Hi,

As I say in "Compiler Options" topic : I'm trying to compile some src ...

I try to make some changes on toga 1.4 beta 5c

Now I have a good speed I can really test correctly the settings and code optimization ...

I'm trying to make a clean and easily readable code first ... (It's very long ... lol)
It will be useful for programming beginners (like me) ... I hope


In the future, my goal is to make my own engine (from A to Z) ... Maybe a dream ... ;-)

----------------------------------------------------------------------------------------

This compile is only a Beta 1.00 and has only very few changes to the original compile.

On my computer, it runs 1% slower than the original compile speed (in kn/s).
And the results seems roughly equals on my first tests.

Code:


Test 1 : 1"+0,01"
====================

Games Completed = 100 of 100 (Avg game length = 1.971 sec)
Settings = RR/32MB/1000ms+10ms/M 1000cp for 12 moves, D 150 moves/EPD:C:\Modules\LittleBlitzer-2.72\openings.epd(4000)
Time = 223 sec elapsed, 0 sec remaining

 1.  Toga 1.4 beta 5c           50.0/100   35-35-30     (L: m=26 t=0 i=0 a=9)   (D: r=21 i=8 f=1 s=0 a=0)   (tpm=16.2 d=6.6 nps=0)
 2.  Toga2012 Beta 1.00         50.0/100   35-35-30     (L: m=20 t=0 i=0 a=15)   (D: r=21 i=8 f=1 s=0 a=0)   (tpm=16.5 d=6.7 nps=0)

Code:


Test 2 : 0.5"+ 0,05"
====================

Games Completed = 100 of 100 (Avg game length = 4.840 sec)
Settings = RR/32MB/500ms+50ms/M 1000cp for 12 moves, D 150 moves/EPD:C:\Modules\LittleBlitzer-2.72\openings.epd(4000)
Time = 511 sec elapsed, 0 sec remaining

 1.  Toga 1.4 beta 5c           55.0/100   47-37-16     (L: m=23 t=0 i=0 a=14)   (D: r=10 i=5 f=0 s=0 a=1)   (tpm=39.5 d=8.9 nps=0)
 2.  Toga2012 Beta 1.00         45.0/100   37-47-16     (L: m=27 t=0 i=0 a=20)   (D: r=10 i=5 f=0 s=0 a=1)   (tpm=39.5 d=8.8 nps=0)

Code:


Test 3 : 1"+0,05"
====================

Games Completed = 100 of 100 (Avg game length = 5.923 sec)
Settings = RR/32MB/1000ms+50ms/M 1000cp for 12 moves, D 150 moves/EPD:C:\Modules\LittleBlitzer-2.72\openings.epd(4000)
Time = 619 sec elapsed, 0 sec remaining

 1.  Toga 1.4 beta 5c           41.5/100   26-43-31     (L: m=26 t=0 i=0 a=17)   (D: r=20 i=7 f=2 s=0 a=2)   (tpm=48.0 d=9.3 nps=0)
 2.  Toga2012 Beta 1.00         58.5/100   43-26-31     (L: m=16 t=0 i=0 a=10)   (D: r=20 i=7 f=2 s=0 a=2)   (tpm=48.0 d=9.1 nps=0)

Code:


Test 4 : 2"+0,05"
====================

Games Completed = 100 of 100 (Avg game length = 8.432 sec)
Settings = RR/32MB/2000ms+50ms/M 1000cp for 12 moves, D 150 moves/EPD:C:\Modules\LittleBlitzer-2.72\openings.epd(4000)
Time = 1463 sec elapsed, 0 sec remaining

 1.  Toga 1.4 beta 5c           54.0/100   39-31-30     (L: m=17 t=0 i=0 a=14)   (D: r=22 i=1 f=4 s=0 a=3)   (tpm=63.0 d=9.9 nps=0)
 2.  Toga2012 Beta 1.00         46.0/100   31-39-30     (L: m=31 t=0 i=0 a=8)   (D: r=22 i=1 f=4 s=0 a=3)   (tpm=62.9 d=9.7 nps=0)

Code:


Test 5 : 3'
====================

Fritz 11 GUI / TB5 on / HS7movesplus.ctg

Toga 2012 beta 1.00 x64  +28/-22/=50  53.00%
Toga II 1.4 beta5c        +22/-28/=50  47.00% 

I need now tests on other computers ...

It's why I'm searching for some nice testers ! ;-)

If you want to help me, please contact me ...


And don't forget : I'm a newbie in programming (last time I program was in 1995(and it was turbo pascal lol)) ... Please be cool ! ;-)

Good luck with your project!

I wish you the best of luck in your project and feel free to contact me if you need any help.

For me, the RobboBases are better than Nalimov-TB...

Toga 2012 Beta 1.03
-------------------------------

- One bug corrected in original code
- some functions optimized
- code cleaned slightly

Code:



Test 6 : 1"+0,01"
====================

Games Completed = 100 of 100 (Avg game length = 2.145 sec)
Settings = Gauntlet/32MB/1000ms+10ms/M 1000cp for 12 moves, D 150 moves/EPD:C:\Modules\LittleBlitzer-2.72\openings.epd(4000)
Time = 242 sec elapsed, 0 sec remaining

 1.  Toga 1.4 Beta 5c           48.5/100   31-34-35     (L: m=21 t=0 i=0 a=13)   (D: r=23 i=8 f=4 s=0 a=0)   (tpm=16.3 d=6.5 nps=0)
 2.  Toga2012 Beta 1.03 x64     51.5/100   34-31-35     (L: m=22 t=0 i=0 a=9)   (D: r=23 i=8 f=4 s=0 a=0)   (tpm=17.0 d=6.5 nps=0)



Code:


Test 7 : 10"
====================

Fritz 11 GUI / TB5 on / HS7movesplus.ctg

Toga 2012 beta 1.03 x64  +30/=45/-25  52.50%
Toga II 1.4 beta5c        +25/=45/-30  47.50%

I'm very slow for the moment but I understand better the engine ... ;-)

No real elo improvment at that time I think but more stable and less CPU cycles used ...

Thanks for the new version, Tennison. Is there a download link for this version?