Clash of Steel - A PC Wargame published by SSI Simulations, 1993

Comment on the document 'Clash of Steel - A Closer Look at WWII Grand
Strategy Simulation' by Jay Purvis, May 2000, available at:
http://www.grognard.com/info1/steel.pdf

Introduction:

I completely endorse Jay Purvis' paper regarding Clash of Steel. COS
really is, in terms of game design rather than graphics, one of the best
WWII grand strategy COMPUTER games ever produced. Please read the above
document for yourself to understand why we think so. Then download the
original game for yourself, completely free, at:
http://www.classic-trash.com/wargames1/clashofsteel.shtml

(The name of the above website is a little unfortunate, as are the not
entirely positive comments made by the webmaster, Thomas Jensen,
concerning COS. Don't be put off - download the game and form your own
opinion. However, full marks to this site and to Mr Jensen for ensuring
that many old games like this one are not lost for ever in the mists of
time.)

Like Mr Purvis, I would really like to see Clash of Steel completely
updated for Windows 98 and re-released. His suggestions for a new
version are excellent. Here are some more:

More suggested enhancements for COS II:

A) Transfer of strength points between ground units of same type and
nationality, and air units of the same nationality.

During WWII, divisions were sometimes transferred from one corps to
another in order to respond to local circumstances. This could be
incorporated into COS as follows:

1. A ground unit may transfer strength points to an adjacent ground
unit. Air units do NOT have to be adjacent to transfer strength points
but must be within 10 hexes of each other.

2. Both the supplying unit AND the receiving unit lose 1 Action Point
(AP) to carry out the transfer. 

3. Transfers may only take place between units of the same type
(Panzer/Tank, Infantry, Air), and both units must be from the same
country. For example, the German Rundstedt Army Group unit may transfer
strength points to the German 4 Armee infantry unit or the German 30
Korps infantry unit, but not to the Guderian panzer unit, nor to the
Fsjgr paratroop unit, nor to the Pioniere engineer unit. German units
may not transfer strength points to or receive them from Italian,
Rumanian or Finnish units, and vice versa. The same applies to British,
French and US units.

4. Each unit may, by unit transfers, be reduced to up to half (rounded
down) of its original maximum (full) strength point value, or increased
to up to 1.5 (rounded up) its original maximum strength point value. For
example, the Guderian Panzer unit could be reduced to 2 strength points,
or reinforced to 6 strength points in value. Obviously, if a unit has
lost strength points in combat, this is taken into account. For example,
if the Hoth unit has lost 2 strength points in combat, it cannot then
transfer 1 of the remaining points to Guderian.

5. Any number of transfers may be carried out by each unit during a
single impulse, provided it has enough strength points and APs to do so.
A unit with no APs may neither supply nor receive strength points. Any
number of strength points may be transferred during a single transfer
action. However, as stated above, the transfer must conform to the
limits of both the supplying and receiving units, as stated in rule 4,
above. 

6. Transfers are absolutely permanent until the player decides to
transfer the strength points back.

7. When an increased or decreased strength point unit is refitted, the
refit conforms to the most recent strength point value of the unit. For
example, a 6 point strength value Guderian will be refitted to 6 points
in strength, not its original 4 point value.

8. If an increased or decreased strength point unit is dissolved or
eliminated, it must be rebuilt at the normal cost (i.e. the PP cost of
the unit is not increased or decreased as a result of the strength point
transfer.) When the unit reappears in reserve, it will have the same
number of (full) strength points as it had when it was dissolved or
eliminated. This takes into account the fact that the reduced unit
stills needs the adminstrative train and command structure of a full
sized unit of its type. For example, Hoth transfers 2 strength points of
its original 4 to Guderian, and is later dissolved. When rebuilt, Hoth
will still have only 2 strength points.

This would add a whole new dimension to play. For example, during the
invasion of France, the leading German infantry unit (e.g. 2 Armee)
could receive 3 strength points from say, 4 Armee, making 2 Armee very
powerful and able to dissolve defending Allied units on its own.
Conversely, the French could tranfer several strength points to the 1st
Army and move that unit into Paris, making it much tougher for the
Germans to capture Paris. Also, the danger exists that weakened units
could be dissolved, and rebuilding them at their normal PP cost is
expensive. During the Russian campaign, tranferring strength points from
the German army groups to the smaller and faster moving Armee and Korps
units will get combat strength up to the front faster. However, the
weakened Army Group is vulnerable to a sudden Russian counterattack, and
if it is dissolved, the Germans will have to rebuild it at its normal PP
cost, not to mention the effect of resulting loss of supply for the
nearby German units. The possibilities for the defence of Moscow are
stupendous - imagine a 11 strength point Russian Army Group defending
the city, which is also a fortress! This would be really tough for the
Germans to crack. It could lead to a real Stalingrad type slugging
match.

B) Naval Warfare

1. The concept of strength points should be built into naval warfare, so
that vessels can vary in strength according to the effectiveness of
their historical armour protection. Strength points for ships should
vary between 2 for a transport vessel or sub to 6 for a tough battleship
like Bismarck. 

2. Ships will also have a firepower rating. The ships with bigger guns
will have a higher firepower rating. Note that three of the US
battleships are old WWI dreadnaughts, worse than Britain's. Carriers
will have good firepower, but only poor protection (strength points).
Firepower determines the amount of damage done when a broadside hits.

Also, naval units will have attack points - each attack point enables
the vessel to fire once per round. Ships with a higher rate of fire
(based on how many shells they can fire per minute) will have more
attack points. Carriers will have the most attack points for their
numerous aircraft. The chance to hit will be lower than in the original
game, but having more attacks per round will compensate and make naval
battles more like slugging matches, involving the exchange of numerous
broadsides (with suitable graphics and sound effects). Damage will be
calculated by firepower rating. Ships with a higher firepower rating
will have a better chance to do more damage when they hit. Also, each
hit will have a small chance, based on firepower rating, to be a
critical hit. A critical hit will destroy the target vessel outright
(equivalent to hitting a major magazine). So events like the Bismarck
destroying the Hood with 1 hit will be possible.

3. Efficiency and morale should also be built into naval vessels -
however, when calculating combat results, efficiency and morale will be
used on their own, not in combination with strength points, or firepower
rating, in order to determine the odds of scoring a hit. Higher
efficiency and higher morale vessels will have a better chance of
scoring a hit. Efficiency for ships will be based on the technical
capabilities of the ship itself (e.g. good optical rangefinders for
German ships, good radar for US ships) while morale will be based on the
historical morale of the crews (Italian ships, for example, are well
designed, with high efficiency, but the crews will have lower morale
than the British. The older British ships like Warspite and Malaya will
have low efficiency because they are only updated WWI dreadnaughts, but
their crews will have high morale.) Morale of naval vessels should be
based on similar figures to their country's ground forces. Also, vessels
without a friendly port in the seazone they are in will have lower
morale, but allowances should be made for German vessels in the
Atlantic. 

4. Naval vessels in the game should have the correct speed for the
vessel they are named for, not a generic value. For example, Hood should
have a 30 knot speed, Prince of Wales 28, and Bismarck 29. The faster
Italian battleships will be able to outrun the older British battleships
like Warspite and Malaya, allowing them a better chance to flee
successfully from combat, but the British vessels will be slightly
stronger than most of the Italian ships (except Littorio and the unbuilt
Italian battleships).

5. A new research item, ship radar, will be added. Each radar research
level above the threshold level of 3 points will increase the efficiency
(and cost) of surface vessels by 10%.

6. When attacking convoys, a ship's raid value will be calculated by
averaging its firepower rating with its attack points and also speed. So
ships with smaller guns will be just as effective against convoys (it
doesn't take much to sink a merchant ship!). Faster ships can chase down
more fleeing targets scattering in all directions.

7. When bombarding shore targets, firepower rating will be the key
factor in the damage done to the target's morale.

8. During amphibious invasions, escort vessels WILL screen the
transports! The intercepting fleet will have to fight it's way through
the escorts first, sinking all of them before getting at the transports.
(Exception: Carriers and subs will not be affected by this rule - they
will have an equal chance to hit any vessel.)

C) Strategic Bombing.

1. Strategic Bombers should be units placed on the board. The Axis
should have 2 strategic bomber units (both German), and the Allies
should have 4 (2 US, 1 British and 1 Russian). Strategic bomber units
will have strength points, efficiency and morale just like air units.
However, players will have the option to be able to buy increases in the
maximum strength points of existing strategic bomber units, similar to
the ground unit upgrade levels, rather than placing a new unit.
Researching strategic bombers increases their efficiency and cost.

Each bomber unit can be rebased around just like air units. Bombers
should be affected by range - the closer the bomber unit is to the enemy
country, the more efficient it is and the more damage it does. There
should be a realistic maximum range for each country's bomber units (the
German ones will have the shortest range (twin engined bombers), the
British one the longest). Each bomber unit can be set a different
mission (attack harbours, factories or oil, and be set to attack
different countries (Allies or Russia) in the case of the Axis.) 

2. Another option is to be able to set the bombers to attack a specific
enemy resource centre or city. Damaging the resource centre or city
enough reduces the PP's it supplies to the enemy production levels.
Repair of the resource centre or city factories will happen
automatically at the end of the turn (i.e. you have to keep pounding it
to keep it out of action, just like in the real war.)

3. Also, attacking cities could interdict supplies. Damaging the city
reduces the supply it gives to nearby units. Damaging it enough will cut
the supply to almost nothing (this will take a LOT of bomb hits to
achieve.) Recovering supply values might take longer than recovering PP
values, optionally. Perhaps a gradual recovery of so many supply points
per impulse.

To do this, each city and resource centre would need to have strength
points.

4. Specific cities and resource centres could be labeled as the sources
of naval production, factory production and oil production for each
country. For example, Kiel would be the main German production centre
for naval vessels, with Koenigsburg as a subsidary. For Italy it would
be Genoa and Trieste, for the UK Liverpool, and Russia Leningrad and
Rostov. Bombing these cities would increase the PP cost of specific
naval vessels for the country concerned. Each unit would have (invisible
to the player) an assigned city or resource centre where it will be
built. Bombing that city could have a chance of increasing the cost of
that particular unit. For example, Bismarck and Tirpitz would be built
in Kiel, while the German transports would be built in Koenigsburg. In
similar fashion, the Ruhr resource centres and other German cities would
be responsible for the German ground and air units.

5. Obviously the US units cannot be affected by bombing. Only the
Russian infantry units should be affected by bombing: the other units
would be built in Siberia, and thus cannot be affected.

6. Bombing oil targets: Certain resource centres, like the ones in
Rumania and Persia, and ones in the Caucasus, should be labeled as oil
production centres. Bombing these reduces the action point levels of the
units of country occupying the resource centre (including air units and
strategic bombers.)

Capturing an oil production centre from the enemy will have the effect
of repairing some or all the existing damage done to the action point
levels of all units (including air units and strategic bombers. Losing
an oil production centre to an enemy will result in damage to some or
all units, depending on how many oil production centres that side
occupies. A unit cannot be reduced below 2 AP by oil production bombing.

7.When intercepted by air units on Guard duty, both the air unit and the
bomber unit can lose strength points. Bomber units will be refitted just
like air units. The intercepting air unit must be within 7 hexes of the
city or resource centre being attacked - the closer it is the more
damage it will do to the attacking bomber unit. The efficiency of the
air unit will also make a difference.

D) Computer AI

The computer AI would have to be much improved over the original.
However, a couple of really good COS players (like Jay Purvis) could
help lay out in detail how the AI should act in many different
situations, and all that SSI need to do is the programming. The basic
gaming system isn't too complicated, the AI can handle that. It does
need a really good sense of attainable strategic goals, select one, and
stick to it without becoming sidetracked by player diversions, and more
tactical skill (able to do deep penetrations as well as head on clashes,
and able to direct a powerful assault team to crack open a strong
position (this was a weakness of the original AI). Not an impossible
task by any means. 

Different AI personalities could be incorporated. For example, each side
could have a 'historical AI personality', where the AI is programmed to
react like Hitler for Germany, Churchill for England, Stalin for Russia,
and Eisenhower for the US. The first three are known to be stubborn to a
greater or lesser extent, and reluctant to accept the fact that they are
losing. All are likely to carry out offensives before it is prudent to
do so. Eisenhower is more cautious and likely to hold back until he's
got a significant numerical advantage before he attacks. 

The 'generic' AI should have different skill levels (but not be
completely useless as beginner level!) and at the highest levels should
make the best possible choices for the situation it's in. Real strategic
'thinking' should be included, for example the computer should note what
units the opposing player is building, and decide what to build to
counter it. It should think in terms of long term objectives as well as
taking advantage of local opportunities. As the Axis it should decide at
the beginning of the war whether to invade England or not, and build and
move its units appropriately. After conquering France it needs to decide
whether to invade Russia in 1941, or wait till 1942 and concentrate on
the Mediterranean theatre first. It should make use of the political
function too, pressuring countries that would make good allies. It
should manage research well too, depending on which side it is playing
and what it's strategy is going to be.

E) Interface

The interface should be updated to SVGA and full Windows 98 compliance
(drop down menus, etc.) Instead of expensive cutscences, the game CD
could have, say, 100 .avi or mpeg video clips of actual WWII film
footage, inc. colour where possible. These will be dated to play at the
correct times (e.g. 1939 footage during the Polish invasion in the 1939
game turns, 1943 footage during 1943 turns. Different clips will play
during combat when a unit is forced to retreat, or is dissolved or
eliminated. The clips will play in windows so as not to obscure the game
board, and will be completely customisable to the user, so they can be
set to play at different game events or not at all.

This would be a lot cheaper than making a few cutscenes, and give just as
much atmosphere to the game (possibly more).

F) Graphics and Sound

Graphics will be secondary to gameplay, but if the publisher is prepared
to put the money in, a lot could be done. For example, in addition to
the default hexgrid map, a really beautiful 'relief' map could be made
(like the ones in the graphical fighting bits of Microprose's Risk
II(tm), or like a full colour quasi 'satellite photo' montage, but with
a hexgrid overlay). This should be movable across the map, and zoomable,
and should be made rotatable in the horizontal plane. Unit counters
could be set to opaque or transparent, or could even be switchable
between the traditional counters and 'animated action figures' - very
small soldiers, tanks or planes, again following the Risk II example.
Suggest one figure, tank or plane per strength point per unit. As
strength points are lost so a soldier dies or a little tank is
destroyed. Because they would be so small (Risk II size or smaller)
there could be lots of them on the screen at once and they wouldn't be
too expensive to design and animate. 

The graphical interface should be as user customisable as possible. It
all depends how much budget the publisher wants to put into the game. 

Instead of expensive cutscenes, the game could have, say, 100 or more
avi or mpeg video clips of actual WWII film footage, or even movie
footage, inc. colour where possible. All will have soundtracks. These
will be dated to play at the correct times (e.g. 1939 footage during the
Polish invasion in the 1939 game turns, 1943 footage during 1943 turns.
Different (random) clips will play during combat when a unit is forced
to retreat, or is dissolved or eliminated. Also there will be ones for
naval battles, air interceptions, amphibious invasions etc. Specially
selected clips will play at key events, such as when a country
surrenders, at the fall of a capital city, and at the end of the game -
these will have audio annoucements (radio annoucements) and could
include 'newspaper headlines'. The clips will play in windows so as not
to obscure the game board, or in full screen if desired, and will be
completely customisable to the user, so they can be set to play at
different game events, play without sound, or not at all. This would be
a LOT cheaper and quicker than employing a team of computer artists for
a year making several '3D' cutscenes, and give just as much atmosphere
to the game (possibly more!) All you need is permission to use the
footage, which shouldn't be a problem if acknowledgement and website
links are given.

Music - good background music is essential. Stirring military stuff
where possible, with lots of variety so people don't get fed up with it.
Imagine different music during different player impulses - Germanic
themes during the Axis phase, and the odd Italian tune during Axis turns
in the North African campaign. Jaunty British music ('We'll hang out our
washing on the Seigfried Line' and 'It's a long way to Tipparary' are
just two good examples) during the Allied phase, with Polish and French
music early in the war, and American stuff added later, and some good
Russian singing during the Russian player's turn. (I think Russian
military music is excellent.) Different music could be played whether
the war is going well or badly for the side in question - dark and moody
stuff if they're losing badly, bright and triumphant if they're holding
their own or winning.

G) Marketing

To save money for the publisher, the game could be marketed following
the Medieval 2 example. The demo could be put on lots of websites and
computer magazine cover CDs, but the full version is downloadable from
the web after the user buys it online. This would include a pdf version
of the manual. (The demo could also be the full version, which is
unlocked by a unique code the user gets once they have paid.) As an
alternative, the full version could also be purchased by post at a
slightly greater cost. This saves the publisher stacks of money by not
having to burn thousands of CDs, design and produce a glossy box and
manual, and they wouldn't have to settle for lower profit margins by
going through all the outlet stores, no transport costs, etc. The game
is targeted at a specific audience anyway, it's not a game that 'every'
PC gamer will buy, like Quake III. All that's needed is a really good
advertising campaign, a well designed website, and lots of magazine and
gaming website reviews. My estimate is that it could all be done for
$250,000, and if only 25,000 people worldwide buy the game at $30 a
shot, that's $500,000 profit for the publisher. If, say, twice the above
budget is put into the game, to make really good graphics and sound, it
could become a real rival for the forthcoming Axis and Allies II, and
sell in far larger numbers!

That's it. Please, SSI, read this and reconsider your decision, if you
can. 

Jay Purvis, if you read this, please get in touch with me.

Tim Smith
Kent, UK
email: tim.s.smith@talk21.com
or: tim.s.smith@dti.gsi.gov.uk

7 July 2000