A Chess forum. ChessBanter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » ChessBanter forum » Chess Newsgroups » rec.games.chess.computer (Computer Chess)
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

Junior 8 - Fritz 8 Complete! 56% for Junior 8



 
 
Thread Tools Display Modes
  #1  
Old July 13th 03, 10:38 AM
Voice of the People
external usenet poster
 
Posts: n/a
Default Junior 8 - Fritz 8 Complete! 56% for Junior 8

Forgive me, I've found the link!


"Philip" wrote in message
...
Great match!


"inTA" [email protected] x=anson y=daylightz.fsnet.co.uk wrote in message
...
Junior 8 wins the 50 Game Match with a score of 28/50.

Time Control:

40/2 hours + 20/1 hour + 30mins rest:
Hardwa Dual AMD 2400+ 1 cpu each
128 MB Hash each.
Permanent Brain on.
Own books with learning.
Complete Set of 3, 4, 5 Piece Tablebases

Current Sco

Junior 8 Test Summer 2003

Junior 8 - Fritz 8 (28-22) +17/-11/=22 (56%) +41 elo

(Btw the Junior 8 - Shredder 7.04 match also ended with 56% for Junior

8)

PGN of the Games and of the ones vs Shredder 7.04 can be obtained from
http://mysite.freeserve.com/intagran...est/index.html






  #2  
Old July 13th 03, 12:09 PM
Scott Hedrick
external usenet poster
 
Posts: n/a
Default Junior 8 - Fritz 8 Complete! 56% for Junior 8

Was the test conducted using the deep junior engine, or with the 'regular'
junior engine? I understand that deep junior is optimized for multiple CPUs,
but it is also supposed to be stronger than junior when on a single CPU.

I would be interested to know which one this is, and if its junior I would
love to see how deep junior fares against shredder 7.04.

Scott

"inTA" [email protected] x=anson y=daylightz.fsnet.co.uk wrote in message
...
Junior 8 wins the 50 Game Match with a score of 28/50.

Time Control:

40/2 hours + 20/1 hour + 30mins rest:
Hardwa Dual AMD 2400+ 1 cpu each
128 MB Hash each.
Permanent Brain on.
Own books with learning.
Complete Set of 3, 4, 5 Piece Tablebases

Current Sco

Junior 8 Test Summer 2003

Junior 8 - Fritz 8 (28-22) +17/-11/=22 (56%) +41 elo

(Btw the Junior 8 - Shredder 7.04 match also ended with 56% for Junior 8)

PGN of the Games and of the ones vs Shredder 7.04 can be obtained from
http://mysite.freeserve.com/intagran...est/index.html




  #3  
Old July 13th 03, 01:47 PM
inTA
external usenet poster
 
Posts: n/a
Default Junior 8 -Shredder 7.04 ended 56% for Junior 8: Testing Hiarcs 8 + h8hbk.ctg now.

I am using the Junior 8 engine (regular). I have already tested Junior 8 vs
Shredder 7.04 and Junior 8 won with the same score of 56% the games can be
obtained from the same link below.

http://mysite.freeserve.com/intagran...est/index.html

I have started testing Junior 8 - Hiarcs 8 now. Games can be seen at
http://mysite.freeserve.com/intagrand/junior8.htm

regards

"Scott Hedrick" wrote in message
...
Was the test conducted using the deep junior engine, or with the 'regular'
junior engine? I understand that deep junior is optimized for multiple

CPUs,
but it is also supposed to be stronger than junior when on a single CPU.

I would be interested to know which one this is, and if its junior I would
love to see how deep junior fares against shredder 7.04.

Scott

"inTA" [email protected] x=anson y=daylightz.fsnet.co.uk wrote in message
...
Junior 8 wins the 50 Game Match with a score of 28/50.

Time Control:

40/2 hours + 20/1 hour + 30mins rest:
Hardwa Dual AMD 2400+ 1 cpu each
128 MB Hash each.
Permanent Brain on.
Own books with learning.
Complete Set of 3, 4, 5 Piece Tablebases

Current Sco

Junior 8 Test Summer 2003

Junior 8 - Fritz 8 (28-22) +17/-11/=22 (56%) +41 elo

(Btw the Junior 8 - Shredder 7.04 match also ended with 56% for Junior

8)

PGN of the Games and of the ones vs Shredder 7.04 can be obtained from
http://mysite.freeserve.com/intagran...est/index.html






  #4  
Old July 14th 03, 09:38 AM
Sir Spock
external usenet poster
 
Posts: n/a
Default Junior 8 -Shredder 7.04 ended 56% for Junior 8: Testing Hiarcs 8 + h8hbk.ctg now.

The results would not be accurate if you play both engines on the same
computer. They can access each other's hash tables, and hence 'see' what
each other think..in a sense... no idea how this affects the results
though...i remember reading this somewhere, please correct me if i am wrong.


  #5  
Old July 14th 03, 11:02 AM
inTA
external usenet poster
 
Posts: n/a
Default Junior 8 -Shredder 7.04 ended 56% for Junior 8: Testing Hiarcs 8 + h8hbk.ctg now.

They can't access each others hash tables nor can they tell what the other
engine is "thinking".

Its a possiblity if the programmers wanted to be crafty but as far as I know
the results are fine.

inTA
"Sir Spock" wrote in message
...
The results would not be accurate if you play both engines on the same
computer. They can access each other's hash tables, and hence 'see' what
each other think..in a sense... no idea how this affects the results
though...i remember reading this somewhere, please correct me if i am

wrong.




  #6  
Old July 14th 03, 06:10 PM
qwerty
external usenet poster
 
Posts: n/a
Default Junior 8 -Shredder 7.04 ended 56% for Junior 8: Testing Hiarcs8 + h8hbk.ctg now.

inTA wrote:
They can't access each others hash tables nor can they tell what the other
engine is "thinking".

Its a possiblity if the programmers wanted to be crafty but as far as I know
the results are fine.


They can't access each other's hash tables, but the result is not fine
because you ran the tournament with permanent brain 'ON,' which means
that two CPUs competed for the CPU power and only god knows who was
better at dominating the CPU power. In other words, it was possible
that Junior got 60% of CPU power and only 40% for Fritz. In such case,
Junior has an advantage over Fritz.

  #7  
Old July 14th 03, 08:06 PM
inTA
external usenet poster
 
Posts: n/a
Default Junior 8 -Shredder 7.04 ended 56% for Junior 8: Testing Hiarcs 8 + h8hbk.ctg now.

Why do people keep saying this?

I ran the tournament on a Dual Computer. I'll make it clear. The computer
has two processors inside (thats right 2 AMD 2400+ chips). Each engine has
access to the equivalent of one cpu. Its 100% fair. Theres no 60-40. Each
engine can only use up to 50% of the complete processing power (2 cpus) as I
speak Junior is out of book and my cpus are being utilised up to 50% because
Junior 8 can only use a single thread the other 50% will be used by Hiarcs 8
when it is its turn to think.

Your argument is correct when the computers have only 1 cpu, but when there
are two cpus present, Permanent Brain can be switched on without any
problems. I have been using dual machines for a long time and I make sure
the settings are fair. I wouldn't waste time to play 100 games at almost 6-7
hours per game with unfair settings.

Regards

inTA

"qwerty" wrote in message
...
inTA wrote:
They can't access each others hash tables nor can they tell what the

other
engine is "thinking".

Its a possiblity if the programmers wanted to be crafty but as far as I

know
the results are fine.


They can't access each other's hash tables, but the result is not fine
because you ran the tournament with permanent brain 'ON,' which means
that two CPUs competed for the CPU power and only god knows who was
better at dominating the CPU power. In other words, it was possible
that Junior got 60% of CPU power and only 40% for Fritz. In such case,
Junior has an advantage over Fritz.



  #8  
Old July 14th 03, 08:15 PM
inTA
external usenet poster
 
Posts: n/a
Default Dual CPUS Permanement Brain ~ Fritz Help file:

"It is possible to conduct autoplayer games on a dual system without any
relevant performance loss. Each program gets a full CPU and has access to
all the resources allocated to it. Permanent brain works perfectly."

Source: Fritz GUI Help file.

"inTA" [email protected] x=anson y=daylightz.fsnet.co.uk wrote in message
...
Why do people keep saying this?

I ran the tournament on a Dual Computer. I'll make it clear. The computer
has two processors inside (thats right 2 AMD 2400+ chips). Each engine has
access to the equivalent of one cpu. Its 100% fair. Theres no 60-40. Each
engine can only use up to 50% of the complete processing power (2 cpus) as

I
speak Junior is out of book and my cpus are being utilised up to 50%

because
Junior 8 can only use a single thread the other 50% will be used by Hiarcs

8
when it is its turn to think.

Your argument is correct when the computers have only 1 cpu, but when

there
are two cpus present, Permanent Brain can be switched on without any
problems. I have been using dual machines for a long time and I make sure
the settings are fair. I wouldn't waste time to play 100 games at almost

6-7
hours per game with unfair settings.

Regards

inTA

"qwerty" wrote in message
...
inTA wrote:
They can't access each others hash tables nor can they tell what the

other
engine is "thinking".

Its a possiblity if the programmers wanted to be crafty but as far as

I
know
the results are fine.


They can't access each other's hash tables, but the result is not fine
because you ran the tournament with permanent brain 'ON,' which means
that two CPUs competed for the CPU power and only god knows who was
better at dominating the CPU power. In other words, it was possible
that Junior got 60% of CPU power and only 40% for Fritz. In such case,
Junior has an advantage over Fritz.





  #9  
Old July 25th 03, 11:22 AM
Emanuele
external usenet poster
 
Posts: n/a
Default Dual CPUS Permanement Brain ~ Fritz Help file:

Hi,
What you say is correct only if the programs are running only one thread a time, if the programs are running more that one
thread at a time nobody can say if it's fair. I don't know how the junior and Fritz code work.
I think they run only one thread and then it's fair, but noone can say it's true apart the authors.
What you paste from the Fritz Help file indicate that probably Fritz use only one thread at a time (or maybe one more thread for the
GUI but this thread don't make really difference).
If a program is developed thinking about multiprocessor computer than it's not fair, because it goes trying to obtain maximal power
from all cpus (in example with some multithreading strategy).
I don't know how the program you used for your prove work, probably they only use one thread at a time to calculate and then it's
fair ... but to tell this you have to be sure that the program are using only one thread.

Have a good day.
Emanuele.


"inTA" [email protected] x=anson y=daylightz.fsnet.co.uk ha scritto nel messaggio ...
"It is possible to conduct autoplayer games on a dual system without any
relevant performance loss. Each program gets a full CPU and has access to
all the resources allocated to it. Permanent brain works perfectly."

Source: Fritz GUI Help file.

"inTA" [email protected] x=anson y=daylightz.fsnet.co.uk wrote in message
...
Why do people keep saying this?

I ran the tournament on a Dual Computer. I'll make it clear. The computer
has two processors inside (thats right 2 AMD 2400+ chips). Each engine has
access to the equivalent of one cpu. Its 100% fair. Theres no 60-40. Each
engine can only use up to 50% of the complete processing power (2 cpus) as

I
speak Junior is out of book and my cpus are being utilised up to 50%

because
Junior 8 can only use a single thread the other 50% will be used by Hiarcs

8
when it is its turn to think.

Your argument is correct when the computers have only 1 cpu, but when

there
are two cpus present, Permanent Brain can be switched on without any
problems. I have been using dual machines for a long time and I make sure
the settings are fair. I wouldn't waste time to play 100 games at almost

6-7
hours per game with unfair settings.

Regards

inTA

"qwerty" wrote in message
...
inTA wrote:



  #10  
Old July 26th 03, 10:10 AM
inTA
external usenet poster
 
Posts: n/a
Default Dual CPUS Permanement Brain ~ Fritz Help file:

Hi,

I know about this also :-) which is why I stated in the orignal post that
each engine used 1 CPU. Junior 8 and Hiarcs 8 can only use 1 thread. The
only multi-threaded engine I used in the test so far was Shredder 7.04 which
lost 22-28 to Junior 8 and I limited Shredder 7.04 to 1 thread.

Well the Junior 8 - Hiarcs 8 match is nearly over and its currently 32-15 in
favour of Junior 8. Some may question my setup after reading this result,
but seeing Hiarcs 8 get 178~250 kN/s in each position, I'm pretty happy that
its getting sufficient cpu power.

Thanks for pointing this out Emanuele. I have been using a dual machine with
Chessbase software for many years now I've become quite verse with the ins
and outs of the SMP setup options.

regards,

inTA
"Emanuele" wrote in message
.. .
Hi,
What you say is correct only if the programs are running only one

thread a time, if the programs are running more that one
thread at a time nobody can say if it's fair. I don't know how the junior

and Fritz code work.
I think they run only one thread and then it's fair, but noone can say

it's true apart the authors.
What you paste from the Fritz Help file indicate that probably Fritz use

only one thread at a time (or maybe one more thread for the
GUI but this thread don't make really difference).
If a program is developed thinking about multiprocessor computer than it's

not fair, because it goes trying to obtain maximal power
from all cpus (in example with some multithreading strategy).
I don't know how the program you used for your prove work, probably they

only use one thread at a time to calculate and then it's
fair ... but to tell this you have to be sure that the program are using

only one thread.

Have a good day.
Emanuele.


"inTA" [email protected] x=anson y=daylightz.fsnet.co.uk ha scritto nel messaggio

...
"It is possible to conduct autoplayer games on a dual system without any
relevant performance loss. Each program gets a full CPU and has access

to
all the resources allocated to it. Permanent brain works perfectly."

Source: Fritz GUI Help file.

"inTA" [email protected] x=anson y=daylightz.fsnet.co.uk wrote in message
...
Why do people keep saying this?

I ran the tournament on a Dual Computer. I'll make it clear. The

computer
has two processors inside (thats right 2 AMD 2400+ chips). Each engine

has
access to the equivalent of one cpu. Its 100% fair. Theres no 60-40.

Each
engine can only use up to 50% of the complete processing power (2

cpus) as
I
speak Junior is out of book and my cpus are being utilised up to 50%

because
Junior 8 can only use a single thread the other 50% will be used by

Hiarcs
8
when it is its turn to think.

Your argument is correct when the computers have only 1 cpu, but when

there
are two cpus present, Permanent Brain can be switched on without any
problems. I have been using dual machines for a long time and I make

sure
the settings are fair. I wouldn't waste time to play 100 games at

almost
6-7
hours per game with unfair settings.

Regards

inTA

"qwerty" wrote in message
...
inTA wrote:





 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT +1. The time now is 03:39 PM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.Content Relevant URLs by vBSEO 2.4.0
Copyright 2004-2017 ChessBanter.
The comments are property of their posters.