I know, you’re already mad. You’re saying “Anders, we want better anticheat and performance and perhaps a new operation!”. And I get it. But I had a crazy thought. I figured it would be fun to put into words trying to think about where Counter-Strike as a game might be going so the question is, what if Counter-Strike had reflections?
I guess before we begin to answer the question we can look at how CS has evolved a bit over the years. To be transparent, the reason why I want to look at this is because I think it will help me guard against the first attack against my idea. The most common feedback to any changes of the game are “that’s not what CS is!” but I think when we look back, we’ll see that the game actually has changed a lot already and still somehow kept its DNA.
Back in Beta 5 point something, the colt, now m4, could zoom, the AWP didn’t have a blacked out scope and grenades did damage through walls.
When CSGO came out we added molotovs and later a pistol like the CZ75. A lot of these to the mind of a Counter-Strike purist just feel absurd. There are plenty of old tweets about how ridiculous the idea of a molotov is and how it would be impossible to play around and so on. In more modern times the volumetric smokes in CS2 are a giant change to how the game functions and could easily be seen as a departure from the traditional path of CS, but it works so well that nobody even thinks about it any longer.
read more
In fairness, there is at least one update I can think of that did actually break the DNA of the game enough that there were something like widespread riots in the community, and that was the riot shield. The riot shield which was added in 1.6 was pretty much universally banned and I think for good reason, it just didn’t feel like playing Counter-Strike any longer and some of us still have nightmares about it.
But what does any of this have to do with reflections? Well, let’s take one more detour before we get there. At its core I think one of the most important pillars of Counter-Strike is information. The most easily understood kind is visual information, as in, can you see someone? But I would argue we have a sub-category of information already that I think we could call information-leaking. The most obvious case of this is the shadow advantage you might have in certain positions, perhaps most noticeable on ancient with the flood lights:
Can I see this player? No. But the information leaks around the corner because of the lighting conditions and we all accept that as an interesting challenge to take into consideration. The other perhaps even more common case is the sound of footsteps which also leak information to opponents.
read more
What I am really trying to claim here is that controlling information is already a major part of the game and something that is so second nature to most of us that we don’t actually ever think about it in those terms. There are many other examples than the ones I have brought up here, but I think the two above serve well enough to bring the point home. So what about reflections?
In a crude sense you could argue that reflections are just another kind of shadow except it can in theory travel in directions that a shadow really can’t. Imagine a puddle like this one on the new cache:
What if you could see a blurry version of a player model in it? What if instead of a puddle it was a blurry mirror on a wall or a reflection in a window? Perhaps in this case you would even be able to shoot the glass causing the reflection, so it would become a bit more of a tactical choice to either leave it or not? This question actually raises another question which is, what would happen if you could shoot the flood lights on ancient? In some cases perhaps you would always just shoot them because it would stop your shadow from showing, but perhaps there are cases where you want to leak that information to distract someone on the other side? Similarly to footsteps where you most of the time try to keep the information leakage to a minimum, but sometimes you will intentionally fake a drop down or run away only to walk back.
read more
I can’t think of a good reason why visual information couldn’t follow the same concept. If you could shoot the flood lights or mirrors to control the information, that just becomes another aspect of the game that I reckon people will quickly take into account when playing.
There is another challenge to this concept which has more to do with performance. Obviously if we tried to do one-to-one renders of player models in full resolution, it would presumably be a challenge for some of the hardware out there. But I imagine that if the quality of the reflection is more of a low-poly blur, it can probably be done in a way more respectful to the GPUs and CPUs.
In in the end I think if we opened up this path it would allow map makers to design some very clever concepts around this idea. Think about a part of a map where you might want to show someone that you’re holding a position, perhaps its an ECO round on the CT side and you have a stack in a bombsite, so you try to show your reflection and pretend you forgot? Perhaps on the T side you shoot out a mirror pretending like you’re worried you’re going to be spotted moving forward, but in fact you fall back immediately. To my mind these subtle mindgames are already a huge part of Counter-Strike and I think adding a few more of them over the years is a great way to make the game feel fresh without changing the DNA.
read more
For a practical example, this window on Inferno reflects the exit from T spawn where the barrel is:
As you can tell I put BOT Frank on the barrel, but he doesn’t show in the reflection of the window. Is this a big deal? Not really, would it change a lot if the player model was reflected? Probably not in this exact case, but I think over time if we started designing maps to incorporate small areas where it was a factor, it could add some very interesting dynamics that lean on already existing gameplay mechanics, just stretched a bit further.
Ultimately I am really curious about what you guys think about this. On the surface I think its easy to dismiss this as being ‘not very CS’, but I think that objection falls apart if you constrain the concept through proper map design. I’d love to hear your thoughts and comments down below!
