

Alice32 tests, WITH 16K expander
For reproducible results, enter spaces exactly as shown!

new
print mem
Result: 23622

cls
new
forj=0to20000:next
Result: 19.16 sec
Result: 18.92 sec
(mc-10 is 20.71 sec)

cls
new
forj=0to60000:next
Result: 59.15 sec
Result: 59.23 sec

printexp(1)
Result: 2.71828183

print tan(exp(sqr(2)))
Result: 1.46440118

cls
new
forj=1to500:a=cos(tan(exp(sqr(2)))):next
Result: 68.86 sec
Result: 68.74 sec
print a
Result: .106194526
(mc-10 is 76.7 sec)

cls
new
forj=1to500:a=sin(log(sqr(int(3.8+2.4-0.05/2)))):next
Result: 56.03 sec
Result: 55.86 sec
printa
Result: .78075907
(mc-10 is 62.5 sec)

cls
new
forj=1to3:sound1,255:next
Result: 52.25 sec
Result: 52.18 sec

=======================================================
CLOADed programs


>>>>>>>>Benchmark
REM DERIVED FROM JIM GILBREATH, BYTE MAGAZINE
REM 1981 AND 1984
REM MODIFIED FOR 16K MICROCOMPUTER MICROSOFT BASIC
REM - NO TIMER. USE YOUR STOPWATCH
REM FOR REPRODUCIBLE RESULTS ENTER THE
REM PROGRAM EXACTLY AS SHOWN WITH SPACES

NEW

(turn computer on)
(cload)

25 S = 2048 : REM SEARCH UP TO S
30 DIM F(S+2)
40 FOR I = 0 TO S-1 : F(I) = 1 : NEXT I
50 C=0
60 FOR I = 0 TO S-1
70 IF F(I) = 0 THEN 110
80 P = I+I+3 : IF I+P > S THEN 100
90 FOR K = I+P TO S STEP P : F(K) = 0 : NEXT K
100 C = C+1
110 NEXT I
120 PRINT C; " PRIMES FOUND"
150 END

cls
run

Result: 40.98 sec
Result: 40.79 sec
 563 PRIMES FOUND



>>>>>>>>Quick test vblank
cls 40
10 s=48928:j=0
20 cls:print
25 print"reset z,n:set z,n"
30 for q=0 to 15
35 n1=0:z1=0:n2=0:z2=0
40 pokes+8,9*16+9:rem remain 0
45 pokes+8,9*16+q
50 forj=0to1999:k=peek(s)
60 if k=0 thenz1=z1+1
70 if k<>0 thenn1=n1+1
80 next
140 pokes+8,9*16+5:rem depend on vblank
145 pokes+8,9*16+q
150 forj=0to1999:k=peek(s)
160 if k=0 thenz2=z2+1
170 if k<>0 thenn2=n2+1
180 next
190 print q;"->";z1;n1;":";z2;n2
200 next

Result:
RESET Z,N:SET Z,N
 0 -> 2000  0 : 15  1985  no change
 1 -> 2000  0 : 15  1985  no change
 2 -> 2000  0 : 15  1985  no change
 3 -> 2000  0 : 12  1988  no change
 4 -> 16  1984 : 13 1987  follow vbl
 5 -> 16  1984 : 15 1985  follow vbl
 6 -> 17  1983 : 11 1989  follow vbl
 7 -> 16  1984 : 14 1986  follow vbl
 8 -> 2000  0 : 2000  0   force 0
 9 -> 2000  0 : 2000  0   force 0
 10 -> 2000  0 : 2000  0  force 0
 11 -> 2000  0 : 2000  0  force 0
 12 -> 16  1984 : 13 1987  follow vbl
 13 -> 17  1983 : 11 1989  follow vbl
 14 -> 17  1983 : 15 1985  follow vbl
 15 -> 14  1986 : 12 1988  follow vbl
Looks like vbl is holding 3-5 scanlines and not 2 as reported
Need more conclusive test
