Returning to
printed.
(C) Nemo 1996
________________________________
"... The end of three months
sample came from Jonathan
Six students - all
Six Exiles, dragging
obtained a new country
idea: fly for rados
Tay flight.
But even easier for them to be
perform the most complex
figure, than to understand what
is hidden
sense of their exercises. "
Richard Bach, "Seagull
Jonathan Livingston. "
Judging by the reaction, the article on
turbirovanie Spectrum machines do not
gone unnoticed. Reviews a wide spectrum. From the simple-"What
was that?" to insinuating-caustic "every frog
my swamp praises. "Indeed it is. extoll its
porugivat products and competitors - it is quite normal
division of labor. For Users this is the only way to get
objective information. Attention
the same should be paid to the extent reasonable and justified
makes the case, and whether they are simply unfounded
statements and general comments.
Retreating somewhat to the side,
I note that the method DMMTS - up
four years of work (the first
Experiments with turbirovanie were still advertising in a book
edited by N. Rodionova "System software for ZX-Spectrum"
(1993), then the results obtained for various reasons no one
arranged). Along the way, to answer
question - why did this brake (ZXFormat N1). The device was
created for laboratory testing methods and degrees of influence
on the program turbirovanie compatibility. The journal material
fell almost by mistake, because of prevailing if the confusion.
However, some readers will frankly admit that
did not understand. The author sees in this a miss.
Pseudo-scientific exposition of the non-scientific topics
turned out to be too science-intensive. The author apologizes
for any inconvenience. Opportunities outlined below will be
clearer.
Leaving aside rassharkivaniya on diplomatic parquet
get down to business and try to
almost compare the actual
speed.
1. Take the watch Montana,
put them in the mode and timing
load the program "TESTSC".
2. Press key "4" (option
RAM SINGLE TEST), while
start the stopwatch.
3. Measure our run-time option to exit to main menu
turbo and normal mode.
4. Summing up.
KAY Scorpion
parameter
norm turbo norm turbo
time
satisfied-20.1 10.6 20.0 14.0
of, s
method M1 M1 DMMTS APOF.
KT 1.0 1.9 1.0 1.43
FT 3.5 7.0 3.5 7.0
Feff 3.34 6.34 3.34 4.77
Efficiency of pro-0.95 0.91 0.95 0.68
core processor
KT - factor turbirovanie;
FT - F clock;
Feff - F effective.
For those who do not quite understand the meaning of the
numbers in the graph K & T, represent them in the form of
relations:
1.9 = 1.47 +1.43
1.9 = 1.33 * 1.43
You can, of course, to pretend
that nothing happened, plugging their ears and bury our heads
in the sand. But it is not so clear is the pump, which are
presented to those same 43%. Neither give nor take - a scandal
in the noble family. So, as the article
again about why it happened.
Now I will explain why other
tests underestimated for KAY
turbirovanie factor. In KAY
all the space port serves neturbirovannym IORQ normal duration.
In other machines IORQ usually short.
This gives a certain winnings in
speed proportional
frequency of treatment to the ports. Strictly speaking, the
question was not worth would have a damn if it does not
Bus. To ensure reliable
service bus, you must have
Long IORQ. Hard to imagine that wants to install on
bus by a specific user or
firm. Mindful of the fact that the sandwich always falls butter
down, you can be sure that
will be chosen just deactivate the solution. To cite one
option. Popular 580 chips at a short IORQ begin
engage in sophisticated sabotage, showing that
signs of artificial intelligence. For example, the path of
conduct port 55 on the bus with a short IORQ prove inscrutable.
Further comments I think
unnecessary.
P.S. Firm Nemo is very flattered by the attention the
company paid Scorpion Concept KAY-bus, having realized her
half-substitute in their new boards. If we are not deprived of
this kind of attention and further, that with pleasure
flatter firm Scorpion, borrowed shadow monitor.
________________________________
