A Path beyond 2020: www.redalert.one

Yoyo folks. Since most of you guys and girls know: Our old website ra1-guides.com had some struggles in the last weeks and went offline. ...

6/11/2019

On Power and Production


by heraclitus


“Whoever produces the most the fastest wins”

Introduction

Most RA players who have read about strategy will know this quote from Milo234, and I think it guides the way the game is being played to a great extent. However, is it always perfectly clear what to produce? Like Andrew Ford calculated earlier, a light tank does more damage per minute than a medium tank, and is also produced faster and cheaper. Would you make light tanks only then? No, because they have less health points. But how much less? And does this mean that just producing the most the fastest isn't enough?

Or how about mammoth tanks? They are useful when you’re being attacked by air, as they possess anti-air defense, and they self-heal up to 50%. Yet their production speed is a little underwhelming, to say the least. In long games, like for example a balanced match on the map "Liquid", some players prepare for the moment when the maximum number of vehicles for the map will be reached by making mammoths instead of heavies. Mammoths have better offensive and defensive powers than any other tank, and this helps these players win the end game. But how many war factories do you need to produce mammoths as fast as heavy tanks? Or is it even possible to make them faster?

Others laid the groundwork for this analysis. Roger Wong wrote an extensive Red Alert Strategy Guide in 1996, wherein he included a list with unit data (like fire rate, health points, attack points) that I use for this analysis. In 2001, JCFarmer (aka ORA-NGUTAN) created a website where he was the first to publish production speeds of units per number of war factories. And Andrew Ford was the first to report on the damage-per-second characteristic of units, writing about it on his blog and the cncnet forums since at least 2015.

Here I aim to make a more comprehensive assessment. How much damage-per-minute does each unit deliver? I call this the offensive power. And how many health points do they have? I call this the defensive power. And how many units of each tank can be produced per minute, per number of war factories? And how does this translate into offensive power produced per minute? And what do we see when we throw both the offensive and the defensive power into the mix and calculate the total power; what tank would deliver the most power, per number of war factories?

Then, for the builders amongst us, how much offensive power from buildings (e.g. teslas, turrets, flames) can be generated per minute, per number of construction yards? And what if we correct this number for their defensive power, to calculate the total power? I suspect a builder with 10 CYs can produce more power per minute than a tanker with 7 wars, but is this actually true? And if so, what is the size of this difference?

Methods and Materials

Firstly I wanted to make sure the data supplied by Wong concerning the unit characteristics were correct. All my test results matched these data (e.g. 15 light tank shots to take out a medium tank, 10 medium tank shots to take out a light tank, 1 tesla shot to take out a light tank, 3 V2 shots to take out a tesla). I have not tested all possibilities however, so some mistakes may have slipped through. It would bring me pleasure if someone could confirm or correct this crucial set of information.

Offensive power
Each weapon inflicts a different amount of damage points on different targets, depending on the type of armor. Most combat structures and vehicles are heavily armored (like the tesla, turret and all tanks except V2s and artillery), and only 2 units are lightly armored (V2 and artillery). All infantry is unarmored and most buildings are made of wood (except for the construction yard that has heavy armor). 

To calculate the offensive power I used the amount of damage a unit or building inflicts per shot on targets with heavy armor. This I multiplied by the shot rate per minute. The result is the amount of hit points that the units/buildings could give per minute; its offensive power.

The hit points for heavy armor (instead of light armor or no armor) were chosen because almost all battle targets that matter possess heavy armor. Choosing this type of armor will closely reflect the game reality.

Defensive power
The unit data sheet from Roger Wong and the cnc.fandom wikipedia page were used to discover the health points for each unit and building. I did not include the type of armor in the analysis for defensive power, as paradoxically light armor in many cases is beneficial. Medium and heavy tanks actually do 35% less damage to lightly armored targets than to heavily armored targets. The main problem of lightly armored vehicles is that they have few health points (V2: 150HP, artillery: 75HP); their light armor is in practice protective, except for fire from enemy infantry or pillboxes.

Total power
The attacking power is multiplied by the defensive power, and the result is uniformly scaled to a more relatable number. Numbers for units and buildings are comparable.

Production speed
I measured the production speeds myself by logging in for an online multiplayer game, and then playing on a random map on my own (speed 7, no AI). Afterwards I compared my results with JCFarmer’s results. JCFarmer measured production speeds not in seconds or minutes, but in units produced per time it took to produce one war factory. I calculated it took a war factory 22 seconds to be built (on 1 CY), and confirmed this in a ladder game I had a recording of. Based on this number I could extrapolate his production speeds. Our results matched in >90% of the cases, however with V2 production speeds we differed significantly. My data was used for this analysis.

It should be said that measuring speed in regular time units like seconds or minutes is problematic, as the game speed depends on several factors, including game settings, type of game (multiplayer vs skirmish), internet connection/lag, et cetera. JCFarmer’s approach to measure it in “war factory production time” makes therefore sense. I did however choose to use a regular time unit here as it makes it more relatable for the reader. All speeds are measured in the same way, so I trust we will be able to make fair comparisons between the different units/buildings, and use the tables and graphs for decision making.

Analysis
All calculations were done in Excel.

Results

Unit Characteristics

*As reported by cousinclaus, may be inconsistent (different guard range in different situation) #Fire rate multiplied by damage per shot, specified per type of target armor

Building Characteristics

*Fire rate multiplied by damage per shot, specified per type of target armor



CLICK IMAGE TO ENLARGE
Offensive strength (attacking points per minute) divided by cost, in the right image corrected for defensive strength (both factors weighed equally)

CLICK IMAGE TO ENLARGE
*Y-axis represents power measured in hit points per minute for targets with heavy armor, X-axis represents number of war factories / construction yards

CLICK IMAGE TO ENLARGE

*Y-axis represents power measured in hit points per minute for targets with heavy armor, X-axis represents number of war factories / construction yards

CLICK IMAGE TO ENLARGE

*Offensive and defensive power weighed equally for this calculation. X-axis represents number of war factories / construction yards

CLICK IMAGE TO ENLARGE
*Offensive and defensive power weighed equally for this calculation. X-axis represents number of war factories / construction yards

Discussion

Interpretation
Unit Characteristics
It is interesting to see how much damage those small artillery vehicles can do. I guess it’s a good thing they have only 75 health points, or else all we would see in this game would be artillery. Other noteworthy things are that the light and medium tank inflict about the same amount of damage per minute (like Ford reported), and that the mammoth tank only inflicts like 11% more damage than the heavy tank per minute. The mammoth nonetheless has 50% more health points.

Value for Money
It fits with the gaming reality that the medium and heavy tanks and tesla are the best "total package" deals in the game. Hence their popularity I guess.

Pure Offensive Power
Up to 6 war factories, producing artillery and light tanks will actually provide you with an army that can deliver the greatest amount of hit points to the enemy per minute. For Allied players, artillery clearly stands out until the 4th war, and light tanks clearly at the 5th war. With 6 war factories, the medium and light tank sort of share the first place. For Soviet players, heavy tanks create most offensive strength from the first up to the twelfth war factory, at which point the mammoth tank takes first place. The advantage from the mammoth over the heavy is only incremental though (56k vs 50k hit points per minute), so one could argue they share first place.

As I expected and – being a tanker – feared, the offensive power for builders reaches a much greater size. With 10 CYs, the maximum offensive power generated per minute is 60% (!!!) greater than a Soviet tanker generates with 7 wars pumping out heavy tanks. From 7 CYs on, building teslas gives more attacking strength than any tanker possibly could.

Strikingly, the power production does not grow linearly with the number of war factories / construction yards. Notably from the 4th to 5th war there is no increased production speed for the medium tank, and the same holds true for the 5th to 6th war for the heavy tank, for the 7th to 11th war for the mammoth tank, and for the 7th to 9th CY for teslas. This non-linear production speed algorithm has implications for gameplay; you probably shouldn't wait too long to build your 6th war after the 5th as an Allied force, or to build the 7th after the 6th as a Soviet, etc. Vice versa, Allies could consider building a service depot after the 4th war as their enemy is not expected to increase their tank production anytime soon, assuming they play as an Allied country as well (on the "V3 Ribbon gems" map for example).

Also noteworthy; at each number of construction yards teslas outperform pills, camo pills, turrets and flames (even glossing over tesla's superior shot range).

Total Power
When we correct the offensive power for the defensive power (weighing both values equally) we see that the artillery and light tanks are outperformed by heavy tanks for the first 3 war factories, by medium tanks at 4 and 6 war factories, and by the mammoth tank at 5 war factories. The heavy tank remains the number one strongest tank to produce in terms of total power from the 7th until the 11th war factory. With 12 war factories the mammoth tank has a clear advantage, being 60% more powerful than heavy tanks. This is due to its production speed doubling from the 11th to the 12th war factory, and its having 50% more health points.

For buildings it's no surprise that tesla retains its number one position, as only the camouflaged pillbox has more health points than the tesla. It’s interesting that the mammoth outperforms tesla here, which can be explained by the mammoth’s higher number of health points (600 vs 400). However, I don’t take this finding serious due to the mammoth’s slow moving speed and tesla’s greater shot range, which will undo the mammoth advantage in game reality. In my view the tesla remains vastly superior.

Limitations
I did the speed tests myself and only once, which will have caused some inaccuracy.

Obviously, economics frequently limit production, which has not been taken into account by this analysis. Applicability is therefore for infinite ore and gems maps, where you want the strongest army the fastest no matter the costs.

Another important limitation is the lack of shot accuracy data. For example a turret is notoriously inaccurate on moving targets, thus inflicting less damage, while a tesla is 100% accurate. Likewise splash damage, unit speed and shot range are disregarded.

About the lack of armor in the assessment: like mentioned earlier, >90% of targets in most games will be heavily armored (light/medium/heavy tanks, mammoths, teslas and turrets), and most offensive units and buildings actually inflict more damage on heavily armored targets than on lightly armored targets.

Naval and aerial units are left aside as well. See here for a list with damage-per-minute information for these forces.

The importance of this subject will be minor compared to things like build order, Qing, clicking speed and accuracy, map knowledge, strategy et cetera.

Strengths
It is shown that certain units (e.g. heavy tanks, teslas) are more cost effective than others (e.g. artillery, pillbox).

Useful unit information is provided which can help in specific situations. For example knowing that a flame has the most powerful attack against wooden structures can give you the edge in certain situations (like in the map "KOTG"). Also knowing that the V2 has a greater shot range than the tesla (10 vs 8.5) can make an important difference in some circumstances.

Graphs are provided for decision making. For example Allies could choose to make lighties with 5 wars, and mediums with 6 wars. Soviets could make heavies until they have 12 wars, when they can switch to mammoths. 

The importance of reaching certain thresholds for production speeds is elicited; like 7 wars for heavy tanks, 12 wars for mammoths, 10 CYs for teslas. Quite frequently that last war factory or construction yard causes a major bump in production speed.

This analysis clearly shows how the early game is a tanker's game, and how from 5-7 CYs on it becomes a builder's game. 

Conclusion

Each combat unit has a unique profile of strengths and weaknesses which, combined with a non-linear production speed algorithm, challenges the player. Strategy needs to be continuously adapted to the opponent's actions, by building and producing wisely. The maximum power output for a Soviet builder is notably greater than that of a tanker, which explains the successful tesla strategies in infinite ore or gems maps. Both skill and knowledge are required for competitive play, and the above supplied tables and graphs can help a player improve his or her game.


Sources

Tables with data that were the basis for the graphs https://imgur.com/a/JBF8HzA
Milo234 Red Alert Guide https://funkyfr3sh.cnc-comm.com/milo/ra.html
Cousinclaus’ report of the Guard Range http://www.redalert.one/2019/05/gq.html
JCFarmer's website with production speeds http://members.tripod.com/red_alert_website/
The Original RA manual with some info on unit armor (although supposedly some is incorrect) https://web.archive.org/web/20160314122728/http://redalert1.com/files/ramanual.pdf
Command and Conquer Fandom Wikipedia with Unit information https://cnc.fandom.com/wiki/Portal:Red_Alert

1 comment:

  1. Thanks for the analysis ;p

    Props to whoever wrote this essay!

    ReplyDelete