| |
BGonline.org Forums
Nacbracs (explained in detail)
Posted By: Nack Ballard In Response To: Nacbracs (explained in detail) (Ian Shaw)
Date: Tuesday, 20 December 2011, at 6:25 p.m.
Nice to have you weigh in, Ian, and yes, an extra character is bearable. It's primarily a question of need.
The single sideways-ish S character ~ represents Snowie 3-ply. No other other ply on Snowie is referenced these days. Using a second character to convey that we mean 3-ply would be overkill for that bot.
In a similar vein, while I did create /|\ for GnuBG 2-, 3- and 4-ply, respectively (as you mentioned), people rarely reference the latter two. We have GnuBG 2-ply rollouts, but little or nothing for GnuBG 3-ply, and for 4-ply there is only the occasional evaluation reference (infrequent enough that I would write out "/ 4p" or "/ 4-ply" if / and \ were not convenient mirror images).
Currently, the only bot for which more than one ply level are routinely referenced is XG. I treat the six symbols <^>=+& as a spectrum of ply settings (3-ply, 4-ply, 5-ply, roller, roller plus and roller plus plus); the second three are not a smooth continuation of the first three (and perhaps 5-ply will get bypassed), but I do think of them as resembling higher plies.
You referred to <^> as "plies" and =+& as "settings" -- seemingly as if they are different aspects. Can you please elaborate?
Note that while X3 and X4 (and X5 if ever used) follow your proposed ply-number scheme, it would still be necessary to refer to X=, X+ and X& (or some substitute) as these "plies" are not numbered. They would just be the same but with a (redundant) X added.
Another issue (arguably only a minor drawback) is that XG1(') and XG2(") are version numbers. Eventually, Xavier will release XG3('"). At that point, X3 meaning XG 3-ply could be confused with version XG3. Likewise, X4 could ultimately be interpreted as version XG4 instead of XG 4-ply. The current scheme of symbols not only saves a character but also creates a separate recognition paradigm, thereby avoiding a numeric conflict.
I'm quite willing to revamp the symbol structure if and when we reach the point that adding a character fulfills a need and I can clearly see what extra-character scheme works efficiently. In the meantime, it seem's difficult to predict in what direction backgammon bot usage will evolve.
Nack
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.