Where is L3 Deep Dwarven Delve located?

The Area Map from L1

In my previous post, I noted that the published version of L3 Deep Dwarven Delve (1997) omits a separate map showing "an aerial view of the mountain and surrounding areas" that the cover of 1979 draft indicated would be included along with maps of the three dungeon levels. In the published module, the map of the first dungeon level does include an outline of the hill at 80 feet, which is the height at which this level is located, and also three surrounding lower topographic lines at 20' increments. This could be considered "an aerial view of the mountain", but there is nothing on it showing "surrounding areas".

The two earlier entries in the L-series, L1 The Secret of Bone Hill (1981) and L2 The Assassin's Knot (1983), each had large area maps. The one in L1, shown above, is drawn in a black & white topographic style similar to the famous one in B2 Keep on the Borderlands, but employs hexes rather than squares, and depicts a much larger area, about 28 miles east-west and 17 miles north-south. The map in L2 overlaps with this to the south, but only adds about half as much territory because much of the map is water. Here is a fan-made splice of these two maps, from a post on the Restenford Project blog, which may be useful if you are running a Lendore campaign:

Area Map combining the maps from L1 and L2

In contrast to the earlier modules, the content in L3 consists of one large dungeon under a single hill, which means it doesn't necessarily need such a large area map, particularly if its location can be referenced using one of the earlier maps. But L3, in addition to omitting an area map, does not even clearly indicate where the Delve is in relation to the maps from L1 and L2...! 

Instead, it keeps the location vague. The "Background" relates that humanoids have been attacking Restenford and Lake Farmin (aka Garrotten), and that after the most recent attack on the former, "a member of the militia, a ranger, tracked the humanoids back to their lair" and thus "[s]omewhere in the dark wilderness nearby lurks a great threat..." (page 3). The module further explains that the town council will "provide the directions to the Delve..." (page 4, "Preparing to Play"); and that "[t]he Delve resides beneath one of the many hills in the area" (page 7, "External Locale"), which could be almost anywhere on the L1 or L2 maps, each of which depicts numerous peaks, or somewhere off the edge of those.

To some degree this vagueness may be because the adventure was introduced by means of an event in Lakofka's campaign. In comments on FB, he wrote: "L3 starts with the attack on Restenford in 576CY. The party follows the retreating humanoids back to their lair: an ancient Dwarven cave complex" (see here); "...But the little road to the front door is not obvious. A ranger tracked them and came back to Restenford to report ... He told the adventuring party what to look for and left a token along the road that tells them where he spotted their movement. Once they know the proper peak and the approximate location of the main entrance they should be able to find it" (see here); and "If it’s too easy to find others will go up and discover the orcs etc" (see here). So, it's clear that he intended for the Delve, in game, to be a "hidden location" that is not accessible until it is "unlocked" by the attack.

But note that while Lakofka didn't want players to find the Delve early, he also didn't intend for the published module to keep it completely hidden from the DM, as evidenced by his original 1979 intention to include an area map, and also more recent comments: "I looked at L3. There is no area map. I would not miss something that basic. But TSR did. The entrance is up 80 feet and the hill at that point is around 2,000 feet. The height of the entire hill is not given" (see here). (Note that on review there's nothing in L3 stating the hill is 2,000 feet high at the entrance; in fact, it clearly states, "The main entrance is some 80 feet above ground level and cannot be seen from the base of the hill" (page 7)).

The lack of a specific location poses some problems for actual play as part of a campaign. In what direction do you tell the players they are headed? What do you do once they get there and want to leave and then return? Or what if you want to include it as a hidden but possibly findable location in the Lendore sandbox from the get-go?

Lakofka's aerial map for L3

Fortunately, in 2018, Lakofka found among his papers a draft of the aerial map of the exterior of the Delve, and shared this on FB, where he wrote:

"And I found the real location! Guardian Peak. On the L1 map it’s labeled Garden Peak!!"

The map he shared, shown above, is possibly the exterior map referred to in the 1979 draft, or a later drafting of it, as to me it looks like something drawn on a computer program in the 1980s. It's not an area map like in the earlier modules, just an immediate location map showing the exterior of the hill up to its peak at 3,145 feet, and with topographic lines at 500 foot intervals. It shows the two entrances to the dungeon, each at 2,000 feet as referenced in Lakofka's quote above, rather than at the 80 feet indicated in the module (and which means that TSR most likely did not have a copy of this map when they produced L3). 

It also depicts the "Humanoid Trail" leading to Entrance A, which Lakofka referenced in one of the above quotes. The published module instead describes this singular trail in the plural, describing that "[t]he humanoid trails leading from the front gate (entrance A) are well hidden. At ground level, the trails can be found only by inspection and are not obvious to casual searchers". 

Most crucially, there is a label added to the map in pencil reading, "Garden Peak aka Guardian Peak". If you look at the L1 area map at the top of this page you will see a Garden Peak near the upper center. The reason for the dual names is that while the peak is labeled Garden on the L1 map, the text of the module refers to it as Guardian (page 8). Lakofka's comment quoted above implies that Guardian is the correct name. In a game, one might use both names if Garden is assumed to be a corruption of the earlier name, Guardian.

There is a discrepancy in the contour lines shown on the aerial map of Guardian, which are at 500 foot increments, and the map in L1, which the key indicates are at 400 foot increments. When he found the map, Lakofka noted this discrepancy, and issued a correction: "Based on the L1 map these elevations are incorrect. The contour lines should be 400 not 500. The very top contour line should be eliminated. So the A [and] B entrances should be on the 1600 foot line and the peak 2145 not 3145" (see here). The L1 map shows only five contour lines between the river and the peak of Garden, so this correction does allow the Guardian map to fit in better. Even adjusting for this, the shape of the contour lines on the Guardian map don't quite match the ones on the L1 map, but are close enough to use. If using this map, make sure to note the direction of north on the Guardian map; it should be turned to line up with the map in L1.

In retrospect, Garden/Guardian Peak is a good location for the Delve, as there are no encounters or areas of interest that are specifically tied to it in L1. It's referenced a single time in the text of the module (page 8), where it is simply given as part of a list of locations, also including Lark Hill, High Top, Low Point and Reddy Forest, that are described as potential sites for encounters with NPCs: "These sites are often used as campgrounds by travelers, and for the purpose of this module they will be sites for special encounters. Other hills and forests may also be added to this list if the DM desires". The text goes on to describe four NPCs that can be encountered in these areas, about 1/3 of the time a random encounter in indicated. While it's a bit difficult to believe that a lone NPC could safely camp on Guardian if it is home to the large force of humanoids found in the Delve, but it is easy enough to remove Guardian from this list and make it more desolate. The module further describes these peaks as "grass-covered, with bushes and rock outcroppings every 50 to 300 feet. Small stands of trees are quite common. The larger copses are shown on the map", which fits well with the aerial map shown above, although there should be one "larger copse" at the north end of the map.

In conclusion, this unearthed treasure provides us with a specific location for the Delve in relation to the other L-series locations. The dungeon in L3 is very linear, but clearly situating it in the L1 Area Map helps to give it more of a feel as just another location in the sandbox.

L3 Deep Dwarven Delve: 1979 draft

Lakofka's photo of the vintage typescript draft for L3

The late Lenard Lakofka is probably best remembered for his long-running, detailed-oriented column in Dragon magazine, Leomund's Tiny Hut (1979-1986), and for writing the classic AD&D modules L1 The Secret of Bone Hill (1981) and L2 The Assassin's Knot (1983). The former is innovative as a small sandbox, the latter as a murder mystery, and they were drawn from the adventures he ran in his home campaign set in the Lendore Isles, which were later incorporated by Gary Gygax into the World of Greyhawk (1980). 

No further L-coded modules appeared during the era of 1st Edition AD&D, and thus for many years that was it for the series, as far as gamers knew. However, after ownership of D&D passed from TSR to Wizards of Coast, they surprised us by publishing a third installment for the 25th anniversary of D&D. Specifically, the 1999 Silver Anniversary Collector's set included the module L3 Deep Dwarven Delve, touted on its cover as "the last 1st Edition AD&D adventure ever to be published!", because it "lain unseen and forgotten in the TSR design vault for twenty years". However, as Shannon Appelcline reports on the DriveThruRPG page for the product, this story may only be considered accurate if you expansively include Lakofka's home as part of "the TSR design vault":

"[Sean K. Reynolds of WOTC] said that all of TSR's copies of the adventure had been "lost or destroyed" over the years. The adventure (apparently) resurfaced only when Lakofka found a copy around his house and sent it to Roger E. Moore in 1997 ... [who] then passed the adventure on to Reynolds in 1998."

Furthermore, as Appelcline explains, even after the original was located, publishing it was not without snags:  

"[Wizards] thought [Delve] needed "depth and clarification" to bring it up to modern AD&D standards. Lakofka was happy to oblige and produced a new version of his adventure… which Wizards again lost. Lakofka says that he didn't hear about the loss until after "Delve" was published, by which time a number of Wizards developers had stepped in to do the required expansion for the adventure ... Lakofka says that "Delve" is about 80% comprised of material he'd turned in two decades earlier."

The reception to L3 was somewhat mixed. While most were grateful for another AD&D module, especially one that written during the original era, some were disappointed that it wasn't as innovative as Lakofka's earlier modules, being a rather linear dungeon crawl. And being a limited edition, copies became increasingly expensive over the years, although now you can get an inexpensive pdf or print-on-demand copy from DriveThruRPG. 

In the years after L3 was published, Lakofka became active in D&D circles again, eventually releasing more Lendore material through Dragonsfoot, including L4 Devilspawn and L5 The Kroten Campaign.

Another twenty years had passed when, in 2018, Lakofka once again located in his house a copy of the draft for L3, in a formatted typescript, and posted a photo of it (shown above) in a comment to a FB group, the Flanaess Geographical Society.

It's exciting to see this typescript draft, particularly because the cover is laid out in vintage TSR format. I don't recall ever seeing a draft of this type for any other TSR module. And with "FINAL CORRECTION COPY" written across the top, it suggests that at one point someone (Lakofka? a TSR editor?) considered it close to finished.

While his photo only shows the cover page and a small portion of one interior page, there are still interesting details to be gleaned: 

 

---The title is "The Deep Dwarven Delve", which became just "Deep Dwarven Delve" as published, although the interior text still refers to it by the original title in several places. 

---The cover has a copyright date is 1979 and uses the TSR Wizard Logo. Lakofka ran Deep Dwarven Delve at Gen Con 12 in August 1979, according to the program book, indicating that the adventure had taken shape by mid-1979. While the draft's 1979 date could just indicate when it was originally written, the Wizard Logo was phased out in 1980, and L1 employs TSR's next logo, the Face Logo. This suggests that this draft, which must have been prepared by an editor at TSR, actually does date to 1979 or 1980. 

---It refers to "one part of a four-part series", whereas the published version, which has different cover text, states that it was "[w]ritten as the concluding adventure of the "L" series". This suggests that Lakofka not only wrote this draft of L3 written in 1979, but also conceived that the series would include an L4 at the time. WOTC in the '90s omitted any mention of further unfinished work, possibly because they wanted to seem like they were bringing the L-series to a conclusion. 

---The reference to "three level maps" matches the published version, but the "aerial view of the mountain and surrounding areas", does not. Either this was never finished, lost, or omitted by WOTC. And this is a big omission, because there's no indication in L3 as to where exactly it is located on the memorable area maps found in L1 and L2. As Lakofka wrote on FB: "i looked at L3. there is no area map. i would not miss something that basic. But TSR did". 

---It also mentions "monster rosters", and the published module does, in fact, contain a section titled "Rosters" on pages 5-7, which has a list of monsters for each of the three levels of the dungeon. 

---It recommends characters of levels 2-6, with approximately 40 levels total, whereas the published module suggests 6-10 characters of levels 3-6, average 4, with about 35 total levels and no more than 45 (page 4). 

---The small portion of the interior page shown in the photo contains text that is close to the same material as published, with a few minor changes. The original reads:

NOTES FOR THE DUNGEON MASTER 

The Deep Dwarven Delve can be played at two levels; first as an orc stronghold (level one), and second as a hidden treasure store and place of great evil (levels two and three). The upper level of the Delve is filled with orcs, bugbears, ogres, trolls an a magic-user. They do not know of the.......................................first level of the Delve 

This was changed on page 4 of the published module to: 

PREPARING TO PLAY  

The Deep Dwarven Delve can be played as two linked adventures; first as a humanoid stronghold (Level One), and second as a hidden treasure store and place of great evil (Levels Two and Three). 

(the heading "NOTES FOR THE DM" was moved to page 3, after the Background)  

---The last sentence in the original paragraph exactly describes the composition of monsters found in the "First Level Roster" on page 5 as published. This suggests the types of monsters on the first level were not changed from draft to publication.

Lakofka later indicated he was preparing a comparison document between the two versions, but unfortunately ended up putting it aside when he moved, and never got back to it or shared what he had completed. He did, however, make the comment that "TSR decided to change the final encounter in the Delve (along with a few minor changes in other encounters)" (here on FB), which together with the details gathered from the draft cover page suggests that overall, the published version of L3 is not too dissimilar to what he original drafted in 1979.

In future posts, I will take a look at where L3 should be located on the L1 area map, and also what Lakofka intended for the original final encounter of L3.

STAR SCHLOCK KS Final Hours!

The kickstarter for the STAR SCHLOCK Battle Game, which I featured in my previous post, ends this evening. 

Based on all of the stretch goals that have been unlocked so far, a $95 Recruit pledge will get you the game plus at least 25 metal minis!

Find it here:

Star Schlock Battle Game

STAR SCHLOCK Battle Game (Kickstarter)

Scrum Club member John S., who blogs over at the 1000 Foot General, is currently running a kickstarter for his skirmish miniatures game, Star Schlock. As the tagline in the graphic says, this game is inspired by the tropes and aesthetics of all of those sci-fi shows and movies we loved as kids - Trek, Planet of the Apes, Buck Rogers, SW, etc. 

I've pledged for the "Recruit" level, which supplies the Star Schlock Battle Game set, including the rulebooks, dice, cards, organizing consoles, and tokens for playing the game, plus a starter set of 12 pewter minis (unpainted) - 6 each of Space Apes and Explorer Corps. Higher pledges levels include more minis. It's already blown way past its funding goal, with all sorts of bonuses for different pledge levels being unlocked.

Space Apes vs Explorer Corps

It's running for about one more week; find it here:

Star Schlock Battle Game

50 Years of Text Games by Aaron A. Reed (new book)

My copy of 50 Years of Text Games, which arrived recently

It was a memorable evening in the early '80s when I first encountered text-based computer games. While staying over at my friend Eric's house, his mom took us into her work place, which I think was a Motorola office, after business hours. 

She left us to ourselves in an office with a computer connected to a teleprinter that printed out everything that we typed and that it outputted, which I never encountered again. Eric already knew some of the games: a Star Trek one where you controlled the Enterprise, which we played briefly, and Adventure, which really struck a chord as we were already heavily into D&D. Eric got us past a few early obstacles, like the snake, that he knew about before we got stuck. Then we went to another office where two guys were also playing Adventure, like a scene out of AMC's Halt and Catch Fire, on a computer with a monitor. They let us watch, and based on my D&D knowledge, I suggested trying to burn the troll with the lamp oil, but it didn't work.

Without a computer at home I didn't get to play Adventure, or any text game, again for a number of years, but eventually we got a Tandy 1000 SL, and I bought Zork at Babbage's in the local mall after playing it at another kid's house and recognizing it was very similar to Adventure. Zork then led me to the myriad other Infocom titles, like Planetfall, the Lurking Horror and the Hitchhiker's Guide to the Galaxy, only some of which I was able finish. I even tracked down Adventure by sending away for a shareware/freeware floppy disk that also included another simplistic D&D-inspired text game, The Beginner's Cave for the Wonderful World of Eamon system. 

These were the only text-based games that I encountered back in the '80s, and after that I've only briefly played them, but always remembered the genre fondly and have been inspired by it when designing challenges for D&D adventures, like the Forgotten Smugglers' Cave, which contains an area like the maze of the twisty little passages in Adventure.

Now I just received a new book, 50 Years of Text Games, by Aaron A. Reed, that provides great context for my early adventures in text, and also shows how they've continued on over the years, and developed in different ways. Subtitled "from Oregon Trail to AI Dungeon and everything ... in between", it profiles one important game for each year from 1971 to 2019, stretching over 600 pages (!). The book started as a blog that was run throughout the year of 2021, and the posts are archived here for anyone interested. I picked up my hardcopy via a subsequent kickstarter, which I believe is now sold out, but a digital version is still available.

I've already read a few sections, including the intro to the 1970s and the entry for the ground-breaking Galatea (2000), and am enjoying the clear prose, and am looking forward to reading more. One takeway, also gleaned from following the Renga in Blue blog, is that there were far more text-based games produced back in the 70s and 80s than I ever realized.

See also: The Renga in Blue, where the author is attempting to play every text adventure game by year of release.

Vampire Bat & Were-Vampire Bat (New Monster)

These are two new monsters for Holmes Basic, the first of which is featured in Area 6 of The Forgotten Smugglers' Cave adventure (which is indexed here), and originally appeared as part of my list of One Hit Point Monsters.

Illustration by Lore Suto

Vampire Bat

Move: 180 feet/turn flying
Hit Dice: 1/8 (1 hit point)
Armor Class: 3 (9 while attached)
Treasure Type: nil

Alignment: neutral

Attacks: 1 bite

Damage: 1 point

A native of subtropical climes, the vampire bat has of late begun appearing in certain temperate coastal towns, apparently escaping from traders importing them for clients unknown. Roosting in colonies of up to 50 individuals, often in the warmer caves of the near underworld, they venture out in smaller groups (2d6 bats) to feed on livestock, other mammals, or even humans. 

When suitable prey is located, the bats will silently swoop down and attack from behind (surprise on 1-5 in 6, attack at +4). A hit inflicts a point of damage and allows a bat to attach and, unless stopped, automatically drain another point for each of the next two rounds (thus, three points total), before detaching and flying away. Attacks on an attached bat are against armor class 9, but a miss with a weapon will automatically hit the victim for a point of damage. Unattached bats will continue attacking in an attempt to distract the prey from disturbing attached bats.

For each round of attachment, there is a 1% chance that a bat will transmit a rare form of lycanthropy, causing the victim to become a were-vampire bat (see below) in 2d12 days.

Established colonies produce a layer of pinkish guano across the floor of their cavern, which which produces an acrid smell that can be detected from a distance, and is prized as a fertilizer (20 gp per bucket). Due to the strength of the odor and slippery consistency, any melee in such areas is conducted at a -4 to hit, with a natural 1 indicating a slip and fall.

The bats sleep soundly in the lair during the day, but any activity in the lair has a 1 in 6 chance per character (e.g., 2 in 6 chance for 2 characters) of waking the bats. Once disturbed, all of the bats of the colony will join in on a frenzied attack on the intruders.

It is rumored that larger forms of the vampire bat exist in the tropics.

Were-Vampire Bat

Move: 120 feet/turn flying (giant bat form) or 60 feet/turn flying or 120 feet/turn walking (man-bat form)
Hit Dice: 2 + 2
Armor Class: 3 (giant bat form) or 5 (man-bat)
Treasure Type: C

Alignment: neutral/lawful evil

Attacks: 1 bite (giant bat form) or 1 bite and 2 claws (man-bat form)

Damage: 1d3 (giant bat form) or 1d6/1d3/1d3 (man-bat form)

The unfortunate victims of the vampire bat may, on occasion, be inflicted with a rare form of lycanthropy, eventually transforming into a were-vampire bat. Like the were-rat, the were-vampire bat may take three different forms: human, giant bat, or horrible hybrid man-bat. As with other lycanthropes, it is immune to normal weapons in the latter two forms. In man-bat form, it is less agile but stronger of limb, and able to attack with claws in addition to biting.

If possible, the were-bat will roost and hunt together with a colony of ordinary vampire bats, favoring old buildings that allow egress in either bat or human form. It can summon these bats to its aid at will, and will hunt together with a group of them in giant bat form.

Agate-Eyed Skeleton (New Monster)

This is a new monster for Holmes Basic, one which is featured in Areas 20 and 24 of The Forgotten Smugglers' Cave adventure (which is indexed here).

Eye Agate. Image Source.

Agate-Eyed Skeleton

Move: 120 feet/turn
Hit Dice: 2
Armor Class: 7
Treasure Type: agate eyes (100 gp each)

Alignment: neutral

Attacks: 1 bony hands
Damage: 1d6

Agates are purported to promote restful sleep, and certain funerary practices incorporate this belief by burying the dead with such gemstones. Some secretive groups go even further, replacing the eyes of the deceased with eye agates, and if these are of sufficient value (100 gp per eye), the strange energies of the underworld may transform the thusly treated remains into an agate-eyed skeleton, an unusual variant of animated skeleton. 

The practices that produce such skeletons means that they are typically found in groups in burial chambers. If undisturbed, the skeletons will remain at rest, but anyone gazing upon their agate eyes must save versus magic or fall asleep for 1d6 turns. Unlike a Sleep spell, characters so affected cannot be awakened prior to this by any means short of a Dispel Magic.

Damaging Skeletons. Any attack that inflicts damage on a skeleton, and does not destroy it, will cause it to immediately animate and attack with its bony claws. Once arisen, characters viewing the eyes no longer need to save versus magic. Furthermore, even if the skeleton is destroyed, 1d6 other skeletons in the same area will animate each round until all in the area have joined in. The assault will continue until the intruders have been destroyed or have fled the area, although a successful turning (as a ghoul) by a cleric will send the skeletons back to their resting places and prevent any further ones from animating.

Moving Skeletons. Any attempt to move a skeleton will, after a one round delay, result in the same effect as damaging the skeleton. However, if during this round the bones are placed in a specially consecrated ossuary, which may or may not be located nearby (DM's discretion), the skeleton will not animate and both agates can be safely removed.

Removing Agates. If a character attempts to remove an agate without disturbing the skeleton, there is a small chance of success. The base chance is equal to the attempting character's dexterity (e.g., a 10% chance for a dexterity of 10). Thieves add their chance of removing traps to this (e.g. a thief with a 15 dexterity and a 15% chance of remove traps would have a 30% chance). If this roll is failed, the eye is removed but the skeleton then animates as if damaged and will incessantly attempt to retrieve its eye, to the point of relentlessly stalking the offending character. Removing a second eye agate from the same skeleton will automatically cause it to animate.

It is rumored that placing such gems in the eyes of skeletons before casting Animate Dead will also produce an agate-eyed skeleton.