Total War Blog

Archive for the ‘Guides’

Multiplayer update Q&A

We have made a number of improvements to multiplayer in Napoleon and The Peninsular Campaign in the latest update, and caught up with the Multiplayer team for the full scoop:

Since the update to Napoleon: Total War, what new units are available in multiplayer?

The whole Spanish faction has been completely revamped for the Peninsula Campaign, and this new Spanish faction is available in multiplayer for everyone, whether or not you buy the DLC. This includes new uniforms, as well as new and revamped units. Spain now gets horse artillery and lancer cavalry, as well as having their heavy cavalry and guard infantry options reworked slightly.

However, the biggest new addition to the Spanish faction is the guerrilla units. These units are experts at ambushing the enemy, and can deploy anywhere on the battlefield, except your enemy’s deployment zone. These units are only available to users who purchase the Peninsula Campaign DLC.

Can you tell us more about the guerrilla units available to owners of The Peninsular Campaign DLC?

The Spanish faction has gained 6 new guerrilla units, which are all capped at one unit each. These can deploy anywhere except your enemy’s deployment zone. This is a totally new concept for Total War, and the possibilities it creates for multiplayer gameplay, and the new strategies that will emerge, are really exciting. We can’t wait to see how players start using these new units and the interesting outcomes that will occur. We’ve been playing a lot with the guerrilla units in multiplayer internally within the studio, and they really are great fun.

What has been done to fix “failed to connect” issues in multiplayer games?

Players hosting games will now host on a fixed port, which is 51914. If this port is unblocked the game will use this first, before trying the old (much larger) port range. Players who are having difficulty connecting to other players should unblock this port and may find their issue fixed.

Previously, players can only do battle against people in the same geographical region. Has anything been done to address or improve this?

We’ve removed the Steam regional division between servers, so now everyone online can see everyone else’s games. This greatly increases the number of games available, and also increases the pool of players in quick battle, drop-in battle etc.

Can you tell us more about the new multiplayer maps that have been added to the update?

We added three new maps: Pyrenees Peak, Galician Ria and Salamanca Province. These are free for everyone, you don’t need to buy the Peninsula Campaign to get them. When making these maps we looked at a lot of the feedback on the forums, and looked at the statistics of which maps people were playing most, as well as what maps we most enjoyed during the making of Napoleon. We tried to make maps that reflected what the community most find fun, and best reflect the emerging gameplay trends. As a result we tried to make them balanced and reasonably symmetrical, to ensure competitive gameplay, but also to give them a strong real-world feel.

What has been done to refine drop-in battle searches?

The drop-in battle UI has been redone so now players can explicitly choose drop-in battles, quick battles or both, and whether they want naval or land battles. You can now search for as many or as few of these options as you want, depending on your preferences. And because of the removal of Steam regions the pool of players you’re searching for will be larger than before, which should make matching faster.

The game is also more generous when allowing you to choose the drop-in battle option during a campaign. Previously it only allowed you to opt for a drop-in battle when the game judged the balance of power to be very even. We’ve relaxed this requirement somewhat (not completely) as we found that, when choosing to enter a drop-in battle, players were more concerned about having fun than having a completely balanced experience (they have ranked games and regular quick battle for that).

Players will also find unit sizes are larger than they were previously in quick battles. Previously it was always choosing small unit sizes, whereas now it will look at each player’s unit size settings and choose the largest size supported by both players.

For all the latest Total War information, stay tuned to, and become a fan on Facebook and Twitter.


Mark O’Connell
The Creative Assembly


Empire: Total War Modding System

Hi All,

Since the mod community seems to be moving full speed ahead and you’ve already got some basic mods up and running I thought I’d explain the way the Empire modding system is intended to work. This should enable you to work with it, not against it, and hopefully avoid breaking things released by CA in the future.

First of all, the ETW data is meant to stay packed. The pack file system and the virtual file system that it builds internally within the game are central to the patching and modding systems, so please don’t ask users to unpack their data in order to run with your mods.

So how should you mod the game? Well, it works like this. Empire loads files of type .pack in the data directory in a certain order determined by the pack’s type. There are three basic types – release, patch and mod. The fifth byte in the pack file header is an numerical id which tells the game what a pack is, boot packs are 0, release packs are 1, patch packs are 2, mod packs are 3. The game then loads release packs first, then patch packs, then mod packs.

At game startup, the file entries in all of these packs are mounted into a virtual file system, and if the game finds an entry in a later pack which has the same name and location as an already existing entry, the new entry replaces the old. This means that a mod pack can replace any file in the vfs and add new ones as well, without asking the person who downloaded the mod to overwrite files on disk and destroy data.
In order to tell the game to use a mod, you do need to feed the game a script command. Currently the easiest way to do that is to create a user script file. The way to do this is to find your preferences.empire_script.txt file (this will be in Application Data on a Windows XP machine), and create a text file next to it called user.empire_script.txt. Inside that file, for each mod create a line which reads “mod my_pack.pack;” (without the quotes, obviously). The user script gets processed on load, and will tell the game to load your mods in that order. The user script can be used for many other things as well, but we’ll leave those for another day.

So basically, a mod is intended to be a single “mod pack” file, and installing a mod is just placing that pack inside the game’s data directory and adding the mod command, and uninstalling it is removing those things.

Now, a note on how to work with the databases. You’ll have noticed that each table in the databases (inside the db folder) is both a folder and a binary file. When the game instantiates the databases, which happens after the virtual file system is built, it scans each directory and parses -all- of the binary files it finds. This means that a mod pack can add a new binary data fragment inside the db/unit_stats_land_table folder which adds several new units, and the game will add them to its land units database. The fragments are processed in the same order as the pack files, and any new database records with the same name as an old record will cause the new to replace the old. This means that you can replace any record within the database without touching the rest of the data or modifying the CA-supplied files on disk.

For example, say the CA packs contain a unit record for ‘french_dragoons’, which is one entry out of hundreds inside the shipped unit_stats_land_table data fragment, which is used to build the unit table that was available at launch. Now say you create a mod pack which places a new data fragment, say called my_mods_unit_stats_land inside db/ unit_stats_land_table, which contains just one data record for ‘french_dragoons’. What will happen is your new record will end up replacing the current record for that unit in the game data, and the rest will stay as it was originally.

So… hopefully you’ll now have a good idea of how the modding system was architected, and how you should build tools to work with it. There’s also another interesting consequence of how the system works, which is that it opens the door for mods built on top of other mods…

One more thing to keep in mind is that we’ve not yet finished testing the mod support fully – that’s part of what’s happening here before the mod dev package gets released. We’re aware that there is currently an upper bound of about 6000 files in a single pack, though it’s very possible to create a mod spanning several packs. But be aware that there is some overhead associated with mounting a pack, and a data directory filled with scores of them will make the game startup time significantly longer.

The other thing I should mention – since you’re going to be experimenting with this – is that the game checks the final database integrity after loading mods. So if you add a pack file which includes a new unit record which doesn’t include a display model or is invalid in some other way, the game will unload that mod pack (though it keeps the rest). This isn’t the main error checking mechanism in the ETW data pipeline, that’s done elsewhere, but as modders you should be aware of it. Unfortunately there currently isn’t any log output on the reasons why the mod failed to load, we’re looking into adding that when the modding dev kit is released.

Armed with this info you’ll at least be able to work with the system as it was intended, without having to wait for the mod dev package. I have to say it’s great seeing you guys making headway so quickly. The pack and database inspection tool was a great start, I think, and really bodes well for the future of the Empire modding scene.

Good luck!


Special Units Guide

Hi guys,

For those of you that have got your hands on either a pre-order exclusive unit or the Empire: Total War Special Forces edition, here is a guide on how the requirements for getting them in-game.

Pre-Order 1

Deaths Head Hussars: Unique unit for Prussia, recruited from Imperial Palace in Brandenburg region.

Pre-Order 2

USS Constitution: Unique unit for America, recruited from Drydock in North America.

Pre-Order 3

Dahomey Amazons: Recruited from Governor’s Residence building in Morocco, Algiers and Tunis regions.

Special Edition Units

Rogers Rangers: Unique unit for Great Britain, recruited from Army Barracks in North America.

Ghoorkas: Recruited from Governor’s Residence building in Kashmir region.

‘Corso Terrestre’ Guerillas: Recruited by Spain from Magistrate building in Spain region.

HMS Victory: Unique unit for Great Britain, recruited from Steam Drydock in England region.

Organ Gun: Recruitable by Ottoman Empire from Cannon Foundry.

To make things clearer, we have also produced two reference tables for you to download:

Special Units 1

Special Units 2

Many thanks,

Mark O’Connell

The Creative Assembly


Empire Total War – Graphics Work Shop

Tuning your Graphics Settings

ETW is built on a brand new rendering engine, internally code-named “Warscape”. The Engine is DirectX9-based, supporting both Shader Model 2 and Shader Model 3 graphics hardware. Supporting Shader Model 2 hardware has proven to be very challenging, requiring scalable content, SM2-optimized alternate shader paths, and a host of careful optimizations to squeeze out the most from SM2 cards. The result is an Engine that scales down to hardware ~4 years old – no mean feat for a game of this complexity.

The focus of this article is on tuning your graphics settings, so I won’t go in-depth on Engine features, unless doing so helps communicate the impact of each setting.

So let’s talk about the Settings UI. Keen players will have already noted that the In-game graphics settings UI (i.e. the UI shown when in a battle or the campaign) is a subset of the Front-end graphics UI, with some options missing from the in-game UI.

Here’s a handy table showing you where these wee beasties live.

Graphics Settings Availability


ResolutionFront End and In-Game

Windowed Front End and In-Game

Vertical Sync.Front End and In-Game

GammaFront End and In-Game

BrightnessFront End and In-Game

Shader ModelFront End

Texture QualityFront End and In-Game

Texture FilteringFront End and In-Game

Anti-AliasingFront End and In-Game

HDRFront End and In-Game

ShadowsFront End and In-Game

Hardware ShadowsFront End and In-Game

Volumetric Effects Front End and In-Game

Depth Of FieldFront End and In-Game

SSAOFront End and In-Game

Distortion EffectsFront End and In-Game

Unit Size Front End

Unit DetailFront End

TreesFront End

GrassFront End

WaterFront End and In-Game

SkyFront End and In-Game

Building DetailFront End

Ship DetailFront End

Particle EffectsFront End

The reason why some options are missing from the in-game graphics settings UI is due to internal engine limitations – in some cases we don’t retain enough internal data to switch settings on-the-fly in-game e.g. for buildings on low quality settings we discard the highest level-of-detail for buildings at scene load, making it difficult to reload efficiently with high Building Detail setting.

So let’s look at each setting and review how it impacts the game.

Resolution (List)

As you all probably know already, the more pixels you draw per frame, the more GPU-power you need, so as a general rule, pick the lowest screen resolution that you can.

The minimum resolution we support is a humble 800×600, clearly this resolution is of no interest to most gamers, but if you’ve got a SM2 card with barely enough texture memory to meet the min spec (256MB) then you may find that playing in 800×600 rez gives you a much smoother frame rate. For most folks though this low rez is of little interest.

Windowed resolutions are pre-picked, full screen resolutions are determined by your graphics hardware. Out of the Box we’ll run in full screen and default to your desktop resolution. We support all common widescreen formats.

Top Tip: Most monitors have a ‘native resolution’, a resolution at which the screen pixels map 1-to-1 with the elements of the display. At this resolution, running in full screen, your game will look crisp; at other resolutions the display hardware will need to scale pixels to map onto the screen elements, which may impact the crispness of the resulting image – though typically you will only notice this on UI text. So have a dig around for details of your monitor and see if you can see any difference at the monitor’s native resolution.

Windowed (Check Box)

I strongly recommend that you run the game in full screen always, you’ll generally get better performance versus windowed mode, and your graphics card will have exclusive access to the available video memory, not such an issue on Vista, but generally recommended and very significant on XP.

If you do run in Windowed mode, then kill off any other DirectX apps in case they’re stealing precious video memory resources.

Vertical Synchronisation (Check Box)

When you are full-screen, enabling this option limits the frame rate to the monitors refresh rate, avoiding tearing that you may see with this option disabled. My recommendation is to run with this off, as the tearing is usually only occasional and minor, and you should get higher frame rates with this disabled.

Gamma/Brightness (Sliders)

These are standard controls that allow you to balance the game’s colour response to better match your monitor or personal preferences.

Shader Model (Drop-down, Shader Model 2, Shader Model 3 (Low), Shader Model 3 (High))

In short, a card that supports Shader Model 3 is capable of rendering many instances of the same geometry very efficiently, compared to Shader Model 2.

If you have Shader Model 3 support then don’t use Shader Model 2; Shader Model 2 is slower, less efficient, and visually less pleasing. If the two Shader Model 3 options are disabled, you have a relatively old graphics card that doesn’t support SM3 – if you can find spare cash then a wise investment would be an upgrade to a SM3 card – of which there are many options to choose from. I won’t make any recommendations as it’s a matter of budget and personal preference.

We’ve provided two flavours of SM3 for your enjoyment. The Shader Model 3 (Low) setting runs the high performance Shader Model 3 instancing path, but does so with a set of light weight Shader Model 2 shaders, so you benefit from the enhanced performance of instancing, but with less visually complex detail across-the-board, so  you win on fame rate – sometimes by as much as 8-10fps, depending on your rig. The Shader Model 3 (High) setting runs with instancing and our high quality shader path, which delivers the most realistic in-game graphics, but uses the most GPU horsepower.

Texture Quality (Drop-down:  Low,Medium,High,Ultra)

This is a key setting both in terms of visual fidelity but also for your texture footprint.

By setting this to anything but Ultra you are telling the engine to throw away detail from the texture maps. A Texture is comprised of a set of 2D images, from the highest detail e.g. 256×256 pixels in size down to the lowest detail – 1×1. Each level of detail is known as a “MIP”, and each MIP Level is a power-of-2 smaller in each dimension e.g. a  256×256 texture has 9 MIP Levels: [256×256,128×128,64×64,32×32,16×16,8×8,4×4,2×2,1×1]. The highest resolutions are seen close up, the lowest resolutions far in the distance. Your Texture Filtering setting determines how the hardware interpolates between MIPS.

So, with this in mind, each Texture Quality level corresponds to throwing away one or more MIP Levels. At Ultra you see the best textures, nothing is thrown away. High settings causes the highest Level to be discarded – the [256×256] level in the example above. Medium discards 2 levels e.g. [256×256 ,128×128] Low discards 3 levels.

Discarding Textures in this manner can free up considerable video resources, but comes at a cost of visual richness – as the lower MIPS can look blocky and very SNES-like.

Texture Filtering (Drop-down Bilinear, Trilinear, Anisotropic 2x, Anisotropic 4x, Anisotropic 8x, Anisotropic 16x)

This setting impacts how your graphics card interpolates a texture across the surface of an object in 3D space. In short, the higher the setting, the better the quality of the resulting interpolation. Of course, like chocolate, there’s a price to pay for over-indulgence, as the higher settings can hit your frame rate.

I recommend that you stick with Trilinear unless you can personally see a qualitative difference in the game visuals when you turn on Anisotropic. The positive visual impact of Anisotropic Filtering is most visible on the Terrain, so the way to figure this out is to run a land battle and look at the terrain in Trilinear vs Anisotropic – with Aniso. on you should see visually cleaner terrain textures, if you don’t see a difference, then leave this option on Trilinear, which modern hardware can handle with little performance hit.

Anti-Aliasing  (Drop-down, values None,2x,4x,8x,16x)

Anti-aliasing is the first setting that I recommend you should tweak to claw back performance.

Anti-aliasing has a big impact on game visuals as enabling it causes the hardware to smooth out the edges of images on screen, so otherwise jagged lines become softer and less noticeable.

The 2x, 4x notation, for simplicity, can be thought of as the additional size of the texture required to enable the effect. e.g. 2x means a texture twice the size, 4x four times etc.

Depending on your other settings turning on AA can have a major impact on your video memory footprint. At 4x setting any full screen render targets are 4x the size in each dimension, which equates to roughly 16x the video memory required, per anti-aliased surface.

HDR – High Dynamic Range Lighting (CheckBox)

This is a big button to push – by which I mean that turning on HDR has significant impact on the performance of the game, especially if you’ve also chosen to enable Anti-Aliasing.

If your hardware supports HDR, and not all hardware does, then you can benefit from the enhanced lighting that enabling this setting provides. HDR delivers ‘bling’ by simulating the eyes response to bright lights – these areas of over-brightness, e.g. the sun glinting from a raised sword, cause a sudden flash of bright light that bleeds into the surrounding area – much the same way as bright light seen through a window appears to bleed around a window frame. The effect is cool, but subtle.

HDR is very spendy on video memory, especially if you’ve also enabled Anti-Aliasing, so spend some time to get the optimum setting for your PC.

Shadows (Drop-down: Off,Low,Medium,High,Ultra)

Rendering shadows of all objects in the scene requires us to render them from the light’s perspective into a texture, which is then sampled when computing occlusion for any object in the scene.

The Quality setting controls two factors, the size of the shadow-map texture (which on the highest setting is 2048×2048, on lowest is 512×512), and it also impacts the range over which the shadows fade out in the scene: 1000 scalemeters on Ultra, 100 meters on Low ( the playable area is 2Kx2K meters).

Shadow artefacts are very visible on the lowest setting – mitigated somewhat by turning on Hardware Shadows.

Hardware Shadows (CheckBox)

Enabling this option causes the engine to use hardware-accelerated texture sampling to improve the visual quality of shadows – effectively softening the edges, and reducing visible blocky artefacts. This works on most recent ATI & NVIDIA hardware, and should be enabled if it’s not too costly on your frame rate.

Volumetric Effects (CheckBox)

A catchy name for a family of graphical effects in the game, which all require a separate render pass for all screen objects to compute and record their depth in the scene. Yep, that means rendering all the scene objects twice per frame, once to record depth, then again to render the lit scene (and a third time to compute shadows!). Clearly this has a significant impact on performance – though the depth pass is heavily optimized and comparatively light weight.

Why do we do this? Well, once we have the depth information we can do cool things. Think back to most games you’ve played where these cool explosions are spoilt by the hard line the particles make with the ground. With depth information at-hand we can alpha out the edges of particles when they intersect with surfaces – leading to a very soft edge with no VPL (Visible Particle Lines).

Building on this effect are Depth Of Field (DOF) and Screen Space Ambient Occlusion (SSAO), both of which require a readable depth buffer to work their magic.

Depth of Field (CheckBox)

This effect simulates the focal range of a camera, objects inside the focal distance are sharp, objects outside are blurred.

You’ll see this in action in the Campaign and also in Battles, where distant objects are out-of-focus. I personally find this very useful, as it helps me focus on the action, and particularly on the campaign map leads to a nice table-top-gaming feel. This is not to everyone’s taste, so take it for a spin.

SSAO (ambient shadow) (CheckBox)

SSAO stands for Screen Space Ambient Occlusion. This is very much a high end feature, it’s the most computationally intensive of the in-game effects, but the results add significant realism to the scene.

Simply put, enabling this option – which is enabled if Volumetric Effects are enabled, and your card is beefy enough – causes the engine to compute local shadowing of objects in the scene.

Imagine you are looking into a white cardboard box in daylight. The inside corners of the box will be slightly darker than the sides, because the sides effectively shadow the corners as light bounces around on its merry photonic journey around the box.

So that’s what SSAO simulates, and the results are instantly noticeable in-game, where subtle shadows appear under objects grounding them to the scene, and shadows appear in the folds of soldier uniforms and in the nooks and crannies of buildings.

In short it’s a big switch that you should throw if you can. The results are a much richer scene, with noticeably more realistic lighting.

Distortion Effects(heat haze) (Checkbox)

This full screen visual effect simulates the effects of drinking 6 pints of Sussex – also creating a hazy distortion affect around heat sources.

It’s relatively cheap to implement too, so you’ll notice the effect on explosions in game, especially the concussion impact of grenadiers doing their stuff – where a shockwave of distortion ripples out from the impact. It’s all very satisfying, and quite cheap to enable.

Unit Size (Drop-down: Small,Medium,Large,Ultra)

This is one of the key settings that you should play with to tweak your performance. Unit Scale is more than just a graphical effect, as it affects the campaign game as well.

Simply put, the lower you put this setting, the fewer soldiers are used to represent your units. As rendering hundreds or thousands of units is one of the trademarks of Total War, and our primary performance bottleneck (closely followed by trees & grass) you should spend some time playing with this setting; and I apologize for this being available only in the Front End (technical limitations), which makes playing with it very time-consuming.

Unit Detail (Drop-down: Low,Medium,High,Ultra)

The close Banjo-playing cousin of Unit Scale, the Unit Detail setting has a huge impact on the game’s performance and visual look.

Units in ETW are hugely improved over Med2, each unit can have variants of each body part modelled- torso, legs, arms, head, hat, cross-belts, hair, cuffs, face hair, and hands. It’s a huge number of variations to render and a challenge to efficient instancing.

The higher your Unit Detail setting the more variation you will see in your units, from coarse level detail such as two chaps with different coloured tunics, to fine detail such as different beard styles close-up.

There are literally hundreds of unique animations for each unit, and Unit Detail setting impacts how we cull out these variations with distance – as units recede into the distance we start sharing their animations, so we have less unique animations to render.

Trees (Drop-down: Low,Medium,High,Ultra)

This setting controls the distance at which those lovely 3D trees turn into billboards. Rendering thousands of 3D trees is very costly, and so arriving at a good setting for Tree quality can have a big impact on you frame rate (and enjoyment).

The lower this setting, the closer those pesky distant billboards become. The sharp-eyed observer will spot that each tree has a number of discrete levels of detail (3 to be precise) that it transitions through on the way to billboards, which helps smooth the transition.

This setting also covers shrubs, which have 3 levels-of-detail, but never render as billboards – you’ll notice that they “fizzle out” as they recede into the distance, disappearing long before the trees billboard.

Lowering the setting also causes as to cull far billboard trees from the outfield (the area outside the playable area) which helps keep frame rates up.

Grass (Drop-down: Off,Low,Medium,High,Ultra)

Grass adds texture and richness to the scene, and the higher you can set this setting, the further away from the camera we draw grass – it’s that simple. Grass clumps are rendered as camera-facing billboards, and pick up the colour of the terrain they sit on.

A note on over-draw as relates to Grass and Trees

When you’re low down to the ground, looking through the grass, or looking through a clump of trees, your frame rate will suffer due to “over draw”, where many screen objects are visually overlapped and the same screen pixel is written to multiple times.

This is going to hurt your frame rate. We do what we can to alleviate this with sorting and culling, but this is only going to go so far. If your rig can handle Volumetric Effects, then enable it because doing so accelerates the process of culling out objects that would otherwise render on top of each-other – I won’t bore you with the details of how/why.

Water (Drop-down: Low,Medium,High,Ultra)

This setting affects the campaign map sea surface rendering as well as rendering of the sea in sea battles, and the river surfaces in land battles.

The 3 features that this setting controls are reflection, refraction and foam effects (sea battles only).

Setting                  Reflection           Refraction                  Foam

Low                         Off                         Off                         Off

Medium                   Off                         On                          Off

High                        On                          On                          On

Ultra                       On                          On                          On

Note that you only see foam effects on sea battles at high wind settings, so to experiment with the impact of enabling foam, start a custom sea battle with gale-force winds.

Sky (Drop-down:Low ,Medium,High,Ultra)

At Low setting the sky is rendered at scene load to create a classic sky cube rendered at 512×512 resolution. With this setting, clouds are pre-rendered into the sky cube and the sky has a rather grainy old-school look.

At any higher setting than Low the sky is rendered using a high resolution pre-computed sky cube overlaid with high res clouds, the resolution of the sky cube and clouds goes up with quality level, to a max of 1024×1024 per face at the highest resolution.

Building Detail (Drop-down: Low,Medium,High,Ultra)

All that geometry in those Star Fort battles and Town battles can put quite a strain on your graphics card. By dialling down your Building Quality you control the maximum level of detail that we load for each building, and you also control the distance at which buildings lose detail.

Like tress, buildings are created with a number of detail levels (typically 3), the lowest detail level being basically a box with some gross detail. Dialling down this setting causes us to throw away detail levels, and decreases the distance to the lowest detail level – making that boxy-building more noticeable.

This setting is perhaps misnamed, it should read “all non-animated objects quality” – but you can see why we decided to simplify the name. The setting effects farm props such as carts and hay bales, as well as buildings.

Ship Detail (Drop-down: Low,Medium,High,Ultra)

This option is currently not wired in, as of the first release, we will be wiring this in as part of a future patch.

Particle Effects (Drop-down: Low,Medium,High,Ultra)

This setting controls our Particle System, which I mentioned earlier in the context of Volumetric Effects. When we talk about Effects, we’re talking about the smoke, dust and fire effects that add so much atmosphere to a battle.

On low quality settings we limit the total number of particles emitted, and we reduce the emission rate of particles. Put more simply you see less dense smoke. On the lowest setting you will probably notice particles disappearing when they hit the budget, otherwise the culling with quality is not that noticeable. Rendering particles introduces loads of over-draw (which I mentioned earlier in the context of grass & trees). Loads of overlapping particles means lot of overdraw, so play with this setting to fine tune performance.

Appendix: a note on Presets – Automatic, Low, Medium, High, and Ultra

The five preset configurations are selected to give a simple coarse-level performance tuning.

The Automatic Setting sets your graphics to a very conservative configuration, which are the settings selected the first time you run Empire.

The higher presets will only be available if your hardware has sufficient video memory.


I hope that you all found the above walk-through useful, and it helps guide your tune-up session.  There’s no single right way to approach performance tuning, as different rigs can perform quite differently. I leave it to the enthusiasts on the forums to formulate and share strategies that may work for others.


Richard Gardner

Graphics Lead, Empire Total War. The Creative Assembly


Superior Tactics #1 – Breaking A Square

If my watch is right then the Empire demo will be live on Steam in a few minutes, but I have something else to show you for now. I know – two TW blogs in a day, crazy times!

Every budding tactician – or anyone who has seen an episode of Sharpe for that matter, knows that you form square against cavalry as troops in a line might as well be queuing up to be slaughtered.  In this first Superior Tactics video the people behind the game will show you how the units in Empire react to attack – and how you get past their defences.  Remember, only one side can claim victory and expand their Empire…

YouTube Preview Image


Kingdoms Battle Map Balance

As outlined in the recent blog by Jason, there has been some significant re-balancing of unit stats for Kingdoms, the expansion pack for Medieval II.  I am Jack Lusted, now a Games Tester at The Creative Assembly UK, but when I helped Jason by contributing to the balancing of Kingdoms I was a modder from the Total War Community.

Although the last blog described the process involved in balancing Kingdoms, it did not say what has changed. In this blog I will aim to explain how the balance has altered from Medieval II, give examples of how specific units have changed, what most unit types should be used for, and how the balance varies between the four Kingdoms campaigns.

The re-balance is not a small one, there have been some big changes. Almost all, if not all of the units have had changes to their stats. Some minor, some major. This is to reflect the fact that the whole balance of the game has been re-examined.

The really big change is to cavalry. Cavalry unit sizes are now smaller, and their stats have been dropped. They are now 30 soldiers on normal unit sizes, which means on huge they are 60. But before you all cry ‘nerf!’, let me explain what this means in-game. The smaller unit sizes makes cavalry more maneuverable as the size of cavalry units in update 1.2 proved to be a little unwieldy at times. Now you can use them more fluidly. And despite having their stats dropped, an increase to the mass of the mounts they ride means that heavy cavalry still pack a devastating punch from the rear or flank. This means that heavy cavalry have moved away from frontal charging wrecking balls, towards how they were used in the original Medieval: Total War – fast moving flanking units who pack a devastating punch. Even with a frontal charge, they can still wreak major damage upon non-spear, pike, or halberd troops.

There have been several global stat changes. All armour values have been reduced by 2, and all shield values have been raised by 2. These changes are designed to increase the importance of shields in combat when charged, but make units more vulnerable from the rear. Most units without shields have also received a boost to their defense skill so they have not been weakened in melee. The exception being missile units but they now have greater accuracy and therefore more deadly projectiles, so it balances out.

An example of one of the units that has had major changes to its stats are the Zweihanders. In update 1.2 this unit had the following stats:

Attack: 14
Charge bonus: 6
Attack attributes: none
Armour: 7
Defense Skill: 4
Cost: 680

In Kingdoms their stats have been changed to:

Attack: 15

Charge bonus: 9
Attack attributes: armour piercing (only counts half target armour when attacking)
Armour: 5
Defense Skill: 8
Cost: 520 (390 in Americas where all unit costs are lower. More on that later)

Quite a big change as you can see. Now they are true shock troops who have a very powerful charge. With the changes in unit stats, there has been a redefining of how some unit types are used. I’m going to go through and try and cover every unit type and how they should be used in Kingdoms.

Heavy cavalry – no longer sweep all before them in frontal charges. Spearmen, pikes and halberds can all stop frontal charges from them, and heavy infantry are better at resisting them. However heavy cavalry still deliver a punch that can’t be beat to the rear and flanks of other units, which combined with the fact they’re cavalry, means they can get into positions which allow them to deal the hammer blow. Rear and flanking charges also come with morale penalties to the enemy so they’re great at routing parts of the enemy line.

This has proved to be a more fun balance, and one that offers better for gameplay. Do not think that heavy cavalry are now underpowered, they are not. For instance in the Crusades campaign, knights are crucial to t he Kingdom of Jerusalem. No other unit can match the maneuverability, impact of charge or morale effect from a charge that heavy cavalry has. They’re not nerfed, they just have a different use now.

Light cavalry – fast and maneuverable – they should be used for dealing with skirmishers, routers, horse archers and possibly rear charges into infantry if the infantry is engaged.

Horse archers – with the improved missile accuracy and smaller more manageable cavalry unit sizes, these guys are as deadly as they should be. Vulnerable to archers and faster light cavalry.

Elephants – no major changes here, still the wonderfully fun point and click weapons of destruction they’ve always been. Just like before flaming arrows, artillery, javelins etc. are the counters to them.

2 handed swordsmen – all 2 handed sword units have gained the armour piercing ability and similar changes to the Zweihanders. They are now perhaps the best shock infantry in the game, but are vulnerable to missles and cavalry, and will suffer heavy casualties in prolonged melee. If used in conjuction with sword and shield infantry to exploit the damage done by the 2 handers charge, they should be able to breach most battle lines.

2 handed axe / polearm units – these guys have been made tougher in melee, and have had slight tweaks to their attack stats. They can now survive better in melee and deal out lots of damage on the charge. Think of them as infantry versions of heavy cavalry. Vulnerable to missiles and cavalry charges.

Spearmen – their main use in Kingdoms should be as the most common anti-cavalry unit type, but with the boost to their attack, they can also take on other infantry a bit better. But as always suffer from the penalties they get from having the spear trait so will be outclassed by other infantry.

Pikemen – the specialist anti-cavalry unit. With much higher mass in Kingdoms no cavalry charge can beat them frontally, and they can also deal with infantry slightly better too. Very weak when flanked and not as good as spears against other types of infantry.

Halberds – they have received boost to their attack values and to mass, so they are better against both cavalry and infantry. Good assault troops, but slow moving and vulnerable to missiles.

Halberds without spearwall – from instance Janissary Heavy Infantry. Have had boost to their attack and defense stats and reductions to cost. Great shock troops but can also do better now in prolonged melee.

Sword and shield infantry – no big changes here, these are still the best prolonged melee infantry unit, and probably the best all round unit type. There is now more variation between units like dismounted Feudal Knights and Dismounted Chivalric Knights.

Missile infantry – have been weakened in terms of their melee abilities slightly, but this is compensated by their increased missile accuracy will become more important due to the higher number of casualties they can inflict with their missiles.

Whilst the overall balance for each of the Kingdoms campaigns is the same, there are differences between each campaign for game-play reasons.

In the Teutonic campaign, all cavalry units are stronger with higher secondary attacks. The Teutonic Order units are also stronger than equivalent unit of other nations, but this is balanced out by the fact that the elite units need to have a cetain percentage of catholicism in a region before they can be recruited. The Order is reliant on those troops to expand and further it’s goals so this balances things and prevents the Order from becoming too powerful, too quickly. The Orders units also cost more because of their higher stats so things are also balanced out this way.

For the Crusades, like the Teutonic campaign all cavalry are stronger with higher secondary attacks. But unlike the Teutonic campaign, the Crusader factions do not have superior troops compared to their Muslim enemies. Even so the Crusader nations will be fairly reliant on their strong cavalry to win the campaign.

With Britannia it is spears that are the unit type that receive a boost. This results in a proliferation of good anti-cavalry units, so infantry will dominate the Britannia campaign. But cavalry are not completely negated, they will still be usable units, just not as powerful as in the Crusades or Teutonic campaigns.

And finally in the Americas campaign, New Spain gets smaller units, but sword armed infantry and cavalry with 2 hit points, and a new generals unit with 3. This is to reflect the small numbers of Spanish troops used in the New World, and the extra hit points prevent the smaller units from being overwhelmed. Unit costs are also adjusted to reflect these changes, so overall most units are cheaper but Spanish units are about the same as in M2TW. The Native units will not be pushovers either, and will put up a strong fight.

That just about wraps up my overview of the balance changes made through the unit stats and hopefulle, gives you an idea as to how you’ll be adapting your tactics to use these changes in each of the four campaigns in Kingdoms.


Jack Lusted


Froggbeastegg’s Guide to the Guides

I find that, of all the questions I am asked, there are certain ones which stand out as more common. An invitation to write a blog entry seems a good opportunity to give answers to those who wondered but never asked, and to explain a few things here and there.

The main focus of these questions is usually “How did you end up writing these guides?” A new patron on the org had been asking questions for a week or two, and I, among others, kept on answering them. He said that if all my posts were gathered up they would make a good newbie’s guide to Medieval: Total War. It sounded easy enough, so I gathered them up and stuck them together in Word. I wasn’t happy with the result – there were so many more things to say about the game! So I said them. When I posted the first draft of my beginner’s guide I didn’t honestly expect much, but the response was incredible. Amid all the praise there were corrections, suggestions, requests, and I did my best to meet them all. It kept on going and going, and the guide grew and grew. Finally there was nothing much left to do with it, and people started asking me what I would do next. Next!? I’d no idea, I’d never considered doing anything else. Which, in a roundabout way, provides the other half of the answer to this question: I wrote because people kept asking and I didn’t want to let them down.

Queries about the writing process, about how much goes into each guide, and about where I get my information are one of the staples of my inbox. Writing the initial version of the guide took me perhaps a week, usually spent writing from 7am through to 11pm or later with barely any breaks. Exhausting. Prior to setting virtual pen to virtual paper there is the necessary research. For my Rome: Total War guide I put in an estimated 200 hours of research spread over nearly two months: reading forums, taking part in discussions, testing things over and over in the singleplayer game, dragging my poor multiplayer partner into games to test yet more theories… Let me tell you, research is tedious and nothing like playing the game for fun. On posting the guide the treadmill of  adding updates begins. Sifting through these many corrections and additions of course adds many, many more hours to the work load.

Because I put these works together in MS Word editing or altering the 200ish pages of text is a nightmare; I don’t have any kind of overview, or a hard copy to consult with, or anything really but my memory of what is where. To make matters worse the manuscript version is full of the coding needed to produce text effects and pictures: it is unpleasant on the eye, and heavily coded areas can be confusing. The RTW guide got so big I had to chop it up into five different documents, containing the five different posts on the forum. Those posts are so big the org software nearly had a nervous breakdown each time I edited them. The final aspect of each guide comes with the messages I get from readers. I try to reply to each and every one; alas time and circumstances do not always make this possible. At this point I’d like to say a big, big thank you to the many people who have sent me kind wishes over the years.

Overall I estimate I’ve spent some 900 hours working on the three guides.

Frogbeastegg consists of one person. Frogbeastegg’s guides are the work of one person only, save for a few technical aspects, such as PDF conversions, which are clearly accredited to someone else. Amusingly I once saw some people labouring under the illusion I was some kind of writing team…

Speaking of misconceptions, I’m not a man. I’m a lady frog. It’s an easy mistake to make, especially when I appear somewhere where I can’t use the custom-made geisha avatar I have on the org. There aren’™t many sites where I would contradict the natural assumption I’m another in the sea of males. This community, however, isn’t populated by people who make crude, lewd remarks and then say I must be a man anyway, because girls don’t play games other than Barbies Happy Horse Fun (now featuring 50% more pink!)

One assumption I encounter a lot is that I have links with CA, or have been employed by them to write my guides, which simply isn’t true. All of the information stated as coming from CA representatives in my work is gathered up from posts made in response to other people on the forums at the org and the .com, or, in the case of MTW, taken from the official strategy guide. So no, I really cannot give you any inside information and/or ask them questions for you and/or do you any favours. Really.What do I gain from writing these guides? There’ve been some truly inventive ways to profit from my work suggested to me; while I’ve taken none of them up I shall remain perpetually curious about how well printed and bound hardcopies equipped with a froggy autograph would have done. That bit of genius makes opening a paypal account so people can donate to my book buying fund seem positively dull.

My gains are, to be honest, likely to sound dull. No money, no writing contracts, my mantelpiece sports no freebie copies of TW games autographed by the CA team members – fun as that might possibly be. I’ve received online awards for my work, and all those messages saying thanks. It means a lot to me. I have learned how to write – that is my major gain. I’m a dyslexic froggy you see. I found that out several months before I started the first guide. Back then I struggled to write anything, no matter how basic. My spelling was so bad Word couldn’t guess at the right words to offer me when spell checking. Grammar was a concept my education had not so much brushed over as swept under the carpet and then pulled the house down on. I was so frustrated I decided that I would do something or die in the attempt. At long last I can say what I want, not what the spellcheckers will allow me to. I don’t need a spellchecker any more. I didn’t get it all from working on the guides, but they helped. Most of my writing these days is fiction: I have my childhood dream of being an author back.

The Beginner’s Guide to Medieval: Total WarThe Complete Total War Unit Guide
(outdated; it doesn’t cover RTW/BI)

Frogbeastegg’s Guide to Rome: Total War and the Barbarian Invasion