Speed test using SB for various i-devices

[MarkP]

I have used the following VERY SIMPLE sB program to determine relative speeds between various iPhone, iPad, and iPod devices.

Let me know if you have additional results that are not in my table supplied below.

I have not been able to test iPhone 3G devices or earlier because they don't support iOS versions that support Smart Basic.

Code: Select all

mm=1e6
option base 1
SET UNDERGROUND ON
OPTION SCREENLOCK OFF
SET ORIENTATION TOP
SET OUTPUT FONT SIZE 17
again:
time reset
for i=1 to mm
j=i       'two simple assignments
l=i       'in a loop
next i

t=time()
kp=mm/t
tot+=kp
k+=1

print "########":kp;"#######":tot/k;k;t
goto again

' 1e6
/*
runs faster w/ wifi off than on ?
why is there variability?

    dev   ios     sB      speed
iph3gs   5      5.1    48.3k
ipad1  5.1.1  5.1    92k
iph4    7.1.2. 5.7     89k,104k,140k
iph4s  7.0.4  5.7     205k
ipad3 9.3.5  5.8     275k
iph5    6,8                412k
iph5s   10                1827k
iph6     10                 2000k
ipadA2  11               2094k
iphSE   10    5.8      2980k
ipadG5  11               2970k
iph8       11   5.8      5840k (ez,new)
*/

Here is the table:

dev ios sB speed
iph3gs 5 5.1 48.3k
ipad1 5.1.1 5.1 92k
iph4 7.1.2. 5.7 89k,104k,140k
iph4s 7.0.4 5.7 205k
ipad3 9.3.5 5.8 275k
iph5 6,8 412k
iph5s 10 1827k
iph6 10 2000k
ipadA2 11 2094k
iphSE 10 5.8 2980k
ipadG5 11 2970k
iph8 11 5.8 5840k (ez,new)

As noted in the code (and the table), I get some repeatable unexplained variability on some devices, as seen on the "iph4" row. I don't know what causes this significant variability.

Also, speed tests sometimes vary depending on whether wi-fi is on or off. I can't explain that either.

Of course, for a truly valid test, there should be no other concurrent active tasks on your device.

I haven't tested an iPhone X yet, but I expect that it will be the same as the iphone 8.

Cheers!
Mark

[rbytes]

Nice test, Mark. My iPad 5G got some speeds over 3000K.

It would be nice to see a running average and an average speed at the end of the test. Or is that what the second column is for?

[MarkP]

Each line of output represents 1e6 (one million) iterations. The 2nd column is the running average, the 3rd column is the time for the 1e6 iteration test.

Of course, I could have put in extra code to get max's and min's, but I didn't want to clutter the code.

[Dav]

NIce program. I have an iPad 2017 model. Here's the output for that:

2851855 2851855 1 0.350649
2962885 2907370 2 0.337509
2977493 2930744 3 0.335853
2964773 2939251 4 0.337294
2978566 2947114 5 0.335732
2975101 2951779 6 0.336123
2971106 2954540 7 0.336575
2974739 2957065 8 0.336164
2982173 2959854 9 0.335326
2979605 2961830 10 0.335615