Mike's Original Base Proposal
Frank,
This was a ton more work than I thought it would be but here
are the preliminary results.
OUTLINE:
SECTION 1: Table of the best times on cdrom 5/19/96 and
comments form their text file if any.
SECTION 2: Proposed new base provided in your note.
SECTION 3: The differentials applied to the current
proposed base.
SECTION 4: How the base could be divided based on the
numerical analysis alone.
SECTION 5: My comments.
---------------------------------------------------------------------------
***************************************************************************
---------------------------------------------------------------------------
SECTION 1:
If the time from COMPET-N is faster for Ultra Violence it is also
included. For maps that were not done by DHT applicants the
COMPET-N time is used as a base. The persons comments (if any)
are included as (xxxxx?) and my evaluation of it as easy, medium,
hard, etc. The differential is: ((par-time)/par)*100
MAP/PAR Times Person/Comments
---------------------------------------------------------------------------
MAP01/0:30 COMPET-N-0:06 syw (80.0% differential)
MAP02/1:30 COMPET-N-0:57 ddl (36.7% differential)
---------------------------------------------------------------------------
MAP03/2:00 jbh-1:48 jbh-easy (pretty simple?)
twm-1:40 twm-easiest (VERY easy?)
mgp-1:27
COMPET-N-0:30 Michael Kren (75.0% differential)
33 sec. differential, 27.5%
---------------------------------------------------------------------------
MAP04/2:00 nbc-1:55 nbc-medium (not as easy as you would think?)
jnm-1:48
ajs-1:47 ajs-medium (a bit hard?)
jrh-1:36
aak-1:32 aak-easy
yzd-1:16 yzd-easy
COMPET-N-0:48 ddl (60.0% differential)
44 sec. differential, 36.6%
---------------------------------------------------------------------------
MAP05/1:30 COMPET-N-0:31 ddl (65.5% differential)
---------------------------------------------------------------------------
MAP06/2:30 trg-2:13 trg-easy
jeh-1:36
COMPET-N-1:02 ddl (58.6% differential)
54 sec. differential, 36.0%
---------------------------------------------------------------------------
MAP07/2:00 jle-1:38 tmd-easy
mjf-1:37
wwc-1:04 wwc-easy (simple?)
tmd-0:59
gib-0:56 gib-easy (all quite easy?)
COMPET-N-1:01 ddl (49.2% differential)
64 sec. differential, 53.3%
---------------------------------------------------------------------------
MAP08/2:00 pnb-1:48 pnb-easy (easy/medium?)
aak-0:57 aak-easiest
jbh-0:56 jbh-easy (not very hard?)
wwc-0:55 wwc-easiest
gib-0:41 gib-easiest
COMPET-N-0:33 kuh (72.5% differential)
79 sec. differential, 65.8%
---------------------------------------------------------------------------
MAP09/4:30 aak-3:07 aak-medium (hardest of relatively easy set?)
gib-3:01 gib-easy (all quite easy?)
COMPET-N-2:02 ddl (58.8% differential)
89 sec. differential, 33.0%
---------------------------------------------------------------------------
MAP10/1:30 mjf-1:06 mjf-medium
pnb-1:03 pnb-easy (easy/medium?)
COMPET-N-0:42 aak (53.3% differential)
27 sec. differential, 30.0%
---------------------------------------------------------------------------
MAP11/3:30 ajs-2:42
COMPET-N-1:10 ddl (66.7% differential)
48 sec. differential, 22.9%
---------------------------------------------------------------------------
MAP12/2:30 trg-2:21 trg-easy
ato-1:59 atp-medium (somewhat difficult?)
jeh-1:49
COMPET-N-0:58 ddl (61.3% differential)
41 sec. differential, 27.3%
---------------------------------------------------------------------------
MAP13/2:30 COMPET-N-1:15 kuh (50.0% differential)
MAP14/2:30 COMPET-N-1:10 ddl (53.3% differential)
---------------------------------------------------------------------------
MAP15/3:30 jrh-2:47
mgp-1:29
tmd-1:07 tmd-easy
COMPET-N-0:44 ddl (79.0% differential)
143 sec. differential, 68.0%
---------------------------------------------------------------------------
MAP16/2:30 jle-2:04
mgp-2:02
COMPET-N-0:46 ddl (69.3% differential)
27 sec. differential, 18.0%
---------------------------------------------------------------------------
MAP17/7:00 jag-6:39
jbh-6:00 jbh-hard (I took lisence here - mgp)
wwc-5:54
COMPET-N-4:25 ddl (36.9% differential)
66 sec. differential, 15.7%
---------------------------------------------------------------------------
MAP18/2:30 avc-1:24 avc-easiest (too easy?)
jag-1:14
yzd-0:56 yzd-easy
smt-0:43 smt-easy
COMPET-N-0:27 Michael Kren (82.0% differential)
107 sec. differential, 71.3%
---------------------------------------------------------------------------
MAP19/3:30 ie-2:35 ie-hardest (super tough!?)
ato-2:17 ato-medium (fairly easy?)
jrh-2:09
COMPET-N-1:17 ddl (63.3% differential)
81 sec. differential, 38.6%
---------------------------------------------------------------------------
MAP20/2:30 jle-2:10
COMPET-N-0:48 ddl (68.0% differential)
20 sec. differential 13.3%
---------------------------------------------------------------------------
MAP21/4:00 smt-2:55
ato-2:47 ato-medium (fairly easy?)
ie-2:33 ie-easy (really easy?)
COMPET-N-0:53 ddl (77.9% differential)
87 sec. differential, 36.2%
---------------------------------------------------------------------------
MAP22/2:30 avc-1:50 avc-medium
yzd-1:51 yzd-easy
COMPET-N-1:20 ddl (46.7% differentail)
39 sec. differential, 26.0%
---------------------------------------------------------------------------
MAP23/3:00 mjf-2:37 mjf-hardest (extreamly tough?)
COMPET-N-1:04 ddl (64.4% differential)
23 sec. differential, 12.8%
---------------------------------------------------------------------------
MAP24/2:30 jeh-1:57
COMPET-N-1:47 ddl (28.7% differential)
33 sec. differential, 22.0%
---------------------------------------------------------------------------
MAP25/2:30 ie-2:09 ie-easy
COMPET-N-0:54 ddl (64.0% differential)
21 sec. differential, 14.0%
---------------------------------------------------------------------------
MAP26/5:00 jle-4:49
jnm-2:58
COMPET-N-1:11 ddl (76.3% differential)
122 sec. differential, 40.7%
---------------------------------------------------------------------------
MAP27/5:30 smt-3:41
ajs-3:26 ajs-medium (medium or perhaps even easy?)
nbc-3:24 nbc-easy
tmd-2:31 tmd-medium (hardest of set, still easy?)
trg-2:13
COMPET-N-0:56 ddl (83.0% differential)
197 sec. differential, 59.7%
---------------------------------------------------------------------------
MAP28/7:00 jnm-2:29
twm-1:58 twm-medium (not to tough?)
smt-1:31 smt-easy
jeh-0:56 jeh-easy (should be in easier set?)
COMPET-N-0:50 kuh (88.1% differential)
364 sec. differential, 86.7%
---------------------------------------------------------------------------
MAP29/5:00 COMPET-N-4:27 ddl (11.0% differential)
---------------------------------------------------------------------------
MAP30/3:00 nbc-2:07
twm-1:57
smt-1:36
yzd-0:32
148 sec. differential, 82.2%
---------------------------------------------------------------------------
MAP31/2:00 avc-1:07 avc-easy (simple?)
jag-1:07
pnb-0:47 pnb-easy
COMPET-N-0:33 kuh (72.5% differential)
73 sec. differential, 60.8%
---------------------------------------------------------------------------
MAP32/1:30 COMPET-N-0:14 kuh (84.4% differential)
---------------------------------------------------------------------------
***************************************************************************
---------------------------------------------------------------------------
SECTION 2:
-----------------------------
Forwarded extract:
-----------------------------
I saw on the DHT5 log that the D1SG level sets were changed and you
were looking for someone to do the same for the D2SG level sets. Well,
I completed all levels on Hurt Me Plenty under the par time (I also
recorded most of them) and I made level sets for them based upon how
tough it was to complete the level alive, how tough it was to complete
it under par, how complicated it is and how many tries it took
me. These are the sets:
Easiest 01 18 32
Easy 03 05 07 08 10 15 21 27 28 30 31
Medium 04 09 11 16 19 20 22 23 25
Hard 02 06 12 13 14 24 26
Hardest 17 29
-----------------------------
---------------------------------------------------------------------------
***************************************************************************
---------------------------------------------------------------------------
SECTION 3:
What the differential times applied to the proposed set looks like.
----------------------------------------------------------------------
Easiest 01(80%), 18(82%), 32(84%)
Easy 03(75%), 05(66%), 07(*53%), 08(73%), 10(53%), 15(79%), 21(78%),
27(83%), 28(88%), 30(*82%), 31(73%)
Medium 04(60%), 09(59%), 11(67%), 16(69%), 19(63%), 20(68%), 22(47%),
23(64%), 25(64%)
Hard 02(37%), 06(59%), 12(61%), 13(50%), 14(53%), 24(28.7%), 26(76%)
Hardest 17(37%), 29(11%)
* Denotes a DHT time faster that COMPET-N, all other times are for
Ultra Violence since they are faster than Hurt Me Plenty times.
---------------------------------------------------------------------------
***************************************************************************
---------------------------------------------------------------------------
SECTION 4:
This proposed set is VERY good based on this data, whoever did it did
a good job! Just using the numbers and with all MAPS going to UV the
sets could be split as follows: Easiest > 80%, Easy 70% to 80%,
Medium 60% to 70%, Hard 30% to 60% and Hardest < 30%.
Easiest 18(82%), 27(83%), 28(88%), 30(82%), 32(84%)
Easy 01(80%), 03(75%), 08(73%), 15(79%), 21(78%), 26(76%), 31(73%)
Medium 04(60%), 05(66%), 11(67%), 12(61%), 16(69%), 19(63%),
20(68%), 23(64%), 25(64%)
Hard 02(37%), 06(59%), 07(49%), 09(59%), 10(53%), 13(50%), 14(53%), 17(37%), 22(47%)
Hardest 24(28.7%), 29(11%)
---------------------------------------------------------------------------
***************************************************************************
---------------------------------------------------------------------------
SECTION 5:
I have not tried to analyze the comments from people's text
files. Currently this is the base partion that I like
based on my research. This is assuming that the difficulty
goes to UV, and I really think it should. I still look at
this and can see possible sets that I don't think would be
`FAIR' (21,26,2 compared to 3,5,13) but I just don't see
any way around this kind of thing, not every applicant will
have the same set of skills going in. So I have also ordered
the sets from easiest in set to hardest in set. So that if
someone got 26 for thier medium they should get lower indexes
for both easy and hard to balance it out. It would take more
work to come up with index factors to ensure that the sets
were always balanced but maybe not much?
Easiest 01 18 *27 *30 32 *28
Easy 03 08 15 31 21
Medium *05 16 *07 *12 04 09 19 *06 20 11 23 25 *26
Hard 13 *10 14 *22 *17 02
Hardest *24 29
* Represents changes to the proposed set below.
Easiest 01 18 32
Easy 03 05 07 08 10 15 21 27 28 30 31
Medium 04 09 11 16 19 20 22 23 25
Hard 02 06 12 13 14 24 26
Hardest 17 29
If you can think of any better analysis process let me know
and I will do it. I doubt that this helps you in making
a decision since it multiplies your options but I really
do beleive in this set now that I have put the work into it.
On the other hand the proposed set only differs very slightly.
I never meant for this to turn into a thesis but I never
seem to be brief.
Hope it is useful!
Mike Pratt
Written by Mike Pratt on 1996 08 11. Footer, and nothing else, edited by fms. Online on the DHT site since 1996 08 14.