > > Sorry You'll have to find an other area to put your data...
> > >
> > > Eric
> > >
> > >
> > >
> > >
> > > On Thu, 11 May 2000, Eric Piel wrote:
> > >
> > > >
> > > > According the famous text of Mattias Lindqvist & Dines Justesen :
> > > > http://www.gbar.dtu.dk/~c958362/ash/text/82-ram.txt
> > > >
> > > > The equivalent data are at $868E but I'm not sure it is as
> safe as in
> > the 83
> > > > (well, try).
> > > >
> > > > Eric
> > > >
> > > > On Sat, 29 Apr 2000, Doug Torrance wrote:
> > > >
> > > > >
> > > > > Does anyone know what the 82 equivalent of the 83's statram (aka
> > saferam2,
> > > > > located at $858f) is?
> > > > >
> > > > > Thanks,
> > > > > Doug
> > > > >
> > ________________________________________________________________________
> > > > > Get Your Private, Free E-mail from MSN Hotmail at
> > http://www.hotmail.com
> > > > >
> > > > >
> > > >
> > > >
> > >
> > >
> >
> >
>
>
References: