Bard’s Tale Speed Run

So, after some analysis, I think there remains a lot of room for improvements to sub-40 minutes. But we’re getting deep into the realm of divine RNG and I’ve yet to complete a full run using these methods.

The first major improvement lies in maxing your gold at level 1 and then using the golems, dragon, and stone giant around Harkyn’s castle. These creatures are easier to access than Ogre Lords and the 2x golems, 1x giant, and 1x dragon yields 11k xp in one round. The order to do these starts at the south end of Harkyn and proceeds clockwise – golem, stone giant, golem, dragon. Then by entering and leaving Harkyn’s, two stone golems can be repoped, this time to be taken in a counter-clockwise fashion.

One cycle with the grey dragon is ~11kXP, each 2x golem cycle thereafter is ~6kxp. This yields ~25k xp/min, significantly faster than Ogre Lords due to the shorter travel time on each repop cycle.

The downside is that the RNG is much harder than the Ogre Lords. The Grey Dragon has a 50% chance to breathe fire, and a 25% chance to attack you, both of which are always fatal until higher levels. The golems also hit much harder than Ogre Lords, so the odds of surviving a hit are very low. The upside is that the pathing and XPM is so much better that even if you die, you can use the temple near the guild to respawn with minimal penalty, and in some cases a slight advantage. A homebase respawn repops the stone giant and the grey dragon, letting you get a better first cycle on return. And, it lets you pass the adventurer’s guild to get a few levelups and more hitpoints to survive future hits.

The second major improvement is at Wiz 1 to combine the 9 wights in Catacombs 1 with a cycle around Harkyns. This is done by farming wights until you have no more spellpoints. Then bumping to Wiz 2, then after healing spellpoints, doing one cycle of Harkyns, then returning to wights. This takes additional advantage of the time spent traveling to heal mana.

A third improvement, but somewhat RNG dependent, happens at Wiz level 9. Wiz 5 through 11 are obtained through the spectre in the catacombs. At Wiz9, spellbind is unlocked, giving a 3/8 chance to spellbind the spectre. This allows you to leave Cats, heal up, level up to 10, and then immediately party attack the spectre as if you had summoned it at level 11, saving one trip down the catacombs. (I am unaware of how to force an aggro’d special slot creature to be peaceful, nothing seems to work except killing or replacing it).

The combination of the above, in addition to some minor optimal pathing improvements from my prior video, should take the time down to sub-40.

Want to push further? A fourth improvement, a minor RNG miracle, and a radical departure from the past approaches, is to start as a magician, level to 7 then go conjurer and level to 13/14 (may be worth the RNG benefit to get another level), do Harkyn/Kylearan to obtain the crystal sword, two silver items, and onyx key. Then go to Mangar to obtain the silver circle and… a mournblade! Put the mournblade on a level 1 hunter, and levelup to Magi 7 via Harkyns and then Wiz 1 to 30 with the hunter.

This has several interesting benefits, first, the time spent in Harkyns and Kylearan and Mangar is well utilized. A conjurer at 13/14 can do the run across Harkyns, Kylearan’s tower, and Mangar’s Circle just as efficiently as a Wiz 30. Assuming they make all of their flees…. which is a big assumption. The real kicker is the RNG to get the Mournblade, which is found on Mangar 2. The most consistent and easiest things to clear on Mangar 2 are the 99 hobits and the 3 bandersnatches and single creatures that try to join your group. Using teleports, the fixed fights can be cleared in succession, and the level reset by hopping down and back.

The upside to this approach is that you have a level drain object for the majority of the levels where you need it the most and get the most time advantage (Wiz 1 to 30). Your level 1 hunter will very rarely, if at all, critical hit. And even if he did, you have infinite gold anyway. You can level Wiz 1 to 11 without going into the catacombs, and 12 to 30 faster than summoned spectres via SPSP! The party attack steps are far less complex and less time consuming with a hunter, and there’s no need to refresh on spellpoints. So, assuming perfect RNG for the mournblade, you’re going to be pushing 37 minutes.

The downside is that the RNG involved is really, really low, and your conjurer has a relatively low mana pool and some flee luck is needed. Oh, and almost all fights on Mangar 2 are invariably fatal. Also, some extra time is spent in Mangar 2 getting the blade, and returning to the guild, and in returning to Mangar when going for the win, which adds up to maybe 1.5 minutes.

So, I did a run in 43 minutes and 42 seconds and without the main hero dying. With improved efficiency and some luck/RNG it might be possible to get under 40 minutes, but it would likely not be an easy task. A 4 minute gain is a 10% improvement at this point, which is difficult from both a RNG/luck perspective, as well as a human performance perspective.

Find the narrated speedrun at this link: 43 minute Bard’s Tale I speedrun

The strategy is fairly similar to the 53 minute run with some improvements. The major improvement being that a level 8 conjurer and a level 1 wizard with high luck/dex and infinite gold can farm the 9 wraiths near the entrance of Catacombs I. They’re like 25k+ xpm, which is a significant improvement from Ogres.

Some possible efficiency improvements might be:

• Add two stone golems to ogres. One round yields 14592 xp, so that 2 samurai and one round of golems/ogres and the grey dragon gets you level 1 to level 7 as a magician. One round of two ogres + stone golems is also enough to get from level 1 to level 6 as conjurer. Adds a lot of risk, saves maybe 1.5 minutes.
• Start wraiths at level 6 (or lower) as a Conjurer, saves 30 seconds or so, but high risk – also, if it takes multiple rounds to kill them, there’s not a lot of time savings, so farming wraiths at low level might be slower than ogres
• Initial Bard creation and use could be skipped, but exceptionally good RNG is required for an actual time saving of 30 seconds or so…
• Ignore all item drops and chests other than eyes and specter snare (saves 1 min or so)
• Enter Mangar at a lower Wizard level – about 1 minute saved for every three levels under 30

The last big area to make speed improvements is to enter Mangar at a lower level. But there are two critical issues: first, flee from everything in Mangar, and second, repel demon lord breath and Mangar/vampire lord instant death spells.

Flee always has a 1/8 chance to succeed in all cases. Almost all encounters in a speedrun in any level in Mangar are going to be fatal if you cannot flee. For example, Mangar level 5 has a fixed encounter with 5 Storm Giants, and they have a 15 to 22 saving throw. A level 16 wizard with no luckshield has a 2+0+3+8+3 = 16 saving throw. This means that you’ll have a slightly worse than 1/4 chance to flee (live) as a level 16 wizard (1/8 random flee chance + (1/8 * 7/8) saving throw flee). But there are also several other fixed encounters (only some are avoidable) like specters and also forced encounters with random critters on the path to the specter snare. As a result, a level 16 wizard has something like a maximum of a 1/64 chance (3 successful flees at 1/4 odds each) to just make it to Mangar.

At level 24, a wizard with no luckshield has a 2+0+3+12+3 = 20 saving throw. This boosts your flee odds up somewhat. A level 24 wizard has a slightly less than 2/3 chance of fleeing. Across 3 encounters that’s a 8/27 chance of making it to Mangar alive.

At level 30, a wizard with no luckshield has a 2+0+3+15+3=23 saving throw. They will always flee from everything.

And then you have to deal with breath and death spells. The demon lords do approximately 40-180 damage with a breath attack. Enough HP can compensate for this damage, and two breaths in the first round of the encounter is somewhat unlikely. But vampire lords also hit hard, and will almost certainly hit you at least once, probably twice or more. As such, it’s risky to rely on being able to tank all the damage.

Also, the vampire lord and Mangar death spells (DEST) are cast relatively often. They can be repelled, but, the general rule is that you have to make a saving throw of 4 or higher to repel the spell, otherwise you instantly die. Vampire lords throw a 13 to 20, Mangar throws a 14 to 21, requiring a saving throw of at least 24 to repel vampire lords and a 25 to repel Mangar. A level 30 wizard has a saving throw of 23, which means that Mangar has a 3/8 chance to one-shot you if he casts. Vampire lords have a 1/8 chance. And Vampire Lords also drain levels (and thus your saving throw) every time they hit you….

In any case, the time difference between level 25 and 30 wizard is just like 2.5 minutes or so. As such, there’s not a lot of time gains to be made by entering in at a low level. And the difficulty of completing the game rises quickly and exponentially with decreasing levels.

So, I did a speedrun at 53 minutes. It would have been sub-50 except that I ran into the stasis trap before getting the onyx key… I get ahead of myself sometimes.

I was on pace for 48ish, but lost 4+ minutes to the stasis trap and then revisiting Kylearan’s tower. Interestingly, the Wizard beat Mangar without dying once…

Overall efficiency was better, but there are some small gains (90 seconds?) to be made. A lot of the actions in the sewers and Mangar levels were unpolished and unnecessary. It’s just a function of having not done any work on this speedrun in some time.

This specific run exploited the interesting gold/item duplication bug in the shop. This lets the adventurer spend about 90 seconds traveling to and from the shop to have essentially infinite gold. Gold economy is no longer a constraint which means even more time (and gold) can be used in leveling as rapidly as possible.

I am near the edge of the limits of this overall strategy (Magi 1–7 to Conjurer 1–13 to Wizard 1 to 30). If I can come under 46 minutes using this path, I’ll be impressed.

The next level of breakthroughs are a function of XPM. There are a few ways to get more XPM (25-70kpm+) that may be accessible through the shop duplication bug.

Currently, Ogres offer a flat baseline of about 17,000 experience points per minute. This is one of the best and most consistent sources for XP and represents a minimal base-line in terms of time spent and XP gained. Walking the Catacombs without APAR is slow as compared to Ogres in terms of XPM for all but the last 2 levels of Magi/Conj. APAR speeds the Catacombs up to 1 level per 75 seconds or so. And finally, SPSP is almost certainly the fastest possible XP in the game for a Wizard level 11 or higher that does not have a draining weapon. (The effort required to get a draining weapon is not materially different from the effort required to beat the game.)

So, XPM in the 25-70kpm+ range—which is more than Ogres but far less than SPSP—represents an opportunity for substantial time gains. Some of the scenarios are interesting but rather complex and difficult to execute. For example, take an army of hobbit bards, level 2, and arm each of them with a firehorn, and then hope for good first-strike RNG. Be sure the hobbit army dies, leaving only 1 entity to gain all the xp, repeat ad nauseum.

Ultimately, the ability to successfully flee, the ability to survive high level breath attacks, and the ability to successfully cast a spell at a high-level target, will trade off time spent leveling. A level 13 conjurer could in theory beat the game in just a few minutes. But the probability of the RNG allowing that to happen is very small (1 in 100,000,000 runs?).

So. Today I did a speedrun. 54 minutes 59.79 seconds. Or maybe even a few hundredths of a second earlier, it’s hard to say. But whose counting?

I used a hobbit bard, a half-orc magician, and several of the original team that loads when the game is first started (e.g., Merlin, El Cid, et al.). The first and the last fights in the game were a team effort.

Lost 3 full minutes wandering around Kylearan’s Tower – single biggest loss of time. (My speedrun depth mastery is fairly shallow, it’s hard to get this deep into it). I think there was probably about 5 minutes of efficiency across the entire run in getting lost, sub-optimal pathing, and sub-optimal Mangar strategy. So, sub-50 minutes is probably doable.

Critical Timings:

• 01:30 – An influx of gold.
• 04:45 – Level 7 Magician (Spell level 4) becomes Level 1 Conjurer
• 16:53 – Level 13 Conjurer becomes Level 1 Wizard
• 19:49 – A brief visit to adventurer heaven
• 40:40 – A level 30 Wizard leaves with his team and begins the assault on the Two Towers.
• 49:34 – Entered Mangar
• 54:59.79(+/- .05) or earlier – We’re home.

Video with post-run audio commentary: 54 Minute Speedrun

So, basically, my runs have been focusing on the Magician -> Conjurer -> Wizard path that skips half the magician levels, gets conjurer to 13, and then levels as wizard to 30. In my initial trials, I have been able to get to Mangar in just over an hour, doing it relatively slowly in a “is it possible” mode. And it is.

I’ve decided to use the default heroes, particularly the firehorn on the bard, to farm the 66 skeletons in the catacombs. I roll a hobbit bard with 18 dex and 17 luck and they work about 2/3 times to wipe the skeletons. This gives the party about 6-12k in gold. This allows us to get a few quick levels as a Magician to build stats and get mystic shield, and then swap to Conjurer (Apar) and then Wizard (SPSP).

The main issue is the lack of the Magician’s REST. This is particularly an issue after the possession trap, when your hp is defaulted to 1. FLERE is not really a quick way to heal up, and it drains a lot of SP. It may be that we’re forced to get REST.

In any case, the bard lets us skip the red dragons, saving time spent healing. The rest of the run is just object collection and pathing. I’m completely confident that a 60 minute run is doable. It’s just a matter of luck.

So, I’ve not done much with speed running Bard’s Tale in late 2020 / early 2021. I have made several attempts to beat my 1:26 time, with fairly decent results, but nothing under the 1:10 mark yet. Simple mechanics and efficiency gains represent at most an additional 5 minutes of time I could shave off at this point. Leveling through all four magic user classes simply takes a lot of time and I’m unaware of any better methods to improve that speed. But I’m determined to win the game in less than 1 hour.

Interestingly, the game is very RNG intensive, and my analysis suggests that in theory, even a level 13 conjurer could win the game (with some very, very low probability, like 1:10,000,000). A level 3 bard and monk might even be able to (again with some very, very low probability, like 1:1,000,000,000).

Otherwise, the game is basically a question how how far to level up in each class so that you have a meaningful probability of winning (1:3?) the last battle. The major tradeoffs as I see it are something like what follows:

• Must start as Conjurer or Magician, but maybe skip one?
  • Conjurer – TRZP and LEVI are high-utility spells, without LEVI you need Ali’s Carpet or a Broom, which are really hard to farm without TRZP.
  • Conjurer’s APAR seems critical and time saving across the entire game. It’s not clear that walking around Harykn and Mangar levels results in significant time savings, if any.
  • Magician – needed only for REST, so the entire class might be avoidable.
  • Magician – STTO is the same as DEST and comes at level 9 / spell 5

• Maybe Skip Sorcerer?
  • All sorcerer levels might be skipped entirely
  • Red Dragon and MIBL are really only needed for harming Harkyn
  • May want to get 4+ levels as a wizard to offset lack of luckshield, loss of 13 sorcerer levels
  • If no STTO, DEST, or MIFI, what are my one hit options? Spectre-Snare as Wizard, ARFI does 30-120, SPBI and DMST are situationally effective
• Wizard is Probably a Must
  • Wizards and Sorcerers get 1-8hp roll.
  • Only Wizard can wield Spectre Snare in PC version
  • SPSP at level 11 opens the door to substantial time savings
  • Must have DISP spell via Wizard or Exorwand to get through possession trap (there are also other options, but they seem less time efficient).
  • Skipping Wizard requires 2 additional levels to make up for the class luck bonus
  • Requires at least two other classes to be level 5 or better, with spells at level 3 or better

As such, some variant of leveling to 13 as Conjurer, to level 5/9 as a Magician and/or Sorcerer, then a lot of Wizard levels may represent the fastest option. We’ll keep experimenting.

So, I completed a 1 hour 26 minute speedrun as a level 36 wizard using ultra pure rules. There were some unfortunate mistakes made, and my overall technique was poor due to cold fingers, but it was by far my fastest run yet.

Started as a Lvl 1 Half-Orc Conjurer, leveled to 6 on Samurai. Then leveled to 10 on Ogre Lords. Then leveled to 13 on Spectre in Catacombs 3. Swapped class to Sorcerer. Leveled to 9 on Ogre Lords, then Spectre until 13.

Swapped class to Magician, leveled to 13 in Harkyn 1. While a Magician, obtained crystal sword and tried Harkyn 2. Obtained a luckshield as the first item drop in Harkyn 2, which was a huge time saver. Farmed in Harkyn 1 until level 13. The goal was to get one or more of Mithril Gloves, Mithril Scale, or Shield Ring, but none dropped unfortunately.

Swapped class to Wizard. Got a one level in Harkyn 1, then immediately went to Spectre to level up until 14. Then went to Harkyn 2 to get the silver square and to do Kylearan’s tower. Unfortunately I failed to bring the bard along, and didn’t realize that until I was at Kylearan’s Tower, so I casted DEST on myself and created a bard to resurrect me. The bard was then killed, the crystal sword equipped on him, and Kylearan’s Tower was finished. The bard was resurrected (I’m not completely mean) and left in the guild.

The Wizard then leveled to 37 on summoned Spectres and the Catacomb’s Spectre. Along the way I accidentally casted DEST on myself instead of a summoned Spectre. The bard was kind enough to resurrect me. I also ran out of mana and gold at the same time, which set me back a minute or two.

Mangar’s Tower went smoothly. The possession trap was a breeze. At the final fight, I targeted a demon lord with DEST. A demon lord opened with a breath cast. Thor then whacked the other demon lord. One vampire lord summoned a wraith, the second hit Thor, and the third wasted a spell. Mangar also wasted a spell. Then my DEST killed the demon lord. After the first round, the remainder of the biggies were disposed of with DEST. Thor eventually died to vampire lords, and a mind blade finished out two wraiths. Kylearan came and the game was over.

Overall, I lost about 7.5 minutes to mistakes and mechanical inefficiencies. And poor item drops in Harkyn also slowed things down. I think future runs using similar approaches could be just under 1:15 with an early luckshield drop. Even earlier if the Wizard is leveled to 18 instead of 36.

I was able to capture the game, find the video with audio-commentary on Youtube at ___________. (to come soon).

Ultimately, I’m targeting a speed run under one hour. To accomplish that, I’ll need a substantially different strategy. I’m actively experimenting with some, and I’ll post any successful results.

So, I ran a 2 hour 15 minute speedrun to beat Bard’s Tale 1 with one hero in the party and without any deaths, inclusive of time spent rolling. I started with a H-Elf Conjurer and leveled him up to a 13 Conjurer, then 13 Magician, then 13 Sorcerer, then 37 Wizard.

I used a level 1 Hobbit Bard to hold only the crystal sword so that the Crystal Golem in Kylearan 1 could be passed and the Onyx key obtained. The Bard accompanied me into Mangar (everyone needs a sidekick) and accidentally obtained the Silver Circle, which he promptly dropped and the Circle was re-obtained properly by the Wizard.

He died to traps before he participated in the last two fights (Red Dragons and Mangar), but I figured his next of kin deserved that 300k xp and gold for his noble efforts in holding a crystal sword in the face of a horrible monster far beyond his powers.

The Possession Trap was gratefully a breeze. And the Mangar fight with Thor + Wizard went very well. Started with DMST on the Demon Lords, then DEST on Mangar, then REDE or DEST to get rid of the remaining nasties. The fight went really well, Kylearan came, and the game was over.

At level 37, the Wizard repels all spells and can only be killed by critical hits or stone touches, or thrown into chaos with possession/insanity attacks. So long as the Demon Lords die quickly, the final fight seems more or less easy.

I think that the game can be beaten solo faster than it can be beaten with a party of 4. With some strategy changes, I think 1 hour 30 minutes would be a relatively easy mark to obtain. Using a level 37 Wizard was overkill, and there is very large amount of room for efficiency increases.

Did a 1 hour 51 minute and 31 second speedrun using pure rules. The party consisted of a lvl 22 Elf Paladin, lvl 14/15ish H-Elf Monks, lvl 9 Wizard, and a lvl 1 Bard. This was an experiment to see the viability of H-Elf Monks. And they’re viable. They get blown apart at Mangar, but they easily hit -5 AC at level 8 which reduces their hp loss (giving some time and gold savings) over the course of the entire game.

The overall approach was similar to earlier speedruns. Level Paladin to 9 and Monks to 8 on Ogre Lords, Wizard to 7 on Samurai. Then I farmed out Catacombs 1 until Conjurer was level 10. Then I turned to Harkyn 1 until the Paladin had -6 AC, and by then Monks were -8 AC or better. The Monks are a sucker for the Grey Dragon, but we mostly lucked out. I’ve also been attempting the Spectre for some early gold when the Paladin is level 10, but I don’t recall if I did this time or not.

Ultimately, the item drops were really good and the Paladin had a full set of Mithril relatively early. Then the party went straight to wights to farm out the party to level 12. Then repeatedly hit the Spectre to bump everyone to 13 or better. Once Conjurer was 13, he was immediately promoted to Magician and the Silver Square and Crystal Sword were obtained. A luckshield dropped early which helped a lot. The Magician bumped up to level 8 or so (50k xp) on wights, and then Kylearan’s tower was attempted and the party got the Triangle and Onyx Key. A luckshield was dropped by the one eyed gatekeeper, bringing the total to two. (but no exorwand was dropped, sadly). And a late dropping pureblade ended up helping the team a lot in Mangar.

After that, the Magician and Paladin alone repeatedly killed the Spectre to get them both extra levels and gold. Once level 13, the Magician was promoted to Wizard. Then Wights were farmed to get the Wizard a level or two, then the Paladin and Wizard repeatedly killed the Spectre until the Paladin was 22 and Wizard was 9. At that point, the party headed to Mangar.

A level 1 Bard was created in the early game to give a bit of gold to the party and a Lak’s Lyre was saved for him. The goal being to use the Bard to get past the red dragon pair which can team wipe a party with ease. Unfortunately, the Bard died on a trap prior to getting there, and the mana cost seemed prohibitive to attempt to resurrect (ANDE/DISP). So, we risked it and the dragon’s died without a breath.

The Mangar fight was rough. The Wizard failed two DESTs on Mangar. The Monks died early, one to DEST, the other to a Demon Lord breath. The Wizard died (I think) to a DEST from a Vampire Lord or Mangar, and things were looking grim. Ultimately, the Paladin and Thor killed basically everything, with their saving throws keeping them alive just barely long enough. Kylearan came for a visit and the game was over.

The early game item drops were consistently good. But there was a full team wipe and several unnecessary deaths that cost precious time. Also, this was my first experiment with H-Elf Monks. Fortunately, the excess gold from farming the spectre made it possible to easily recover from the deaths.

It seemed really important to level the Paladin and Mag/Wiz together to the exclusion of the Monks (once the Monks were 14 or higher). There seems to me to be a greater benefit to give that time/levels to the Paladin and Wizard over sharing it with Monks and slowing the Paladin/Wizard growth. In the end, I think there’s about 10 minutes or so of additional improvement to be had in avoiding deaths and optimizing the balance between killing wights and letting the spectre bump team members up a level.

There are also some high level strategies I continue to explore, if they turn out to be useful I’ll write about them.

Ran a 2h 10m run. Used a very similar approach to the 2h 21m run. This time I was able to get screen recorder software working and I posted the video to YouTube with commentary. Find it at https://www.youtube.com/watch?v=yh0EEJ5iwIk

This attempt included some additional refinements. The item drops were better than average overall.

Some timings:

• Farming Samurai/Ogre Lords
• Catacombs 1 – 19:30
• Harkyn 1 – 34:30
• Wights – 47:45
• Harkyn 2 for items – 1:13
• Kylearan’s Tower – 1:26
• Farming Wights/Spectres
• Mangar Attempt #1 – 2:00
• Mangar Attempt #2 – 2:06

There is still substantial room for improvement as several minutes were lost to team wipes and bad pathing/event ordering. I will be analyzing the various timings to see where things can be reordered/adjusted to save time. In any case, with this general approach, a sub-2 hour run is to be expected.

I think moving forward I might experiment in creating a level 1 bard just before entering Mangar to avoid the Red Dragons. In the worst case it’s time neutral if it avoids the fight. Average case it probably saves 15-20 seconds. And in all cases, it avoids a team wipe.