Dragon Warrior Run

2013-08-30 04:25:56 pm

Thanks very much BJ! Actually, if the first Metal Slime victory occurs around 4:30 regardless of the saved game, it opens the possibility of using that fight to reach level five at the start of the game. Surviving the walk to Zone 11 would be a pain, though. Even without that fight, I think it's plausible, if extremely difficult, to use three Metal Slime kills (in three segments) to start a speedrun.

DrHoofington, I'm glad you were entertained. You should check out Ryan8bit's Formula FAQ on GameFAQs, because it's invaluable for code diving. The same RNG is used for every purpose (in contrast to, say, Final Fantasy), and it's actually a two-byte operation, cycling through 32,678 values. The RNG is called twice per step when moving continuously. It's not the same number of steps before the fixed fights, because they can be shorter or longer depending on how long the previous fight takes -- it wasn't uncommon to miss a Metal Slime kill (and get the Demon Knight instead) because the previous fight went terribly. I haven't tried counting, but I don't think it's the same number of fights between them either. On the plus side, it is consistently about four minutes between them, provided I always hit buttons promptly (the RNG permutes much faster whenever you are idle).

Segment time: 2:00:06
Total time: 4:39:31

This one is all about grinding to level eighteen, the minimum where I can beat the Dragonlord. After testing, I found that Zone 10 (south of Hauksness) remains the best for grinding all the way up to level seventeen -- it's a big step above Zone 9 (north of Hauksness), and the subsequent ones have time-wasters like Magiwyverns or Starwyverns. For level eighteen, Zones 10, 12 (southeast of Hauksness), and 13 (inside Hauksness) were very close (from 112 to 118 XP/min), with 12 barely winning, but they were so close that I wouldn't be surprised to be wrong. Anyway, I spend some of level seventeen in a convenient section of Zone 12 hills, resting in 10, but after running out of spare MP I had to spend the remainder in the old Zone. I ruled out Zone 13 because effective grinding there requires multiple HEALMORE casts, and thus a return to the castle, and I can save a minute and a half by making my MP pool last the entire segment.

Using the Metal Slime trick described higher in the thread, I was able to get eight (out of a possible fourteen) Metal Slime kills. For reference, the Metal Slime fights occur around the following times: 4:30, 8:20, 23:40, 40:10, 56:40, 1:05:10, 1:09:20, 1:13:40. Not included are a few times when I thought I would get a Metal Slime and got a Magiwyvern or Wraith Knight instead. There were also familiar Demon Knight fights around 19:40, 44:00, 52:30, 1:25:40, and 1:34:00, whose corresponding Metal Slime fights would have been losses. I also successfully avoid a nasty Demon Knight at 12:10 by pausing for a moment, letting the RNG proceed past the fight. This one would have been a preemptive strike for maximum damage, so I was happy to be able to miss it.

I'm on track for a final time of 4:50. The final segment is next, but expect it to take a while, as I can't devote as much time to it now that the semester has started.

Thanks very much BJ!  Actually, if the first Metal Slime victory occurs around 4:30 regardless of the saved game, it opens the possibility of using that fight to reach level five at the start of the game.  Surviving the walk to Zone 11 would be a pain, though.  Even without that fight, I think it's plausible, if extremely difficult, to use three Metal Slime kills (in three segments) to start a speedrun.

DrHoofington, I'm glad you were entertained.  You should check out Ryan8bit's Formula FAQ on GameFAQs, because it's invaluable for code diving.  The same RNG is used for every purpose (in contrast to, say, Final Fantasy), and it's actually a two-byte operation, cycling through 32,678 values.  The RNG is called twice per step when moving continuously.  It's not the same number of steps before the fixed fights, because they can be shorter or longer depending on how long the previous fight takes -- it wasn't uncommon to miss a Metal Slime kill (and get the Demon Knight instead) because the previous fight went terribly.  I haven't tried counting, but I don't think it's the same number of fights between them either.  On the plus side, it is consistently about four minutes between them, provided I always hit buttons promptly (the RNG permutes much faster whenever you are idle).

[spoiler="Segment Nine"]http://www.youtube.com/watch?v=n9qc19of2dY[/spoiler]

Segment time: 2:00:06
Total time: 4:39:31

This one is all about grinding to level eighteen, the minimum where I can beat the Dragonlord. After testing, I found that Zone 10 (south of Hauksness) remains the best for grinding all the way up to level seventeen -- it's a big step above Zone 9 (north of Hauksness), and the subsequent ones have time-wasters like Magiwyverns or Starwyverns. For level eighteen, Zones 10, 12 (southeast of Hauksness), and 13 (inside Hauksness) were very close (from 112 to 118 XP/min), with 12 barely winning, but they were so close that I wouldn't be surprised to be wrong. Anyway, I spend some of level seventeen in a convenient section of Zone 12 hills, resting in 10, but after running out of spare MP I had to spend the remainder in the old Zone. I ruled out Zone 13 because effective grinding there requires multiple HEALMORE casts, and thus a return to the castle, and I can save a minute and a half by making my MP pool last the entire segment.

Using the Metal Slime trick described higher in the thread, I was able to get eight (out of a possible fourteen) Metal Slime kills.  For reference, the Metal Slime fights occur around the following times: 4:30, 8:20, 23:40, 40:10, 56:40, 1:05:10, 1:09:20, 1:13:40. Not included are a few times when I thought I would get a Metal Slime and got a Magiwyvern or Wraith Knight instead. There were also familiar Demon Knight fights around 19:40, 44:00, 52:30, 1:25:40, and 1:34:00, whose corresponding Metal Slime fights would have been losses. I also successfully avoid a nasty Demon Knight at 12:10 by pausing for a moment, letting the RNG proceed past the fight. This one would have been a preemptive strike for maximum damage, so I was happy to be able to miss it.

I'm on track for a final time of 4:50. The final segment is next, but expect it to take a while, as I can't devote as much time to it now that the semester has started.

anonymous

2013-09-15 10:23:27 pm

Hurry up and do eeet!

Lhexa

2013-09-18 06:07:59 am

The semester just started, so I haven't been working on it the past two weeks. I'll try to finish up this weekend, though.

anonymous

2013-09-20 10:56:16 am

These "speedruns" take so damn long lol

Lhexa

2013-09-22 12:32:27 am

And it's done!

Segment Ten: 7:37
Total time: 4:47:08

After many tries I was able to get a very brisk descent into the bottom level of Charlock. The Dragonlord's first form goes down in two hits, which was a first for me. I can normally get twelve hits on the Dragonlord's second form, when I alternate healing and fighting. But it typically takes fifteen hits to kill him, so I need to get three extra hits at some point. Since I was still short by two of these halfway through the fight, I took a risk and hit him with 45 HP remaining -- there was about a 1/4 chance of dying right there. I won with barely any HP left, but that's how the victories usually end up.

Way back in segment three, I bought one more Magic Key than I actually needed. Here was the reason:

"Wilt thou take me to the castle?" Sure! I was already headed that way.

Time with princess: 4:49:18

I doubt this would count as another category, but it amused me to do. Anyway, now that I'm done, here's what I think an ideal segmented run would look like:

1 -- Grind to level 7 by killing four Metal Slimes (one segment per Slime). This is extremely difficult, but I think it is possible.
2 -- Acquire Erdrick's Armor and Erdrick's Sword at level 7, using the same strats I do.
3 -- Grind to level 12, buy Silver Shield shortly thereafter.
4 -- Grind to level 17, then kill the Dragonlord using the SLEEP spell. This is the most difficult part, unless some new trick is found.

Now to put together my run notes, and submit.

And it's done!

Segment Ten: 7:37
Total time: 4:47:08

[spoiler="Segment Ten"]http://www.youtube.com/watch?v=lcP5y2LVav8[/spoiler]
After many tries I was able to get a very brisk descent into the bottom level of Charlock.  The Dragonlord's first form goes down in two hits, which was a first for me.  I can normally get twelve hits on the Dragonlord's second form, when I alternate healing and fighting.  But it typically takes fifteen hits to kill him, so I need to get three extra hits at some point.  Since I was still short by two of these halfway through the fight, I took a risk and hit him with 45 HP remaining -- there was about a 1/4 chance of dying right there.  I won with barely any HP left, but that's how the victories usually end up.

Way back in segment three, I bought one more Magic Key than I actually needed.  Here was the reason:

[spoiler="Bonus Segment"]http://www.youtube.com/watch?v=yiZfZrAOIGY[/spoiler]
"Wilt thou take me to the castle?"  Sure!  I was already headed that way.

Time with princess: 4:49:18

I doubt this would count as another category, but it amused me to do.  Anyway, now that I'm done, here's what I think an ideal segmented run would look like:

1 -- Grind to level 7 by killing four Metal Slimes (one segment per Slime).  This is extremely difficult, but I think it is possible.
2 -- Acquire Erdrick's Armor and Erdrick's Sword at level 7, using the same strats I do.
3 -- Grind to level 12, buy Silver Shield shortly thereafter.
4 -- Grind to level 17, then kill the Dragonlord using the SLEEP spell.  This is the most difficult part, unless some new trick is found.

Now to put together my run notes, and submit.

anonymous

2013-09-22 03:28:03 am

Great work and colossal speedrun to one of my favorite old school RPGs.

And beating the game with the princess - fucking epic!

bj_waters

2013-09-23 11:12:49 am

Woohoo! Good work Lhexa. We'll finally have this classic(?) game up on SDA! Unless someone finds a pretty serious trick, I doubt anyone will challenge this anytime soon (though I could be wrong). I wonder what I'd have to do to be a verifier for this run . . . . .

Maybe getting the princess counts as 100%? I mean, you've already got the best stuff in the game and I can't think of anything else that would really count (except maybe finding and reading Erdrick's Tablet).

Crow!

2013-09-23 10:46:19 pm

What's that gemma?

I think Gwaelin should have specified which castle she wanted to be taken to.

Lhexa

2013-09-25 08:50:32 pm

It's submitted.

Thanks all!

jw

2013-11-20 09:47:04 pm

So where the hell is this? It's been submitted for two months now.

Lhexa

2013-11-21 08:15:59 pm

The run made it through verification a couple of weeks ago. Presumably there's still a queue of accepted runs ahead of mine, which will be processed and posted before mine makes it up. It tends to be a long process.

jw

2013-11-22 07:06:46 am

Ah, okay. Glad it made it.

Fionordeguester

2013-11-26 08:17:47 pm

So in what way did you "cheat" on that Mad Knight fight?

Lhexa

2013-11-26 08:42:17 pm

Quote from Fionordeguester:

So in what way did you "cheat" on that Mad Knight fight?

Do you mean the Axe Knight battle? No cheating there, just a strategy with very low odds of success.

Fionordeguester

2013-11-26 08:44:11 pm

Oh. In your description, you said that J "cheated" to beat the overpowered Axe Knight, so I thought you meant that literally.

ShinerCCC

2014-05-25 01:31:07 am

Yes, a worthless avatar riding my posts.

Big necrobump but I'd like to announce my trials and tribulations of going for a single segment run for SDA.

I thought it would be a simple, one-time affair trying to get 6:03:37 again. But as feasel's streaming these days has proven, I had no idea what a rare beast I had achieved.

Here are the runs I've gotten and subsequently deleted:
- 6:12:54 but I forgot the Stones of Sunlight on the way to Charlock at level 13 so I had to go back. Looked bad so I decided to have a do-over
- died EIGHT times to the Axe Knight
- got perfect Cantlin, second try Axe Knight, first try Erdrick's Sword, on pace for 5:58:xx...lost 4 minutes farming to 14 somehow, then lost all my time gains trying to powergrind the Axe Knight station. Charlock was unkind twice in a row, had to give up.
- Died twice before level 6.
- Got a Death Necklace, putting me 8 minutes ahead of the 6:12 !!!! but bad luck and a few mistakes added up over the course of the run and I finished with 6:16:54. Currently sitting on this run.

Is it okay if I just submit the 6:16 and call it a day? Running this game is pure luck and feels like running an autoscroller. I'm sick wasting 6-hour chunks of my life on this.

Big necrobump but I'd like to announce my trials and tribulations of going for a single segment run for SDA.

I thought it would be a simple, one-time affair trying to get 6:03:37 again. But as feasel's streaming these days has proven, I had no idea what a rare beast I had achieved.

Here are the runs I've gotten and subsequently deleted:
- 6:12:54 but I forgot the Stones of Sunlight on the way to Charlock at level 13 so I had to go back. Looked bad so I decided to have a do-over
- died EIGHT times to the Axe Knight
- got perfect Cantlin, second try Axe Knight, first try Erdrick's Sword, on pace for 5:58:xx...lost 4 minutes farming to 14 somehow, then lost all my time gains trying to powergrind the Axe Knight station. Charlock was unkind twice in a row, had to give up.
- Died twice before level 6.
- Got a Death Necklace, putting me 8 minutes ahead of the 6:12 !!!! but bad luck and a few mistakes added up over the course of the run and I finished with 6:16:54. Currently sitting on this run.

Is it okay if I just submit the 6:16 and call it a day? Running this game is pure luck and feels like running an autoscroller. I'm sick wasting 6-hour chunks of my life on this.

Darkwing Duck

2014-05-25 01:40:51 am

Highly Evolved

I think 6:16 isn't a great time, though. Error got essentially a flat 6:00. I would think a 6:05 would at least be a requirement.

HavocProdigy

Edit history:

HavocProdigy: 2014-05-25 01:46:06 am

2014-05-25 01:44:27 am

Quote from ShinerCCC:

Is it okay if I just submit the 6:16 and call it a day? Running this game is pure luck and feels like running an autoscroller. I'm sick wasting 6-hour chunks of my life on this.

This is how I now feel about running Milon's Secret Castle and that's only a 9 minute run. If you're not enjoying it, I'd just stop. I'm trying to remind myself to do the same.

There's a run that's 16 minutes ahead of your 6:16. My humble opinion is that if you don't find the game enjoyable enough to get a run that is as good as you can possibly deliver, I'd let someone else handle it. Because there is at least 1 run out there that is comfortably ahead of yours and because several people have recently been working on this game, I don't feel that 6:16 would stand for very long on the site anyway. Again, just my opinion. Best of luck with whatever you decide.

ShinerCCC

2014-05-25 01:49:35 am

Yes, a worthless avatar riding my posts.

The knowledge of this run has existed on this site for about 3 years and nobody else seems willing to do it for SDA. I'm obviously not trying to get a world record, Twitch moves faster than SDA ever will. Error72's 6 hour run came about due to some very risky strats that I also don't agree with. Truthfully the only reliable run to submit is a level 18 Str/HP run but that requires even more ridiculous luck.

The 6:16 has a lot of hilarious bad luck and recovery, which is relatively entertaining as far as Dragon Warrior runs go. SDA's mission statement has always been to collect entertaining videos, not necessarily records. They probably won't get a more entertaining run from me, since this one gets a Death Necklace...

Lhexa

2014-05-25 02:44:09 am

6:03 was indeed an amazing time for the strats I think you were using, to an extent that I wondered whether you might have mistimed it. But at this point both Feasel and Error are using a riskier strategy (Erdrick's Armor at 12), and it's only a matter of time until one of them gets a sub-6:00. While SDA runs aren't necessarily records, they should at least be comparable to the records in terms of route and execution, and "fifteen minutes slower than the other two serious RTA runners" doesn't seem comparable.

Anyway, doesn't Feasel submit runs to SDA? And I don't think Error has ever said he wouldn't submit; rather, he's unhappy with his 6:00 because it still had three unintentional deaths. If he doesn't want to submit, I'll try to talk him into doing so.

ShinerCCC

2014-05-26 12:28:13 am

Yes, a worthless avatar riding my posts.

I can't stream while recording in AmaRec so I suffer in solitude while making these attempts while these guys get to stream and have fun. Well, it's not all suffering, I get to watch operas while attempting these, but still... I just started a new job so my opportunities to make attempts are farther apart.

I didn't get a straight answer from feasel as to whether he records in AmaRec while streaming, and he never said if he would submit to SDA when he gets a good run. I saw bits of Error72's run and I don't think I can beat 6 hours unless everything goes perfectly. My sum of best segments is actually 5:36:56 but that includes the Death Necklace run, the first-try gear run and a first try Charlock. (to elaborate on the lost 5:58:xx run: The first attempt at Axe Knight was a surprise round where I failed to escape and died, so I didn't lose any Herbs; the rematch was an Excellent Move on the first turn, so it was REALLY fast)

I may have a silly audio commentary file to use but having listened to it again, it IS from 2011 when gamers were, myself included, much more foul-mouthed...and I don't just mean swearing, I felt uncomfortable with all the slurs that were thrown around...if I get this run I'll still use a censored version of it, even though it most likely won't sync up with the events that are transpiring. But if someone else submits a run first, I won't bother donating it.

I can't stream while recording in AmaRec so I suffer in solitude while making these attempts while these guys get to stream and have fun. Well, it's not all suffering, I get to watch operas while attempting these, but still... I just started a new job so my opportunities to make attempts are farther apart.

I didn't get a straight answer from feasel as to whether he records in AmaRec while streaming, and he never said if he would submit to SDA when he gets a good run. I saw bits of Error72's run and I don't think I can beat 6 hours unless everything goes perfectly. My sum of best segments is actually 5:36:56 but that includes the Death Necklace run, the first-try gear run and a first try Charlock. (to elaborate on the lost 5:58:xx run: The first attempt at Axe Knight was a surprise round where I failed to escape and died, so I didn't lose any Herbs; the rematch was an Excellent Move on the first turn, so it was REALLY fast)

I may have a silly audio commentary file to use but having listened to it again, it IS from 2011 when gamers were, myself included, much more foul-mouthed...and I don't just mean swearing, I felt uncomfortable with all the slurs that were thrown around...if I get this run I'll still use a censored version of it, even though it most likely won't sync up with the events that are transpiring. But if someone else submits a run first, I won't bother donating it.

presjpolk

Edit history:

presjpolk: 2014-05-26 07:57:57 am

presjpolk: 2014-05-26 01:41:04 am

2014-05-26 01:39:23 am

HELLO!

Error doesn't record offstream, such that he'd have something that would be usable for SDA submission.

Edit: I had no idea ERROR was registered here which is why I chimed in. Smiley

ERROR

2014-05-26 07:52:26 am

The route I'm taking now takes 3 very significant risks.

1) Setting up the treasure chest glitch at level 4. As this occurs fairly early in the run (Dragon Warrior standards wise XD), I don't think this is too horrible for going for.
2) Getting the armor at 12. Since I'm using a STR/HP name for the possibility for level 18 Dragonlord attempts in the future. It makes alot of sense going for the armor at 12. STR/HP, iirc, does not get the defense break on wyverns at 12 without the armor, so it would make the grind from 12 to 13 painfully slow.
3) Getting the sword at 13. Twenty extra attack is nothing to scoff at, so getting the sword ASAP is definitely desireable. I'm waiting until 13 now because at 13, I would have enough money to buy the silver shield, so best case scenerio I would pick up the shield and the token at the exact same time, and then if I died in Charlock castle, it's not the absolute end of the world.

I know that armor at 12 is super risky. I believe it's around a 35% shot per attempt of succeeding assuming you reach the axe knight with full health and 6 herbs. Not particularly good odds considering this occurs around the 2 hr 30 mark in the run, but I go for it anyway for a couple of reasons.

1) As mentioned above, I'm using a STR/HP name, so armor at 12 makes sense for trying to be as good as grinding Wyverns at 12 as the STR/AGI name.
2) I had good luck getting the armor at 12 in previous attempts. I know this isn't really a good reason, but I believe my first 3 or so attempts at getting the armor at 12, I got first try armor once and second try armor twice. That certainly made me alot more willing to continue attempting the armor at 12 for the forseeable future. Of course, in the 6:00:10 run and then my attempt yesterday, it took 4 attempts and then 6 attempts, and the 6 attempts I died before I even got out of the town, which killed any possibility of a level 19 sub-6 was well as my motivation to finish XD.

I don't know what strats you are attempting to use. Using a STR/AGI name and then going for the armor at 13 doesn't seem like too much of a bad idea, since I don't think I'd ever go for level 18 Dragonlord in offline mode. I don't know what else I would really consider changing from the route though, since I feel like grinding to 14 without both the sword and armor would considerably slow down the run.

And, I wish you luck Shiner. I don't think I could put in offline attempts...this is only bearable because I can interact with viewers while streaming. Unfortunately, I currently don't have a setup that would allow me to both stream and make a recording that would be SDA submittable, so I don't see myself doing a SDA submission of this game in the near future.

The route I'm taking now takes 3 very significant risks.

1)  Setting up the treasure chest glitch at level 4.  As this occurs fairly early in the run (Dragon Warrior standards wise XD), I don't think this is too horrible for going for.
2)  Getting the armor at 12.  Since I'm using a STR/HP name for the possibility for level 18 Dragonlord attempts in the future.  It makes alot of sense going for the armor at 12.  STR/HP, iirc, does not get the defense break on wyverns at 12 without the armor, so it would make the grind from 12 to 13 painfully slow.
3)  Getting the sword at 13.  Twenty extra attack is nothing to scoff at, so getting the sword ASAP is definitely desireable.  I'm waiting until 13 now because at 13, I would have enough money to buy the silver shield, so best case scenerio I would pick up the shield and the token at the exact same time, and then if I died in Charlock castle, it's not the absolute end of the world.

I know that armor at 12 is super risky.  I believe it's around a 35% shot per attempt of succeeding assuming you reach the axe knight with full health and 6 herbs.  Not particularly good odds considering this occurs around the 2 hr 30 mark in the run, but I go for it anyway for a couple of reasons.

1)  As mentioned above, I'm using a STR/HP name, so armor at 12 makes sense for trying to be as good as grinding Wyverns at 12 as the STR/AGI name. 
2)  I had good luck getting the armor at 12 in previous attempts.  I know this isn't really a good reason, but I believe my first 3 or so attempts at getting the armor at 12, I got first try armor once and second try armor twice.  That certainly made me alot more willing to continue attempting the armor at 12 for the forseeable future.  Of course, in the 6:00:10 run and then my attempt yesterday, it took 4 attempts and then 6 attempts, and the 6 attempts I died before I even got out of the town, which killed any possibility of a level 19 sub-6 was well as my motivation to finish XD.

I don't know what strats you are attempting to use.  Using a STR/AGI name and then going for the armor at 13 doesn't seem like too much of a bad idea, since I don't think I'd ever go for level 18 Dragonlord in offline mode.  I don't know what else I would really consider changing from the route though, since I feel like grinding to 14 without both the sword and armor would considerably slow down the run.

And, I wish you luck Shiner.  I don't think I could put in offline attempts...this is only bearable because I can interact with viewers while streaming.  Unfortunately, I currently don't have a setup that would allow me to both stream and make a recording that would be SDA submittable, so I don't see myself doing a SDA submission of this game in the near future.

puwexil

2014-05-26 10:07:31 am

Professional Second Banana

For everyone having issues recording while streaming, give these Amarec settings a try - https://kb.speeddemosarchive.com/Amarectv_configuration

Crow!

Edit history:

Crow!: 2014-06-01 02:01:02 pm

Crow!: 2014-06-01 02:00:44 pm

Crow!: 2014-06-01 01:57:41 pm

Crow!: 2014-06-01 01:57:08 pm

Crow!: 2014-06-01 01:55:26 pm

Crow!: 2014-06-01 01:54:25 pm

Crow!: 2014-06-01 01:53:15 pm

Crow!: 2014-06-01 01:50:09 pm

Crow!: 2014-06-01 11:47:08 am

2014-06-01 11:42:50 am

What's that gemma?

Since all the cool kids have been doing Dragon Warrior RTAs, I've been doing some theorizing and calculating for the game. I'm probably not going to sit around to run a 6 hour game, though, especially not one which frequently dies at the end for no reason other than bad RNG.

Anyway, here's my idea:
Buy Small Shield.
Buy Dragon Scale.
Equip Dragon Scale.
Sell Torch, sell Dragon Scale.
Buy Herb.
(Edit: faster order)

Now you're set to dive for the Death Necklace chest at level 1. Each attempt takes up to 5 minutes (about 3 minutes is the usual death to a Skeleton on the way to the Cave; once you're in the cave things are much easier). You have something like a 25% chance (purely "guesstimating" on my part, from my test attempts) to reach the chest, and a 1/32 chance to get it once you arrive, as usual. So it would be around 8 hours of attempts, on average, to get one 5-minute Death Necklace.

You'll have to navigate the Mountain Cave blind, and you'll have to take some risks with your two Herbs (one from shopping, one from inside the cave itself). The biggest threat is a Skeleton, which can deal over 10 damage and has a high enough Strength to make running difficult; keeping yourself safe from its attacks would often mean blowing your herb too early to get through the area overall. The final floor also has Warlocks.

I'm assuming that having 1250 gold, a Small Shield, and the Dragon Scale bonus at 5 minutes into the game is a significant time boost, though it's not 100% clear to me what the next step would be. Perhaps a trip to Garinham for Hand Axe + Chain Mail, followed by grinding near there until level 3, followed by grinding at Kol until 1500 gold, followed by a Broad Sword + grinding at Rimuldar?

EDIT:
You can also do the dive at level 2, arriving at the chest at ~8 minutes usually rather than at ~5 minutes rarely. If I again "guesstimate" that this has a 75% success rate, then this means 5.5 hours of attempts for one early Death Necklace, though that necklace is 3 minutes later. That route goes as follows:

- Purchase Bamboo Pole + Shield
- Fight in the Hills of Hubris (running from Drakees), and fight 3 Red Slimes.
- Depending on your HP and how many non-Red Slimes you fought, either return to zone 0 (possibly at the one desert square to the east), and finish your level, or stay in the Hills of Hubris for the higher encounter rate.
- Once you have 30 Gold, buy the Dragon Scale, sell it and the Torch, and buy an Herb.
- Once you have both level 2 and the requisite shopping, death warp to heal yourself.
http://www.twitch.tv/iicrowii/c/4377218

-----

I've made a spreadsheet for predicting EXP/min of various grinding strategies. So far I've focused on trying to figure out what is best in the level 16-18 portion of the game, but the spreadsheet could be adapted for other portions of the game.

The spreadsheet requires a bunch of data from the battle simulator, and it has a lot of roughly estimated durations for various portions of each battle which could be improved. It has the ability to account for differences in the average length of a turn from one enemy type to the next (Wizards might be rather different than a Rogue Scorpion), but I haven't put in that level of precision yet. I'm overestimating the actual EXP/min by ~10%, but I hope it's accurate enough to compare different strategies with each other.

When trying to adapt this to a different portion of the game, the hardest columns to adapt will be the "Healing time." Some average time cost associated with taking each point of damage has to be assigned. At present it assumes 8 HP gained per battle due to Erdrick's armor, plus some time cost to cast HEALMORE, assuming that on average that spell will recover 70 HP. This calculation would have to be very different for a HEAL + SLEEP + Inn farming strategy.

Anyway, here's my findings:
Assuming you don't run out of MP, farming in Hauksness is about 4% faster than farming in the desert south of Hauksness in level 16, and about 3.5% faster in level 17. Running away from Starwyverns is not profitable in either level, but it nearly breaks even in level 17 whereas it's at a significant cost in level 16.

Farming in Charlock's Blue Dragon areas at level 17 is about 5% faster than farming in Hauksness, assuming you never run out of MP, don't get killed in an individual unlucky fight (SLEEP lock), and run away from Stonemen. This does not account for the amount of time you spend walking to Charlock. Using SLEEP makes Stonemen safe to kill, but it also makes the battles take so many rounds that it doesn't pay off anyway. I had no way to calculate the cost of a SLEEP+Run strategy, but I suspect just Run is probably faster.

Since all the cool kids have been doing Dragon Warrior RTAs, I've been doing some theorizing and calculating for the game.  I'm probably not going to sit around to run a 6 hour game, though, especially not one which frequently dies at the end for no reason other than bad RNG.

Anyway, here's my idea:
Buy Small Shield.
Buy Dragon Scale.
Equip Dragon Scale.
Sell Torch, sell Dragon Scale.
Buy Herb.
(Edit: faster order)

Now you're set to dive for the Death Necklace chest at level 1.  Each attempt takes up to 5 minutes (about 3 minutes is the usual death to a Skeleton on the way to the Cave; once you're in the cave things are much easier).  You have something like a 25% chance (purely "guesstimating" on my part, from my test attempts) to reach the chest, and a 1/32 chance to get it once you arrive, as usual.  So it would be around 8 hours of attempts, on average, to get one 5-minute Death Necklace.

You'll have to navigate the Mountain Cave blind, and you'll have to take some risks with your two Herbs (one from shopping, one from inside the cave itself).  The biggest threat is a Skeleton, which can deal over 10 damage and has a high enough Strength to make running difficult; keeping yourself safe from its attacks would often mean blowing your herb too early to get through the area overall.  The final floor also has Warlocks.

I'm assuming that having 1250 gold, a Small Shield, and the Dragon Scale bonus at 5 minutes into the game is a significant time boost, though it's not 100% clear to me what the next step would be.  Perhaps a trip to Garinham for Hand Axe + Chain Mail, followed by grinding near there until level 3, followed by grinding at Kol until 1500 gold, followed by a Broad Sword + grinding at Rimuldar?

EDIT:
You can also do the dive at level 2, arriving at the chest at ~8 minutes usually rather than at ~5 minutes rarely.  If I again "guesstimate" that this has a 75% success rate, then this means 5.5 hours of attempts for one early Death Necklace, though that necklace is 3 minutes later.  That route goes as follows:

- Purchase Bamboo Pole + Shield
- Fight in the Hills of Hubris (running from Drakees), and fight 3 Red Slimes.
- Depending on your HP and how many non-Red Slimes you fought, either return to zone 0 (possibly at the one desert square to the east), and finish your level, or stay in the Hills of Hubris for the higher encounter rate.
- Once you have 30 Gold, buy the Dragon Scale, sell it and the Torch, and buy an Herb.
- Once you have both level 2 and the requisite shopping, death warp to heal yourself.
http://www.twitch.tv/iicrowii/c/4377218

-----

I've made a spreadsheet for predicting EXP/min of various grinding strategies.  So far I've focused on trying to figure out what is best in the level 16-18 portion of the game, but the spreadsheet could be adapted for other portions of the game.

The spreadsheet requires a bunch of data from the battle simulator, and it has a lot of roughly estimated durations for various portions of each battle which could be improved.  It has the [i]ability[/i] to account for differences in the average length of a turn from one enemy type to the next (Wizards might be rather different than a Rogue Scorpion), but I haven't put in that level of precision yet.  I'm overestimating the actual EXP/min by ~10%, but I hope it's accurate enough to compare different strategies with each other.

When trying to adapt this to a different portion of the game, the hardest columns to adapt will be the "Healing time."  Some average time cost associated with taking each point of damage has to be assigned.  At present it assumes 8 HP gained per battle due to Erdrick's armor, plus some time cost to cast HEALMORE, assuming that on average that spell will recover 70 HP.  This calculation would have to be very different for a HEAL + SLEEP + Inn farming strategy.

Anyway, here's my findings:
Assuming you don't run out of MP, farming in Hauksness is about 4% faster than farming in the desert south of Hauksness in level 16, and about 3.5% faster in level 17.  Running away from Starwyverns is not profitable in either level, but it nearly breaks even in level 17 whereas it's at a significant cost in level 16.

Farming in Charlock's Blue Dragon areas at level 17 is about 5% faster than farming in Hauksness, assuming you never run out of MP, don't get killed in an individual unlucky fight (SLEEP lock), and run away from Stonemen.  This does not account for the amount of time you spend walking to Charlock.  Using SLEEP makes Stonemen safe to kill, but it also makes the battles take so many rounds that it doesn't pay off anyway.  I had no way to calculate the cost of a SLEEP+Run strategy, but I suspect just Run is probably faster.

Attachment:

DW1 EXP guesswork.xlsx