Guilty Gear -Strive- Closed Beta was amazing, except for its matchmaking

Guilty Gear -Strive- has just wrapped up it’s first limited closed beta, and I can honestly say this is one of the best looking, best sounding, best playing fighting games I have ever played in my life. Now if only it’s janky lobby system ever let you play it.


Let’s start where it counts, gameplay. Guilty Gear -Strive- is certainly a “modern” guilty gear, in that it’s following a lot of modern-day fighting game design sensibilities. That is to say, it’s vastly simplified itself, at least compared to other Guilty Gear titles. Combos are shorter, damage is higher, there are fewer moves that cancel into each other, and most of the wonky mechanics like “Danger Time” from Xrd have been completely removed.

That’s not to say there isn’t a nice dash of the good old Guilty Gear complexity for some spice. The tension gauge and roman cancels operate the same way as before. Lots of moves have specific special properties, like Ky’s “shock state.” The RISC gauge makes a return, punishing you for being too defensive by making all hits count as counter hits, and there’s tons of finicky rules on what normals and specials do and do not cancel from each other.

Speaking of counter-hits, the game is practically designed around them now. Landing one with a light move will cause a small hit-stop effect and increase your damage. Landing one with heavy attack or special move creates a HUGE hit-stop effect and vastly increases your stun and damage. Potemkin landing a counter-hit hammerfall into any normal takes off about 40 percent of a character’s health.

This makes for some really hype matches though. Whenever you see that big “COUNTER” flash across the screen, you not only have time to figure out what to do next, whatever it is that you do will be absolutely insane. It’s almost like there is a built-in hype-man right in the U.I.

Combo structure is a bit worrying. The reduced cancel options make neutral game feel a little bit stiff. Once you land a hit, especially in the corner, things open up, but that’s just because everything seems to link into everything and there seems to be no voluntary air-tech.

The thing is, links don’t quite click in low to intermediate players’ minds as well as cancels do. Everyone can kind of understand that you hit buttons in the order of lightest to heaviest. Linking involves frame data and hit-boxes and special hit-states, and while people with a little bit of experience can jump in and start futzing around and developing their own combos, newbies will be left mashing buttons for a real long time and losing because of it. I’m not entirely sure I want to see something like an easy mode or auto-combo integrated into the game but opening up the basic combo system would probably make the game a little more inviting to players of lower skill levels.

But other than that, matches are fast and full of hype, just how I want my fighting games to be. You constantly feel like you are on a knife’s edge, looking for a way to turn the game around. I cannot tell you how satisfying it is to land a move, see “COUNTER” flash across the screen, pop a roman cancel, and then just go to town on the opponent, taking off more than half their life in one combo that you just sort of pulled out of nowhere on the spot. The feel of battles, from the ebb and flow of momentum to the weight of every single strike, is just perfect. It certainly doesn’t feel like any Guilty Gear in the past. It’s more like Guilty Gear and Samurai Shodown were smashed together, but hey, that’s a combination I’m willing to play.

P.S. Potemkin’s Potemkin buster removes over half of your life, and as a Potemkin player, I find that very satisfying.


When I said that this was one of the best looking and best sounding games I have ever played, I wasn’t lying. If you thought that Arc System Works’ other fighting games looked like living anime, think again. The models are so incredibly detailed. The animations are so remarkably fluid. The intros look like… well like something out of an anime, with a dynamic camera highlighting each character’s snarky quips and shining personality. In fact, the camera works overtime to highlight each counter-hit, each super, each throw, and much more.

Characters are just bigger this time around. They take up more of the screen space with more of their detail, and frankly that’s a really good choice. Mobility in general has been reduced in -Strive- with air-dashes having a bit of lag at the front of them, super jumps not traveling as high, dashes starting up and traveling a bit slower, and so on. While your opinion on what that does to gameplay might vary, graphically this has allowed Arcsys to focus on each animation just a little bit more, cramming in more detail to moves that we already recognize from previous iterations. Heck, most moves simply have more frames of animation this time around, which isn’t to say that they are slower. Rather, if you did a 15 frame move in Xrd, it might have seven-ish, frames of unique animation. Now it feels like all 15 are unique and detailed.

Stages are a treat to look at. I’m not 100 percent up on my Guilty Gear lore, so I couldn’t tell you where the characters were fighting, but regardless “sky garden looking place,” “creepy forest,” “steampunk magitech gears” and so on are all great stage concepts that come to life in the background. Like many other fighting games, Guilty Gear -Strive- has integrated stage transitions into its gameplay, which causes you to go to another part of the stage whenever you have been smacked against the corner too often (getting the attacker a meter boost in the process). What this means is that each stage is actually three or more stages, and every time you go to a stage you appear to start at a different location, which constantly gives you variety that you just wouldn’t otherwise see in a traditional fighting game.

And then there’s the music. You can’t see me doing a chef’s kiss through text but trust me I’m doing it as hard as I can. Guilty Gear has always been known for its fantastic music and -Strive- is no different. I’ve been playing the main theme over and over on loop, and sure enough it’s one of the great tracks that you can fight to. But no matter what track you hear you are going to be hit with some sick guitar, belted-out broken English, heavy metal screams, and much more. Now of course, you might not be into weedily guitars and loud thumping drums, and if you’re not, I feel for you, but you have to admit that this style of music is just perfect to fight to. I mean, what style of music would you put to a fast-paced fighting game? Light elevator jazz? I mean who would do that?


Now I have to take a break from all my true and genuine glowing praise to tell you that Guilty Gear -Strive- has one of the dumbest, most broken, most frustrating online systems in fighting game history.

Let’s get the good out of the way first. There’s actually a decent idea here. Players are sorted into towers and each tower has floors that correspond to your skill level. You can always go to higher floors to push yourself but you can never go to lower floors. This ensures that people of similar skill level always play each other, with people who are just learning protected from getting their face curb stomped by veteran pros. At the same time, it allows people to still find matches should the servers one day become barren. It’s a really great way to keep people finding matches.

But everything else, EVERYTHING ELSE, is so bad, so stupidly complicated, and so full of bad decisions that try to reinvent the wheel I barely know where to begin.

Lobbies are all two dimensional worlds and you control a small psuedo-pixel art character that you have a small amount of customization control over in these 2D worlds. To challenge someone to a match you hold a button to draw your weapon and walk toward another player who also has their weapon drawn. The game is then supposed to recognize that you want to challenge this payer and immediately take you into matchmaking.

There are so many problems with this system it's mind-blowing. First of all, the servers just did not stand up to the load. During the most popular times I would find myself waiting 40 minutes for a match, in a room full of other people who were also waiting 40 minutes for matches. There’s nothing sadder than seeing tons of avatars, weapons drawn, begging for a fight, and the servers won’t let ANY of them play.

Second, it’s a 2D lobby, so characters will frequently overlap each other. That means that you can’t really tell who you are challenging when players are all bunched up in a group.

Third, it’s easy to forget you have your weapon drawn, which means just about anyone you touch will be challenged to a fight when all you are really trying to do is get to a different area of the lobby to, say, change your controls or edit your avatar.

Fourth, it’s unclear if walking away from someone during matchmaking cancels matchmaking, but the servers update so slowly you can’t tell when you are attempting to match with someone until you get the message at the bottom of the screen “attempting to connect to the opponent.” At that point you have to freeze in place, like some perverse game of red-light green-light just to hope that your connection goes through.

Fifth, connections so rarely work out. Even in times of light server load I would have to wade through ten or so failed connections just to get one that works. I was paranoid that it was my connection that was the problem, so I tried it on TWO SEPERATE NETWORKS, one wired, one wi-fi, and then tested both networks on another Arc System Works game (Granblue Fantasy Versus) and found that, no, my connection works just fine, it’s this lobby system that is just borked in every possible conceivable way.

Sixth, there is no penalty, whatsoever, for rage quitting or griefing, and it was rampant in the beta, so you can imagine that it’s only going to get worse in the full version.

Seventh, even if people are not griefing you, the way the lobby is set up almost ENSURES that people will accidentally grief you! Since it takes so long to find a match, you might walk away from your computer with your weapon drawn to do something, maybe take a bathroom break. If ANYONE TOUCHES YOU they will be brought into a matchmaking prompt that will hang forever because you just aren’t around! You weren’t trying to cause problems but the nature of this wonky lobby system caused them for you.

And the tragic thing about this? The netcode itself wasn’t that bad! Arcsys games have always used wonky delay based netcode in a world that has moved on to rollback netcode, and this beta test was still going to use delay based netcode even though the final product was going to use rollback. The thing is, even with delay based netcode the matches ran really well. I could barely notice the delay, except in the off circumstance the game hung up here or there. It’s not the netcode that is the problem. It’s the matchmaking and lobby system, which turned one of the greatest fighting game experiences I have ever had in recent days into one of the most frustrating online experiences in my life.

Fix it Arcsys. Just fix it. Stop trying to be fancy. Just give us something that works. You know what works? A menu. A menu where I can say “I want to play against this guy.” That works. All the pixel-art and 2D lobbies and emotes don’t mean anything if your matchmaking doesn’t work!

The verdict

My Guilty Gear -Strive- beta experience was incredibly bizarre. I’ve never had such an incredible gameplay experience that was just so often denied to me by the game’s online system, even in other betas.

Now, the good news is, Guilty Gear -Strive- isn’t planned to come out until late this year and none of the problems I had with it were core to the gameplay system. A half a year is more than enough time to fix up a janky lobby and matchmaking system, and I think Arcsys will do it. If they do, this is going to be one of the biggest fighting game releases of this year, or really of this generation. So I’ll say keep your eye on this one, because great things can come of it.

But I also think now, more than ever, we need to stop making excuses for fighting games that are great in every other aspect but their online capabilities. It’s 2020, every other game genre has figured out online matchmaking, and now more than ever we need the capability to play with each other over the internet instead of next to each other on a couch.

This doesn’t go out to just Arc System Works, this goes out to every fighting game developer out there. You have to stop assuming you can just make a good game without a good online infrastructure to lift it up, and expect the fanbase to just deal with it and schedule their own meetups to get around bad matchmaking or netcode. Making your fighting game work online is not just a necessity now, it’s a responsibility, and you shouldn’t take that responsibility lightly.

Am I being a bit harsh, yes, but I’m just very tired of talking about Arcsys games as “an amazing game but the online leaves something to be desired.” I hope that you got the data you needed from this beta test to make the next beta, and the eventual full release, work smoothly online. If so, I could see giving this a 10/10. If not, then it probably isn’t even worth a second look on the shelf.