I have been looking for a successor to Steel Panthers for a while. Many rumors of a Steel Panthers remake just fizzled out over the decades. I owned the Matrix version and the free version from Shrapnel Games. I thought their games were brilliant, but the UI was horrible. At least the Shrapnel version had arrow keys you could use, which the older version did not. But the worst offender to me was that it was hard to make out the shadows where your units can move in any version I played. It was a game I badly wanted to get into. All the vast scenarios, the military historical picture images with unit stats when you pressed the unit info button. The huge mega campaign on the Matrix version was intriguing too. 

A few days ago, I woke up and saw someone on Facebook mentioning Steel Company. I played the demo’s tutorials and was so intrigued I immediately had to contact the developer. 

Enter the interview with Nick Emery from GYR Software.

Chris: To start off, tell me about your background in gaming.

Nick: Sure! I’ve played PC games ever since I was little, of pretty much every kind. Definitely a lot of RPGs and strategy games. Mount and Blade is probably the game I have the most hours in (somewhere around 10,000 hours at this point across all the different mods), but also the Civilization series, Command & Conquer, Elder Scrolls, a lot of Paradox games, etc., and of course Steel Panthers, and a bit of Combat Mission. I’ve always loved games with lots of mods and lots of depth and customization, but also ones that had a really tight core loop, like Mount and Blade.

Chris: Great! So how many years do you estimate you’ve been playing Steel Panthers?

Nick: I was first introduced to Steel Panthers by my dad back in probably like 2010, and he had been playing it since it first came out (around the time I was born). It took me a while to come back to it, though. It happened tangentially, as I was learning more history about WWII and digging through old US Army manuals, which can be interesting. I guess the most recent time I came back to Steel Panthers was from playing Foxhole and getting deep into combat engineering and defenses, etc. This led me back towards wanting a game that depicted combat in a much more grounded way. 

I remember he had a disk of SPWAW in one of those big binders with CDs.

Chris: You sure sound like you’re one of the younger Steel Panther fans.

Nick: I am 30, so I guess it depends on the crowd, and for Steel Panthers fans that’s almost certainly true. 

Chris: What motivates you to finally make a Steel Panther follow-up? Has the wargaming scene been begging for this?

Nick: Well, I wouldn’t claim to be THE follow-up, as there have certainly been some interesting games in the same vein released over the years, but I’m not aware of anything that truly tries and succeeds to achieve the same scale and packages it up in a nice UI that runs well on modern systems.

I suspect part of the reason is that it’s an awful lot of work, and wargamers are a discerning audience. So, it requires getting a LOT of things right, and then at the end of the day you sell it for as much as any other game (or less).

I think it will probably take the “modders/fans/community first” approach to accomplish, where a handful of crazy people on the internet just decide to make it happen and pour in the necessary work, and then hopefully modders come in to make it even better and expand it more. 

Chris: How many games have you made before this, and how did that help you prepare for this task? Did you do any modding as well?

Nick: Nothing approaching the size of this game, and nothing that I’ve sold or distributed widely. I had made a few small games and mods for myself and my family/friends over the years. I have modded Mount and Blade, Morrowind, Oblivion, and a few others.

It helps a lot being a professional software engineer, though… So, I have a lot of experience thinking about technical problems and learning new tools, trying to design the data model right, adjusting to user feedback, etc.

I work in biotech, and that’s my main background, which is a very finicky field where there are a lot of complicated details that are easy to get wrong, so I think some of those skills transfer over.

Chris: So far, from what I played, the AI seems impressive. Is it scripted well or is there actual AI?

Nick: It’s not like a machine learning model or something, if that’s what you mean. It’s mostly a bunch of “soft” rules (most of which are moddable, by the way!) about what to do in different kinds of situations. I try to make the AI “see” the game kind of like the player sees the game. So, for example, there’s a “threat map” of where the AI thinks your troops might be, even if they’re not spotted, and I use that for all kinds of things. So, it will say “well, here’s my main axis of advance, and there’s that threatening clump of trees and trenches over there, I think I’ll bomb or smoke that as I advance.” The main thing is just lots and lots of playtesting. And I have the ability to run headless AI vs AI runs, so any time I tweak it I can see whether it’s getting better or worse.

Chris: Chris: Has AI implementation been easier last few years compared to, like, the early 2010s, or is it the same type of system you think? For me, a good sign of a good AI is if it attempts to outflank you and doesn’t use illogical use of units like making your Battalion HQ charge the front line.

Nick: It’s basically the same kind of system. Maybe there’s some way to get an LLM or something to make it smarter or more “humanlike”, but I don’t really think so honestly for this kind of game, at least not at any cost/speed that makes sense. I also want it to be moddable/tweakable, which would be hard if it was all one big machine learning model/LLM.

And yeah! There are a couple different kinds of flanking maneuvers the AI will try to pull on you in Steel Company today, like sending a couple light vehicles to infiltrate your back lines or doing a full single/double envelopment. It still does dumb things sometimes, but I also don’t want it to do the “perfectly optimal” thing every time because then it’s too predictable (not to mention unrealistic).

Chris: Did I read you plan on making a campaign? If so, is it story-driven in any way or just text briefings? By story-driven, I mean something like Decisive Command Barbarossa.

Nick: The full game will have both “scripted” campaigns (e.g., a historical command for the 101st Airborne or whatever, with various decisions along the way) as well as custom/randomly generated campaigns. Both types will have units carrying over between missions and gaining experience, etc.

I also have some crazy ideas for a whole operational map layer… I’ve got an entire map of Syria in 2013 for a little “Syrian Civil War” mod I’ve been making myself, with hundreds of little neighborhoods and roads and real terrain/elevation data… but I probably shouldn’t talk too much about that because I’m not sure if it will make it into the final game. It’s hard to do well.

Chris: That sounds awesome! I hope that follows through. Where do you base your OOBs and terrain data?

Nick: They’re from various different sources stitched together. The OOB data isn’t pulled from any one game/database, because, for one thing, the engine has its own system, so not everything is directly comparable. The OOB definitely still needs several more passes for accuracy. Already people have found and corrected several little things. I also like to do my own reading of primary sources for things like the effectiveness of different weapons, etc., which is part of the fun.

Chris: Do you make the units behave differently or have different stats based on the type of military unit? Like one country may have higher morale or superior doctrine with infantry tactics?

Nick: They definitely behave differently depending on their stats. I try to make the AI smart enough to “use whatever it has” so that you can drop a random modded unit into the game with an unexpected combination of stats and the AI will use it intelligently.

As for differences between countries, a little bit, mostly in experience/morale levels and force composition. I haven’t given them different “personalities” beyond that, although the engine certainly allows for it, and I probably will at some point. 

Chris: If you move to a more modern period after this, would you try to make the battle play differently than WW2? For example, the presence of Helicopters in Vietnam or Drones in Ukraine? I’m just thinking readers will love this game and expect the follow-up past WW2 like what Steel Panthers did. After this game proves successful, of course.

Nick: I don’t feel like I know enough about modern-era combat to really do it myself (at least not right now, maybe in a few years with a lot of reading). However, I very much hope to make the engine moddable enough to support it. There’s a “flying” flag on units, for example, which can support on-map hovering units, even though there are no helicopters in the game today. And there’s a full-on scripting API that can do all kinds of crazy stuff, so I hope someone takes it on eventually.

Chris: One last question for tonight so you can go to sleep. What got you into WWII and wargaming? Most young people these days seem less interested. 

Nick: Probably mostly my dad, who has a whole library of WWII books and games like SP, TOAW, HOI, etc., has been the biggest influence.

Chris: Well, I’ll let you sleep now. Thank you so much for the amount of dedication you’re putting into this game; it sounds like something wargamers are looking for!

Well, there you have it! An interview with the mind and the hands behind one of the most promising wargames! Go to Steam and wishlist it!

Support Strategy and Wargaming

I do what I do in Strategy and Wargaming because I love to do this, and I’m never going to stop. If you would like to support me with that, you can buy me a coffee for a dollar if you’re feeling generous. If you can’t, no worries, Strategy and Wargaming will always be free, and I’d love to have you around!