BB(7) - Revision history

ADucharme: /* Top Halters */ add prurq halter to unique list, eliminate >20th rows

2026-11-09T18:11:25Z

Top Halters: add prurq halter to unique list, eliminate >20th rows

@@ Line 209: / Line 209: @@
 |-
 |{{TM|1RB1LA_1LC0RE_0LD1LB_1RD0LA_1RF0RA_0RG0LA_1RB1RZ|halt}}
 |10 ↑↑ 192.67
 |Terry Ligocki
@@ Line 259: / Line 247: @@
 |data-sort-value="10 ↑↑ 9998"|<math>2 \uparrow^{11} 2 \uparrow^{11} 3</math>
 |Pavel Kropitz
 |-
 |{{TM|1RB1RZ_0RC0RE_1LD1LA_1LC0LG_0RF1LF_0RD1LF_1LB0LE|halt}}
@@ Line 327: / Line 319: @@
 |10 ↑↑ 116.05
+|
 |}

ADucharme: /* Top Halters */ TMify

2026-11-09T18:09:02Z

Top Halters: TMify

← Older revision		Revision as of 18:09, 9 November 2026
Line 164:		Line 164:
	\|Pavel Kropitz		\|Pavel Kropitz
	\|-		\|-
	\|1RB1RC_1RC0LA_1RD0RB_1LE0RG_1LF1LE_1RD0LC_1RZ1RB		\|{{TM\|1RB1RC_1RC0LA_1RD0RB_1LE0RG_1LF1LE_1RD0LC_1RZ1RB\|halt}}
	\|10 ↑↑ 10^134,678,540		\|10 ↑↑ 10^134,678,540
	\|@prurq		\|@prurq

ADucharme: /* Top Halters */ add prurq 4th place halter

2026-11-09T18:07:12Z

Top Halters: add prurq 4th place halter

← Older revision		Revision as of 18:07, 9 November 2026
Line 163:		Line 163:
	\|data-sort-value="10 ↑↑ 9998"\|<math>2 \uparrow^{11} 2 \uparrow^{11} 3</math>		\|data-sort-value="10 ↑↑ 9998"\|<math>2 \uparrow^{11} 2 \uparrow^{11} 3</math>
	\|Pavel Kropitz		\|Pavel Kropitz
			\|-
			\|1RB1RC_1RC0LA_1RD0RB_1LE0RG_1LF1LE_1RD0LC_1RZ1RB
			\|10 ↑↑ 10^134,678,540
			\|@prurq
	\|-		\|-
	\|{{TM\|1RB1RZ_0RC0RE_1LD1LA_1LC0LG_0RF1LF_0RD1LF_1LB0LE\|halt}}		\|{{TM\|1RB1RZ_0RC0RE_1LD1LA_1LC0LG_0RF1LF_0RD1LF_1LB0LE\|halt}}

ADucharme: /* Phase 3 */ update gpu enumeration progress

2026-11-08T10:52:59Z

Phase 3: update gpu enumeration progress

← Older revision		Revision as of 10:52, 8 November 2026
Line 1,605:		Line 1,605:
	\|-		\|-
	\|268,680		\|268,680
	\|72,~~617~~		\|66,136
	\|72.92%		\|75.38%
	\|~~13.05~~ hours		\|~500 hours
	\| style="text-align:left" \|FAR RWL_mod maxT 100000 H [1-~~2] mod [1-6] n [1-10]~~		\| style="text-align:left" \|FAR RWL_mod maxT 100000 H [1-4,6,8,12,16] mod [1-6] n [1-10]
	~~FAR RWL_mod maxT 100000 H [3,~~4,6,8,12,16] mod 1 n [1-2]
	\|}		\|}
	Only 301 TMs found to halt by @finlay required enumeration, adding 8,428 holdouts. Reductions have been performed by Andrew Ducharme. Not all deciders between the point of enumeration and the current holdout list have been run yet.		Only 301 TMs found to halt by @finlay required enumeration, adding 8,428 holdouts. Reductions have been performed by Andrew Ducharme. Not all deciders between the point of enumeration and the current holdout list have been run yet.

ADucharme: /* Phase 3 */ crediting GameDungeon + decider runners

2026-11-07T16:35:56Z

Phase 3: crediting GameDungeon + decider runners

← Older revision		Revision as of 16:35, 7 November 2026
Line 1,588:		Line 1,588:
	\| ---		\| ---
	\|}		\|}
	Of the 870,000 some TMs found to halt by David SG, 12,378 had multiple undefined transitions. As each transition can represent one of 28 possible rules in BB(7), this meant a growth in the holdout list by 346,584 TMs. Roughly 22.5% were removed by the reverse engineer filter, i.e., were trivial to solve. Not all deciders between the point of enumeration and the current holdout list have been run yet.		Of the 870,000 some TMs found to halt by David SG, 12,378 had multiple undefined transitions. As each transition can represent one of 28 possible rules in BB(7), this meant a growth in the holdout list by 346,584 TMs. Enumeration would have taken significantly, significantly longer without the [https://discord.com/channels/960643023006490684/1369339127652159509/1566834448803045478 assistance of @GameDungeon]. Roughly 22.5% were removed by the reverse engineer filter, i.e., were trivial to solve. Reductions have been performed by Andrew Ducharme. Not all deciders between the point of enumeration and the current holdout list have been run yet.
	{\| class="wikitable sortable" style="text-align: right"		{\| class="wikitable sortable" style="text-align: right"
	! colspan="2" \|Holdout TMs		! colspan="2" \|Holdout TMs
Line 1,611:		Line 1,611:
	FAR RWL_mod maxT 100000 H [3,4,6,8,12,16] mod 1 n [1-2]		FAR RWL_mod maxT 100000 H [3,4,6,8,12,16] mod 1 n [1-2]
	\|}		\|}
	Only 301 TMs found to halt by @finlay required enumeration, adding 8,428 holdouts. Not all deciders between the point of enumeration and the current holdout list have been run yet.		Only 301 TMs found to halt by @finlay required enumeration, adding 8,428 holdouts. Reductions have been performed by Andrew Ducharme. Not all deciders between the point of enumeration and the current holdout list have been run yet.
	{\| class="wikitable sortable" style="text-align: right"		{\| class="wikitable sortable" style="text-align: right"
	! colspan="2" \|Holdout TMs		! colspan="2" \|Holdout TMs

ADucharme: /* Phase 3 */ Enumeration tables and explanations added

2026-11-07T16:20:54Z

Phase 3: Enumeration tables and explanations added

@@ Line 1,538: / Line 1,538: @@
 == Phase 3 ==
-Phase 3 is characterized by significant reductions hiding TMs improperly dropped from the holdout list. It began with @David SG using novel GPU-based direct simulation to eliminate halting TMs with step counts under 10^11. However, when a TM with multiple undefined transitions was found to halt, this decider did not enumerate new TNF children, causing the accepted number of holdouts (~16.9M) to be an undercount. Andrew Ducharme applied more mxdys C++ code to eliminate further non-halting TMs. @finlay ran the mxdys inductive decider on the generally accepted holdout list, eliminating a very impressive three million holdouts, a 19.4% reduction. This run also generated, but did not expand, new halting TMs with multiple undefined transitions. This finally triggered an effort reunify the holdout list.
+Phase 3 is characterized by significant reductions hiding TMs improperly dropped from the holdout list. It began with @David SG using novel GPU-based direct simulation to eliminate halting TMs with step counts under 10^12. However, when a TM with multiple undefined transitions was found to halt, this decider did not enumerate new TNF children, causing the accepted number of holdouts (~16.9M) to be an undercount. Andrew Ducharme applied more mxdys C++ code to eliminate further non-halting TMs. @finlay ran the mxdys inductive decider on the generally accepted holdout list, eliminating a very impressive three million holdouts, a 19.4% reduction. This run also generated, but did not expand, new halting TMs with multiple undefined transitions. This finally triggered an effort reunify the holdout list.
 The three main deciders were performed on three different computer clusters with different (and in the case of the GPU run, unknown) number of cores, so the TMs/sec/core columns have been removed. This brought the overall reduction in Phase 2 is so far 28.78%, from ~17.8M to ~12.7M TMs. The details are contained in this table:
@@ Line 1,587: / Line 1,587: @@
 | ---
 | ---
 |}
-==References==
+== References ==
 [[Category:BB Domains]][[Category:BB(7)]]
 <references />

ADucharme: Phase 3 summary text, updated main table

2026-11-07T16:03:12Z

Phase 3 summary text, updated main table

← Older revision		Revision as of 16:03, 7 November 2026
Line 1,538:		Line 1,538:

	== Phase 3 ==		== Phase 3 ==
			Phase 3 is characterized by significant reductions hiding TMs improperly dropped from the holdout list. It began with @David SG using novel GPU-based direct simulation to eliminate halting TMs with step counts under 10^11. However, when a TM with multiple undefined transitions was found to halt, this decider did not enumerate new TNF children, causing the accepted number of holdouts (~16.9M) to be an undercount. Andrew Ducharme applied more mxdys C++ code to eliminate further non-halting TMs. @finlay ran the mxdys inductive decider on the generally accepted holdout list, eliminating a very impressive three million holdouts, a 19.4% reduction. This run also generated, but did not expand, new halting TMs with multiple undefined transitions. This finally triggered an effort reunify the holdout list.

			The three main deciders were performed on three different computer clusters with different (and in the case of the GPU run, unknown) number of cores, so the TMs/sec/core columns have been removed. This brought the overall reduction in Phase 2 is so far 28.78%, from ~17.8M to ~12.7M TMs. The details are contained in this table:
	{\| class="wikitable sortable" style="text-align: right"		{\| class="wikitable sortable" style="text-align: right"
	! rowspan="2" \|Done by		! rowspan="2" \|Done by
	! colspan="2" \|Holdout TMs		! colspan="2" \|Holdout TMs
	! rowspan="2" \|~~<math>*^1</math>~~		! rowspan="2" \|% Reduced
	! rowspan="2" \|~~<math>*^2</math>~~		! rowspan="2" \|Compute Time
	~~! colspan="2" \|TMs/sec/~~core		(core-hours)
	! rowspan="2" \|Description		! rowspan="2" \|Description
	! rowspan="2" \|Source		! rowspan="2" \|Source
Line 1,550:		Line 1,553:
	!Input		!Input
	!Output		!Output
	~~!<math>*^3</math>~~
	~~!<math>*^4</math>~~
	\|-		\|-
	\| style="text-align:center" \|David SG		\| style="text-align:center" \|@David SG
	\|17,823,260		\|17,823,260
	\|16,949,557		\|16,949,557
	\|4.90%		\|4.90%
	\|~~TODO~~		\|Unknown
	\|		\|Direct simulation by custom GPU code
	\|
	~~\|TODO~~
	\|[https://discord.com/channels/960643023006490684/1369339127652159509/1512553945837605016 discord]		\|[https://discord.com/channels/960643023006490684/1369339127652159509/1512553945837605016 discord]
	\| rowspan="3" \|		\| rowspan="3" \|[https://drive.google.com/drive/folders/1YZNpzqhmVuopX5Qgz8vvcpnY1c7EXax4?usp=sharing Google Drive]
	\|-		\|-
	\| style="text-align:center" \|Andrew Ducharme		\| style="text-align:center" \|Andrew Ducharme
Line 1,568:		Line 1,567:
	\|15,743,264		\|15,743,264
	\|7.12%		\|7.12%
	\|~~TODO~~		\|~120,000
	\|
	\|
	\| style="text-align:left" \|FAR RWL_mod maxT 100000 H [1-4,6,8,12,16] mod [1-6] n [1-10]		\| style="text-align:left" \|FAR RWL_mod maxT 100000 H [1-4,6,8,12,16] mod [1-6] n [1-10]
	\|[https://discord.com/channels/960643023006490684/1369339127652159509/1516364912375627816 discord]		\|[https://discord.com/channels/960643023006490684/1369339127652159509/1516364912375627816 discord]
			\|-
			\|@finlay
			\|15,743,264
			\|12,693,129
			\|19.37%
			\|29,000
			\|decider --maxT 100000 --exploop --arithseq
			\|[https://discord.com/channels/960643023006490684/1369339127652159509/1563950872000467005 Discord]
	\|-		\|-
	\| style="text-align:center" \|'''Stage 7 Cumulative'''		\| style="text-align:center" \|'''Stage 7 Cumulative'''
	\|'''17,823,260'''		\|'''17,823,260'''
	\|'''15,~~743~~,~~264~~'''		\|'''12,693,129'''
	\|'''11.67%'''		\|'''28.78%'''
	\|'''~~TODO~~'''		\|'''>150,000'''
	~~\| ---~~
	\| ---		\| ---
	\| ---		\| ---

Polygon: /* Phase 2 */ updated text

2026-11-07T15:46:13Z

Phase 2: updated text

← Older revision		Revision as of 15:46, 7 November 2026
Line 705:		Line 705:
	Phase 2 began with Andrew Ducharme's initial decider runs (see Stage 1 in the table). The goal of Phase 2 is to run deciders on the current holdout list (the last in the table) and produce an accessible new holdout list which can then be added to the table. Input Holdout TMs may be unequal to the next row's Output Holdout TMs because more TMs are TNF-enumerated in the process of executing the next row's filtering. Accessing the holdout list can be done through the link "Holdouts" in the "Data" column. The link "Details" leads to additional data/information pertaining to each step. If there isn't a link there it means the data is still being put somewhere everyone can access it.		Phase 2 began with Andrew Ducharme's initial decider runs (see Stage 1 in the table). The goal of Phase 2 is to run deciders on the current holdout list (the last in the table) and produce an accessible new holdout list which can then be added to the table. Input Holdout TMs may be unequal to the next row's Output Holdout TMs because more TMs are TNF-enumerated in the process of executing the next row's filtering. Accessing the holdout list can be done through the link "Holdouts" in the "Data" column. The link "Details" leads to additional data/information pertaining to each step. If there isn't a link there it means the data is still being put somewhere everyone can access it.

	Stage 1 reduced the number of holdouts by 33.3% from ~86.1M to ~57.5M TMs using the Ligocki's C++/Python codes. In Stage 2, Andrew continued using mxdys' C++ code and reduced the number of holdouts by 51.0% from ~57.5M to ~28.2M TMs. Terry Ligocki then pushed this further using additional deciders/parameters to generate Stage 3 which reduced the number of holdouts by 17.3% from ~28.2M to ~23.3M TMs. Andrew then ran the Ligocki's C++/Python code with larger parameters which reduced the number of holdouts by 2.6% from ~23.3M to ~22.7M TMs. This formed Stage 4. Terry then switched back to the mxdys' C++ code with different deciders/parameters for Stage 5 where the number of holdouts was reduced by 10.2% from ~22.7M to ~20.4M TMs.		Stage 1 reduced the number of holdouts by 33.3% from ~86.1M to ~57.5M TMs using the Ligocki's C++/Python codes. In Stage 2, Andrew continued using mxdys' C++ code and reduced the number of holdouts by 51.0% from ~57.5M to ~28.2M TMs. Terry Ligocki then pushed this further using additional deciders/parameters to generate Stage 3 which reduced the number of holdouts by 17.3% from ~28.2M to ~23.3M TMs. Andrew then ran the Ligocki's C++/Python code with larger parameters which reduced the number of holdouts by 2.6% from ~23.3M to ~22.7M TMs. This formed Stage 4. Terry then switched back to the mxdys' C++ code with different deciders/parameters for Stage 5 where the number of holdouts was reduced by 10.2% from ~22.7M to ~20.4M TMs. Stage 6 was Andrew Ducharme running a mix of mxdys and Ligocki deciders, reducing holdouts by 12.7% from ~20.4M to ~17.8M TMs.

	This brought the overall reduction in Phase 2 to 77.0% from ~86.1M to ~19.8M TMs. The details are contained in this table:		This brought the overall reduction in Phase 2 to 79.3% from ~86.1M to ~17.8M TMs. The details are contained in this table:

	(done to reduce column size:		(done to reduce column size:

ADucharme: /* Phase 2 */ fix phase 2 table after Stage 7 removal

2026-11-07T15:42:17Z

Phase 2: fix phase 2 table after Stage 7 removal

← Older revision		Revision as of 15:42, 7 November 2026
Line 1,527:		Line 1,527:
	\|'''86,129,304'''		\|'''86,129,304'''
	\|'''17,823,260'''		\|'''17,823,260'''
	\|'''81.72%'''		\|'''79.31%'''
	\|'''93019.8'''		\|'''93019.8'''
	\| ---		\| ---
Line 1,533:		Line 1,533:
	\| ---		\| ---
	\| ---		\| ---
			\|
	\|-		\|-
	\|}		\|}

ADucharme: Creation of Phase 3 from former Phase 2, Stage 7. Existing decider information moved to new Phase 3 table

2026-11-07T15:39:42Z

Creation of Phase 3 from former Phase 2, Stage 7. Existing decider information moved to new Phase 3 table

@@ Line 1,522: / Line 1,522: @@
+|
 |-
-| colspan="20" style="text-align:center"|'''Stage 7'''
+| colspan="20" style="text-align:center"|'''Overall'''
 |-
-|style="text-align:center" |David SG
+|}
 |17,823,260
 |16,949,557
@@ Line 1,533: / Line 1,561: @@
 |TODO
 |[https://discord.com/channels/960643023006490684/1369339127652159509/1512553945837605016 discord]
 |-
-|style="text-align:center" |Andrew Ducharme
+| style="text-align:center" |Andrew Ducharme
 |16,949,557
 |15,743,264
@@ Line 1,541: / Line 1,570: @@
+|
+|
-|style="text-align:left" |FAR RWL_mod maxT 100000 H [1-4,6,8,12,16] mod [1-6] n [1-10]
+| style="text-align:left" |FAR RWL_mod maxT 100000 H [1-4,6,8,12,16] mod [1-6] n [1-10]
 |[https://discord.com/channels/960643023006490684/1369339127652159509/1516364912375627816 discord]
 |-
@@ Line 1,553: / Line 1,582: @@
 | ---
 | ---
 |}