Periodically I like to re-read The Killer Angels.
In the time before Netflix, I also would frequently re-watch Gettysburg, the film based on Michael Shaara’s book, which because I happen to have it on DVD, was a choice when I didn’t have any rentals available.
The film Gettysburg was originally developed, over 25 years ago in 1991, as a mini-series for ABC. That network pulled backing when another Civil War-themed miniseries was a commercial flop. As the developer, Ted Turner then targeted the mini-series to his own, newly minted cable network, TNT. However, when he saw the near-finished project, he decided it was too big for television, and had it released to theaters.
Due to the running time, the theater release was limited and the box office take was modest (although it did turn a profit). The main problem was the running time of the film, clocking in at 254 minutes (growing to 271 minutes for the Director’s Cut), which required an intermission. If my recollection holds up (and who knows these days), it was then re-targeted back to TNT and the originally-proposed mini-series format. Cable channels, at the time, were rather a niche venue and again the “success” was modest.
Once released on DVD, the film finally came into its own. In the 24 years since theatrical release, it has become an all-time best seller. It’s popularity is for both its entertainment and its historical value, as it is now used as a classroom lesson augmentation. Still striking, to me, to this day is the use of the actual battlefield in the film. This was the first time that the U.S. National Park Service allowed filming at the Park. The battle scenes were reconstructed with reenactors and has been praised, except for the obviously well-fed soldiers, for their authenticity.
Similarly, the source book also took some time to come into its own. It won a Pulitzer prize in 1975, the year after its release, but it was not a commercial success. Appreciation for the book has built over the years, probably in no small measure due to the movie. For myself, I watched the movie first on TNT and only years later read the book when it was left at the take-a-book/leave-a-book library at work.
While re-reading the book, there are many passages that were reproduced religiously in the film. It is impossible to read and not see Tom Berenger, rather than James Longstreet, reciting some of the lines. In contrast, my minds eye cannot abide with Marty Sheen; I see only Robert E. Lee himself as I read. Naturally, the book has much more depth and is ultimately more satisfying than just another play through of the movie.
Also, to add to the experience, I decided to break out some Gettysburg games that I have.
While we are reminiscing, such a discussion would not be complete without considering Sid Meier’s Gettysburg!, the 1997 portrayal of the battle. While hardly the first computer-game treatment, it offered a unique take on the battle and the design of this kind of game.
At the time, I had a couple of games addressing this battle, including Battleground Gettysburg (itself with reenactors) and Civil War Generals (2, I think). I didn’t pick up the Sid Meier games until years after release. It was in some sort of bargain offering that I got after I had played Austerlitz: Napoleon’s Greatest Victory, a later game based on the same engine. When I finally experienced these games I thought, this is what computer games should be about!
In the traditional hex-and-counter treatment, numbers and die rolls have to be used for the cohesion of units. As troops move quickly, or change facing, or perhaps move from column to line, one needs a numerical way to deal with the negative effects of that action (and how it diminishes over time). In Sid Meier’s Gettysburg!, by contrast, you order a unit to move and deploy in the line, and then have to wait, watching, while the orders are executed. It made so much sense.
Unfortunately, even by the time I got the game, it was already looking a little long in the tooth.
Move Out Men, and Spritely!
Step into the breech a series initially known as Take Command. In 2004, partnering with the History Channel, a game called Take Command: Bull Run was released. It was a continuous-time strategy game covering, naturally, the First Battle of Bull Run with regimental-sized units. It was to be everything that SM: Gettysburg was and then some; more realism, more control. Plus, an Artificial Intelligence that could not only challenge the player, but also play cooperative with the player as a subordinate, a commander, or a fellow officer fighting on the players flanks. The game was followed up in 2006 with a 2nd game, this time covering the 2nd Bull Run.
From the get-go, the development was fraught with minor controversy. I recall extensive arguing about the use of 3D graphics versus sprites, the latter chosen by the developers but being seen by some as out-of-date for 2004. Another brouhaha erupted when the game’s forums devolved into an argument about copy protection, and the moderator threatened to ban anyone who argued against the technology. In the end, as the developer was working on a follow-up based on the Battle of Shiloh, the company fell apart amidst internal disputes.
Half of the team, however, rose again in the form of NorbsoftDev and a series called Scourge of War: Gettysburg. Given the previous episode, I hesitated to jump in with the new series but eventually was lured by a discount to pick up a game and an expansion.
When I first picked it up, I started in with the tutorials. Even though I had played both Take Command iterations, I wanted to re-familiarize myself with the game system. Somehow, I didn’t get very far before I moved on to other games. Nevertheless, I recall being very impressed with the earlier series. I remember how it rewarded the use of realistic tactics – one of the first scenarios was only winnable by redeploying a regiment, unobserved, by using a depression. I also remember delegating an attack to an AI Stonewall Jackson and being blown away by his execution of it. AI Jackson split his forces, using part of his command to fix the union troops in front of him, then hit the flank with the remainder, to devastating effect.
So here’s where I started when I wanted to relive that first day at Gettysburg.
Good ground: Col. Gamble has deployed his brigade.
Having been quite a while, I went back to the tutorials. Now I remember why I moved on from this game when I played it before. The tutorials are not quite what I’d expect. Yes, they introduce concepts slowly building to the whole scope of the game, but something seems off. Usually, I expect a tutorial to be pretty easy (if not foolproof) to get right, particularly if you follow directions to demonstrate the concept being covered. Here, I often have trouble accomplishing my goals. In the movement tutorial, I have to do a lot of extra fidgeting to triggered my “goal” and in the second one I just quit rather than wait for everything to trigger. In the battle scenarios, I got stuck fighting in the wrong place against a superior force, and no way (seemingly) to set things right (although I remember it going better when I played some years ago). In another, I was supposed to fight alongside an artillery battery, but I couldn’t find them.
Even taking screenshots has its glitches. It’s mercifully not as clear at the reduced resolution above, but all the screenshots have these 1 pixel blue dotted lines across them. This from an in-game screenshot system. What gives?
Finally, I gave up on the tutorials and went straight into the first scenario, Buford’s initial encounter with the Army of Northern Virginia. I wound up restarting the scenario 3 times (probably necessary because I hadn’t made it through all the tutorials). In my first game, as I tried to deploy the cavalry into dismounted defensive positions, they would mount back up and charge. On the second try, I wound up detaching all my units and commanding them directly, which made for a huge mess. I finally realized I was missing setting on the commanders which tells them whether to attack or defend – by default they are set to attack.
Although, by everything I’ve read, the system is improved over the previous iteration, it all seems harder to get right than I remember. Trying to get a brigade deployed in a defensive position at the crest of the hill seemed way too frustrating. I no sooner get everyone in the right place pointing the right direction when one unit starts wheeling around 90 degrees because an enemy is coming up. Then I realize another unit actually doesn’t have line-of-site over those crops. Add to that, an inordinate amount of time seems spent trying to figure out which way to the enemy! Switching between units always starts you facing the same direction as the unit you’ve navigated to. So as everyone gets turned this way and that, once the battle is started, you wind up in your unit view staring at an empty field. Is it because you’ve moved away from the action, or is the enemy closing in and just happens to be right behind you? The fact that the map shows camera position but not direction doesn’t help.
Additionally, I struggle with the command system to give higher level commands. While one can set positions, formations, and facing directly, there are also commands to, for example, tell units to advance or retreat. The user also has choices about whether to give commands to at the officer level, or to issue them to units (or subordinate officers) at a lower level. The problem is, I see inconsistent results. For example, if I wish to reposition an artillery battery forward, with direct sight of the current battleground, I can simply tell the commander where to go. However, I often come back to find that the unit hasn’t moved, or has moved very little. By contrast, if I go to each cannon and order it directly to limber and then, once read, give it an exact location to move to, I’ve seen far better results. I’ve also found artillery stacked up at fords, trying to figure out to cross. Most frustrating is the uncertainty; I’m never sure what is going to work.
I’ll come back to the Scourge of War experience, but before I do, I’ll take a look the other game.
Ultimate General: Gettysburg is the creation of Darthmod, the modder of the Total War series that had been doing major overhauls of those products for years. In 2014, the standalone product was released through Steam, and in many ways is a spiritual successor to the Sid Meier game.
The game is scaled up a level, as compared to either the Sid Meier version or the Scourge of War games. The lowest-level unit of maneuver is the brigade, rather than individual regiments. Aside from that, it brings back that old feeling. The art style is bright and colorful, and obviously meant to be more representational rather than realistic. The commands are both intuitive and aesthetic.
Pettigrew’s Brigade is moved in to support Davis’ left flank at around 10:20 AM on the first day. Giving commands is as simple as drawing in the arrow.
Another novel UI element. Selecting multiple units can be done by drawing a circle around them.
Graphics and interface wise, this really is a thing of beauty.
Purely from the wargaming perspective, however, it is a step in the other direction. The basis for this game was the Real Time Strategy genre, albeit the historical RTS world of Total War. The developer said his goal was to de-emphasize the dominant rock-paper-scissors mechanic of most RTSs and give more weight to factors like elevation, morale, and fatigue. Thus one expects an improvement on Total War. Compare that to the design philosophy of Scourge of War, however, where they are trying to factor in the effect of each musket, and it’s obvious this is more of a wargame-lite. That comes through in the gameplay. The pace is very rapid, and the strategy feels much more superficial.
I Could Feel Them Breaking
So getting away from the controls, and back to the overall feeling of the wargame, we find Scourge of War starts to show its quality.
My line of cavalry has completely broken and nothing stands between Rebel infantry and my artillery positions.
When I focus on experience of the game itself, something different emerges. In my third playthrough as Buford, most of the details were left to AI Gamble. Chaos ensued and eventually all my defensive lines were broken. In the screenshot above, we see the Confederates advancing on my artillery positions, with nothing to stop an overrun. AI Gamble has positioned the 8th Illinois across the road, but they are so exhausted that they won’t survive contact with the enemy. I was resigned to losing again.
But then an amazing thing happened. The Rebels did not advance.
Very much mirroring the actual battle, with each attack the Southern commanders assumed they had an easy victory. With each repulse, they pulled back, and reformed and reinforced to deal with the situation they encountered. Having been dealt heavy losses, and seeing my 8th Illinois waiting for them, they take the time to prepare to make a proper attack. Just like Gen. Heth, AI Heth does not know what he is facing, and he does not appreciate the weakness in my position.
Just as the enemy has reformed and is prepared to sweep me from the ridge, help arrives in the form of the infantry of the 1st Division of the 1st Corps. In the nick of time.
The enemy only took another 5 minutes to prepare a coordinated attack, but it happened to be enough. As they formed their attack, friendly infantry was arriving to fill in my lines.
The result was a major victory, mirroring Buford’s own success on the field. Like the real situation, the result wasn’t pretty. Buford’s victory was not that he successfully defeated a superior enemy, but that he delayed the enemy just long enough for Reynold’s Corp to come up.
For a more direct comparison between the two games, I played the next hour of the battle from the Confederate side. With a slightly larger force, I had to rely on the AI command structure all the more. The battle is chaotic and often frustrating, but a good amount of that frustration reflects the reality of this battle.
For a quick explanation, the battle puts the player in the situation where he is advancing along the road in the face of retreating infantry/cavalry. I gave my commanders orders to attack along a broad front, but the terrain makes this difficult to follow. The road itself is narrow, and the surrounding ground is broken by woods, fences, creek-beds, and a large ditch created for a railroad that hasn’t been completed. It makes it very difficult to shift units from one side of the battle to the other.
Roughly midway through the game, I realize I’m in a sticky situation. On my right, Archer’s Brigade has advanced well ahead of my left flank, and is now engaging superior numbers in some open farmland. On my left, the defense has all but collapsed, but I have one regiment still, barely, holding a line in the face of two full brigades. I know that I need to get one of those brigades across the battlefield to reinforce my right, but to do that quickly enough, I need to use the road in front of me. I have enough force to knock aside that regiment and then blast through reinforcements, but how can I actually convince my AI subordinates to do it that way? I am forced to sit back and watch as the brigade slowly deploys, engages, and finally clears the road.
At long last, I’ve managed to reinforce Archer on the right with infantry and artillery.
After all of that, with my reinforcements moving forward, I find Archer’s Brigade in a fine mess, with Archer himself having been captured (I wonder if that was scripted in?). At this point, my right begins to do quite well for me, but on the left some Union reinforcements find me spread out and under-strength, and my problem of half-an-hour before has reversed itself.
In a more traditional wargame (and, in Ultimate General), it would be easy to see where reinforcements are needed, and plot a path to get them there. I suspect the frustration and confusion of this game comes a lot closer to the reality of Gettysburg where the lack of knowledge about what was going on around was a significant factor in the outcome of the battle.
Scourge of War, at its most difficult settings, has modes where the view is limited to that actually seen by the player’s commander and orders are sent, not by the UI, but only by courier. While I’m pretty sure I’m not ready to handle it, my initial experience says this is the way the game is meant to be played. The question, then, as with any of these games with competent friendly AI, does it get to a point where the game essentially plays itself and you’re left (much like a Lee or a Meade at the time) simply observing as the battle plays out without you?
By many objective measures, the Gettysburg Campaign, if not the battle itself, is seen as the turning point of the war. Pickett’s Charge, and its reaching of the Union lines at the Stone Wall, is described as the “high water mark” of the Confederacy. This makes the battle ideal for considering the “what-ifs.” This explains the popularity of the movie, and the number of games focused on the battle. The Killer Angels exposed many amateur historians to the Longstreet versus Lee controversy, which I think is a major factor in its popularity.
To this day, I am not aware of any game, however, that explores the real question posed by the book (and the movie). What if Lee had taken Longstreet’s advice and disengaged at Gettysburg to seek a defensive position elsewhere? Would the South have won that battle? Would that have ended the war?
And why are we so fascinated with asking questions like these to which we can never know an answer?
A quick update follows in the next post, with considerably more detail added later.