Speed Test

grsbanks
Posts: 1122
Joined: Tue Apr 25, 2017 11:23 am
Location: Preston, Lancs, UK
Contact:

Re: Speed Test

Post by grsbanks »

HPMike wrote:
Thu Oct 15, 2020 4:28 pm
Is this true for all modules, or only some ? From what I can tell, program POLY in module MATH works fine with the DM41X set to FAST mode.
So does any other.

Didier was only pointing out that even in FAST mode, the DM41X will always be slower than the DM42 because it is an emulator, not a simulator, which is what gets you the level of compatibility that the DM41X offers.
There are only 10 kinds of people in the world: those who understand binary and those who do not.
dlachieze
Posts: 613
Joined: Thu May 04, 2017 12:20 pm
Location: France

Re: Speed Test

Post by dlachieze »

grsbanks wrote:
Thu Oct 15, 2020 4:35 pm
Didier was only pointing out that even in FAST mode, the DM41X will always be slower than the DM42 because it is an emulator, not a simulator, which is what gets you the level of compatibility that the DM41X offers.
Exactly ;)
DM42: 00425 - DM41X: β00066 - WP43: 00042
User avatar
salvomic
Posts: 186
Joined: Sat Dec 30, 2017 10:09 am
Location: Ragusa, Sicily
Contact:

Re: Speed Test

Post by salvomic »

HPMike wrote:
Thu Oct 15, 2020 4:28 pm

Is this true for all modules, or only some ? From what I can tell, program POLY in module MATH works fine with the DM41X set to FAST mode.
also the routines for triangles (SSS, ASA...), complex root (Z↑1/N), INTG, INTEG from MATH and Advantage seem to work fine in FAST, and also the routines in Navigation.
Good.
∫aL√0mic (IT9CLU) :: DM42 (SN: 00881), DM41X (SN 00523), DM16, HP Prime, 50g, 41CX, 42s, 71b, 15C, 12C, 35s, WP34s -- Free42
amafan
Posts: 87
Joined: Tue Sep 29, 2020 11:40 pm

Re: Speed Test

Post by amafan »

salvomic wrote:
Thu Oct 15, 2020 3:49 pm
HPMike wrote:
Thu Oct 15, 2020 2:56 pm
...

The difference is even more dramatic if you run the DM41X in SLOW mode, the DM42 is almost 300 times faster !
about this, I tested a short integral from an old thread in HP Museum:
in DM41X with Advantage module create a function FX:

Code: Select all

LBL "FX"
1
X<>Y
-
SQRT
1/X
END
then (accuracy 0.01)

Code: Select all

FIX 02
ALPHA: FX
0 ENTER 1
XEQ INTEG
Slow: 01:34:00
Fast: 00:08:26

In DM42 (accuracy 0.01):

Code: Select all

00 { 19-Byte Prgm }
01▸LBL "FX"
02 MVAR "X"
03 RCL "X"
04 +/-
05 1
06 +
07 SQRT
08 1/X
09 RTN
10 END
then

Code: Select all

∫f(x) -> LLIM 0 ULIM 1 ACC 0.01, ∫
DM42: less then 1s
DM42 (acc 0.001): 4s

:-)

But we already knew that.
What is the result of the original HP41CX?
DM42 (#6476), DM41X (#458), DM15, 12 - HP 12, 17BII, 35(2), 45, HP 27S, 28S, 30B, 41CV, 41CX, 42S, 48S, 35S, HP10bII+, 12C PLAT
I may have a problem!
User avatar
salvomic
Posts: 186
Joined: Sat Dec 30, 2017 10:09 am
Location: Ragusa, Sicily
Contact:

Re: Speed Test

Post by salvomic »

amafan wrote:
Thu Oct 15, 2020 8:26 pm


What is the result of the original HP41CX?
With FIX 2 -> ALPHA FX 0 ENTER 1 XRQ INTEG, in the HP41CX: 01:41:17 here. About as SLOW DM41X (01:34:00), only 3 s diff.

Salvo
∫aL√0mic (IT9CLU) :: DM42 (SN: 00881), DM41X (SN 00523), DM16, HP Prime, 50g, 41CX, 42s, 71b, 15C, 12C, 35s, WP34s -- Free42
User avatar
Walter
Posts: 3070
Joined: Tue May 02, 2017 11:13 am
Location: On a mission close to DRS, Germany

Re: Speed Test

Post by Walter »

salvomic wrote:
Thu Oct 15, 2020 8:44 pm
With FIX 2 -> ALPHA FX 0 ENTER 1 XRQ INTEG, in the HP41CX: 01:41:17 here. About as SLOW DM41X (01:34:00), only 3 s diff.
Challenged by elementary math? :?
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
User avatar
salvomic
Posts: 186
Joined: Sat Dec 30, 2017 10:09 am
Location: Ragusa, Sicily
Contact:

Re: Speed Test

Post by salvomic »

Walter wrote:
Thu Oct 15, 2020 11:39 pm
salvomic wrote:
Thu Oct 15, 2020 8:44 pm
With FIX 2 -> ALPHA FX 0 ENTER 1 XRQ INTEG, in the HP41CX: 01:41:17 here. About as SLOW DM41X (01:34:00), only 3 s diff.
Challenged by elementary math? :?
only for fun :)

cheers
∫aL√0mic (IT9CLU) :: DM42 (SN: 00881), DM41X (SN 00523), DM16, HP Prime, 50g, 41CX, 42s, 71b, 15C, 12C, 35s, WP34s -- Free42
User avatar
Walter
Posts: 3070
Joined: Tue May 02, 2017 11:13 am
Location: On a mission close to DRS, Germany

Re: Speed Test

Post by Walter »

salvomic wrote:
Fri Oct 16, 2020 10:07 am
Walter wrote:
Thu Oct 15, 2020 11:39 pm
salvomic wrote:
Thu Oct 15, 2020 8:44 pm
With FIX 2 -> ALPHA FX 0 ENTER 1 XRQ INTEG, in the HP41CX: 01:41:17 here. About as SLOW DM41X (01:34:00), only 3 s diff.
Challenged by elementary math? :?
only for fun :)

cheers
Well, but 7:17s > 3s with wine as well as without. ;)
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
User avatar
salvomic
Posts: 186
Joined: Sat Dec 30, 2017 10:09 am
Location: Ragusa, Sicily
Contact:

Re: Speed Test

Post by salvomic »

Walter wrote:
Fri Oct 16, 2020 1:12 pm

Well, but 7:17s > 3s with wine as well as without. ;)
it was a lapsus :)
∫aL√0mic (IT9CLU) :: DM42 (SN: 00881), DM41X (SN 00523), DM16, HP Prime, 50g, 41CX, 42s, 71b, 15C, 12C, 35s, WP34s -- Free42
Amphitryon

Re: Speed Test

Post by Amphitryon »

salvomic wrote:
Thu Oct 15, 2020 3:49 pm
[...] in DM41X with Advantage module create a function FX:

Code: Select all

LBL "FX"
1
X<>Y
-
SQRT
1/X
END
Where did you find this routine? Or is it a formula you found yourself? Lines 02..04 are for what? Just some burden for the speed test? They may be erased with no effect to the result regarding maths only. Mirroring at line x=1/2 does not alter the outcome.
But we already knew that.
Yes, sure we did. Anyway, on faster emulators like the "PC version of the HP 15C calculator" your benchmark is instantly finished. So I suggest to rewrite the routine, like this (HP15C-like):

Code: Select all

 1-42.21. 6  LBL 6
 2-   45  1  RCL 1
 3-      14  y^x
 4-   43 32  RTN
 5-42.21.11  LBL A
 6-       0  0
 7-      36  ENTER^
 8-       1  1
 9-42.20. 6  INTEG 6
10-42. 5. 0  DSE 0
11-   22 11  GTO A
Now put in R0 how often to repeate the iteration drudgery, and in R1 an adequate exponent, then excute A. For your original 1/X SQRT sequence set R1 = -0.5, where minus stands for 1/x and 0.5 for SQRT. (But we already knew that.)
Changing the exponent in R1 gives besides the requested accuracy of the integration an additional adjusting screw to put the equation towards a "pathological case" in respect to numerical integration. For example, for R1 = -0.8 the result should be 5. (In this case R0=0 is fine.)
R1 must be > -1. Set to a/b the result is b/(a+b). In the example above a=-8, b=10, so the integration returns about 10/(10-8) = 5.
Have fun.
Post Reply