New feature in my ECM applet

[alpertron]

I added a feature to my factorization applet which can be useful for some people here. The applet now can execute for loops in order to factor several numbers without typing too much.

For example, entering the line

x=10^36;x=x+1;i-100;x

in the batch factorization box (scroll down the page in order to find it) and pressing "Factor expressions" you can see the factors of the numbers from 10^36 to 10^36+99. The syntax is explained in the same page.

Please let me know if it works for you or if you have further suggestions.

[henryzz]

Originally Posted by alpertron

I added a feature to my factorization applet which can be useful for some people here. The applet now can execute for loops in order to factor several numbers without typing too much.

For example, entering the line

x=10^36;x=x+1;i-100;x

in the batch factorization box (scroll down the page in order to find it) and pressing "Factor expressions" you can see the factors of the numbers from 10^36 to 10^36+99. The syntax is explained in the same page.

Please let me know if it works for you or if you have further suggestions.

i had noticed that
i looks good
would it be possible to show the form tested or factored in the results box
for example:
x=2;x=x+1;i-100;2^x-1
currently the output is:

3 = 3
7 = 7
15 = 3 * 5
31 = 31
63 = 3^2 * 7
127 = 127
255 = 3 * 5 * 17

what would be quite nice would be:

2^2-1 = 3 = 3
2^3-1 = 7 = 7
2^4-1 = 15 = 3 * 5
2^5-1 = 31 = 31
2^6-1 =  63 = 3^2 * 7
2^7-1 = 127 = 127
2^8-1 =  255 = 3 * 5 * 17

showing the expression used
my only reservation about that is that it would use more space

also something that has always bugged me is that you cant do M(5) instead of 2^5-1

you have done a brilliant job with your factoring applet to make it a convenient quick factoring tool thanks

[alpertron]

I optimized SIQS on the applet so now the number 10^59+213 is factored in 35 seconds in a Core 2 Duo 1.86 GHz. This includes 7 seconds of ECM. In the same machine msieve needs 9 seconds.

[MatWur-S530113]

Originally Posted by alpertron

I optimized SIQS on the applet so now the number 10^59+213 is factored in 35 seconds in a Core 2 Duo 1.86 GHz. This includes 7 seconds of ECM. In the same machine msieve needs 9 seconds.

Very good, thank you very much Dr. Alpern. The testet number needed 24sec with Core 2 Duo E6600 2.4GHz. Some other numbers I testet are much faster now, too. Keep up the excellent work!

Best regards,

Matthias

[alpertron]

Thanks, but I'm not a Doctor. Mathematics is a hobby for me.

[MatWur-S530113]

Originally Posted by alpertron

Thanks, but I'm not a Doctor. Mathematics is a hobby for me.

Oops, my fault, sorry. I already thought, that the applet is to good for a professional mathematican.

[alpertron]

Well, it appears that I'm listed at the Number Theory Web with other important people in the field as if I were a professional mathematician. So you were not the only one who did not know I'm amateur.

[R.D. Silverman]

Originally Posted by alpertron

Well, it appears that I'm listed at the Number Theory Web with other important people in the field as if I were a professional mathematician. So you were not the only one who did not know I'm amateur.

Interesting! Note that I am not listed!

[alpertron]

Originally Posted by R.D. Silverman

Interesting! Note that I am not listed!

Well, these pages are named Number Theorists' Home Pages/Departmental Listings. What is your Home Page URL?

[10metreh]

Originally Posted by R.D. Silverman

Interesting! Note that I am not listed!

I wonder what the reason for that was? You should certainly be there!

I guess you don't have enough Facebook pals, Bob - that makes you "unimportant"

That's discrimination.

[alpertron]

Going back to the thread topic, I uploaded a new version of the factorization applet, where the SIQS routine was optimized again.

                             10^59+213                    10^71-1
                       ECM   SIQS   Total           ECM    SIQS   Total
Jan 29th version       7s     28s    35s           2m48s   4m57s  7m45s
Feb 3rd version        7s     22s    29s           2m48s   4m3s   6m51s

These timings were done with a Core 2 Duo 1.86 GHz computer.

There is still more room for optimizations on the SIQS algorithm.