Elder Scrolls Daggerfall & Morrowind revisited (Pt2)

Our story so far

In Elder Scrolls Daggerfall & Morrowind revisited (Part 1) I talked about some of the problems with The Elder Scrolls (TES) games. Just to be clear, I am a huge fan of TES. If I am critical of the series it’s only because I want it to be better. I also wish to bring back aspects of the games removed to make them more marketable to the general public.

Much of my motivation to publish articles on the subject of TES series have come from Youtube videos. In particular, The Elder Scrolls: The Dumbing Down and Zaric Zhakaron’s series “What if ____ was good?” Both of these talk about how Bethesda has made sacrifices in order to produce games for a wider audience. After Daggerfall, TES games were made more limited in scope and easier to manage. In addition, Bethesda has increasingly adapted games to be more multi-platform friendly. This meant that it was necessary to remove elements that did not lend themselves to simplistic controls and limited processing power. In effect, newer TES games were limited to the lowest common denominator, the console.

The reason I bring these things up is to make it clear how amazing it was when Bethsoft created Daggerfall. Daggerfall was one of the most highly anticipated CRPGs ever produced. Daggerfall did what no other CRPG before or since has ever done. Developers of the game made a real effort to create a simulation of a full sized, immersive world. This attempt was far from perfect. As I have said before, the computing power and technology to do what they were trying to do simply did not exist at the time.

Morrowind, the good, the bad, and the ugly

Morrowind was the next step in the evolution of the series. This game sought to correct the problems in Daggerfall by handcrafting all aspects of the game. While this produced a visually appealing game, the drawback was a severe reduction in emersion and realism. As stated in part 1, the world was drastically reduced in size. This was a disappointment to fans of the series who had hoped to see a large world with improved generation of locations, terrain, NPCs, etc. Instead, we got the Disneyland version of Morrowind’s island of Vvardenfel; a country-sized island shrunk down to an easily walkable size with some rides to make it feel like you were really getting that Morrowind experience.

Rarely considered, is how much of the “handcrafting” of impassable terrain and obscuring flora as well as the addition of game mechanics like the silt striders, teleporting, boats, etc. are less about making a better, more immersive world than you might think. Rather, these things help conceal the true size of the game space while at the same time steering the player in the direction the game designers wish them to go.

How can this be fixed?

As it happens, most TES fans did not start out playing the early games in the series. Consequently, they have little sense of loss when the large, 1:1 world of Daggerfall went away never to return. It’s much the same with the RPG aspects of TES games. When skills, stats, and magic got watered down in later games, few players felt that anything had been taken away. The same goes for the open world. Many see no need for areas in the game that do not directly impact gameplay. For me, the ability to just wander about is a wonderful part of the game.

Blow it up

Okay, first, the play space needs to be enlarged by a couple of orders of magnitude. The tiny islands surrounding Vvardenfel need to be islands, not rocks. Cities on Vvardenfel need to be cities and not just a collection of houses. Take the city of Vivec. Where do people live? There are places for the craftspeople etc. but virtually no one else. If such a city existed, there would need to be houses and farms and people to support such a city. Vivec is mostly a temple complex. There would need to be a town of people nearby to call it a city. The same is true for the other cities on the map. These need to be true cities. None of the cities in the game have anything resembling an infrastructure, making daily life possible.

Many areas on the map exist merely to separate other areas with terrain features, etc. This can be fixed by stretching the map like the skin of a balloon. Give cities room to expand. Provide large areas of farmland to feed a realistically sized populous. Of course, some areas should be left much as they are. The Caldera mines should remain near Caldera but should not be just as close to Balmora.

Petting zoo for n’wahs

Next, remove static respawning monsters. Seeing the same monsters over and over in the same area, again and again, is tedious. OpenMW with its greatly increased view distance and ability to see statics far away makes this flaw very obvious with dozens of cliff racers hanging motionless in the sky or flying in simplistic patterns. Of course, once you make the game space large enough there is no longer a need to have static spawns. Instead, it’s possible to have a percentage chance based on environment type that particular creature will spawn. A spawned creature may come from a burrow, nest or some similar place and possibly be part of a pair, pack or family group. Creatures should not necessarily be automatically aggressive unless defending an area, nest, young, etc.

All creatures should be limited by ranges, food sources, competition, and predation. An aggressive creature that ranges near a settlement would likely be hunted down. Hunting will cause wildlife to diminished around population centers and thoroughfares. For players, hunting creatures should be a challenge as wildlife learns to avoid people. An ecosystem should be self-balancing and feel that way.

Bringing sexy fast travel back

In the real world travel between cities takes a long time. In the land of Tamriel, this can be many days or weeks. Travel in Daggerfall consisted of both land and water travel. A number of factors determined travel time. Would you be sleeping in inns? Did you own a horse? Would you be traveling by ship? All of these factors affected your speed and the amount of money it would take to travel. In Daggerfall, you were able to purchase a ship. This allowed you forgo the expense of paying passage. Morrowind travel is by foot, silt strider, boat, ship, teleportation. Often, getting to a location will involve a combination of these.

Bring back the open world

Give the player the option to use fast travel to move to anywhere on the map. Fast travel in the Baldur’s Gate series was a good mechanic. As the player fast travels down a particular road an encounter happens specific to that location. You might exit fast travel as you: are waylaid by bandits, meet a strange traveler, are handed a mysterious note, etc. Of course, there is no reason that a triggering event needs to be a particular location. Perhaps you have a bounty and someone spots you. It could be a particular time or any of a number of things.

Your ability to travel in certain ways could expand as you discover perhaps secret means of teleportation requiring a special item or become a member of the mages guild, etc.

Why is this a good time to talk about this?

The OpenMW project has produced a game engine without the map size, view distance and other limitations of the original game. As a consequence, it now possible to address the issue of the cramped game space. The technology to produce realistic terrain via procedural generation has advanced. This technology can produce virtual land, rivers, roads, mountains, farms, towns, etc.  All without giving up the hand-crafted aspects of TES series.

About the featured image:

Balmora. Aerial view
by AlexeyRudikov
An artist’s re-envisioning of the city of Balmora on the island of Vvardenfel

Elder Scrolls Daggerfall & Morrowind revisited (Pt1)

An exciting time for old games

Lately, I have been revisiting The Elder Scrolls series. In particular, I have been playing TES III, Morrowind. What makes this decade and a half old game really enjoyable to play, once again, is the work of the various fan projects that are now starting to bear fruit. Among these, is a project that has produced a game engine that’s, in ways, better than the original (more later).  The unfortunate side effect is the capabilities of this engine now reveal the real limits of the original game. Fortunately, the same sort of engine update is now starting to show off the efforts of the creators of Daggerfall with its unprecedented 1:1 scale.

A world compressed

When Morrowind came out I was rather disappointed. Not by the gameplay or story but by the world itself. Bethesda rescaled the island of Vvardenfell making it too small by a couple of orders of magnitude. From one city to the next is a 10min jog. If one is able to fly the time is even shorter. An area that should be about the size of France is, in fact, smaller than Paris; Paris, Texas.

The Elder Scrolls Morrowind island of Vvardenfell scale size comparison with Paris, Texas

The game hides its true scale by limiting view distance and providing various means of fast travel to different locations on the map. Somewhat like taking an Uber to the corner store while being charged like they are driving you to another city.

A world before

The series’ first game “Arena” was good but largely unremarkable. It introduced the lore, the continent of Tamriel and it had an interesting story. It was also the only Elder Scrolls game I’ve ever actually finished. I have played all of the main games, up through Skyrim. However, in most of the TES games, completing the main quest has seldom been my goal. For the most part, I just enjoyed playing and exploring.

This brings me to the subject of the previous game, TES II Daggerfall. Daggerfall was a game I very much looked forward to playing. The game limited the player to just the area of the Iliac Bay. When I say “just the area of the Iliac bay” this should not be taken to mean the area was small but rather it was the biggest area ever put into an RPG before or since.

Daggerfall was a monumental undertaking. The problem was, that at the time, the technology was unable to really handle it. The 3D world of Daggerfall was entirely populated by low-resolution 2D sprites turned to face you like billboards. Some of the people in the game were animated sprites and some were even animated from multiple angles to present the illusion of 3D.

A world with problems

On top of this, there were a couple of big issues. Number one was a bug that would sometimes cause you to fall out of the dungeon or world entirely. The other was the vast number of adventure locations were procedurally generated. This in a time before procedural generation could be said to be at all developed as we think about it now. Dungeons were often random spaghetti bowls that made no sense, with a quest item randomly placed somewhere inside. You might go to a dungeon, kill a monster in the first room, and be done or crawl through for an hour until you find a chest in some obscure unguarded room, behind a secret door, you missed somehow.

A world renewed


Thanks to the ever-increasing popularity of TES games there are a number of projects that have produced remarkable results. The Unofficial Elder Scrolls Pages website is among the oldest. UESP has been around since the mid-90s and is one of the best repositories of TES lore and other information anywhere. There are a couple more sites like this, but UESP has been around since the very beginning.

Tamriel Rebuilt

Next on my list is Tamriel Rebuilt. This project was organized when Bethsoft announced it would not be including the whole of Morrowind in TES III Morrowind but only the island of Vvardenfell. The project seeks to expand the game to include not just the island but the mainland of Morrowind. At one time, they planned to eventually include the whole of Tamriel (hence the name). To say the project was, and is still, ambitious, is something of an understatement. This project is now over 15 years old and still only about half or a bit more finished. Still, what TR has published is a truly impressive expansion to the original game.

Project Tamriel

Project Tamriel is a sort of extension to TR, adding landmass to the original game. Basically, it addresses expansions for TES III outside of the Morrowind province. It also provides for the sharing of assets between these fan-made expansions. Right now, the three expansion mods under construction are “Province: Cyrodiil”, “Skyrim: Home of the Nords”, and “High Rock 427”. The sharing of assets is a pretty big deal. It minimizes duplication of effort and allows the maximum number of people to create assets since assets from one mod get shared with all other mods.

Open Morrowind

Lastly, as I talked about in a previous post, there is the Open Morrowind Project.  After ten years of work, it’s nearing full, 1.0 release and is at this point just being polished. OpenMW is a free, open-source, replacement engine for the original Morrowind engine. The project’s goal is to release a game engine, virtually indistinguishable from the original, capable of playing Morrowind using all the power of a modern PC: widescreen, multi-core processors, multithreading, GPUs, shaders, textures, bump mapping, etc. One exciting thing about the project is it eliminates many limitations of the original, now 15-year-old, engine. Also, because it is open sourced, it can continue to be updated into the foreseeable future. After the 1.0 release, the team plans to start adding additional features not present in the original Morrowind.

A world revisited

I’m going to talk about the future of my two favorite games in TES series, Elder Scrolls Daggerfall and Morrowind. I will be talking about Morrowind in part two because there is a lot to cover just with Daggerfall. Both of these games tended to stress the role-playing part of the action RPG. They tended to be less scripted than later TES games, (though Morrowind moved a bit more in this direction) and had a more free-form and open world.

We’ve already discussed the problems with these two games, as they were released. In short, Daggerfall suffered from Bethesda’s lack of the technological ability to generate a truly immersive world with complex, procedurally generated, details and connections. Morrowind, on the other hand, lacked Daggerfall’s 1:1 scale, with immersion into a world that was truly real in size. It also lacked a lot of the details that Daggerfall implemented. These details increased immersion but didn’t really benefit gameplay.


How can we fix this? Let’s first start with Daggerfall. The size of the game needs no fixing, it’s huge. What needs to be fixed is the lack of connection that the randomly generated people and locations have to each other. The game is filled with tens of thousands of NPCs; people without lives. In addition, Daggerfall has 15,000 locations. Most are remote and unconnected with the world around them. Thousands of dungeons, crypts, farms, towns, etc. are random, uninteresting, populated with monsters that exist merely as something to kill or people with nothing to say, with no links to anything.

Fortunately for TES fans, Bethsoft has released Daggerfall for free and there are projects that seek to update the primitive, DOS-based, 3D into something a bit more modern. Hurray! Hopefully, soon we will see a more playable and enjoyable Daggerfall and possibly more.

Daggerfall II

Let’s for a moment imagine a fan-based effort to not only fix the bugs, etc. in Daggerfall but really update it with modern technology. We’ll call this mythical project Daggerfall II or D2. Of course, it should have persistent terrain, true 3D objects and the ability to see things just as we do in the real world. The Daggerfall Workshop has brought this closer into being. Take a look at this video to see an example of this.

A living world

Wouldn’t it be wonderful if instead of a stagnant world we had a world in which people had jobs and lives etc.? There exist, now, algorithms that predict with good accuracy out of a city of say 10,000 how many of those would be cobblers, weavers, tailors, dyers, potters, plumbers, blacksmiths, jewelers, brickmakers, bricklayers, carpenters, masons, prison guards, town guards, tax collectors, nobles, bankers, glass blowers, cabinet makers, shipwrights, fletchers, roof thatchers, woodcutters, school teachers, homemakers, basket weavers, bakers, butchers, etc. Every person you stopped on the street would be going somewhere to do something. There would likely be a market square in any sizable town where shops would be, along with carts of goods for sale lining the streets.

Possibly, goods would make their way to the cities via roadways. Imperial guards would sally forth from cities, forts, and depots to patrol the roads of the empire and ensure the smooth continuance of commerce. Farms would cluster around large towns. Trade caravans would move between cities on highways and rivers. Fishing towns would provide a constant supply of seafood nearly year round.  Ports would provide exotic goods from places near and far.

A created world

In our imagined game, all these things become part of the rules that help generate our procedurally generated world. Importantly, areas within half a days ride of a large city are filled with farms and villages. These places create goods to be consumed in the city. It’s not difficult to calculate the area an individual family could sow with an ox or horse-drawn plow. We have seen the result from hundreds of years of planting and the moving of stones to the edges of the field results in built up hedgerows. Importantly, where you find farms, you find farm to market roads. You will also find farriers, smiths, and general stores, etc. Farms would have roads connecting them with farm towns or villages and those would have roads connected with highways and other towns.

Can it be done?

Of course, this might seem a massive undertaking. In game terms, these rules would tend to govern who the NPCs are what they know and where and when you would likely to meet them. For example, traveling out of a city in the morning you would most likely meet farmers and craftspeople moving into the city to sell their goods. At evening they would likely be returning home. Along a highway, you might meet a trade caravan or an Imperial patrol. In the woods, you might meet a hunter or a woodcutter. Villains and monsters would range out from nearby dungeons. Bounties might be posted in towns plagued by nearby ruffians. Migrant workers would journey to farmsteads in spring and fall to make a few septims during planting and harvest season and home again after.

A player might, for instance, meet a farmer traveling in a cart. “I’m taking sacks of grain to the mill at Lomton to be ground into flour.” the farmer might say. That flour would then be used to fulfill the existing orders of a city or town some distance away. The player might have thousands of encounters like this with no two alike. Each NPC would have limited knowledge centered around local lore. The NPC might have a small chance of producing a breadcrumb leading to knowledge of a local dungeon or quest item.

Other things to fix

Most important, locations, especially underground locations should be specific and make sense. Places should have histories and purpose: Mines converted to underground lairs; A natural set of caverns expanded with passwall spells or forced labor or dwarves, etc.; Creatures in hiding from the Empire; A criminal organization; The headquarters of a forbidden cult or wizards looking to research in secrecy. Moreover, in each of these scenarios, the location should follow a logical plan. Items, rooms, monsters, traps should all follow the theme along with a physical structure that makes sense. Additionally, quest items should be in logical places. Are you on a quest to recover mummy wrappings? The mummy could be the man or woman who constructed the place and they have laid about them a number of traps and monsters to guard their tomb.

Secondly, commerce is wide open in Daggerfall. Future versions of the game could allow this to be expanded upon. Travel about the Iliac bay in your ship buying and selling: Alik’r silks, Daedric weapons, herbs, potions, magic, etc.

Finally, co-op play has been toyed with in of TES games where you become a party of two or more adventures. With some effort, this could be added to a revised Daggerfall. In future, this would allow a single player type game with companions who are PCs. Perhaps companion characters could be played or left on autopilot. This would allow people to come and go from the hosted game as they wish.


Because there is no deadline to get D2 out, in future, the game can continue to expand, adding more: features, detailed quests, improved graphics, etc.

Part 2 Morrowind 

About the featured image:
Screenshot from the Daggerfall Unity replacement engine
The image shows the tremendous view distance now possible with this update to the old DOS-based engine

The Elder Scrolls continues to be updated after Bethsoft moves on

TES Revisited

I’ve been revisiting some of the earlier The Elder Scrolls (TES) games lately. I keep an occasional eye on one of the oldest and largest projects, Tamriel Rebuilt. This project started in 2001 before Morrowind was released. For the amount of work that has gone into Tamriel Rebuilt, they have been very slow to release content. This is largely due to TR’s commitment to quality. Their rule is to release only material as good as Bethsoft would have, had Bethsoft created it.
The scope of the project is to release material that is about twice the size of Morrowind when it came out. In essence adding TR triples the world space of Morrowind. The team is also working with other modders doing projects around Morrowind with the thought that these things will eventually mesh. The team has also joined with Project Tamriel to establish a unified asset creation depot. Assets from other TES mods are available to the TR team and vice versa. Assets can be used in other TES mods. This is done with only one file to download. This is sort of modding Holy Grail.

An Open-source Replacement for the Morrowind Executable

In related news, OpenMW is coming along nicely. OpenMW is a Morrowind replacement engine. At this point, Morrowind is totally playable with OpenMW. The other day I cracked open my copy of MW and installed it. Afterwards, I installed OpenMW and found that there was virtually no discernible difference in play experience between vanilla and OpenMW. This was until I started playing with the CFG file; turning on the landscape options. BANG! My view distance became HUGE. 
In Morrowind, view distance is very restricted. With Vanilla MW, you walk around in a fog. As you look at parts of the sky, those parts disappear; covered when a building spawns. I find this tends to break my suspension of disbelief. With OpenMW loading lots of cells no longer bogs down game speed to unplayable levels. The executable uses a modern, multithreaded, engine. As a result, OpenMW can handle the added load. Looking across the map is amazing. Have you ever wanted to look at the whole of Vivec? Now you can. Of course, the side effect of this is it shrinks the perceived size of the world space.

TES Episode Two: Daggerfall

This brings me to the topic of Daggerfall. While messing with all this MW stuff I got to wondering ‘Is anyone doing anything similar with Daggerfall?’. Well, they are. Daggerfall Workshop has ported DF to the Unity engine. Unity is a well-supported engine with 3D, etc. Terrain in Daggerfall can now be better rendered. View distance is dramatically increased.  Currently, Tamriel is still populated by sprites, but textures can now be bump mapped, caused to glow, given shine, etc. Dungeon mapping has improved significantly. The sky looks amazing.
In terms of playable world space, Daggerfall is the largest RPG ever produced; 62,394 square miles, over 15,000 towns, cities, villages, and dungeons for the player’s character to explore. Daggerfall was essentially the “No Man’s Sky” of 1996. The problem was then in ’96 procedurally generated content was still very primitive. One of my dreams is that the assets of Daggerfall could be upgraded with more sophisticated content generation and some more handcrafting. I still want TES games to be huge. The games should have human-directed, procedurally generated, content included. I feel certain, this will be the future of content creation.
Daggerfall was like a human-defined canvas. Specific things were placed by hand. Then a machine sort of randomly splattered bits of paint on that canvas. Conversely, Morrowind had the creators hand stitch the canvas and then each one dipped hatpins in paint and placed every dot of color. I would have greatly preferred that the canvas of Morrowind been like that of Daggerfall with the handcrafting saved for dungeon locations, cities, etc.

About the featured image:
Screenshot from the OpenMW replacement engine
The view distance has been greatly increased. The downside is this reveals some real flaws in the game.