Dota artifact

The pre-order of Valve’s card game in the Dota 2 universe started yesterday. The game immediately became the bestseller.

First mistakes

Now it’s clear that the information in open sources was simply not enough for Dota 2 fans to understand how difficult the game can be. A week before the release of BeyondTheSummit, the first tournament was held: for most viewers, Artifact seemed too complicated and incomprehensible.

Suffice it to say that about 80 thousand people watched the beginning of the broadcast, and about 20,000 remained by the end of the event. People were frankly bored and could not figure out how the mechanics work. Some were so frustrated by this that they abandoned the Artifact forever. Others were wary and did it absolutely right.

The release finally dotted all the i’s. If game critics were more cautious about Artifact and rather praised it, then ordinary gamers on the very first day threw negative reviews on the game page and gave it a sad 2.8/10 on Metacritic.

Along with the high entry threshold, the economy has become a big problem. Doters accustomed to free-to-play now had to pay 1400 rubles for the game, and this was just the beginning: paid modes and, most importantly, an internal model that allowed donators to get cards and decks that were too strong.

Metacritic review (November 2018):

Welcome to pay-2-pay! Excellent game quality, great mechanics, great lore, but all this was created with one goal in mind — to squeeze more money out of you. Everything here requires payment, including competitive modes. At the moment there is no ladder, ratings or progression system. Not even achievements, just imagine! If you’re not going to spend hundreds of euros on a game, don’t even touch it.

Later, game designer Richard Garfield called the monetization model a mistake. He and his team didn’t want to make Artifact a free-to-play game, as that would require them to attract new users with daily login rewards and encourage people to spend money in various ways. But the authors of Artifact could not establish the necessary relationship with the players.

There were other problems as well. Weak balance, no ladder, no daily quests or events. Difficult adaptation to game mechanics. Too long matches. Yes, there were those who were able to understand and love the Artifact, but most of the players were disappointed and closed it forever.

If on the day of release TCG was in fourth place in terms of popularity on Steam, on the second day Artifact’s online was lower than that of Team Fortress 2. Despite the fact that Valve and the community (pro-players and tournament organizers) tried save the game, popularity fell at a catastrophic rate.

Lack of direct control is annoying.

In auto-battlers, you are the commander of your army, making important decisions. Sports simulators have a coach mode where you don’t play directly. The lack of direct control makes sense: the coach doesn’t catch the ball in real life either.

In Artifact, what you have control over and what you don’t doesn’t seem logical. Why do my creeps, soldiers, part of my army choose random lanes? If I’m in charge, shouldn’t I decide where they go? If I can get my hero drunk with a potion, put an item on him, i.e. essentially make a decision at the micro level, why can’t I choose where the character will attack? It’s the same solution at the same level. I choose the lane where they go, but I don’t choose where they stand and who they attack, although I can choose what potion they drink and what they wear.

It may sound like a nitpick, but often times, systems get disliked because they’re mishandled. Players easily accept game restrictions if they follow the logic.

The game will have paid in-game events

There will be following modes, in fact they are the same arenas from Hearthstone/Elder Scrolls Legends, with special rules. 5 event tickets will cost $4.95

Games go up to 5 wins or 2 losses. The prizes for them are as follows:

Similar to packs, there is no way to get tickets for these events for free, they will not be sold on the market.

It’s not fun at all

There is a popular idea in game development that you can make a game fun by adding good things to it or removing bad things.

A lot of good games only focus on one of the two: Killer 7 and No More Heroes have a bunch of fun, interesting, but unpolished ideas, and Blizzard games usually try to get rid of everything bad, shy to add something innovative and new.

Unfortunately, Artifact can’t handle either. When I say a game isn’t fun, what I’m saying is that there are few «wow moments» in the game, and instead of removing the bad stuff from the game, the developers make the bad stuff the pillars of the gameplay.

Looks too much like a spreadsheet for accounting

It’s certainly a scandalous statement, but I don’t like eurogames. Too many of them are transparent optimization problems, seasoned with some topic. In other words, tables in Excel with graphics. Artifact also suffers from this in the combat phase, which is important in many games. In Artifact, the numbers on both sides are counted against each other, and give the result, very suspiciously resembling the solution of a mathematical equation in which the cards are just coefficients.

In Hearthstone, combat is «my venomous snake bites your taunt unit and kills it, and then 3 of my wolves hit your bear.» In Artifact, this is «my combination of numbers is subtracted from your combination of numbers, but before that I will play a card that will increase the second number in my combination by 2.»

Keeping the player from participating in combat is a working idea. This is how Autochess/Underlords/Teamfight Tactics work. But in these games, combat is visually captivating and a bit unpredictable. Observant players can look at how the fight went and make adjustments to their unit formation to improve their performance. Combat doesn’t look interesting in Artifact, and even before it happens the game will tell you what the outcome will be. It doesn’t even make sense to look at it, the fight is a formality.

-Yes, I met a lawyer, we went to dinner, I ate lobster soup, we went to my place, la-la-la, we never met him again. “But you sla-la-la-blahed the best part. — No, I said about the soup.

The fight in this conversation is the same la-la-la. It should be the most important, but instead the most boring thing in the game.

Very. A lot of. Random.

Implosion was hated in Hearthstone. This card dealt 2-4 damage and summoned that many 1/1 imps. The problem is that if you count a 1/1 Imp as a unit of damage, then the card is capable of dealing from 4 to 8 damage. The spread is incredibly large. If you’re lucky, this card could win you the game on its own.

Random in Artifact is almost entirely inspired by Implosion.

Bounty Hunter sometimes deals 4 additional damage per turn. Sometimes not. There is no counterweight to outcomes. You just toss a coin and it has a 50% chance to deal extra damage. In MOBA games, characters attack each other hundreds of times, and in such a game, critical damage just obeys the law of large numbers. In Artifact, the Bounty Hunter will hit a maximum of 8 times per game, and the law of large numbers does not apply here. If you’re lucky in the first three moves, you’re on a roll. If not, instead of a hero you have a useless unit.

Ogre Magi has a 25% chance to make a copy of any cast spell. 75% of the time, nothing that seems unpleasant happens when this Ogre Magi is yours. And 25% of the time it copies spells, which feels terrible when it’s not yours. You can cast 6 spells per game, which Ogre Magi can copy, which means that the law of large numbers does not apply here either.

Sometimes your Bounty Hunter stands in front of an enemy hero (which is also determined by random), he gets bonus damage, you apply Track (a card that increases the gold reward for killing an enemy hero), and then Payday (a card that doubles your gold ). Spend a bunch of gold on buying a Helm of the Dominator (an expensive item that wins you the game most of the time, especially in drafting) because it ended up at the top of your item deck by accident and you win the game in a couple of turns.

This is how my last game in Artifact went, after which I deleted it.

Of course, someone might say that the best player in the world could win against Helm of the Dominator on the second turn, but matchmaking should match players of the same skill. Other players will say that this can happen on both sides. The only question is that no one will play a fighting game where a low kick has a chance to kill an opponent with a probability of 1%. People hated tripping in Smash Brothers. It doesn’t matter if such things are rare, or that they can happen to both players with an equal chance, and therefore they are «balanced». They annoy, not entertain. Losing due to this mechanic is not due to lack of skill, just like winning.

Artifact is packed full of random things: where the hero will appear, where the creeps will appear, where the creeps will stand, in which direction they will hit (the infamous «arrows»), which card you will take from the deck, which cards will lie in 3x decks in the store, two of which were not made by you Eclipse selects targets randomly Chain Frost bounces in a random direction Roar scatters targets in random lines Ravage can stun everyone or none All of this is bad randomness, like Implosion Instead of two approximately equal, but different options you either get lucky or not.

The strangest thing about this random is that the game positions itself as a complex entertainment that requires maximum brain effort from players with high IQ. But at the same time, it has a dozen mechanics, performed in the style of Mario Party.

Heroes are primitive and boring

Heroes should be the center of the game, but are often just sets of stats. They have weak or no abilities at all, and their key abilities have been removed to separate cards. (When you add a hero, you also add 3 copies of the «signature» card).

One of the happy moments in card games is opening a new set of cards, looking at which you have to think about how to add them to your deck. In Artifact, you might think heroes are the foundations around which you should build the rest of your deck. But the design of the characters themselves is quite conservative. What can you think of around a hero that deals 1 damage to a random target every turn? Or around a hero with good stats but no abilities?

The fact that heroes themselves are either good or bad says a lot. They either show up in the best decks or don’t show up anywhere, instead shaping strategies around their features. If you’re making a deck of red cards, there are some heroes you want to add no matter what, and others you don’t want to have under any circumstances, ignoring the idea of ​​your deck. Ax, one of the best heroes in the game, has no abilities. He has good characteristics, thanks to them he kills enemies. His signature card also helps him kill enemies. He is interesting in the most boring way.

We can say that signature cards and heroes should be considered together, but signature cards are also very weak. Sometimes, having to add them to the deck feels like a punishment. And using a spell from the deck is completely different in relation to using the abilities of the heroes.

Another bitter pill lies in the similarity of the characters to each other. Green cards have a hero that gives +1 defense to nearby units, a hero that gives +2 defense to nearby units, a hero that gives +2 regen to nearby units, and a hero that gives +2 attack to nearby units. There are 4 heroes in the game that increase the base stat of neighboring units, and one of these effects is better than the other.

Hero Debbie, deals extra damage to heroes and buildings. Phantom Assassin deals bonus damage to Heroes. Bounty Hunter deals additional damage to buildings and units, but with a 50% chance. These are all cards of the same color, slightly different from each other. Crystal Maiden and Outworld Devourer both restore mana, while Kanna, Prelix and Venomancer summon additional units. All of them are blue characters.

Card games often use cards that are similar to each other to create a difference between cards of different colors. In Magic, Healing Salve is a white spell that heals for 3, while Ancestral Recall is a blue spell that draws 3 cards from the deck. The cards hint at the philosophy of their colors, highlighting the differences between them. In Artifact, similar cards are also in the same color, but instead of emphasizing the differences, they tell us about the simplicity and one-sidedness of each color. Greens have 4 heroes that increase stats, probably this should tell the player that green cards increase stats, but such a coincidence of mechanics is not interesting, and the best spell that increases the characteristics of red, not green.

The philosophy here is reminiscent of a fighting game where there are many characters with the same models painted in different colors. But such shenanigans are needed in order to add an interesting idea, spending less time on development. In a card game, if you already made a character that gives +2 defense to neighbors, why add a character with the same effect, but gives +1 defense to neighbors? You’re doing the exact opposite of what fighting games do: skimp on unique ideas.

Perhaps the logic is to reduce the difficulty a little and make the game easier to understand, but in this case it is a mistake: of all the ways to simplify, turning heroes into dull copies of each other is the worst choice.

Even the character art looks dull. Each art individually is good, but it creates the feeling that the artists were ordered to work in a pair of identical poses, with the same angles. And many arts do not make it clear the nature of the characters and their abilities. In Dota 2 Ax, he’s a cool badass, while Ogre Magi is more of a comedian. In Artifact, they both stand in imposing, stoic poses.

I’m not a racist, but they all look the same.

Criticism of art may seem out of place when we talk about game design, but the point is that conservatism in decision making goes even beyond game design.

The game encourages inactivity

Very often, the best strategy (or even the only one) is not to let your opponent do anything, or not to do anything yourself. Maintaining the initiative in Artifact is incredibly important, even though the mechanics themselves seem invisible (which makes a non-intuitive game even less intuitive). If you have the initiative and you finish your turn, then on the next line you will start walking first, keep the initiative, and so on ad infinitum. Often, the best move is to pass the turn to the opponent, sometimes three times in a row to keep the initiative until the right moment.

In most card games, your hand has the maximum card size and you start with more cards. In Artifact, you start with a small hand and draw two cards per turn, but you must use them on 3 different lines (3 full different tables). Sometimes, doing nothing is not just a winning strategy. Sometimes, doing nothing is a mathematical pattern that you come to if you use at least one card on each of the lanes per turn.

“What if we make the game more difficult for players to play by adding a nasty RNG?” you might ask. Well, the developers of Artifact thought the same thing. There is a mechanic here that allows you to block enemy cards for a while. It’s like «discard a random card» in Magic, but Magic has mechanics that are revealed by interacting with discarding cards, and the graveyard adds even more interaction, while Artifact doesn’t offer anything.

Sometimes you will block 3 opponent’s key cards and win. Sometimes not. You will never know which cards you have blocked, and therefore both players are unhappy: one of them randomly blocked a card, and the second does not know whether the blocked card is useful or not.

Best of Worst Game Design

A lot of the design of this game feels like it was added to answer the question “what can we do to keep the game from falling apart?”

Why is the location of heroes on lanes random? Because with the ability to choose lanes, you could hunt weak heroes and put your own, stronger ones.

Why do 3 creeps randomly enter different lanes on the first turn, and only 2 on other turns? I guess because the other options were game breaking.

Why are the «arrows» (direction of attack of units) random? Because if the units attacked directly, then the game would be faster and less interactive. And if units always attacked other units, it would be too tedious. The ability to choose would have added a clunky interface, and the game would have become even slower. Therefore, so that the game does not fall apart, the arrows are placed randomly, based on some data. But if the data can tell you how to set the arrows optimally, it will not necessarily tell you how to set them in a way that is interesting and fun. And in fact, they don’t.

All these solutions are solutions to already existing problems. All alternatives are certainly worse, but making the best choice from the worst does not make a good game. Choices are limited to fundamental game mechanics that the developers couldn’t get away from.

Hope before release

After the first announcement in August 2017, the Dota 2 community was waiting. Against the backdrop of the weak use of the Dota universe, a full-fledged game that could theoretically become a worthy competitor to Hearthstone seemed like something breakthrough, something incredible.

News and inside information was at a minimum for the next year, as Valve chose the path of private beta testing (artifact invited commentators and players from Hearthstone, Magic: The Gathering and Dota 2) and sparingly shared information.

But the excitement around the game kept growing: in the summer of 2018, Virtus.pro signed three players at once in a new discipline, and later other organizations from China, Europe and North America followed suit. There were great prospects ahead, and it was simply impossible to believe that the game would be a failure.

Roman Dvoryankin, general manager of Virtus.pro (July 2018):

Valve hasn’t released a game in years, which means the next one should be powerful. I am sure that the company will make every effort to ensure that this is the case. Therefore, we believe that Artifact will be the very next big project.

In our opinion, there is room in the market for another card game like Hearthstone or Gwent. Given that Blizzard is completely focused on Overwatch, and things have not been going well in Hearthstone for several years, Artifact has every chance to shoot.

Three lines are unusual. And uncomfortable.

Artifact’s most unique game mechanic comes from DOTA. This, of course, is about three lines, or three game tables. The tables are not related to each other, although the player has only one hand, from which he plays cards on all three tables. And these three lines are one of the biggest problems of the game.

At any point in the game, you either look at one of the three lines, or move the camera away and look at all three, although in this form it is incredibly difficult to see anything. This means that a lot of information will be hidden from you. Not in the sense, as in games with «hidden information», but in the sense that you will not see it on the screen. If you’re playing a match in Artifact, you’ll always have to jump between lanes to keep the whole picture in your head: maybe you want to play an item on one lane, but you need to double-check — what if it’s better to play it on one of the remaining two? The game makes it feel like you are playing a board game and you are looking at it through a porthole window that only allows you to see a third of the game table. Because of this, even the best players will make mistakes, each time forgetting about the state of the other lines, and the process of eternal rechecking itself increases the duration of the games.

For the viewer, it’s even worse. If the player needs to refresh his memory, he can switch lanes when he needs to. The spectator does not have such an opportunity. If the viewer has forgotten the state of one of the lines, he will have to hope that the player has forgotten the same thing, and switch the camera to the line that the viewer needs. One of Artifact’s stated goals that Gabe Newell talked about is a deeper understanding of esports in DOTA, but Artifact is a bad fit for esports as the experience for the viewer is guaranteed to be terrible. (The Million Dollar Tournament never happened.)

When we try to port Magic: The Gathering to PC, the game becomes slow and clunky. This, in turn, tried to avoid Hearthstone. One of the strengths of Hearthstone is that it was developed as a computer game. Artifact, on the other hand, looks like a game that was conceived under a large wooden table in real life, but for some reason the game ended up in a computer format for which no one was going to adapt it.

The argument for three lanes may be based on the fact that this structure is taken from DOTA, but Artifact is not a DOTA card game, it is a DOTA card game. In DOTA, all lanes lead to the main building, which must be destroyed in order to win. In Artifact, the lines never converge at one point, and the conditions for victory are different. Therefore, three game tables is an idea that connects the game with DOTA, but it is poorly implemented, has many shortcomings, although hypothetically it could be a strong point.

There is no way to get packs for free.

Generally. There is no in-game currency in the game. For the games you get absolutely nothing (now there is not even a ladder, everyone who has access to the game at the moment has all the cards). Card packs can only be bought or won in paid events. Hence the following fact:

Deceptive unimportance of long-term investments

The unifying characteristic of tabletop «eurogames» such as Dominion or Ticket to Ride is the management of one’s potential and kinetic energy. In these games, the player’s actions either increase his future potential, or spend resources to get a positive effect here and now. An important element of strategy for serious play is that you choose when to be a tortoise and when to be a hare. In others, you choose between building up strength and running to the finish line at all times, overtaking other players.

At first glance, Artifact looks like such games. Even before the release of the game, the developers focused on the modification system. Mods are a long-term investment that improves the map for the rest of the game. The item buying system seems to be a variation on the same theme, allowing the player to buy a cheap and useful item now, or save up some money and buy a powerful item later. For 3 gold you can buy a Traveler’s Cloak which will give you +3 health, or you can save up 7 gold to buy a Fur Lined Mantle which will give you +8 health. The latter gives you more per card/item slot, but is slower — the turtle approach.

But there is one problem: Artifact is too short for the turtle approach to work in this game. Even in the draft, which is a bit slower, games end within 5-7 turns.

When a hero dies, he waits 1 turn to return to the battlefield. If they die before doing anything, they are useless for 2 turns. Modifications and expensive items remain on the character after death, but by the time the hero returns to the battlefield, the game will already be nearing completion.

As a result, many cards that apply modifications are bad. And the good ones would remain good even if they acted for a short time. It says a lot that the best mod card is Time of Triumph, which costs 8 mana and is used as a finisher. They win the game right away, and have no «long-term» effect.

Midrange (not early, but not late) items are almost always useless. Saving money on turn 4 for an item with a long-term value does not make sense if the game ends in the next 3 turns. The mechanism for buying items also works against «average» items. From your item deck, you are offered 1 random item. In order for you to be offered another, you must either buy this one or wait one turn. Therefore, a strategy that allows you to only get cheap items will dominate, perhaps with a couple of very expensive, game-winning items that you quickly get to by buying cheap ones.

Before the release of the game, the developers emphasized that value (the value that the card generates) is a key concept. But playing trying to squeeze out the maximum value is often a mistake.

Unlimited hand and table size sounds cool. But it just sounds.

One of the features of the game that the developers focused on was the unlimited size of the table and hands. Like three lanes, this mechanic sounds good on paper, but it has serious drawbacks. U I does not do well with large tables and hands — the player will have to scroll across the entire table to get to creatures that do not fit on the screen. So you are forced to look at only one line out of three, but if there are too many creatures on the board, you can’t even see that single line completely.

Even if we leave out the clunky UI, it’s still not clear if the unlimited table size is a plus or a minus. The wider the table, the more difficult it is to predict where the units will be located. This makes the game even less predictable and increasingly random. Also, it’s not very pleasant to realize that blue cards are both good at creating a large number of creatures on the table and clearing the table. Other colors do not have tools for interacting with the big board.

Right now 6 units are not visible on the screen, and this is only one of the three tables

Unlimited hand size allows you to build on a strategy where you first accumulate a large number of cards, and then play them in one big mega move. In most card games, however, this behavior is usually punished by the loss of initiative on the table (tempo) and mechanics that discard cards (for example, when you collect too many of them). Both Magic and Hearthstone have penalties for drawing too many cards. In Artifact, while players rarely draw their entire deck (I don’t even know what happens then — I’ve never encountered this), often the right strategy is to do nothing, and accumulate cards aggressively. A hand size limit would force players to do something a little more often, as well as force them to make tough choices about which cards to keep and which to play or fold.

The cost of a complete collection of cards

Many streamers have said that the full collection of cards will cost $300 (in addition to $20 for the game itself). But as it turned out, all players who had access to the game before 11/10/2018 received all the cards at once, there are simply no buttons for buying packs in the game at the moment.

But the figure of $300 for the entire collection is the subjective opinion of streamers and players. There may be a situation when a ton of duplicates will fall out of packs for this amount, and let’s say the required rare cards will not fall out. What will you do? That’s right, go to the market. And there this card will cost at least $40-60 dollars. You will sell duplicates for a penny, because the market will be saturated with them. And there are situations when the prices of cards from the strongest meta decks will skyrocket to absurd values.

In this video, experienced Hearthstone player Kripparrian talks about prices in both games. But as it turned out, he does not have access to Artifact

And all these expenses will not be the last. In the official FAQ in the description of the packs, it says «Each contains twelve random cards from the «Call to Arms» set». I.e. already at this stage, additions are planned that include new cards (haha, as if we are not talking about tcg)

And this means that when the next expansion is released, old cards will most likely «leave the meta», their price on the market will drop, and it will be impossible to spray them like in Hearthstone. And if you want to continue playing, you will have to buy new packs or cards from them on the market.

Steep dive

It soon became clear that this was a resounding failure. Another problem was that Valve almost did not react to this, acting in their own style: a minimum of communication with the community and their own line, from which one cannot deviate a single step. Artifact did not receive global updates, and even the discount on the Valve Complete Pack (you could get the game at half the price) simply disappeared the next day for no reason.

The players responded in kind. On December 9, Artifact left the Steam top 30, and in a month, the daily online rate will drop below five thousand people for the first time. Valve is firing Garfield’s team and «feeding» the community with common phrases. Thanks for your interest in Artifact. Your support is very important to us. We are working on updates.

Message from the Artifact team (January 2019):

Now we are working tirelessly. We still have a lot of work to do. We have big plans for Artifact, but it will take time to realize them. Now we are not ready to say what awaits the game next.

Message from the Artifact team (March 2019):

Thank you so much for playing and we’re glad to hear that you like it. We are still working hard to improve the game, and there are many changes ahead of us.

Quality updates are our way of building trust with the community, and we plan to keep releasing them. Sometimes a lot of time passes between updates, and if we are silent, it means that we are fully at work. I hope we can win your trust.

However, users did not see real results. By the end of March, online Artifact fell below 100 people for the first time, and the game section on Twitch returned to the top 10 only when dozens of channels launched streams of gacha, movies and anime on it. By the way, after that, the platform introduced two-factor authentication for new streamers.

By the fall, the situation had not changed, but there was hope: the authors allegedly talked about a big update that would be just about. It didn’t quite fit with the fact that two or three people worked on the game, but the belief in the restart remained. As a result, by March 2020, Valve announced a new version of the game Artifact 2.0.

Monetization aside

This is a blog about game design, but it would be irresponsible to discuss the failure of Artifact without mentioning monetization. If you have heard criticism towards Artifact, most likely it is about this. For this reason, I don’t see the point in developing this topic, but I will only note that the game combined 4 different monetization models: Buy the game, pay for the cards, pay the commission for selling the cards, pay to play the big guys mode. I once heard that King of Fighters 95 is a big experiment on fireballs and uppercuts. Artifact, in this case, is a big experiment on monetization. Frankenstein’s monster, consisting of different monetization strategies.

Some might say that monetization is fine, or even that it is «generous». In this game, it’s not difficult to get rewards so often that you never have to pay, and the price for all the cards combined is pretty low. These people are wrong, and a separate large detailed blog could be written on this issue, but even if they were right, such punishing monetization still remains a problem. It’s not the same problem that killed the game, but it’s still a problem. Quite a few people still bought the game and started playing it, despite the monetization. But even they stopped playing very quickly.

What is the future of Artifact?

It’s useful to look at an example of a game that managed to recover from its failure: Final Fantasy Realm Reborn. (I worked at Square Enix when this was happening but wasn’t involved in the development of XVI, only helped a little by sharing my opinion with other developers)

One of the turning points was the change of the main producer of the game. They became Naoki Yoshida. If an item in the game has a bad scan, the 3D artist is to blame, if in some part of the game everything seems to be repeated over and over again, the level designer may be to blame. If the game is filled with problems to the top, and these problems are pouring in from a cornucopia, you either have problems with your entire team, or with whoever is at the very top of it.

The development team suffered from decision making problems and the new producer was able to prioritize correctly.

Naomi Yoshida, after taking his position, posted this blog entry.

Sometimes I consult other developers, and the first thing I do is sit down to play a game and write down all my impressions. This can be as high-level reasoning «The two game modes look unrelated to each other», and «this effect has small artifacts around the edges.» The idea is to find all the problems, fix them, and end up with a good game.

And Naomi Yoshida’s blog post is a great example of this approach. Some things like «learning system for new players» are critical, and some things are «fanfare sound when you level up is too quiet/not cool enough».

I would like to note the contrast of this approach in relation to trendy roadmaps, which serve only as a tool to create promises and the illusion of communication with the community. A lot of what Naomi has said in his blogs is simply admitting mistakes. «We see that the system is bad, we will figure out what is wrong with it, and try to come up with a solution that will be better.» Sometimes they already have a solution, and then they tell the players about it. For players, this approach is more refreshing than cold water. Reading stuff like this makes me want to say out loud «this guy knows what he’s doing», and this approach inspires confidence.

Square Enix made it clear that they are serious about this game. Perhaps the restart of the game for them in some sense was even more serious than its original.

I’m afraid that Artifact 2.0 is going in the opposite direction. Apparently, some of the developers have already left for other projects, incl. in Underlords. It is not clear whether this is just a shake-up in the team, or whether the developers have changed their priorities. Nothing tells us that Valve’s approach has changed since the failure.

Valve is very focused on statistics, playtesting, and other objective indicators. But they also led the game to failure.

The company has already invited streamers to try the new Artifact. The only problem is that the same people were touting this game before it came out, saying that it had «minor flaws». They gave positive feedback for months, only to reveal after the release of the game that the audience had a completely different opinion.

And the worst thing is that it is not clear whether Valve knows what the problem is with the game. The whole communication is a short blog where they said that the game is really flawed. But it’s not clear if they consider three lanes to be vital to the success of the game, or is it a curse for them that only spoils it? The combat system — what is needed, or what needs to be reworked?

I understand the logic of the «speak calmly and always carry a big stick» approach. Instead of big speeches, they prefer to go off the radar and release a game that will turn out to be great, impressing everyone with its quality. But Square Enix’s approach lets him make sure they’re in touch with reality. If they released a blog with a controversial idea, then the players would say «no guys, your decision is wrong». Locking yourself in your basement and working on a game for years is a very dangerous undertaking, especially when you’ve already failed miserably once by doing the same.

If you’re not familiar with Artifact’s rules, it might be hard for you to follow the thought, but I think you still caught the spirit of criticism even if you didn’t catch the details.

So if you’re wondering what’s in store for Artifact, I can tell you that nobody knows. Maybe 2.0. will blow everyone’s brains, or maybe the game will be quietly left to die in the backyard of history. But I, as a player, would like to see two things from the developers: specifics about what the developers consider the strengths of the game, and what are the weaknesses. And the willingness to question any game mechanics, even the most key ones in this project.

The game costs $20

The pre-order of a card game from Valve in the Dota 2 universe started yesterday. The game immediately became the bestseller. At the same time, the Beyond The Summit Preview Tournament for this game began (Playoff starts at 20.00 today, you can watch it on steam.tv and on the btsartifact twitch channel) . And it sounded a few facts about the business model of the game. I wanted to discuss them.

There will be no regional prices. The game costs 1,390 rubles. No trials, no demos. If you just want to check if you like the game, welcome to YouTube. Yes, for $20 dollars you will get 2 starter decks, 10 packs of 12 cards, and 5 event tickets. Those. in fact the game is free, but you have to buy 10 packs. And from here comes the next paragraph:

Last try

I must say that Artifact 2.0 didn’t have much chance anyway. The card game was long forgotten, and only very high-quality changes could save the game. They were, yes: Artifact began to treat newcomers better, pleased with new gameplay and interesting mechanics. Even the economic model has become very pleasant.

But not everyone could see it. Valve chose the path of a closed beta test, and this time they sent invites randomly. Several professional players and commentators failed to receive invitations, but Lex, who refused to launch the game, received a letter of happiness.

Yes, in the summer the number of invites increased. Yes, there have been several major balance updates. But Artifact 2.0, for some reason, remained a closed club where you can’t invite friends. Where there are no popular streamers and bloggers who could promote the game.

Even if the game turned out well, is it any wonder that Valve failed to attract new players? Perhaps the company has a different opinion on this issue, but the result, as they say, is obvious. Today, Valve has completely closed this issue; and I would like to think that we have drawn some conclusions. Farewell, Artifact.

Evaluate by what we see

Generally speaking, when it comes to Artifact, there is a popular thought on the Internet: “The game may have problems, but you have to admit that it has great design!”. The consensus looks like this: the game design of this game is a strong point. But the game failed for some obscure, vague reasons, like «doesn’t fit in the market» (The thing they say about any failed product). I want to offer you something different: Artifact looks like a game with good game design. But in fact, this game design is disgusting.

Artifact is truly a game made by professionals. The game has the feeling that it was made by smart people and smart girls — those who passed the exam with a good score and went to Carnegie Mellon University. The game looks like a quality product. And I think it’s tempting, especially among other game designers, to look at this game from a distance and say that it has great design. Because that’s what it really looks like.

To get anything out of this blog, you have to acknowledge that a game can have quality production, strong developers, and other high quality metrics, but still be deeply mediocre. And throughout the rest of the text, I will prove to you that this Artifact is exactly that case.

The customer is always wrong

I suspect that many game developers are familiar with such game designers: In response to criticism from players or testers, these game designers begin to scold the players for their shortsightedness, instead of considering that there may indeed be flaws in their design.

As a person who likes to take unpopular positions, I think that “trying to satisfy testers 100%” is wrong. Tomonobu Itagaki once said this about Ninja Gaiden: «In the beginning this game was easier, but when the testers said it was ‘too heavy’ I made it even harder.» I can’t help but admit that this statement makes me smile. But if most players find your game confusing or repulsive, then you should admit to yourself that something is wrong.

UPD: earlier the text was cut off here, but in the comments they threw off the version of the article from the web archive, and I immediately ran to finish off the material, enjoy, and thanks to Alexei Saenko from the comments.

Players are very bad at complaining in such a way that they can sort out and solve the problem. Almost always you, as a developer, have the opportunity to find fault with words and shut a person up. But you can also try on a detective’s cape and still figure out what’s wrong with your game mechanics.

I used to spend a lot of time on the Star Wars Galaxies forum. It’s not that I’m a fan of the game — I just enjoyed reading how the developers respond to player complaints. Almost always, they wanted to prove to the players that they were wrong, instead of understanding what problem the players were facing. If the player writes: «this weapon is weak», then the developer will answer «according to our table, this weapon has above average DPS, you are wrong.» It is clear that this approach is not productive, and if you want to make a good game, then you should get to the root of the problem: why does the player think that the weapon is weak? Perhaps some bug is to blame for this, perhaps the formula in your table is written with an error, perhaps the practical damage differs from the theoretical one, because the weapon hits too rarely, but strongly, leading to a serious overkill. Maybe the weapon has little knockback and stun on enemies, and it’s more important to know «how much damage will I deal before the enemy approaches me?». Maybe the weapon is too bad at dealing with a very common enemy, or maybe the players just don’t understand how to use it correctly at all? Even the sound of a gunshot can create that feeling. The problem is that the players are not invested enough to spend their energy on such a problem.

Unfortunately, the interview with Three Donkeys is more about denying the problem than dealing with it. For every claim, they always have a prepared answer up their sleeve, why the person is wrong.

Bad monetization? No, what are you, collecting the entire collection can be cheaper than in Hearthstone and Magic. Of course, this comparison doesn’t make sense, because Magic is a game with a long history, physical cards, and a whole bunch of card sets that came out and complemented the game. And in the online version of the game, Magic Arena, you get so many free cards that you don’t know what to do with them. But this game is free, Artifact asks you to pay for the game first, does not provide quests or other reward system, and also asks you to pay for the «big boys» mode (Prize play). Yes, Artifact is cheaper to play, but saying it out loud requires some serious mental gymnastics in your head.

Are there many RNGs in the game? Well, the game has a very wide spread of ELO between all players, which means it is demanding on your skills. Case closed and no appeal! The problem is that we can’t even look at this ELO, and therefore confirm or refute this statement. We can confirm that good players win more often than bad ones, but what does this give us? It looks like matchmaking is bringing the bad players together with the good, and it’s not clear if this is due to the lack of players, or normal matchmaking, but even when the game came out, the players were fighting with the beta testers who were given the keys a year ago! Even if the players did not have a «wide spread of ELO».

You can refer to your internal statistics as much as you like, which says that the game is difficult, and there is little randomness in it. And by some measure it will, but you can’t point your finger at a player’s ELO ratings and say your game is fun to play. But the fact that Artifact failed and collected negative reviews is exactly the data that says that it is not fun to play. Worst of all, the developers attribute the negative reviews to «review bombing». Except that «review bombing» happens when gamers find more SJWs than they can handle, or when a game becomes an Epic Games Store exclusive. Getting negative reviews is not review bombing. The man on call 2 was not a victim of «review bombing». «Man on Call 2» was the victim of a simple circumstance — it’s a bad movie.

Difference between review bombing and negative reviews

The developers do not give any explanation for the failure, except «the players did not appreciate how close the game is to the ideal». And «close to perfect» is not my mockery, the developers believe that they have made the best card game in their entire history of game development. Why then did she fail? Even when it comes to actual issues like monetization, they quickly pull back, saying that the players are confused and did not understand how the monetization in the game is actually very OK.

The rest of the people — developers, players, journalists, in turn, explain the failure by the fact that the game «did not fit into the market.» But then again, saying «didn’t fit in the market» is tantamount to saying «it failed.» It doesn’t explain anything. Obviously, the market for computer card games exists, so if the game failed, then the point is in its implementation.

This part of the text could be a little rough, but it reflects our reality: if you are a developer, you should be able to work with criticism. You can’t come up with clever explanations for why the critics are wrong. Because those who criticize the game are not your enemies or rivals. If a person does not like your game, you cannot prove to him that he is wrong, and that he should love it. That’s not how things in this world work.

Overly «smart» creativity

The final pillar of this game’s gaming philosophy is the emphasis on «smart» card interactions. In games like Magic: The Gathering, a «smart» interaction would be when a player casts a spell that causes them to discard a powerful creature and then cheaply resurrect it with a second card.

In Artifact, «smart» interactions are related to mathematics and the formal description of maps, and not to the interaction of interesting systems with each other. For example, take the Rend Armor card. Its text reads «change the unit’s armor to -X armor, where X is the current armor». Intuitively, this card seems to destroy armor; after all, that’s what the word Rend means. But here’s the trick: if the unit has negative armor, then we will subtract the negative number from the negative number, i.e. fold. Therefore, you can use this card on a unit to gain armor.

The second example is the «Shop Deed» card. Its text is «Each item in the Secret Shop costs X less, where X is its current value.» This is a very complicated and very formal way of saying «everything in the Secret Shop becomes free.» But what if you have TWO Shop Owners? Then, the cost of the item is (X-2X), i.e. — X, i.e. when you buy, you get money, and not vice versa.

And although it makes sense and quite logically follows from what is written, intuitively it does not lend itself to logic. Having one shop owner will allow you to get things for free, that’s fine. But why would two owners of the owner pay you? What is it — two owners in one shop? If you own a Subway franchise, can you buy it again so you get paid to eat there?

All this may seem too petty to discuss, but I’m focusing on this, because that’s what the developers have focused on. This is what they are proud of. But I don’t understand why, because all this «smart» design reminds me of SAT exams, with questions like «1/X less than X — always, sometimes or never?» Once you learn how to read carefully, all these ornate descriptions of cards will stop surprising you, and they will have nothing more to offer you.