OpenEuphoria

Dredge benchmarks

• Posted by jbrown105 at speedymail.org Oct 14, 2002
• 350 views

I've renamed my rparse program to Dredge, and renamed rparse.ex to
rpcomp.ex I've haven't had time to add many changes yet, but I plan to add
these:

Refrenced types
Classes
Improved internal error handling (catch multiple declarations, handle
	problems where function and variable have same name, etc.)
Make Dredge programs more optimizable (i.e., improve alias handling,
	add native keyword to speed up variable handling, etc.)
Try-Catch Exceptions

Also, besides from being buggy, I've noticed that Dredge can be slow.
I'll try to optimize it, but I'll guarrenette nothing. (It'd be easier
to bloat the language by letting the programmer choose to optimize it
via keywords such as native, which allows sequence variables to be
accessed faster, at the cost of not being able to make pointers of the
variables.

jbrown

PS Here is the file for rparse benchmark comparisions:

Note that rparse is SLOW!!! Especially when compared to Euphoria. rparse is
EXTREMELY SLOW (this is due to the way variables are handled internally via
rparse).

For example, here is the sequence.ex benchmark under Euphoria:

                Euphoria Sequence Benchmarks

* Initializing a length-100 sequence ---> 475200 initializations per second

* Adding two length-100 sequences ---> 116466 sequence-adds per second

* Appending to a sequence ---> 3218333 appends per second

* Slicing a sequence ---> 1599166 slices per second

* Name look-up ---> 338400 look-ups per second

And this is the sequence.ex benchmark under rparse:

                Euphoria Sequence Benchmarks

* Initializing a length-100 sequence ---> 114190 initializations per second

* Adding two length-100 sequences ---> 54733 sequence-adds per second

* Appending to a sequence ---> 7807 appends per second

* Slicing a sequence ---> 8023 slices per second

* Name look-up ---> 295066 look-ups per second

A staggering difference!

Shell.ex benchmark under Euphoria:

shell sort benchmark ...
6993.3 sorts per second
{1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,
26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,
49,50}

and under rparse:

shell sort benchmark ...
  62.8 sorts per second
{1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,
26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,
49,50}

Finally, sieve.ex benchmark under Euphoria:

prime sieve benchmark ...
3816.7 sieves per second
   2   3   5   7  11  13  17  19  23  29  31  37  41  43  47  53  59  61  67
  71  73  79  83  89  97 101 103 107 109 113 127 131 137 139 149 151 157 163
 167 173 179 181 191 193 197 199 211 223 227 229 233 239 241 251 257 263 269
 271 277 281 283 293 307 311 313 317 331 337 347 349 353 359 367 373 379 383
 389 397 401 409 419 421 431 433 439 443 449 457 461 463 467 479 487 491 499
 503 509 521 523 541 547 557 563 569 571 577 587 593 599 601 607 613 617 619
 631 641 643 647 653 659 661 673 677 683 691 701 709 719 727 733 739 743 751
 757 761 769 773 787 797 809 811 821 823 827 829 839 853 857 859 863 877 881
 883 887 907 911 919 929 937 941 947 953 967 971 977 983 991 997

and under rparse:

prime sieve benchmark ...
  44.6 sieves per second
   2   3   5   7  11  13  17  19  23  29  31  37  41  43  47  53  59  61  67
  71  73  79  83  89  97 101 103 107 109 113 127 131 137 139 149 151 157 163
 167 173 179 181 191 193 197 199 211 223 227 229 233 239 241 251 257 263 269
 271 277 281 283 293 307 311 313 317 331 337 347 349 353 359 367 373 379 383
 389 397 401 409 419 421 431 433 439 443 449 457 461 463 467 479 487 491 499
 503 509 521 523 541 547 557 563 569 571 577 587 593 599 601 607 613 617 619
 631 641 643 647 653 659 661 673 677 683 691 701 709 719 727 733 739 743 751
 757 761 769 773 787 797 809 811 821 823 827 829 839 853 857 859 863 877 881
 883 887 907 911 919 929 937 941 947 953 967 971 977 983 991 997

as you can see, Euphoria is far better optimized for speed. (Note that Euphorian
benchmarks were run with rpex.ex unmodified - rparse has ref/alias/deref to
allow the programmer to optimize slices, and the native keyword will allow
you to bypass rparse's conversion of variables, thus letting you manipulate
it at max euphorian speed (after sacraficing the ability to make a pointer of
it).