Nan Duan

Hades (2020)

by

I’ve had an interesting relationship with Hades. I purchased in Early Access in late 2019, on the Epic Games Store, when I felt it was already very hyped. I played about 20-30 runs (no clears) on 2 different saves (1 at work and 1 at home – this was before they integrated Epic Game Services for cloud saves), and moved on to other games.

When the game moved out of Early Access in 2020 – I was quite surprised at the big reception it had, perhaps even more so amongst game developers (at least those that show up in my Twitter timeline). This revealed a bias I had developed: I had subconsciously discounted the marketing opportunities of the “real launch” and the subsequent (well-deserved) critical acclaim.

I would though, love to see an updated sales figure, which would help better quantify what portion of lifetime sales are from “Early Access”.

Somewhat related: as of Mar 10, Valheim has sold 5.7M copies in 5 weeks of Early Access on Steam; what are the odds that it will have another massive adoption bump when it removes the “Early Access” label?

In any case – I got back into Hades back in January, with another fresh save. And I was truly hooked. As of this writing, I’ve spent roughly 200 hours, and have completed all achievements (it’s been a long time since I cared about doing that in any game – the last game I aimed for 100% was Red Dead Redemption a decade ago). I’m still doing 1-2 runs per day, mostly trying to beat my speedrun personal best – which, as of Mar 17, is 9m35s (the world record is 5m57s; my very first clear was over 40m).

My personal best time in Hades

Design systems

Let’s talk about the design systems that Hades employs to deliver a 200h gaming experience.

Within a “run” – leveraging randomization to create a unique experience every time:

  • A full “run” consists of 30+ rooms across 4 biomes1, with 4 boss fights.
  • Randomized room types – a “run” is constructed of a random collection of rooms (with some assembly parameters) :
    • Majority of rooms are combat oriented, with some variation in enemies composition: standard enemies rooms, mini-boss rooms, boss rooms. There are also some special objectives that can be applied, for example “survive 45s” or “compete with Thanatos to see who has more kills”.
    • Non-combat rooms: shops, fountains, story rooms (one per biome in the first 3 biome), Chaos Gates (which gives players the chance to take on some additional risk for some reward).
    • Parameters: typically, biome 1 has 14 rooms; biome 2 is 10 rooms; biome 3 is 12 rooms; the last room per biome is the boss room, while the preceding one is a shop.
  • Crafted room layouts and runtime enemy spawn behavior:
    • Rooms are picked from a handcrafted pool (not procedurally generated).
    • There are a (large) number of spawn points from which enemies can spawn; enemies spawn in waves; and spawn location is influenced by the player’s location.
    • The result is that even when the same room is played repeatedly, the flow of combat is rarely exactly the same.
  • Randomized room rewards – after the player completes a typical room, they are presented with 1-3 choices for the next room, with the corresponding reward (and sometimes indication of room type) displayed:
    • “Boons” are usually the most sought-after, and directly shapes a player’s build each run. There are 8 “standard” Gods that offer roughly 150 Boons in total, including 28 “Duo Boons” (which require any 2 of the 8 Gods as prerequisites, 8*7/2=28). There are also 2 special Gods, Hermes (with a set of Boons unrelated to other Gods), and Chaos (from Chaos Gates, who grants bonuses after inflicting debuffs).
    • “Hammer upgrades” significantly alter your weapon’s effects, providing another layer of build customization.
    • Other items that impact the run: “Poms”, which level up your Boons; gold, which can be spent at the shops in the run; “Centaur Heart” which grants bonus HP.
    • 4 meta-game currencies for permanent progression (which can impact a run, after certain meta upgrades).
  • Heat system – as a significant end-game system, the Heat system offers players a variety of choices to increase the difficulty and modify a run’s experience, for example:
    • Quantity of enemies, their HP, damage, attack speed.
    • Armored enemies have special additional effects.
    • Boss fights are modified and upgraded.

Outside a “run” / permanent progression systems:

  • “Mirror of Night” – this is the main permanent power progression system, with 12 passive ability slots, and 2 choices per slot. These passives unlock a lot of power very early on – for example the 3rd ability Death Defiance grants 1 extra revive per upgrade.
  • “Fated List of Minor Prophecies” – this is a quest system with 50+ unique quests. A typical quest is to acquire each Boon of a God at least once, and as such, quite directly encourages exploration.
  • Weapons and Weapon Aspects – starting out with one weapon, players eventually unlock 6 different weapons, and each weapon has 4 “aspects” (variants with different passives or movesets).
  • Keepsakes and Companions – “Keepsakes” are items that grant passives during a run (for example, the first keepsake listed grants bonus HP). “Companions” are a special type of keepsakes that grant a consumable active (e.g. 5 charges of an big AOE damage spell). Both are gradually unlocked by meeting with the game’s vast cast of NPCs, and progressing the narrative. In the end-game, keepsakes are integral to players crafting their desired builds in a run, as after each biome you can change your equipped keepsake, so there’s a light min-max game with keepsake management.
  • Achievements – this is the distribution platform (Steam / Epic etc.) achievements system. Functionally it overlaps with the quest system above, but provides a different set of goals. In my experience this list was easier to finish than the quest system.

I realize this is a pretty dry list to read through. But there are a few things that jumped out. The first is that these systems are quite independent, and as a result multiplicative in their impact on replayability. The second is that each system is scalable in isolation, and once scaled up in content, gives a nice boost to overall replayability. I suspect this pattern is partly a reflection of the game’s public development in Early Access: new systems were layered on over time, and content per system were also added over time. (I haven’t gone back to check the patch notes to validate this guess.) In any case, this is an elegant set of systems, and collectively the game feels surprisingly content-rich given the small size of the dev team.

Experiential recap

Recapping my 200 hours with Hades so far, the game has consistently given me a smooth experience with clear progression goals to pursue.

In the very beginning, the game is primarily about exploration: pushing deeper into the biomes, overcoming the major bosses for the first time; seeing new Gods and their Boons; trying out different weapons. At the same time, the early meta progression was generous and directly lifting the player’s power, which meant the player was making deeper runs even without necessarily growth in skill (but usually the 2 were organically intertwined).

At some point, I ran into the final boss Hades for the first time (and beat him!). This significantly moves the narrative forward – without giving away too much, this is why I embedded the soundtrack at the top. The narrative momentum compelled me to keep playing – I needed to find out what happens next, and with every full clear, I learnt a bit more. I can’t emphasize enough how surprisingly strong the narrative hook was during this phase.

After beating Hades enough times, the main narrative concluded. At this point, I kept playing because there were still more gameplay content to experience – weapon aspects to unlock and upgrade; and the Heat system which rewarded the rare currencies needed for weapons. There was also a narrative epilogue, that essentially required grinding more conversations with NPCs and increasing the relationship with them.

There were also still quests and achievements to chase, and interestingly, chasing these quests often meant playing sub-optimally, for example picking a Boon option I never played before (to move towards completing a quest), even though there’s a more powerful option.

As an aside, after about 50-100 hours with the game, I switched my input device from keyboard+mouse to ps5 gamepad. I was motivated to do so because I was seriously hurting my wrists playing this game (the game’s frantic spamming is quite excessive). One lesson of this switch-up: at least in my case, understanding of game state mattered a lot more than execution ability, as I did a full clear in my very first gamepad run, despite noticeably fighting the controls as I tried to retrain muscle memory.

Up to this point, I hadn’t really done any theory-crafting, and was not thoughtful about builds. After venturing into some Youtube content, I found myself accidentally diving into speed-running. The 2nd 100 hours was thus spent on a mix of farming Heat levels, pushing high Heat to accomplish a challenge, and dabbling in speed-running. I found that I liked the game enough to always give myself a new immediate goal, just as I was getting tired of the grind.

Speedrunning

A section about speedruns specifically. Hades has an active speedrun community, and the game seems aware of the fact that speedrunning is a key source of community content (UGC). There are some nice nods in-game towards this: for example, right after beating your own fastest time, your character will say “I bet I can go even faster”.

As an aside, I’m the type of player that never theory-crafts builds, and rely on “netdecking” (to borrow a CCG genre term), i.e. referencing the community-discovered meta builds and strategies. So my venture into speedrunning was accidental, out of a desire to understand the meta. Once I tried executing some meta builds, I got some very addictive positive feedback as I saw my clear times dramatically decrease. At least for me, this was a very enjoyable loop, and I see it as a type of engagement lever for a pure PVE experience.

More broadly, I’ve found it interesting how the Hades speedrun community has kept engagement up, by constantly adopting new speedrun formats. What do I mean here? The main Hades speedrun leaderboard is the “Unseeded AnyHeat”, which translated to plain English is a competition to see who has the fastest time for a single clear (based on the in-game clock), at any difficulty, without manipulating the run’s RNG seeding. However, this leaderboard has gotten quite stale, as the meta has largely been solved (there is one particular weapon that is clearly the fastest), and the current best record has already been set to under 6 minutes.

If you are a speedrunner / video content creator, trying to improve your standings on this leaderboard is an effort in diminishing returns. Instead, you could switch your attention to the other leaderboards, for example, the high Heat leaderboards, or “Fresh File” (fastest time to beating Hades for the first time, from a brand new save), or 3 Weapons (3 consecutive clears, with 3 different weapons). These different leaderboards spawn their own unique metas. And it’s not hard to imagine new formats being invented – as long as there is a common interest in playing the game, and competition based on it, the community can make up new rules to play by.

In the grand scheme of things, this community is quite niche. The amount of people who are good enough, and care enough to record and submit their speedrun videos for this game is probably just a few hundred to thousand. But I think there are some useful insights here for game developers making PVE experiences and worried about long-term engagement.

  1. For those familiar with the game, I’m cheating a bit as I’m mentally treating each of the branches in Styx as 1 “room”. ↩

2020 year in review

by

My recap posts on 2019, 2018, 2017, 2015. I don’t remember writing with such consistency (my son was born Jan 2016, I guess that explains the gap?). Skimming these prior posts, I do see I’m repeating myself an awful lot, but it’s still rewarding to see how the same themes (and my attempts at framing them) have evolved over time, and quite gratifying when some predictions I’ve made turn out to be on the money.

As usual, I’ll write about a few topics that I personally found fascinating.

Genshin Impact and the coming “Industrial Age”

At the end of my 2019 post, I called Genshin Impact an aspiring blockbuster, which was not that bold a claim given the viral hype (and controversy) it already enjoyed in China at the time. It has easily surpassed my expectations, even more so in western markets.

I’ve already written a long post about the game. In the aftermath of the game’s explosive launch, much of the Chinese industry chatter was about “industrialization”. I’ve not seen a clear definition of the term in this context (it seems taken for granted), but loosely, the logic is that as consumers demand higher fidelity games (rivaling PC/console AAA in quality) and ever more content, Chinese developers will have to embrace the flywheel of “bigger teams (and more specialization of talent), more sophisticated production pipelines, and more advanced technologies and tools”.

In other words, Genshin Impact is seen as a landmark game, one that has permanently shifted consumer expectations higher, and subsequently started a new industry-wide arms race in China. In my view this is quite overblown – the dominant market leaders Tencent and Netease have for years chased higher budget productions made by “armies of developers” – but Genshin captured the zeitgeist with the audacity of its vision.

What’s next? A lot of UE4 projects, for one thing. To name a few: Tencent Quantum Studios’ Dawn: Awakening is an open-world survival game (with 3rd-person shooter gameplay) made in UE4. Lilith Games recently announced Farlight 84, another UE4 project with a post-apocalyptic theme, Battle Royale PVP gameplay (perhaps amongst other game modes) and mobile-PC cross play. (I don’t know if either of these will take off – their themes lack the easy viral appeal that Genshin Impact had.) Meanwhile, Tencent Timi Studios recently posted job ads for a “AAA-grade” realistic racing simulation, built in UE4 for mobile; and miHoYo has been recruiting UE4 developers as well.

When China meets the world

Chinese game developers have for years studied and learnt from their global industry peers – whether it’s GDC talks, studio visits, academic studies or direct talent acquisition. There still doesn’t seem like a lot of information flowing the other direction – language and culture are big barriers (for Chinese developers to share outwards – the “supply” side), but lack of interest on the “demand” side has been a deterrent as well.

In this aspect, the games industry seems a step behind the broader tech sector, where Silicon Valley now clearly pays a lot of attention to trends in China. When there is more interest, and deeper exchange of knowledge about China’s game development practices, I suspect there will be a good amount of bemusement and shock from the outside.

I’m reminded of an ancient news piece – when the first MacBook Air was announced in 2008 (by Steve Jobs memorably pulling it out of a manila envelope), a group of Japanese engineers did a teardown and expressed surprise at the “wasteful” and “expensive” internal design:

“If I proposed such a design, our company would never approve it,” said one of the engineers. “I can’t find anything that is technically superior. We can make the same computer at a lower cost,” said another.

In hindsight this was obviously missing the forest for the trees – the Japanese experts weren’t necessarily wrong, but their points were irrelevant in the big picture. Game developers should avoid making the same mistake when they examine Chinese game development – the sausage might be made in an ugly and wasteful way based on your perspective, but don’t neglect the end results or their growth trajectory.

Chinese developers have been self-reflective about the gaps. For instance, in this recent interview (in Chinese, but the Google translate is well worth a read) with the head of Timi J3 (the team behind Call of Duty: Mobile), he called out investment in tooling as one area where China still has much to learn:

姚远：… 再就是欧美厂商对工具化的实践比我们强太多。之前和《幽灵行动：荒野》的团队聊，他们说花5年时间做了个编辑器。这个编辑器强到什么程度呢？基本上随便拉一下，所有村庄、道路、人物、动物、植被全都出来。这就是育碧的工业化能力。

葡萄君：国内厂商不一定会做类似的事情。

姚远：是，国内很多项目因为开发周期限制，没时间做工具，但欧美厂商不一样。育碧的编辑部团队会在预研阶段做各种军事、历史的考证，去相关地点采风，完善工具，再慢慢把项目做起来。这个流程非常值得我们学习，如果真的要追求效率，还是得一开始就准备好。

Yao Yuan (head of Timi-J3): …Furthermore, European and American manufacturers’ practice of tooling is much stronger than ours. I talked to the team of “Ghost Recon: Wilderness” before, and they said it took 5 years to make an editor. How strong is this editor? Basically do some drag & drops, and all the villages, roads, people, animals, and vegetation will come out. This is Ubisoft’s industrialization capability.

Interviewer: Domestic manufacturers may not do similar things.

Yao Yuan: Yes, many domestic projects have no time to build tools due to development cycle constraints, but European and American manufacturers are different. Ubisoft’s editorial team will do various military and historical research in the pre-research stage, go to relevant places to collect features, improve tools, and then slowly start the project. This process is very worth learning. If you really want to pursue efficiency, you still have to be prepared from the beginning.

(English via Google translate with light edits.)

Earlier in the interview, Yao made this comment about their production capabilities:

举个例子，我们和一年前相比，同样是两三百人的规模，产能却翻了3~4倍。通过和近300人的外包团队协作，现在一个月能做若干玩法和地图，上百把武器和几十个角色。虽然有些内容总体所需的生产周期比较长，比如一个人物从概念设计到监修要3个月，但生产流程、管线都非常强大和成熟。

For example, compared with a year ago, we have the same size of two or three hundred people, but our production capacity has increased by 3 to 4 times. By cooperating with an outsourcing team of nearly 300 people, we can now make several game modes and maps, hundreds of weapons and dozens of characters in one month. Although some content requires a relatively long production cycle, for example, it takes 3 months for a character to go from conceptual design to IP-stakeholder approval, but the production process and pipeline are very strong and mature.

(English via Google translate with light edits.)

So the picture here is, this team is consistently churning out vast amounts of live-ops content, despite relatively immature tooling (compared to their western peers), and their efficiency is rapidly improving. And they have stayed on top of the organizational challenges of running such a large team. And there is still a lot of productivity upside if they do seriously tackle tooling – that’s the scary part.

5 years of Honor of Kings

Honor of Kings launched in late 2015. SCMP did a profile recently, and their graphs painted the picture succinctly:

Let’s be clear: the “real” lifetime revenue is a lot higher than this $7.8B figure from Sensor Tower, as it does not include China Android revenue (understandably hard to model), and Honor of Kings has very low revenue outside China. Indeed, I would say you can double that figure to $16B and possibly still be low. Coincidentally, $16B is a cool 100B RMB, a nice round figure for half a decade.

Beyond these eye-popping (and speculative) numbers, it’s hard for me to talk about Honor of Kings without doing some soul searching. Professionally I had a ring-side seat to this spectacle – I was a part of the China team at Riot Games, based in Hong Kong in 2015/16. I played the game when it launched, and was 1) amazed by how it recreated some of the high satisfaction moments of PC MOBAs, but 2) also confident that it was not a major threat to League of Legends as the gameplay was still too shallow for core players. My main takeaway was that Riot should absolutely look into making a mobile MOBA as well.

To show my thinking then: in March of 2016, I wrote a post titled “Are mobile games disruptive?“, and the disruptive game I was talking about was Clash Royale (which took up every second of my day when I wasn’t taking care of my newborn):

I believe mobile games have so far followed the [disruption] theory here:

– They have focused on catering to previous non-gamers / casual gamers, and most of the early successes reflected this (Angry Birds, Candy Crush Saga, Flappy Birds)

– These games were simpler to play, and offered less complexity in the gameplay

– These games were generally looked down upon by core gamers

What gets interesting is what happens next. The disruption theory says that from this low market position, the new entrants are able to mount an attack on the establishment thanks to both product evolution (so they catch up in product experience) and their new attributes which the power users (core gamers) previously didn’t care about.

While my memory is fuzzy, I believe I largely stopped playing Honor of Kings for fun after the initial few months. However, by Chinese New Year 2017, it was clear that a disruption was playing out according to Clayton Chistensen’s theory. League of Legends players were being pulled into Honor of Kings en masse: it turns out social ties and bragging rights were more powerful motivations to many (if not most) people than gameplay depth and mastery. But really, the bigger story was how Honor of Kings activated so many non-gamers.

So that’s what it feels like to be disrupted.

I’ve often described Honor of Kings as an attack from below – if you think of the hierarchy of MOBA players as a pyramid, with the very pinnacle being esports players, Honor of Kings successfully activated the bottom tier first. There was little organic endorsement or word-of-mouth from the establishment influencers. (The game did try to piggyback on the popularity of League‘s esports celebrities, with ambush marketing like getting Faker to do a livestream.)

League of Legends: Wild Rift, in contrast, will be an attack from above. I played a modest role in getting this project off the ground (and I’ll shamelessly overstate it on my Linkedin page), so it’s something quite close and dear to me. The existential question for Wild Rift has always been: is there any chance against Honor of Kings?

Sentiments aside, I think the answer is yes, even in China. The League “establishment” that shunned Honor of Kings have been dying to play a League mobile MOBA, and perhaps this echelon of esports pros and streamers can create a big enough beachhead. And there is still a brand premium in my opinion, though that picture is nuanced as Honor of Kings has leaned into Chinese culture – in a way, it’s a bit like Apple versus Huawei in China. At the end of the day, players across all tiers of the pyramid will try Wild Rift – the question is can Riot get them to stay.

As an anecdote, I’ve been lurking in a wechat group of League influencers who have overcome formidable obstacles to play the game on Asian servers. Some of them are already organizing pro teams and recruiting players at the top of the ladder. Their enthusiasm is infectious, and gives me cautious hope about the next chapter in this saga.

Lastly, one other personal reflection from these 5 years is that, even with the personal experience of how Honor of Kings disrupted League of Legends, I was able to repeat the mistake when it came to thinking about PUBG Mobile. I had seen an early build before its launch, and I was impressed. But my gut again told me that it would not satisfy PUBG players, and its input complexity would be overwhelming for casual players. I was definitively wrong on both – I guess that speaks to how strong one’s biases can be.

(Optional extra reading – this piece in Gamesindustry.biz shows Timi leadership’s reflections about Honor of Kings and their views on industry trends.)

China’s dynamism

Taking a step back from games for a moment. I came across this year in review letter by Dan Wang, who is a tech analyst based in Beijing. It is thought-provoking and beautifully written, and honestly I envy his prose. (Seriously, you should stop here and go read that letter instead.)

This small bit particularly resonated with me:

This year made me believe that China is the country with the most can-do spirit in the world. Every segment of society mobilized to contain the pandemic. One manufacturer expressed astonishment to me at how slowly western counterparts moved. US companies had to ask whether making masks aligned with the company’s core competence. Chinese companies simply decided that making money is their core competence, and therefore they should be making masks.

This “can-do spirit” is Chinese game developers’ biggest (and perhaps most overlooked) strength – it’s the rising tide that lifts all boats. While studios elsewhere debated first principles about whether MOBA / FPS were viable on mobile, Chinese studios simply hacked away at it. When the Chinese government tightened the publishing license process, companies rapidly pivoted to overseas expansion. And with this “industrialization” wave, Chinese developers are again just diving head-first.

Having recently lived in the US for almost a decade, I feel the stark contrast. As I was writing this, a mob of Trump supporters stormed the Capitol building. (It may well have been a scene from a dystopian video game like The Division 2.) There are bitter divides and deep social justice issues. It feels trivial, and perhaps disrespectful even, to be discussing game development against such a backdrop – but I feel the need to argue that Americans must find a way to rekindle a similar can-do spirit, and just build stuff. Build institutions, social welfare, infrastructure, housing, startups… Whatever it is that motivates you, which hopefully for some would be video games.

This is obviously a huge topic that was discussed in the US in 2020, when Marc Andreessen wrote his “It’s time to build” post. I didn’t follow the rest of the discourse closely, but I enjoyed reading this essay “On cultures that build.” I’m not well equipped to really add more to the conversation, but I will say this: China faces huge (if not bigger) societal challenges as well, but part of the dynamism is rooted in people’s belief that they can (and must) improve their livelihood via hard work. They have the lived experience of the dizzying growth – the building of everything – of the past 40 years. For Americans, more cultural exchange and economic ties with China – not less – may have a nice side-benefit in combating the complacency.

M1 Macs

In my 2018 post, I briefly speculated on “the beginning of the end of the PC (x86) platform”. With the arrival of the M1 ARM-based Macs (finally), I’d like to raise my bet.

I haven’t used a M1 Mac yet, but from everything I’ve read so far, it sounds like a generational leap in objective performance as well as subjective user experience. Now the question becomes, is this strategy and capability unique to Apple, or will others attempt to follow suit? I’d argue yes, if not simply because of the gravitational force of the mobile ecosystem. In particular, if Nvidia’s acquisition of Arm is cleared, Nvidia will be in an enviable position to attempt some big integration plays here.

If there is such a foundational migration on the Windows side of PCs, one clear worry is app inter-operability during the transition. It’s hard to imagine the Windows side offering as smooth a transition as Apple is doing with Rosetta 2, thanks to the much more fragmented hardware stack. And games as a special category of applications may suffer the worst of any transition. Again – a gaming-focused company such as Nvidia could be particularly motivated to navigate a path through this.

There is some irony, to me, if at the end of this, it’s consoles (due to their long generational cycles, and current commitment to backwards compatibility) that become the laggards that stick with x86.

Personal stuff

2020 was the first full year I’ve lived in mainland China for over a decade, and the first time I’ve lived in Shanghai. My timing was pretty good, in retrospect. When the strict lockdown started in late January, we thought we had the worst luck, but by May, life was mostly back to normal (even masks were mostly gone, except where mandated such as in public transit).

In the summer, many friends went vacation traveling again; we didn’t as we had very young children. But we couldn’t resist booking a short vacation trip for December. Alas, the weather in Xishuangbanna wasn’t warm enough to take advantage of the private pool we had in our villa, but it was still a pleasant trip.

Shanghai is a very livable city. I say that as a proud Beijinger. The summer is still too hot and humid for my liking, but the city has a good balance of culture (with a dash of western influence), urban planning, and pace of life. Beijing feels too bureaucratic, and it takes too long to get anywhere. Shenzhen feels too rushed, and the hot humidity is just as bad as nearby Hong Kong. Los Angeles – I love the climate, and the parks, but I don’t miss the driving.

The pandemic also gave me some new perspectives about effective governance. The US response has been appalling to see from afar. I wonder how much of it is uniquely the failings of the Trump administration, and how much is reflective of the general state of decay and complacency in US institutions. To be clear, I’m not looking at this from a lens of US versus China as superpowers, or other sorts of macro-economic debate. I’m much more concerned with the micro-economic life decisions we make as a family – where we should spend our precious years together, and can offer us the best mix of professional fulfillment, income, education, and life experiences. And for me the US fell a lot in the rankings this past year.

Investing: some years ago, there was a popular startup catchphrase about seizing the big trends, coined by Xiaomi founder Lei Jun: “even a pig can fly if it is in the middle of a whirlwind.” (Jack Ma, who’s living through some interesting times himself recently, apparently had a witty response: “when the wind stops blowing, it’s the pigs who fall to their deaths.”) The stock markets certainly made me feel like a pig facing a hurricane, torn between FOMO and having a nasty fall. It was quite surreal to see the market movements in contrast to pandemic life.

Remote work was a much discussed concept, and a collectively forced experiment. In my case, with 2 young kids running around the apartment, working from home simply did not work. When my older son’s kindergarten re-opened, it was marginally better. I’d probably need a private office away from the office to make remote work viable.

To wrap up with the games I played in 2020. I played various mobile games due to professional interest, but the one that stuck with me, surprisingly, is Merge Mansion. (Disclosure: my current employer, Supercell, invested in the developers.) I’m not a puzzle game player, and this is a game that’s still very early in development with a lot of rough edges. But it became the perfect time-killer game, and I’ve averaged 20 minutes of play every day for several months now.

On PC/console, I played Hades more during early access in 2019 than I did in 2020, but I should go back and play it some more to experience the complete game. During the depths of the spring lockdown, I occupied myself with Ghost Recon Wildlands and The Division 2. Fall Guys probably brought the most joy and laughter, and it was eye-opening to see how much it resonated with my 4 year-old. Later in the year, Ghost of Tsushima was an easy crowd-pleaser, which I spent more time on than I should have.

The game that resonated the most with me though, without a doubt was The Last of Us Part II. Its harrowing discussion of trauma, empathy and perspective-taking was particularly fitting for these times we live in. And the toxic fandom around the game felt like an inadvertent meta commentary that echoed the game’s core themes. Months after my playthrough, I still think about my experience with the characters. Perhaps it’s time to pop the disc into that new PS5…

Genshin Impact (2020)

by

Genshin Impact has had a great global launch – indeed, I struggled to come up with some good comps on Sensor Tower as it has really stormed out of the gates. In terms of launch revenue, I actually couldn’t recall a better game than Pokemon Go. See below launch-aligned revenue graph:

If we just look at China, where it’s easier to compare apples-apples (at least for iOS only, since Sensor Tower doesn’t track China Android), we have an early 3-way tie across AFK Arena, Brawl Stars and Genshin Impact – as some of the best launches of 2020:

As a side note – AFK Arena’s $60M launch month on iOS in China (in January before Chinese New Year), is the best new mobile game domestic launch this year as far as I can tell. (The usual caveats about lack of China Android estimates apply…)

But, Genshin is not just a mobile game – in China, it launched on PC first on Sep 15, a full two weeks of early access. I still find this an unusual choice – the most affluent, hardcore gamers rushed in on day 11, but the negative reviews came in almost instantly. Snobby PC/console gamers mocked the game’s lack of polish and lower graphics fidelity compared to premium AAA titles; whereas mobile gamers looking for a progression head-start via PC immediately raged at the poor gacha loot table. They didn’t hold back their emotions on Taptap:

The poor reviews didn’t seem to impact sales much (and the Taptap crowd is hard to please). It’s also interesting to contrast this reception with the western audience reception – there seems to be a lot of voices expressing surprise that the game offers so much content and is free.

In any case – it’s very early days yet as we are barely two weeks into the official launch, but all things considered it’s a great start for Genshin Impact. It will be fascinating to see how the game trends over the next few months.

Core gameplay

The game is quite well-reviewed on Metacritic (though a small sample size), and deservedly so. The scope of the open world, the combat system, and the character roster (and their visual presentation) are impressive.

The game is most fun (I’m currently Adventure Rank 27) when you are doing the dependable open-world loop: you start out with a particular objective (maybe a quest, or just a point-of-interest you spotted in the horizon), and along the way you get side-tracked by numerous side content. There is a lot of side content: collectibles, side quests / daily missions, environmental puzzles, loot chests that respawn periodically… You get the picture.

Genshin Impact certainly takes a lot of inspiration from Zelda: Breath of the Wild‘s open-world formula, but ventures far enough to end up in its own place. The biggest departure is combat: Genshin referenced the elemental interactions from BOTW, and converted it into a catchy combo system that still feels intuitive enough. It can get repetitive, but it’s still satisfying to set up an explosive fire-lightning combo (for example), and there are hints in the equipment system (I haven’t gotten far enough yet) of intriguing build possibilities.

Personally, I’m not too concerned with the graphics fidelity on PC – I’ll always take smooth frame-rate over graphics quality, and this is where Genshin does not fully deliver. On mobile, I can’t run reliably run 60fps on an iPhone XS Max (even when I toggle everything down); and the game’s min-spec on iOS is iPhone 8 Plus, which suggests dev challenges with performance optimization (in comparison – PUBG Mobile‘s min-spec is iPhone 6s).

I do want to talk a bit about gameplay feel and polish, where Genshin is behind its PC/console peers in some areas. As a player, I found myself often wrestling with the game’s character, camera and controls (the “3Cs”), on mobile (more egregious) as well as on PC. Some examples of jank:

  • Ranged aiming feels finicky in general, even on PC with mouse; sensitivity settings are too coarse in my opinion. And for mobile / PS4 there should have been some aim-assist support (even if they can be turned off).
  • The camera has janky movement at times – as a tiny example, when you fall and roll forward, the camera takes too long to recover from facing downwards, and requires a manual adjustment2.
  • Some characters have attacks that dash through the enemy, which would require you to rotate the camera 180 degrees to see the enemy again. This is in my opinion dangerous design space for a 3rd-person mobile game with virtual joysticks.
  • Similarly, the Traveler’s ability that creates a giant rock is also dangerous design space in combat: this is a climbable rock that can cause unintended player interactions; and it often displaces enemies to the top of the rock, where they don’t seem to know how to get down (and thus severely disrupting combat pacing). The fact that the ability is aim-able is also stress-inducing on mobile.
  • Enemies who are displaced from the combat area (for example, falling off a cliff) get reset (with full health), which is often frustrating.

I’ve also found the game’s boss fights tend to have more jank and annoyances. For example, the first major boss fight below:

There’s a bunch of things here that irritated me (and yes, my skills are probably below-average…):

  • The level requirement stated upfront was a bit of a misdirection, as you are given a trial character with their own level, and the whole fight is primarily designed for that character.
  • Unskippable cut-scenes, which is a pretty big no-no if players have a chance of having to replay this fight several times (which I did…).
  • The flight combat sequence starting around 1:30 has readability issues, with the backdrop that is quite static and the boss always center-screen – the first 2 times I played this fight, I didn’t understand I could actually fly towards the pick-ups (I tried maneuvering and felt I couldn’t change direction).
  • The final phase has a custom camera angle (side-scrolling), which is a bit jarring as most of the game you are not driving your character’s movement primarily with left/right input. Combined with the ledges, this created a level where I fell off quite a few times – while not lethal, it was very annoying for combat pacing.

Over time, the player learns to work around these problems, but there’s clearly a gap in terms of the developer’s capabilities, sensibilities, and/or priorities – these are the 20% issues that can take 80% of the time to solve to get that AAA polish.

Progression and monetization

When it comes to progression and monetization, Genshin at a high level shares a lot of the generic Chinese mobile RPG template (of which AFK Arena, mentioned earlier, is the current best-in-class example).

The basic formula of such games is a deep progression system (with layer upon layer of different stats to chase), with stringent upgrade gates interlaced with periods of relatively smooth leveling. The stringent upgrade gates provide heavy incentive to do the daily/weekly grind for resources (certain key resources can only be farmed on specific days of week). And the sheer amount of dimensions to progress (amount of characters + depth of each character) converts into aggressive monetization design, where ultimately cash can be turned into characters (through gacha), upgrade resources (directly purchased), stamina for grinding resources and so on.

This is the rinse-and-repeat formula that hundreds of Chinese games have used – Soul Hunters, Naruto, Honkai Impact 3rd, Onmyoji, Arknights, AFK Arena, to name just a few of the biggest over the past decade.

The marriage of such a formula to the open-world gameplay in Genshin is at first jarring – the early leveling experience of a player who immediately spends several hundred dollars on gacha is going to be very different (and arguably for the worse – as all sense of early pacing is out the window) from a non-spender. I know such a gamer – he is trained to plunking down a few “648”s (by convention the most expensive SKU in the cash shop, roughly $100) any time he starts a Chinese mobile game – and were it not for social peer pressure, he would have churned several times by now (despite spending almost $1k already…).

After 20 hours in, when I’ve largely picked up the various complex systems and are somewhat invested in some characters, the disconnect starts to go away. It becomes very clear that access to a lot of fun gameplay is gated behind monetization – the 5-star characters that everyone is enamored with are not going to come easily (and even if you unlocked them – you need so many duplicate copies to fully level up their powers). You can still have a good time – but you will be missing out on a lot of gameplay possibilities.

Regardless of whether you monetize or not (or how much), the grind is still somewhat egalitarian3. That’s the other funny part of this RPG formula – it demands both money and time.

Usability pains

One part I’d like to complain loudly about is the game’s UI/UX and usability issues – not only because I suffered lots of irritations here, but also I feel there’s a hard-to-measure (but perhaps material) impact on the game’s engagement. Genshin is already overloaded with design complexity (as is typical with Chinese RPGs), and the usability issues amplify the cognitive load.

I already discussed some issues in the gameplay section above, but here are a couple of examples specifically about the UI. This part of the discussion is quite tactical.

First, I found it baffling that the map and quests UI were not integrated. They are activated via separate buttons on HUD, and don’t link to each other.

On the world map, you actually can’t see available quests (with the exception of the 4 daily quests). You have to “track” a quest in the Quests UI for it show up in the map. This would be much less of a problem if the Quests UI were available as a pop-up / side-bar in the Map UI. But currently, you have to jump back and forth across 3 UI screens to complete a simple action of “select a quest and find nearest teleport point”.

The Map UI is also lacking functionality in some other basic areas. For instance, you can place custom nodes on the map to keep track of points-of-interest (players use it to tally the important collectibles, for example). But you can’t “navigate” to a custom node, which feels like a pretty useful interaction.

For an open-world game, the Map feature should be something that really emphasizes ease of use – help you make decisions about what to do next, and get out of the way as fast as possible, so you stay immersed in the world. But in Genshin it currently is subpar compared to most contemporary open-world games (e.g. Ghost of Tsushima for a very recent example). I honestly can’t remember the last time I was so confused at the map feature of a game in this genre.

The second UI issue I’d like to talk about is the various screens related to character and team management – I did a quick navigation flow to help illustrate my point:

Broadly speaking there are 2 distinct needs: roster selection and character management (leveling, gear management, skills upgrades etc.). So having separate menu entry points for “Team setup” vs “Character” does make sense, even though they could be combined in an alternative flow. But I find the “Characters” vs. “Character” flows puzzling – they are largely made up of same/similar screens (just with reversed navigation), but there are some weird UI inconsistencies:

  • The ordering of characters is inexplicably different – in “Characters”, sorted by level by default; while in “Character”, the active roster is shown first, then the rest by levels.
  • The “Characters” screen and the “Character selection” pop-up also make for an interesting comparison: these two screens have largely the same layout and high level purpose (view list of characters, select one to navigate to), but have numerous small UI differences (list of 4 in a row vs 3 in a row; blank space vs. right bar of character details; “X” vs “back” navigation buttons…).

I might sound nitpicky here, but these small inconsistencies add up to unnecessary cognitive load (forcing the player to actively think), which begs the question of why do these 2 largely redundant flows exist?

To me, this is a reflection of lack of holistic game polish, and perhaps related to the production culture, which I’ll do some extrapolation and speculation below.

Closing thoughts

I wrote a post last year “Assessing China’s game development capabilities.” I think Genshin Impact is a continuation of the trends I discussed there, but it should also be proudly celebrated by Chinese developers as a product breakthrough in original IP on the global stage.

It succeeds in part due to its sheer audacity in vision and content scope – original IP, open-world, cross-platform (with mobile as the core), and years of live-ops content runway. miHoYo is well positioned to tackle this, having honed its IP creation skills in the Honkai franchise, and with good access to China’s “industrial scale” mobile production capabilities.

I do think it’s a “quantity over quality” approach, as I feel the game clearly prioritized volume of content (and future expansibility) over polishing details. But again, this is the proven formula for Chinese devs – getting the fundamentals barely good enough, then production scaling like crazy. Whereas western devs tend to be wary of the content treadmill (e.g. WoW’s expansion cycles), Chinese devs seem unfazed about embracing it. They don’t enjoy it – but they are more willing to grind it, and for the successful games, the economics seem to pay out well. While we don’t know for sure currently, I expect Genshin to have a stream of major updates planned already (the next one is probably close to completion by now), and the update cadence may again surprise the global audience4.

(UPDATE: the update schedule was actually announced, and it looks like initial reactions from the hardcore community globally was disappointed. See this reddit thread.)

Where Chinese devs should/need to grow further, in my opinion, is the discipline, thoughtfulness, (and frankly) prioritization of better UX5. This is not easy to do – in my personal experience, I’ve found Chinese devs’ strong production-scaling tendencies and general haste to be big barriers for holistic game polish. But as the market, and gamers, get more demanding, I expect higher emphasis here in future, which may force shifts in development models.

Before I forget, a couple of things to highlight that are part of the game’s breakthrough:

  • First is localization: I played the English version for a while before switching back to Chinese, and I thought the English localization had very high production quality, made by a veteran team of writers and VO cast.
  • Second is the music: I love it. It’s clear no expense was spared in music production, and the soundtrack is lovely. However, sometimes the music transition triggers seem a bit ungraceful (again, perhaps one of these polish cases).

At a macro level, I think it’s safe to speculate that Chinese devs are going to have even bigger ambitions post-Genshin, despite significant external headwinds (China-domestic regulations, state of economy, global geopolitics). The NA/EU market is the last frontier geographically. It will also be interesting to contrast East/West approaches to cross-platform: Chinese devs will be grounded in mobile-first (otherwise they leave a lot of money on the table with the China-domestic market), whereas western devs will tend to prioritize PC/console à la Fortnite.

  1. As a sign of Chinese gamers’ constant vying for status (which is a huge part of their motivation for gaming), gamers actually complained loudly that miHoYo opened the servers a couple of hours ahead of schedule – they felt betrayed about missing out on snatching a sexy low-digit UID. ↩
  2. I don’t remember specifically, but this was probably one of the ear flicks that caused me to give up on using a controller on PC. ↩
  3. If you spend for stamina, you earn the privilege to grind more. ↩
  4. While simultaneously being criticized by Chinese players for being too little too slow – such is the diminishing returns. ↩
  5. by UX I’d like to refer to this particular framework here. ↩

Epic vs. App Stores

by

The ongoing fight between Epic and Apple / Google is one of the biggest tech stories of the year. The situation is very fluid, with a lot of developments since last week, and a ticking time bomb by end of August.

Not surprisingly, there’s been a lot of “takes”, most of which is candidly not too useful, and a small amount that have covered the situation from insightful angles. Instead of regurgitating these insights, I thought I’d just list a few here (most of these are usual suspects if you surf a lot of tech punditry):

The Chinese Android app stores example

I think it’s rather futile to debate the abstract merits of “open” vs “closed,” which at the ideological level is the heart of this fight. Tim Sweeney has been very consistent over the years – his public criticism of UWP is spiritually similar to his stance against Apple / Google, and I believe it’s stemming from not merely a business interest calculation (though he is often accused of such), but a genuine belief in “open.” 1

Instead, I think it’s more useful to discuss the Chinese Android app distribution landscape, as a real example of why Epic’s desired state (open up iOS to 3rd party stores and alternative payments) may not be good for consumers. (The linked Chinese post above is a great read on this, below is my brief summary of the same topic.)

When Google abruptly exited China in 2010 (and along with it, the Google Play store), there was a gold rush to fill in the vacuum left in the Android ecosystem. At a 30,000 ft level, a series of things happened:

  • In the beginning there was a flood of independent stores, with notable ones like Wandoujia (funded by ex-Google China head Kaifu Lee’s Innovation Works) and 91 Assistant.2
  • In a landmark deal at the time, Baidu acquired 91 Wireless (which owned the 91 store) for almost $1.9B in 2013.
  • As of 2013 Tencent also had an Android app store MyApp. After Tencent leveraged WeChat’s popularity to promote MyApp (“if you wanted the latest version of WeChat, go to MyApp”), MyApp gradually became one the most popular stores.
  • In 2014, prominent Chinese Android handset brands (with the exception of Xiaomi) formed a coalition called the “Mobile Hardware Alliance”. A major goal of this coalition was to exert influence in the distribution of games (which was recognized as the key cash-cow in app stores) in the Chinese Android ecosystem.

The current state of stores, at a high level, is this:

  • All the Chinese Android brands have their own stores, and because of the coalition, these stores have significant weight.
  • Tencent MyApp is the biggest non-OEM owned store.
  • The once prominent independent Android stores (without backing of OEM or a major social app like Tencent’s QQ/WeChat) are greatly declined in presence.
  • Collectively there are still dozens of stores.

How about the economics – let’s talk about that 50%?

  • There isn’t a unified rate – everything is negotiated. But indeed, if you are a game publisher not Tencent or Netease, the 50% store cut is the common term you will get.
  • Strictly speaking, this isn’t an “Apple-apple” comparison, as these Chinese Android stores call this “joint operations” of games where in theory they are providing more value-add (funneling more traffic etc.).
  • The prevailing rate for Tencent and Netease have been pushed down to 30%. (And of course Tencent keeps 100% in its own MyApp store.)

To summarize, the Chinese Android app store landscape is very much objectively a worse state than the Apple / Google monopoly Epic is complaining about:

  • Consumers have a confusing user-experience (overwhelming amount of store choices, fraud / security / malware concerns, inconsistent UX of the same app across different stores).
  • Developers are typically giving up a much higher share of revenue.
  • Developers have a lot more development costs / headaches (support dozens of app stores, SDKs, builds).

To be clear, it’s not a certainty that we will see a similar end-state if the Apple / Google “app distribution market” and “payment market” is opened up by regulation. (For one thing, the Hardware Alliance thing is clearly suspect to anti-trust scrutiny.) But it is clearly a possibility with strong factual support.

Problems that Apple should address

Having argued why “the grass isn’t greener” on the other side that Epic desires, let’s briefly talk about issues that Apple should tackle. This part is focused on gaming specifically.

For the 30% rate, I do believe (and clearly I’m biased with a vested interest here…) that this should be pushed lower with how the ecosystem has grown and evolved, even if purely arguing from an economies of scale perspective. Ultimately though, economics are a reflection of “who owns the customer”, so Valve’s model of volume-based tiers (starts at 30%, drops to 20% for sales above $50M) isn’t a bad reference. (This is also the common logic in retailer / wholesaler agreements.)

(Alternatively, Apple can continue to make confidential deals with the biggest partners, offering rev share discounts on a case-by-case basis.)

Apple also should update its strategy (and thus policies) regarding emerging services like cloud gaming. The rejection of Microsoft xCloud on iOS feels short-sighted, and untenable in the long-run if cloud gaming does take off. (It’s also a bit silly that at the same time thousands of HTML5 games are available directly within WeChat, which seems like a much bigger violation; arguably xCloud is offering much better games that would enrich the user-experience of iOS gamers.)

To end on a light-hearted note. Every time I write about Apple and mobile gaming, I will bring up my dream for an Apple-designed controller peripheral. I don’t think that will ever happen, but one can dream…

  1. Conversely, Apple, like Nintendo, like Disney, have been decades-long champions of the “closed” side of the debate. Just for transparency, at at the abstract level I lean closer to this camp, because I idolize seamless user experiences (which are typically easier to realize in a “closed” ecosystem). ↩
  2. As a sign of the times, a popular feature-set back then was a PC client that was a storefront and also a manager for the download and installation to the phone, similar to using iTunes to manage iPhone apps. ↩

The Last of Us Part II

by

Some notes on the controversial blockbuster sequel.

***Spoilers warning*** I’ll be liberally discussing all aspects of the game, so please do not read further if you want to avoid spoilers.

First a quick note about my play experience – I got the disc a week after the launch date, by which time there was already a massive storm online. Despite my efforts, I did have some key story points spoilt, so going in I was somewhat prepared mentally.

I could only afford to play a couple of hours a day, a bit more on weekends, so it took me a good two weeks to get through. My final session was a 5-hour binge on a weekday night, ending at 3/4am (that day at work wasn’t very productive).

I started the game late, was armed with some knowledge of the drama, and had the play sessions paced out – I felt all these were positive factors that helped me enjoy the game. To me this game is a flawed modern masterpiece, that deserves to be remembered as one of the most ambitious narrative games of the decade.

Narrative structure & theme

TLOU2 follows a very rigorous structure, which I’d break down as follows:

TLOU2 acts

The structure showcases the game’s risk-taking ambitions. In my view there are 2 primary risks taken: 1) the story decision to kill off the predecessor’s protagonist, Joel, as the inciting incident; 2) in a surprise switch at the half-way point, forcing players to play as Joel’s killer, Abby – and revealing that this is a game with dual playable protagonists on the opposite ends of a revenge plot.

In retrospect, to me the game’s real theme is about how people deal with trauma, via a story of hate-driven vengeance. The game delivers a traumatic event to players (Joel’s death), then forces players to go through the stages of grief (both in the game as Ellie, and in real life with their own feelings towards Joel). The perspective change to Abby is an experience in forced empathy, which to me is a secondary theme.

The perspective shift is not new as a literary device – Game of Thrones clearly leveraged this to great effect with memorable characters like Jaime and Tyrion Lannister. But this feels like the most ambitious example in a video game I’ve played, and the effects were fascinating. In the climatic fight between Ellie and Abby, like so many players, I did not want to hit the attack button. But, just like the predecessor’s climatic surgery room scene (which you revisit so many times in this game), the game does not offer you a choice. Thankfully, the game ends the fight mercifully.

I’m not going to go deeper on the narrative and theme – that would be a huge endeavor, and many people have already offered lots of great content. I’ll link here one video I particularly enjoyed.

Game loop & “level” pacing

The game’s narrative beats (the bullet points structure above) serve the long-term and mid-term motivations. At times this can feel ham-fisted: I felt Ellie’s 3 days in Seattle was a bit repetitive in its use of “go to point X to find the next clue about Abby’s whereabouts”. Anyhow, if we zoom in 1-2 levels further, we get to the layers of the “core loop” below:

  • Long-term goal, e.g. find Abby
    • Mid-term goal, e.g. go to Hospital
      • A series of “levels”, or set-pieces

My loose definition of a “level” here is a 5-20 minute section of gameplay made up of elements from the following:

  • Combat, stealth or non-stealth (sometimes forced non-stealth)
  • “Walk and talk”, the most basic way to deliver the story
  • Exploration, which is a lot of ambient storytelling (reading notes etc.)
  • Scavenging and crafting
  • Environment puzzles (some light platforming gameplay)
  • On-rails set pieces, e.g. car chases
  • Mini-games, like guitar simulator
  • Cut-scenes

From a player perspective, I wouldn’t say there’s any crazy systemic design innovations – these are the proven gameplay elements of Naughty Dog action adventures. The craft comes from the thoughtful sequencing & arrangement to create great pacing, and the insane polish (and the vast technical investments to deliver that polish).

What I thought the game did particularly well for pacing, was keeping players on their toes with surprises. Examples:

  • Have you grabbed by an enemy (either transitioning into combat or a cut-scene) as you go through a level transition (“squeeze through this space”, exit this door)
  • Give you a clear environment puzzle, then as you are moving towards the solution, have the floor collapse under you into a mini-boss fight
  • Give you a workbench (for equipment upgrades), as you start reviewing your upgrade choices, have enemies rush you and grab you from the bench

The game does these surprises very sparingly (like only once) – but they are very effective at making you second-guess yourself and stay alert. Is there going to be a jump-scare at that next workbench? (No.)

Also, they are done in a fair way – in the workbench example, these enemies didn’t spawn out of nowhere; they came out of a locked room in the apartment. So if you had planted some mines in front of that door before you engaged the workbench, you would have had the jump on them instead. This is the level of detail and polish that surpasses player expectations.

Transitions – in my view, any time where you go through a transition where you cannot backtrack (e.g. going through a door/gate and blocking it behind you, going down a sliding slope, jumping down a vertical), that’s usually a sign of a level transition, which serves pacing and possibly technical goals. There are also occasionally hard transitions after cut-scenes (teleporting you to a location), some of which I felt created dissonance (after having you struggle mightily for a few hours to get to a place, it seemed trivially easy how you got to another location).

There are some issues for me with the basic gameplay formula. Resource hoarding is a pretty big problem (at least at moderate difficulty). Thematically as a post-apocalyptic survival adventure, the game encourages players to engage in stealth through tight resource constraints. This is in conflict with utilizing the fun combat skills that players unlock. (The player could tweak the difficulty settings very granularly, for example increase the environmental resource amount to encourage more open combat.) To be clear, I actually agree with the game’s trade-off here: the combat is less fun, but thematically more immersive. But it’s one element why some players find this game un-fun to play.

Another problem created by the scavenging gameplay and ambient stortytelling is backtracking. That is, after clearing an area, combing back through it to open every last drawer, and trying to find every ambient story point / collectible. This is again in conflict with the desired pacing. And it also creates narrative dissonance – “I gotta hurry to rescue my friends… but after going through every drawer.” Even in levels/sequences where there was clear urgency, I couldn’t help but think – hey, maybe there’s some rare collectible here, I should take my time.

Lastly, the concessions in the buddy AI was at times immersion-breaking. As a Naughty Dog convention, there are many parts of the game where you have an AI companion. This serves important story goals (after all, without a buddy, it’s hard to have “walk and talk” sequences), and they can assist in puzzle and combat gameplay as well. But in stealth gameplay, they still land in uncanny valley too often – they can sneak around, and will look for positions of cover, but it feels like overall they are still treated as invisible to enemies. I’m not 100% sure of this – there was one occurrence where it felt like the companion was detected; but on the hand there’s probably a dozen occurrences where the companion should have been spotted, but was completely ignored (sometimes comically).

Problematic game length

This was the biggest issue I had with the game. If we take each bullet point in the 4-act structure above as a “chapter”, and each chapter roughly takes up 2-3 hours of game time, then we get to a 20+ hour game length. In my own experience, I got to the end of Act 2 cliffhanger after roughly 18 hours, and ended the game after about 32 hours. For a linear action adventure, it’s both an astonishing feat and an excessive over-indulgence.

I feel it’s the product of compromises – it was set that the game would have dual protagonists, and each protagonist’s arc demanded a experience that couldn’t be too compressed. But the end result is a journey that is both too long and still too rushed. There wasn’t space to flesh out the numerous side characters, and I’d loved to see more of the Seraphites’ story, for example.

I can’t help but think, what if this game was broken into 2 parts, and released episodically? This is most probably a terrible idea, with lots of risky questions – how will the episodes be priced, how far apart would the releases be? How many players would purchase the first episode but not the second? But to me it would seem to be a better match with the game’s ambitions, and could perhaps help position expectations better.

Alternatively – what if the “chapters” were unlocked at an announced schedule? Like an episode a week (more practically, maybe one chapter every couple of days)? There might be something here, if a narrative-focused game’s content release factored in the social media cycle – e.g. the weekly reddit discussion/reactions of the latest Westworld episode, and the community activity leading up to the next episode. Again, probably a terrible idea still…

Player expectations / toxic fandom

I feel we also have to talk about the massive community controversy since the game’s release. In hindsight, the marketing misdirection was probably too clever and came across toying with players’ emotions. And the overly-strict spoiler guidelines to reviewers was also a major lost opportunity to align player expectations. For example, I don’t think there would have been a significant downside for reviewers to discuss the dual protagonists setup – yes, it would have been less surprising in that moment, but the forced experience in empathy would still hold (and players would be less distracted wondering how long the Abby section would last).

I think I can empathize with much of the community angst, especially the most fervent fans who dived in on launch day and were shocked. That moment of shock, and initial grief, became a rallying call online, and took on a momentum of its own. In contrast, professional reviewers under embargo had to process that moment in isolation, and were obligated (professionally) to finish the game and reflect on the whole experience. This is perhaps one factor contributing to the gulf of opinion between professional reviewers and players.

However, this raw emotion of anger / denial is in no ways justification for the massive abuse (racism, sexism, homophobia, transphobia and various other forms of prejudice / bigotry) hurled at the makers of the game. (Unfortunately, this is all too common these days – The Last Jedi and the 2016 Ghostbusters come to mind.) The sense of entitlement is out of whack. The industry, and the fans, need to reflect on this.

And I guess there’s some meta irony that a game about hatred (it’s futility and overcoming it) is the subject of so much futile hatred. It is after all, a video game, a work of fiction. Perhaps this was the 5D chess that Naughty Dog was playing all along. But at the end of day, as a developer, and as a player, I hope that we can see more games take risks like TLOU2, and I hope the controversy doesn’t discourage game-makers.

设计师的执念

by

把最近读到、听到的一些东西串起来写篇感想。

最近在听的一档游戏播客，接连两期都是很赞的访谈。先听的一期是访谈Control的Game Director Mikael Kasurinen。Game Director，我想通常翻译为游戏总监，不过我认为游戏导演在这里更为贴切——Kasurinen在被问到职责时，大意说“我是捍卫游戏愿景的人”。

Control是去年相当出彩的游戏，我写过英文点评。Kasurinen整个访谈最令我回味的是，他提到在项目一开始，他就提出了游戏里一切物件皆可被玩家操纵的核心概念，这给技术团队带来了大量的挑战。我想，当时团队大概率是有反对声音或者质疑的，技术同事很可能会问，“真的需要所有场景物件都可被抓取吗？为什么不能是一部分定义好的物件？这个玩法到底是怎样的？”

的确，这里很容易被扣上“策划需求不明确”的帽子，尤其是这样一个似乎找不到参考玩法的主意（并且，也只是玩家若干技能中的一条）。对这样一个需求兴师动众，似乎性价比极低。然而Remedy的这个团队就是把它原原本本地做了出来，把连带的各种细节雕琢到了极致。

我不知道这个团队的氛围是怎样的：这个需求是导演以强大的意念强压出来的，还是团队从一开始就有共鸣去给自己挖这样一个坑。仅从结果看，他们显现出了我标题所谓的“设计师的执念”。

听的第二期节目是访谈Valve的设计师Robin Walker，他参与了VR游戏新标杆Half-Life: Alyx的设计。Walker有几段分享很有见地。

他提到了团队为了找到VR独有玩法的一个研究过程：他们早期做的一件事，就是把当年Half-Life 2的资源在VR场景里加载出来，然后邀请玩家试玩，他们去观察玩家的行为。他们注意到，玩家会走到敌人很近的地方，去全方位近距离观察角色模型——这是在非VR中玩家少有的猎奇行为。于是，团队就试图以此作为启发去打造一个体验，结果是一个玩家不得不近距离接触的boss。他说这个boss战，在很长的时间里面都是有严重问题的，他们也差点放弃了。似乎又是一处“执念”的例子。

他还有一个很实在的项目观点，那就是项目做到了前中期，其实这个阶段你开发的内容越多，离终点就越远。（开辟的战线越多，打磨的工作量就越大。）

Walker还说了这样一句：novel game design is a search problem. 新奇的玩法设计是一个搜索问题。也就是说，新鲜玩法是要通过大量尝试搜寻出来的。他说这是为什么Valve坚持开放的生态：有更多的modder和独立团队去创作，就有了更多的实验尝试，也就离穷举出下一个新鲜玩法更近了一步。（这的确是Valve的游戏创新理念——他们绝大多数的项目，都是把社区里的团队转正而来的。）

他这个讲法，也让我想到了另外一个观点。我前阵子终于读完了Stephen King那本讲写作的书On Writing。King认为，故事不是构造出来的，而是像考古挖化石那样慢慢挖掘出来的。也就是说，故事本身是客观存在的，创作者是在按照蛛丝马迹把这个故事搜寻出来。听起来有点玄学，但和Walker那个搜索玩法的描述不谋而合。

(Early) Thoughts on Valorant

by

I’ve wanted to write this post for a few weeks now, but have not yet had time to extensively play the game. Finally I decided I should just jot my current thoughts down (or these thoughts will just be lost in time), noting that it is founded on a dangerously shallow understanding of the core game.

A quick disclaimer: I used to work at Riot Games, the developer behind Valorant, the game I’m about to discuss. My tenure at Riot overlapped quite a bit with this game’s development, but I was never affiliated with the project. My discussion below is based on public info.

Valorant is Riot Games’s new FPS currently in closed beta for PC platform, and its first new game IP since League of Legends a decade ago. There’s a lot riding on this game: in the short to mid term, this game will largely determine whether Riot is an multi-IP games studio,1 or “just” the League of Legends company (which to be clear is an extremely enviable position). It’s also a major test for Riot’s R&D process, as the game has been in development for over 6 years.

Savvy beta marketing

Marketing-wise, Valorant has had a great start. Its Twitch beta key strategy (keys randomly drop by watching Valorant streams, initially with designated partners, later with all channels) has overall been a resounding success.2 This is a mechanic that CS:GO players are familiar with, as CS:GO tournaments have often used in-game drops as rewards for watching streams. One criticism of such tactics is that they inflate Twitch engagement numbers; that certainly happened with Valorant, though I don’t think it’s Riot’s goal to hit specific viewership goals, but rather, to have optimal visibility / hype around the game’s beta launch – and that goal was more than fulfilled. (Possibly over done, even – for a while, Reddit was filled with complaints about not being able to get a key despite watching dozens of hours.)

Riot has also deployed its community engagement best practices to great effect. I’ve skimmed the subreddit over the past few weeks, and the community has generally been very appreciative of Rioter engagement. The “devs vs streamers” showmatch (where the devs won by a landslide) also earned the team a lot of street cred.

Gameplay

The core game (“5v5 character-based tactical shooter”) can be crudely described as 80% Counter-strike and 20% Overwatch. Counter-strike lends the main structure of the game: the 5v5 rounds-based format (with its economy macro play), the map objectives (bomb plant / defuse), and even the broad strokes of the weapons and gunplay feel. It even brought over the esoteric mechanic bunny hopping.3

The limited selection of grenades in CS is replaced with an expansive character system (the “20% Overwatch“), and character abilities are mostly about utility – detection, blocking vision / movement, mobility, and so on. Abilities are not free to use; instead, charges are purchased with hard-earned cash at the beginning of every round, which suggest their origins in CS-grenades. (Ultimate abilities are the exception: they are charged up by kills, deaths, planting / defusing, or collecting power-ups.) Damage-dealing abilities have been contentious within the community, partly due to Riot’s own marketing statements.

Based on very limited game time, I would say the core game works. It builds on the proven foundations of Counter-strike, and adds variety and depth with the characters system. It’s a very strong execution of a clear game thesis.

Bull & bear cases

This is where I make some wild speculations of the game’s future. This is done in earnest as a thought exercise, but take it for it is – subjective predictions and guesses. I’ve also intentionally pushed myself to plant some stakes in the ground, instead of hedging – so there’s a higher chance I look like an idiot in a few years time when I look back at this.

The game environment – cheating, toxicity, etc.

The seedy underbelly of a competitive online game. Riot has had a lot of experience manage this aspect in League of Legends, but it remains a challenge. In particular, anti-cheat is a never-ending war of attrition, and FPS games on PC seem to have the worst of it – PUBG, Apex Legends, Call of Duty: Warzone, and of course CS:GO. Riot made a big promise – “a commitment to anti-cheat from day one”, and promptly walked into a big, ongoing, controversy with its anti-cheat software, Vanguard. (A quick search will turn up lots of articles discussing this.)

As an aside – the situation around Vanguard tells you a lot about the PC platform. Vanguard asks for very high level system privileges, and raises legitimate concerns about privacy / malware / digital surveillance – the fact it can do so, and needs to do so4, is a problem unique to PC gaming (I’d guess Android is close). There are cheaters on console and iOS, but the scale / prevalence does not compare – for example, see the recent story about console Call of Duty players turning off crossplay to avoid PC cheaters. (And the compatibility headaches it is running into, with all sorts of hardware / software configuration edge cases, is also unique to PC gaming.)

Anyways – some of the Vanguard controversy is founded in conspiracy-theory land – singling out Riot for its ownership by Tencent, and thus leaping straight to concerns over Chinese hacking. Unfortunately, it is a sign of the times, and the trajectory of worsening US-China relations. But I won’t delve into that here.

The bear case here is that the security drama severely hampers the game’s growth, or even sinks it. But I think that would be extremely unlikely.

I am more concerned about toxicity, and how it reduces the addressable audience. Here I’m more pessimistic. I don’t expect Riot to do much better than it did in League – which is to say, the game will have a male-dominant (like, ~90% male) community that is frequently toxic, and often prejudiced and hostile against female (and other minorities) gamers. This bleeds into my next point.

Audience

This is the biggest variable to Valorant’s future (and encapsulates many other variables, so this is not a MECE analysis). To start with my conclusion – if you were to ask me right now, I’d guess that Valorant stays safely within the confines of the existing PVP-shooter audience, and carves out a playerbase from various existing shooters; it will have a loyal following, but it will not challenge battle royale’s position as the leading PVP-shooter sub-genre globally.

The bull case for Valorant is where the game goes beyond converting its bulls-eye target of Counter-strike players, and attracts players of other adjacent PVP shooters – Overwatch, Rainbow Six Siege, Call of Duty, PUBG, Fortnite, Apex Legends, Escape from Tarkov… Going even more broadly, it could also appeal to players of other types of real-time multiplayer PVP games, such as League of Legends itself – this poses a mild cannibalization risk (which I wouldn’t lose sleep over).

It’s hard to imagine the game converting a lot of non-PVP gamers. I would guess that Riot does not have much interest in targeting them (at least not for this game), in contrast to, say, Fortnite‘s efforts at building a digital lifestyle brand (and catering to a wide gamer demographic). A lot of this is rooted in the company culture, which for years was “HARDCORE GAMER”, but this has been relaxed/widened a bit in recent years. Still, Valorant‘s Game Overview section on its beta website is pretty telling about the intended audience:

Here’s what we think it takes for you to trust a game enough to invest: 128-tick servers, at least 30 frames per second on most min-spec computers (even dating back a decade), 60 to 144+ FPS on modern gaming rigs, a global spread of datacenters aimed at <35ms for players in major cities around the world, a netcode we’ve been obsessing over for years, and a commitment to anti-cheat from day one.

Shooting in VALORANT is precise, consequential, and highly-lethal – we want you to win on your skill and strategy alone.

This is a laser focus on CS players, and disgruntled players who’ve complained about the “shitty netcode” of just about every shooter with a PVP mode. (Maybe it’s just for the beta phase, where they are prioritizing veterans above all else.) I’d argue this language is alienating to players less familiar with PVP-shooter games, who don’t necessarily understand jargons such as “128-tick”, and thus this marketing actively reinforces the existing male-dominant audience stereotype.

The arguments for a bear case come in a few flavors. The first is where CS players churn and flow back to CS, because at the end of the day, Valorant is a different game. There’s some premature indication of this on reddit, where CS veterans would demand certain types of mechanics (that are present in CS). This is a delicate balancing act, and looking at Riot’s early days with League and Dota veterans, I’m not too worried that Riot would over-cater CS veterans. (But the League / Dota analogy would also suggest that loyal CS players will stick with CS, and even be antagonistic to this new game which poses a threat to their community – this would limit the efficacy or targeting CS players to begin with.)

The second bear case argument is where Valorant fails to capture players other than CS die-hards. This does not seem to be the case so far, but I would guess the ceiling here is not high. My negativity here is largely emotions-based: since Valorant’s inception 6 years ago, we’ve witnessed some dramatic new entrants to the PVP shooter space – Overwatch, PUBG, and Fortnite, to name just a few. These games all brought some genre-defining “fresh” factor. I couldn’t help but feel that Valorant in comparison feels too old-school, too familiar (“I know exactly what I’m getting into”). There’s a market premium for novel experiences – for example, that first chicken dinner was unlike any game experience I’ve ever had before – and Valorant judged by its cover is treading on familiar ground.

The last argument is about overserving player needs. I recently came across this excellent article on fy_iceworld – and vivid memories of playing CS1.5 in college in China came roaring back. I was the snob that begged classmates to play the “real game” (as in, play 5v5 bomb defusal mode), and we did very occasionally; most of the time though, we were “messing around” in fy_iceworld or playing 20-person PUGs (with max economy every round, of course). My point here being, if my cohort of CS players 15+ years ago is any indication (highly anecdotal, and a long time ago, for sure), the majority of players around me were playing CS “casually”. (Just like the vast majority of soccer enthusiasts around the world are not playing 11vs11 games on full-sized grass pitches with FIFA rules.) If Riot is too strict on the game modes offered, and don’t provide “casual” outlets in-game, it could cause these “bottom of the pyramid” players to churn, which could also pull away their social connections.

Esports

I don’t have much to say here, except that Valorant is clearly built as an esports title (in the proud tradition of CS), and it should have a vibrant esports scene that helps with maintaining the game’s player engagement. I also think that for spectators, shooters are much easier to understand and follow conceptually (vs MOBAs), and thus the bull case could be as big (or bigger) than League esports today. So I’m personally quite bullish here, and think that Valorant could enjoy disproportionately higher esports popularity relative to its active playerbase.

One bear case argument is societal attitudes towards video-game violence, and how much that impacts a shooter like Valorant when it comes to sponsorships or broadcast coverage. This may be an issue in North America.

Winning the Chinese market

I’m quite bearish here. Valorant will have a difficult road to launch in China (could be delayed by years), and even then its prospects are murky.

There is a strong bull case to be made. Firstly, Riot is owned by Tencent, which has market-leading publishing capabilities in China, and did a phenomenal job publishing League. Secondly, the PC PVP-shooter landscape is much less crowded (and more stagnant) than it is in North America – PUBG, Fortnite, Apex Legends, Call of Duty franchise are all not officially available in the market. Based on one source (tracking PC cafe consumption), the market leader remains Tencent-published CrossFire (launched in China in 2008), followed by Tencent’s self-developed Assault Fire as a distant second (10x engagement gap), with Overwatch and CS:GO in 3rd and 4th place. One could argue that the market is ripe for a new entrant, and Riot + Tencent is a fearsome combo.

But the bear case is quite stacked as well. To begin with, there is the regulatory uncertainty – for years, the trend has been in one direction, and that is tightening. And US-China relations are at historic lows, which makes getting the license approval for a US-based IP that much harder. (There’s a reason the big names above are all absent.) There’s a low (but non-zero) possibility that Valorant never gets a license.

Second of all, the CrossFire audience may not be interested in Valorant, despite the superficial similarities. This links back to my earlier point about fy_iceworld and the audience motivations.

Thirdly, the initial Chinese player reaction has been mixed, which reflects some brand gaps and taste differences. On NGA (a popular forum for Chinese hardcore gamers), Valorant‘s gameplay has been labeled “缝合怪” (stitched-up monster), which is a common term to describe video games that mash-up mechanics from different games5; and the visuals were unfortunately derided by some as “browser-game quality” (not understanding or refusing to acknowledge that it’s a conscious art style choice). These comments partly stem from hostile rivalry between Riot’s supporters and supporters of Blizzard and Valve. Blizzard is clearly the biggest and most beloved studio brand, and their Chinese supporters seem a tad unhappy about Valorant possibly taking players from Overwatch; meanwhile Valve supporters are still holding a grudge from the League – Dota2 rivalry, which has always felt much more intense in China. However, I do think the art style is an acquired taste to many Chinese players.

And last but not least, there’s the question of mobile.6

Mobile

I’ll try to be concise here: my take is Valorant needs to have a mobile version, but it will be very challenging to get it right.

First, the most popular PVP-shooter globally, by a long margin, is a mobile game. PUBG Mobile announced 100M MAU last May. Its sibling game in China, Peacekeeper Elite (rebranded for regulatory reasons), was estimated to have had 197M MAU this March. So it’s plausible that the combined PUBG Mobile franchise currently has over 300M MAU – about the population of the US, or comparable to Twitter’s MAU.

In the China context, what this means is “all gamers are hardcore gamers”, if you define “hardcore” by genre played. To put this into a picture: Chinese moms are playing mobile battle royale with their children.

These Chinese moms will likely never play PC games, if they don’t already. A fraction of these kids will, but I’d bet majority of them will be mobile-only gamers. Clearly, Valorant is not a game made for them (and not every game needs to be made for the widest audience/platform); but I can’t help but feel Valorant cannot be a truly global game (which matters for its esports aspirations), without at least trying to accommodate such players somehow.

So should Valorant make a mobile version? The core game’s methodical play and precision aiming does not translate well to current mobile shooter control schemes (or console either). PUBG Mobile can get away with it, and retain the spirit of the original PC game, because the maps and the encounters are so open-ended – it’s only during close quarters combat where the gameplay feels like a parody at times. Perhaps CrossFire Mobile could be a reference here: the game superficially resembles its PC ancestor, but I’ve heard the engagement with the content is notably different from PC.

In closing…

As a meta comment: this post probably both took me the most time to write (10 hours over 3 nights, as I debated endlessly with myself), and left me least satisfied with the results. I hope you find it marginally useful. If I were to do it again, I would break it up into a couple posts, so I can have the energy and the space to mull over a specific point.

For the game discussed, I guess my overarching sentiment is moderate pessimism over product-market fit. Valorant is strong execution against a clear game thesis – I just don’t know how big that audience is, versus other possible opportunities.

  1. This doesn’t take into account Riot’s studio acquisitions – Radiant in 2016 and Hypixel in 2020, where there’s scant public info about their projects. ↩
  2. Side note – I was surprised that beta keys could only drop on Twitch – I would have thought Riot would have enabled other streaming properties, such as Youtube & Mixer, to also participate. I speculate this is due to a lack of infrastructure (APIs etc.) on these partners, rather than lack of interest on Riot’s part. ↩
  3. I don’t claim to be an expert on FPS games; I couldn’t understand why this mechanic is needed, aside from making CS players feel at home. It reminded me of creep-stacking and denying in Dota, and League of Legends choosing not to implement them. ↩
  4. This is one of the hotly debated points in the controversy. ↩
  5. Star Wars Jedi: Fallen Order is another recent game that got this meme label. Not bad company to keep. ↩
  6. For folks who’ve read some other posts on this site – do I sound like a broken record about mobile yet? ↩

Platform alliances and fragmented user experiences

by

As a bit of funny context, I’m writing this post amidst one of the biggest stock sell-offs in recent decades. US stock markets actually triggered their circuit-breakers after opening 7% down, and paused trading for 15 minutes. I thought I’d try to be a bit more constructive with my time (while sipping some scotch), instead of just fidgeting in front of my portfolio dashboard.

Nvidia’s GeForce Now has been a steady thread of gaming news in recent weeks. Despite (or rather, exactly due to) what feels like a more compelling end-user value proposition (stream the games you previously purchased), numerous publishers large and small (Activision Blizzard, Bethesda etc.) have pulled their content from this service. Publishers’ perspective is simple: they really don’t like it when a new service generates revenue off their content, without their permission or share of profits. (This The Verge article gives a good overview and analysis of the situation.)

One interesting development today – on one hand, another large publisher, 2K Games, pulled their content; on the other, Epic Games endorsing the platform with the full weight of their store.

The Epic decision is interesting partially because how transparent Tim Sweeney is with his intent. To quote him directly:

Epic is wholeheartedly supporting NVIDIA’s GeForce NOW service with Fortnite and with Epic Games Store titles that choose to participate (including exclusives), and we’ll be improving the integration over time.

It’s the most developer-friendly and publisher-friendly of the major streaming services, with zero tax on game revenue. Game companies who want to move the game industry towards a healthier state for everyone should be supporting this kind of service!

Cloud streaming services will also be key players in ending the iOS and Google Play payment monopolies and their 30% taxes. Apple has decreed that these services aren’t allowed to exist on iOS, and therefore aren’t allowed to compete, which is megalomaniacal and won’t stand.

Some quick observations. First of all, Sweeney caveats/acknowledges that this support still requires publishers’ participation. So the actual impact / enrollment still needs to be seen. (When I first saw the headline without reading Sweeney’s original tweet, I thought it was an unilateral move and wondered how that would work.) And if it does not result in a great majority of the store’s titles becoming streamable, it could mean a very fragmented (and confusing) end-user experience – for example, would Epic Game Store (EGS) add a platform tag on each title that shows which ones are compatible with GeForce Now?

Second, Sweeney discloses his motive is at least partially (if not primarily) about forming an alliance against iOS / Google Play. This makes theoretical sense, and is the sort of classic partnership that gets discussed in business strategy classes. The reason Epic is so eager to rally support against iOS / Google Play platform rules is simple – across the entire video-games vertical, mobile (due to its sheer scale) is generally the biggest opportunity for revenue growth, and the 30% platform fees are likely the biggest opportunity for profit (bottom-line) growth. This logic is the same whether Epic sees itself as a games developer or a publishing platform for 3rd parties.1

For this particular partnership, I’m a bit dubious whether it will meaningfully move the needle. I’m generally bearish on streamed gaming – I think it’s cool technology in search of a problem / audience, and if the strategy is to reach multi-billion gamers, I’d argue mobile-first (or mobile-only) remains the best path over the next 5 years.

Anyway, the other part I wanted to write about is the fragmented user experience in this age of multiple co-existing / overlapping games platforms. What do I mean? To use AAA games as an example – to date, publishers have all built their own launchers / platforms (Uplay, Origin, Rockstar Social Club) alongside distribution platforms such as Steam, GOG and EGS. Players are all familiar with the experience of launching a Ubisoft game on Steam / EGS, just for it to pull up Uplay and the game itself. Putting aside the obvious overlapping functionalities (duplicate social features, overlays, platform achievements, cloud saves), I’ve also seen cases of weird hand-offs – for example, the initial game download was handled by Epic, but during the first launch Uplay initiates a huge patch.

Also – alliances can be temporary, and every break-up is a chance for fragmentation. When Bungie exited its partnership with Activision on Destiny, and moved over to Steam, what did that mean for players’ friends-lists? (They stayed on Battle.net.) And how long will Bungie keep its PC migration service up? (Not that it matters that much, as inventory rapidly devalues with new expansions.)

There can also be player-caused fragmentation. Again using Ubisoft as an example – I bought The Division 2 on EGS, but purchased the Warlords expansion directly on Uplay. Would this cause any weird experiential issues on EGS? (I don’t know because I’ve not launched the game via EGS since, which I’m sure is exactly what Uplay prefers.)

Similarly, there are (and will be more) fragmented experiences when we talk about “cross-platform-play”. Experiential fragmentation is not going away (at least without a huge fight), because it’s not fundamentally a technical problem, but rather a business model / strategy problem – just as it is in the above case of GeForce Now. In theory, the combination of cross-play and free-to-play monetization is exactly opposite what large platform incumbents like Microsoft and Sony want: they want to have a monopoly on network effects, and they certainly don’t want out-of-network spending. In practice, there are complicating motivations,2 and some games like Fortnite have a big enough audience to force the issue, though as we have seen these games remain the exception.

From the end-users perspective – a lot of this is exactly an argument for a vertically integrated platform with monopoly market share, in exchange for an user experience guaranteed to be seamless. For PC, that sounds a lot like Steam, which I guess partially explains the player anger towards EGS. But Steam’s monopoly was always built on shaky grounds (the underlying attributes of the PC platform), and Steam’s margins were Epic’s opportunity.

  1. By the way, if it’s not clear, I am not being critical of Epic’s motives. ↩
  2. Microsoft as the also-ran to Sony, has an incentive to open up and share the network. ↩

漫谈内容渠道与平台

by

This is a backup of a wechat account post, dated Feb 5 2019.

这篇与游戏无关。

这些天，因为2019-nCoV疫情，想来中文网络媒体流量都是井喷的状态，也在微博、微信上都看到了不少的“强时效”内容（短暂存在散播，之后因各种原因被删除）。结合我自己公众号发布的粗浅体验，便有了以下感想。

微信公众号是个极度受限的内容平台

此平台对个人创作者的限制是非常大的，内容不能添加“外链”（常规的html链接）、个人公众号没有评论功能（2017年以后注册的），单这两条就已经和主流的国外内容平台形成了巨大的反差。

制造如此苛刻的条件，其一当然有个极大的环境因素，就是中国的内容审查，无须多言。其二，也反映了中国网络商战的丛林法则，即正面战场上腾讯、阿里、头条系的严酷流量之争，以及地下灰产的各种奇葩游击战。

（顺带吐槽一句，公众号平台的内容编辑器的交互体验。。。可能是我用不惯吧，不过真有种15年前web1.0时代产品的感觉。）

那为什么大量内容创作者还是会牺牲这么多，选择公众号平台呢？因为另一边读者的体验的确不错：app内阅读体验、社交分享都是极其便捷的，而围绕公众号订阅的内容推送、发现也是对读者与作者皆有价值的体验。简而言之：流量啊！

平台的权力

说回到这些天大量的内容举报、删除。首先要说：私有的商业平台对内容进行审查、乃至删除，不管出于何种动机，在商业与法律规则上绝对是平台的权力。

有趣的是，美国的平台（如Facebook，Twitter和Youtube）过往是躲在“希望提供绝对的言论自由的平台”的挡箭牌后，而不去履行一些其在社会价值层面应扮演的审查角色——比如，对散播各种谣言、阴谋论、对受众进行极端化洗脑的内容放任自流，乃至2016年大选被境外势力（俄罗斯）利用操纵了选情。因此，近年来美国各界对这些平台的压力是迫使其更多地承担内容审查的义务，以消除其平台内容对社会的负面影响（经济学原理中的负外部性）。

中国的情况当然是有点反过来的：平台都是有强大的自审压力的。前面也说了，逻辑上我也尊重平台的权力。不过，作为一个创作者，我也当然会想有保护自己声音的权力。这便回到了经典的内容创作者（媒体）的两难选择：依附平台，有流量、没控制；自建渠道，有控制、没流量。

过去15年的起承转合

以下思维有点发散、跳跃。

我在网上写博客已经至少15年了。最早是在水木BBS的博客（那时写的都是些没有营养的扯淡灌水）；经历了2005年初的接管事件后（巧了，今年正好15周年），曾经把内容东搬西搬到几家境内外的博客平台上。不过那些平台也各种不稳定，便萌发了自建博客的想法，一直持续至今。

15年前正是自建的黄金年代：开源的LAMP技术栈以及基于此栈的开源Wordpress，大幅拉低了入门门槛与运营成本；RSS的普及以及各类RSS聚合应用的兴起，努力引导着读者用户去定制自己的阅读列表。

其后的演变：一来是社交网络兴起带来对内容消费习惯的争夺，二来是海量媒体内容井喷后面临内容发现等问题（这也是为什么会有类似今日头条这样的智能推荐切入点），三来是PC到移动终端迁移引发阅读场景剧变。2013年Google Reader关闭，现也可视作RSS作为一种面向大众消费者的技术的墓志铭。

对于个人创作者，主观感受就是：还在自己的网站上写文章的人已经很少了，内容都在一些平台上（比如国内微信公众号、知乎专栏等等，国外硅谷科技圈似乎很多人都在Medium上写）；内容发现则几乎完全是在社交网络上。

有趣的是，科技圈似乎总是在经历一些轮回：国外近年来新兴的阅读习惯是订阅newsletter，即以邮件订阅的形式获得一些创作者新内容的直接推送。不得不说这有点搞笑：这不是RSS的推送体验，以更古老的email协议卷土重来吗？然而它的确颇为流行，比如我常看的Benedict Evans的周邮件，始于2013年，如今有超过13万邮箱订阅。这里还有一个付费的分支，即比如Stratechery、Above Avalon这样的内容创作者自建的付费会员推送。

说到newsletter订阅，也是有平台和自建之分的。其实古董的wordpress，原生就支持邮件订阅，但这种推送貌似很容易被邮箱平台的算法认定为垃圾邮件。故此邮件推送需要接一个mailchimp之类的管理服务。另外，如果希望对内容进行收费，也就需要整合一个支付系统。上面提到的Stratechery等，把自建渠道走到底，也算是走出来了，是其创作者的全职主业。而大多数创作者可能没有这样的时间与毅力——于是随着邮件订阅的流行，也就出现了比如substack这样的一站式平台：内容在上面发布，邮件订阅、推送，乃至付费，由其提供支持。

（中国大众用户的邮件使用习惯和国外差别很大，商业上来说，上面这些在国内会是另外的玩法。）

为什么要写作

该收尾了。自我一点，说说为什么要写作？先做一个自我定性：我是一个内向的人，但又有闷骚的虚荣心，因此写作是一个获得关注与赞美的方式，这是15年前早期动力。动力之二，是写作是对逻辑思维与沟通能力很好的锻炼，一个复杂的事情，如果不能用简单的文字描述出来，也就是还没有融会贯通。我是很功利的，思维与沟通是平时工作的核心技能，这是为什么断断续续这些年也坚持写了下来。

这是公众号上我写的第三篇。本来开号也主要是希望把中文写作重新练起来，另外从语言上把博客聚焦为英文内容，公众号为中文内容。不过这个平台体验的确不太友好，后面我可能还是以发布到博客上为主，公众号同步为辅。打个广告，我的博客是bayjinger.com:)

Valve的平台困境

by

This is a backup of a wechat account post, originally posted Dec 5 2019.

（楔子：写这篇，起因是上周Steam Controller手柄停产的消息，在我看来也算是Valve此前的一套策略的一个完结，故作此文。）

2011年我在加州刚刚进入游戏行业的时候，同事们聊起Valve这家公司都是带着万分敬意的：

  • 它不只是一个创造过数款划时代游戏的顶级游戏工作室；

  • 它也是一个有着令人敬仰的独特企业文化的公司（我们都认真拜读过其员工手册，并将其与Netflix著名的culture deck相提并论），小而美的数百人公司规模，在行业里四两拨千斤；

  • 另外在当时，其数字平台Steam已经初步确立了其PC游戏数字渠道霸主的地位，这也是一个令人艳羡的公司战略转型，之后的走势更是主导着欧美PC游戏行业的潮流（譬如近年来独立游戏的井喷，与其密不可分）；

  • 延伸一点，既运作一个大型软件平台，又直接掌控了平台上最有价值的应用（Dota2和CS），这与微软当年左手Windows、右手Office两大利润中心如出一辙（不愧为从微软出来的创始人）。

在过去的近十年里，Valve这种奇特的小公司、轻资产、平台化的模式表面看起来依然靓丽，活跃用户数不断创新高，其它的指标（营收、平台游戏发行量等）画图大概也都是向右上方迈进。不过，我倒是觉得Steam陷入围城已久，种种破局的尝试也基本都宣告失败，未来十年的走势不容乐观。

对Windows强依赖

我不看好Steam的最大原因，就是因为其对Windows平台的强依赖。而Windows平台的影响力的式微，用A16Z的Ben Evans 2015年的一张经典的图可以概括（推荐拜读其原文”Microsoft, capitulation and the end of Windows Everywhere“）：

（这里的点当然不是说Windows/PC业务崩盘了，而是在巨大的移动端增长面前，Windows平台已经退居次席，不再是整个生态的中心，也就与未来创新、新增长点大体无关了。在可预见的未来，Windows还会存在，缓慢衰老——这里可参考上上一代行业霸主IBM，大型机业务每年还在卖几十亿美元。）

过去十年里，Valve并非没有去积极尝试”去Windows化”。2013年的SteamOS（基于linux的操作系统），15年的Steam Machine（基于SteamOS的硬件），以及配套的Steam Controller手柄，概念上来说是已经垂直整合了整个终端体验，而且处处彰显着Valve借力打力、四两拨千斤的经典思维。

只是，在严峻现实面前，这套操作步步皆输。拥抱linux固然自由了，但作为面向大众终端消费者的操作系统，linux本身就是海市蜃楼（已经快20年了，linux消费终端爆发的春天似乎总是就在眼前）；SteamOS得不到开发者重视，做不起来也就毫不意外了。（前两天看到某外国开发者吐槽，没记错的话大概是这样说的：”linux版本是我们游戏0.5%的销量，却占用了我们50%的客服资源。”）

而Steam Machine所期望的轻资产模式（让有意愿的硬件商来制造），也是很难走得通的。20年前微软进军主机市场，也曾寄希望于有外部厂商能负责硬件，但无人问津（商业模式上这是不成立的亏本买卖），微软只好自己砸钱。

至于Steam手柄，这本身倒是一个大胆的想法（怎样用一个手柄取代键盘、鼠标，以便于在沙发上对着电视玩游戏？），但也可以说是蛮典型的以公司痛点为出发点（而非消费者痛点）的产品——玩家并没有在电视上玩文明6的刚需，而是Valve抓耳挠腮在想怎样把自己的平台拓展到客厅里。

PC渠道的红海残局

回到Steam的主战场，本来似乎波澜不惊的残局（各大发行商的自有平台都只做得马马虎虎，Steam的大哥地位看起来很稳），但突然杀出了个Epic Games这样的程咬金。

Epic的路子迈得很大很野，而且似乎很清晰地就是在模仿Valve当年的策略，以一个现象级游戏堡垒之夜的用户量和利润为依托，强势拓展平台业务：

  • 打分成价格战、花重金签独代来吸引开发者，这其中还自然利用了其UE4引擎的垂直整合；

  • 坚持每周免费送一款游戏去拉玩家；

  • 最野的一步是在安卓端，借堡垒之夜安装包自建渠道，无视Google Play。

面对Epic的搅局，以及一个没有什么大增长点的红海市场背景，Steam的市场份额与利润几乎必然下降（平台分成下调恐怕只是时间的问题）；如果应对不当，得罪了开发者或玩家，甚至可能让出龙头地位。

破局思路？

那么Valve面对这样的局面，有哪些宏观策略选择呢？

若继续坚持平台为核心的策略方向，那还是要回到”怎样在windows之外有存在感”这道题的解法，而在微软、索尼、任天堂、苹果和谷歌的主机/手机生态里面似乎都看不到什么明显的空间。（Steam的iOS app，存在很久了，基础的社交、购买体验实在是乏善可陈；而且稍微大胆一点的想法，比如通过app来云游戏玩自己steam上的游戏，也极易触及苹果平台规则的禁区。）哦，或许欧美安卓生态上，随着Epic对Google Play商城的公然挑战，会有更多的文章可做（会引来更多的效仿者），但Steam并不像Epic那样有个安卓爆款游戏/应用作为天然的流量切入点。

如此这般，平台策略里好像也只剩下垂直整合、开辟新硬件平台这条路了。而VR，看起来是Valve的一个重点布局，毕竟连当年发家的Half-Life IP都拿出来为VR站台了。只是，VR概念热闹了这么多年，能看到的杀手级游戏好像也就是Beat Saber（开发团队刚刚被Facebook Oculus收购）；对于硬核游戏，单单是输入方式上就还有很多基础科研要做。

抛开平台策略不谈，其实作为一个玩家当然是希望看到Valve能继续推出好的游戏。回归游戏研发的本位，Valve是有巨额的IP财富的，但近年来的AAA开发的团队规模又已经上了一个量级。在玩家膨胀的预期面前，Valve的中小团队能否对经典IP交出满意答卷，恐怕并不乐观（卡牌游戏Artifact的雪崩就是实例）。