I have 200 versions of IvanHoe. I plan on using the strongest and also the weakest version in my clone wars. 10 strong clones will be included. All will be plus 2900 elo. 10 top engines will be included.
5/3 time control / 1000 games.
Best,
Gerold.
Strongest version of IvanHoe???
Moderator: Ras
-
- Posts: 10121
- Joined: Thu Mar 09, 2006 12:57 am
- Location: van buren,missouri
-
- Posts: 786
- Joined: Sun Apr 16, 2006 9:09 pm
Re: Strongest version of IvanHoe???
Hi, Gerold.
In my tests the best versions of this engine are Ivanhoe 50kQ x64s by Quocvuong (in my 6 cores computer) and Ivanhoe B50kb by Peterpan (in my 4 cores computer) . But the difference with other compiles is not big. Other testers have obtained different results. I suggest you to try both compiles in your computer and decide which one is better under your settings.
Regards,
Tom.
In my tests the best versions of this engine are Ivanhoe 50kQ x64s by Quocvuong (in my 6 cores computer) and Ivanhoe B50kb by Peterpan (in my 4 cores computer) . But the difference with other compiles is not big. Other testers have obtained different results. I suggest you to try both compiles in your computer and decide which one is better under your settings.
Regards,
Tom.
-
- Posts: 10121
- Joined: Thu Mar 09, 2006 12:57 am
- Location: van buren,missouri
Re: Strongest version of IvanHoe???
Thanks Tom. I will include them in my testing. I know PP did some nice work. I have run most of his. It will take quite a while to run all these versions to get the best ones for the final testing.
Best,
Gerold.
Best,
Gerold.
-
- Posts: 4790
- Joined: Sat Mar 11, 2006 12:42 am
Re: Strongest version of IvanHoe???
gerold wrote:Thanks Tom. I will include them in my testing. I know PP did some nice work. I have run most of his. It will take quite a while to run all these versions to get the best ones for the final testing.
Best,
Gerold.
Maybe this will help. This was after a few thousand games at 40/4 repeating. All played the same opponents- but none played each other. I just don't believe there is anything to gain from that. And Peter Osterlund was kind enough to give a ranking to them for me. All were 64bit ,1 CPU. To arrive at these rankings took a couple months of steady testing.
Code: Select all
Ivanhoe Top 10
--------------
Rank Name Elo
1 Ivanhoe 46h x64 41
2 Ivanhoe B46fE.02 x64 28
3 Ivanhoe B46fE.01 x64 21
4 Ivanhoe B46e x64 18
5 Ivanhoe B46fC x64 10
6 Ivanhoe B49jAx64p 2
7 Ivanhoe B46a x64 0
8 Ivanhoe B47f0.2 x64 -6
9 Ivanhoe B50kBx64p -8
10 Ivanhoe B47e x64 -12
&( Ivanhoe-B PP-0.1 x64 -13
&( Ivanhoe B47cCx64 -16
This was before IvanHoe 50kQ x64 existed. As Tom said, it is quite strong. From my testing it may be the strongest in head-up competition. But it does not seem to do quite as well as 46h ag. a number of diff. opponents, ie in a RR. But again, seems to be awfully close- again as Tom said.
The only mistake in running the Ivanhoe versions, any of them- you DO NOT WANT TO MAKE: Always run 46h, 46e, 47c, etc. Do not run 9.46h, 9.46e, 9.47c, etc. They are okay, but they are all Franklin's compiles- and I can promise you each is weaker than the comparable one by PP. I have no idea where anyone came up with the ludicrous idea that Franklin's were the "official" release. There wasn't- and NEVER HAS BEEN an official release of any versions- nothing but modified source codes have been released. Anyone's compile is just as official as any other. All you want to be sure is you go with the strongest. My advice, for what it is worth- is take my advice.
Best,
george
-
- Posts: 1600
- Joined: Mon Feb 21, 2011 9:48 am
Re: Strongest version of IvanHoe???
If you are looking for sources, you will find the official version, with readme, compilers linux, windows compilers, Pdf, ect, ect.geots wrote: NEVER HAS BEEN an official release of any versions- nothing but modified source codes have been released.
Best,
george
No one who has compiled, but the officer
and sources.
http://chess.cygnitec.com/engine/ivanhoe/
-
- Posts: 4790
- Joined: Sat Mar 11, 2006 12:42 am
Re: Strongest version of IvanHoe- One Last Item!
Gerold, I am sure what I say now is going to be hotly disputed- but that is ok with me. I can live with it. Different people just have different ideas. Personally, I would never use 4 or 6 cores per engine when running blitz. Possibly I "might" at a control like 40/12 repeating. That is borderline to me. I would be much more comfortable in trusting the results running those numbers of cores at 40/20 repeating, or slower.
Best again,
george
Best again,
george
-
- Posts: 4790
- Joined: Sat Mar 11, 2006 12:42 am
Re: Strongest version of IvanHoe???
velmarin wrote:If you are looking for sources, you will find the official version, with readme, compilers linux, windows compilers, Pdf, ect, ect.geots wrote: NEVER HAS BEEN an official release of any versions- nothing but modified source codes have been released.
Best,
george
No one who has compiled, but the officer
and sources.
http://chess.cygnitec.com/engine/ivanhoe/
That was never my point, Velmarin. I don't care where he gets the version from- but KLO's compiles are not "official" compiles in the sense they are no more official than Izak's or anyone else's- including yours if you compiled the code.
gts
-
- Posts: 1600
- Joined: Mon Feb 21, 2011 9:48 am
Re: Strongest version of IvanHoe???
I'm not going to try to convince you.geots wrote: That was never my point, Velmarin. I don't care where he gets the version from- but KLO's compiles are not "official" compiles in the sense they are no more official than Izak's or anyone else's- including yours if you compiled the code.
gts
But it's official site,
There is little argument against it.
But as Ippolit is Public domain and free version, you can interpret it.
But that's the official site, there is no argument.
-
- Posts: 4790
- Joined: Sat Mar 11, 2006 12:42 am
Re: Strongest version of IvanHoe???
velmarin wrote:I'm not going to try to convince you.geots wrote: That was never my point, Velmarin. I don't care where he gets the version from- but KLO's compiles are not "official" compiles in the sense they are no more official than Izak's or anyone else's- including yours if you compiled the code.
gts
But it's official site,
There is little argument against it.
But as Ippolit is Public domain and free version, you can interpret it.
But that's the official site, there is no argument.
Velmarin- let me try to get you to see what I am saying. The real programmers may very well release their next version on the site you speak of. That very well may be the official site. However, on that official site where they release the next version- you cannot install and play what they release. ALL THEY RELEASE is a modified source code. It is left up to whoever chooses to compile the code. Without someone compiling it- it is just improved code. Izak's compiles are not the official compiles. KLO's compiles are not the official compiles. There ARE NO official compiles. If you compiled an Ivanhoe version from the next release- yours would be as official as anyone else's. What I said is that where they release the modified code doesn't make a dime's worth of difference to me- and has not one thing to do with what I am talking about.
Good night, or day- my friend,
I am heading to get some sleep.
george
-
- Posts: 10121
- Joined: Thu Mar 09, 2006 12:57 am
- Location: van buren,missouri
Re: Strongest version of IvanHoe???
Thanks for your reply.velmarin wrote:If you are looking for sources, you will find the official version, with readme, compilers linux, windows compilers, Pdf, ect, ect.geots wrote: NEVER HAS BEEN an official release of any versions- nothing but modified source codes have been released.
Best,
george
No one who has compiled, but the officer
and sources.
http://chess.cygnitec.com/engine/ivanhoe/