The MASM Forum Archive 2004 to 2012

General Forums => The Laboratory => Topic started by: hutch-- on November 20, 2010, 10:07:22 AM

Title: Benchmark 4 unsigned to ascii algos.
Post by: hutch-- on November 20, 2010, 10:07:22 AM
This is the generated log text file content.


Results 8 pass average

timing utoa       291 ms
timing utoa_ex    91 ms
timing udw2str    568 ms
timing vcrt ustr$ 400 ms
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: frktons on November 20, 2010, 10:12:07 AM
When the CPU/OS version display? Are you planning to do it as well?

(http://www.masm32.com/board/index.php?action=dlattach;topic=15381.0;id=8484)

Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: hutch-- on November 20, 2010, 10:27:44 AM
The CPUID string and related info is not that important for this particular task, what I am inteested in is the timing of the 4 algos.

Same question again, what OS verion are you running, your display is excluding the later common control manifest.
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: frktons on November 20, 2010, 10:59:02 AM
Quote from: hutch-- on November 20, 2010, 10:27:44 AM
The CPUID string and related info is not that important for this particular task, what I am inteested in is the timing of the 4 algos.

Same question again, what OS verion are you running, your display is excluding the later common control manifest.

Same OS as I said before.
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: hutch-- on November 20, 2010, 11:40:04 AM
 :bg

I got it in the other post.
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: dedndave on November 20, 2010, 11:47:56 AM
prescott w/HTT, XP SP2 at the moment   :P
Results 8 pass average

timing utoa       542 ms
timing utoa_ex    197 ms
timing udw2str    1722 ms
timing vcrt ustr$ 1795 ms
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: dioxin on November 20, 2010, 01:45:26 PM
Phenom II x4 3GHz
Results 8 pass average

timing utoa       353 ms
timing utoa_ex    105 ms
timing udw2str    378 ms
timing vcrt ustr$ 744 ms
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: clive on November 20, 2010, 01:58:38 PM
AMD Athlon II X2 215, 2700 MHz, Win 7 x64 Home Premium

Results 8 pass average

timing utoa       390 ms
timing utoa_ex    123 ms
timing udw2str    425 ms
timing vcrt ustr$ 842 ms


Atom 270, 1600 MHz, XP Pro

Results 8 pass average

timing utoa       1464 ms
timing utoa_ex    537 ms
timing udw2str    1242 ms
timing vcrt ustr$ 3085 ms
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: FORTRANS on November 20, 2010, 03:01:33 PM
Hi,

   Windows 2000 and PIII.

Regards,

Steve N.

Results 8 pass average

timing utoa       1642 ms
timing utoa_ex    653 ms
timing udw2str    2482 ms
timing vcrt ustr$ 4543 ms
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: FORTRANS on November 20, 2010, 03:31:22 PM
Hi,

   JFF, Win98, P-MMX.  Took a while to run.

Cheers,

Steve

Results 8 pass average

timing utoa       14780 ms
timing utoa_ex    6983 ms
timing udw2str    38370 ms
timing vcrt ustr$ 41871 ms
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: jj2007 on November 20, 2010, 05:37:14 PM
Quote from: hutch-- on November 20, 2010, 10:07:22 AM
This is the generated log text file content.

Celeron M:

Results 8 pass average

timing utoa       679 ms
timing utoa_ex    302 ms
timing udw2str    1201 ms
timing vcrt ustr$ 1255 ms
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: hutch-- on November 20, 2010, 09:31:47 PM
gratsie all,

I think the algo that Paul Dixon wrote clearly has the legs so the strategy of adding two different algos to the library makes sense. I have verified both Ray's shorter algo and Paul's long one and they produce valid results for the entire DWORD range.
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: frktons on November 21, 2010, 06:31:20 PM
Quote from: hutch-- on November 20, 2010, 09:31:47 PM
gratsie all,

I think the algo that Paul Dixon wrote clearly has the legs so the strategy of adding two different algos to the library makes sense. I have verified both Ray's shorter algo and Paul's long one and they produce valid results for the entire DWORD range.

That's great  :U
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: hutch-- on November 21, 2010, 10:35:29 PM
I have found an unusual effect with the last algo that is in MSVCRT, there seems to be a reasonably large timing difference from one version of MSVCRT to another. Its at its fastest on my XP sp3 box, on the same box with a VM running Win2000 its noticably slower against the others and on my win7 box its slower on an i7 quad.
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: dedndave on November 21, 2010, 10:37:55 PM
may be due to added security to prevent buffer over-run
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: hutch-- on November 22, 2010, 01:51:21 AM
Yeah, its probably something like that or making it thread safe but it definitely justifies the "roll your own" approach.
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: dedndave on November 22, 2010, 02:04:47 AM
in truth, i would prefer using the masm32 lib functions over CRT....
1) we have control over when and how the code changes, and are aware of the precise implications
2) usually faster and better   :bg

i will say, the CRT is handy - sometimes you can grab a function without having to re-invent the wheel
Title: Re: Benchmark 4 unsigned to ascii algos.
Post by: frktons on November 22, 2010, 02:08:30 AM
Quote from: dedndave on November 22, 2010, 02:04:47 AM
in truth, i would prefer using the masm32 lib functions over CRT....
1) we have control over when and how the code changes, and are aware of the precise implications
2) usually faster and better   :bg

i will say, the CRT is handy - sometimes you can grab a function without having to re-invent the wheel

Agreed  :P