Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Q&A with Feng-Hsiung Hsu

Author: Uri Blass

Date: 08:08:19 10/17/02

Go up one level in this thread


On October 17, 2002 at 10:46:42, Robert Hyatt wrote:

>On October 17, 2002 at 08:01:56, Uri Blass wrote:
>
>>On October 17, 2002 at 07:21:24, Jeremiah Penery wrote:
>>
>>>On October 17, 2002 at 07:04:42, Uri Blass wrote:
>>>
>>>>On October 17, 2002 at 06:46:50, Jeremiah Penery wrote:
>>>>
>>>>>On October 17, 2002 at 06:13:26, Johan Melin wrote:
>>>>>
>>>>>>If somebody tells you that "the storage capacity of this harddrive is 20 GB, 5
>>>>>>GB is just the linux partition", then what is the storage capacity? 25 GB?
>>>>>
>>>>>To use your hard drive example:  If 20(5) is the question, and I answer, "20GB
>>>>>total in terms of Windows.  5 is just the Linux partition," what would you make
>>>>>of that statement?  IMO, it's possible to interpret that in more than one way.
>>>>>The problem is that it's unclear what "in terms of Windows" means, just as it's
>>>>>unclear what "in terms of brute force" means in Hsu's statement.
>>>>
>>>>The difference is that windows and linux seem to contradict when there is no
>>>>contradiction between brute force and hardware.
>>>
>>>It does not say, "The Windows partition is 20GB," or "Windows says the disk is
>>>20GB, but 5GB is given to Linux." Just like Hsu doesn't say, "12 ply is the
>>>software partition," or "12 ply is the total (hardware+software) depth, but 6
>>>ply is given to the hardware."
>>>
>>>There is not a clear definition of what the phrase "in terms of Windows" means,
>>>just as we don't have a clear definition of what Hsu means when he says "in
>>>terms of brute force".
>>
>>We have a clear hint
>>
>>Copied from previous post:
>>
>>--------------------------------------------------------------------------------
>>CrazyBird(DM) kibitzes: 12(6) means 12 plies of brute force (not
>>counting the search extensions & quiescence).
>>CrazyBird(DM) kibitzes: 6 means the maximum hardware search depth
>>allowed.
>>CrazyBird(DM) kibitzes: this means that the PV could be up to 6 plies
>>deeper before quiescence.
>>-------------------------------------------------------------------------------
>>
>>Brute force means not counting search extensions and qsearch.
>
>How to conclude that?  Of course it doesn't count q-search because that is
>_heavily_
>based on forward pruning.  But not counting extensions?  We don't refer to
>anybody's
>12 ply search as "12 plies of brute force + extensions"...
>
>That kind of misuse of the term brute-force would not make it past any journal
>reviewer, because the term "brute-force" is very specific in the world of AI.
>
>
>
>>
>>In terms of brute force means without counting qsearch and extensions.
>
>Not _anywhere_ that I know of.  No book.  No Journal.  No Nothing...
>
>
>>The discussion was about the output so he also explained the importance of the 6
>>by the words that the pv could be up to 6 plies deeper before quiescence.
>>
>>It seems clear to me that he meant to compare the length of the real pv and the
>>length of the pv in the logfiles.
>
>Maybe, except for one thing.  "up to 6 plies deeper".  If the hardware searches
>6 plies,
>then that statement is _wrong_.  He would have said "at _least_ 6 plies deeper,
>since
>the hardware also does the normal extensions (singular extensions, recaptures,
>check,
>etc.).  So "up to 6" doesn't mean what you think...  Had he said "at least 6"
>your interpretation
>might be reasonable...

I do not know if the hardware also does the normal extensions but I did not read
his book.

Even in case that the hardware does normal extensions,
"at least" is wrong because the hardware did not search a constant number of
plies ,and there were cases that it searched less plies than the maximal plies.

Uri



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.