Re: Q:prg slowing down-why?
[Prev][Next][Index][Thread]
TI-Basic programs slow down as free memory decreases, it's because of the
way variables are allocated in the lookup table (VAT - Variable Allocation
Table). It's similar to the way if you have a lot of files in one directory
on a PC, it takes longer to access them....
If you want more info on this, let me know and I'll post more...
At 06:01 8/27/96 GMT, you wrote:
>Subject: Re: Q:prg slowing down-why?
>Newsgroups: bit.listserv.calc-ti
>References: <32135AC4.1F3F@hrz.tu-chemnitz.de>
><4vnl90$6dq@newsbf02.news.aol.com>
>Distribution:
>
>CalFolds (calfolds@aol.com) wrote:
>: There are a few resonse why your calc slows down. 1 : The longer a
>: program is the slower it gets because the processor has more instructions
>: then it can handle so the program slows down waiting for processor speed
>: sence the OS takes up so much processor time the only way around this is
>: to use asmbly (Wich does'nt all way help!) or optimise your code to be
>: smaller which will greatly increase the speed of you prog. 2 : The lore
>: memory you use the slower it gets. same thing with a PC or Mac. If you
>: have 16 megs of mem youe system would be much faster than say one with 8
>: or 4 megs. and sence the calc uses the same memory for storage and the
>: execuction it'll slow down untill you run out of mem where the calc will
>: slow down to unbearabl speeds. 3 : As the bateries grow weaker there will
>: be less power going to the processor and the processor will slow down
>: because of a lack of power.
>
>: Hope that helped
>: Cal
>: P.S. I'm pretty sure this is correct
>
>Mmm....
>
>1) The longer the program, the more time CPU uses to find sub-routines if
>it searces them from the beginning of program. To minimize the speed loss
>caused by this, place all frequently called sub-routines to beginning
>of program. This hint I've read as a hint of C64 BASIC programming.
>
>2) Speed loss of Mac or PC, when there's no free RAM left, is caused by
>use of virtual memory. Writing/reading to/from disk is slower than
>to/from RAM. TI doesn't have virtual memory, so, amount of free memory
>shouldn't affect to calc's speed. However, as I tested my TI-85 with my
>own speed test program, speed with only that basic test program in memory
>was, say, 1.00, with memory almost full, about 2 kB left, speed was about
>0.80. Slowing down has also been noticed when calculating
>time-consuming calculations. Speed loss could be explained, if one knew
>TI's BASIC interpreter's way of functioning down to the smallest detail.
>
>3) TI's CPU's clock generator is a RC (resistor-capacitor) one, that
>varies its speed while operating voltage varies. CPU's operating voltage
>itself doesn't effect on speed.
>
>-A
>
>--
>* archy@rhea.otol.fi
> Vintousin kdyttvd, osa 1:
>DLDKD _KOSKE_ SIIHEN ENTERIIN!!! EIEIEIEIEI!! .... Noniin, pitihdn sitd
>kuitenkin painaa. Otahan se uudestaan auki ja tee kaikki alusta.
>
>
+--------------------------------------------------------+
| Peter Kolbus +1.810.474.0872 |
| ae607@detroit.freenet.org #include <disclaimer.h> |
+--------------------------------------------------------+
| "640K should be enough for anybody" |
| -Bill Gates, 1986 |
| |
| Who is General Failure, and why is he reading drive C? |
| |
| ***Written using 100% recycled electrons*** |
+--------------------------------------------------------+
References: