Meanwhile there have been precisely zero patches and zero communication regarding Redfall since launch.
I have it 3 hours and found it non-atmospheric and boring, unfortunately. So maybe it is best if they move on quickly.
There was something else I found interesting in the Bloomberg article:
Along the way, Smith and other leaders assured the staff that the game would get exponentially better once the final art was implemented and the bugs were fixed, promising that “Arkane magic” would manifest at the last minute as it had with previous games. (Other developers, such as Electronic Arts Inc.’s BioWare, have been widely criticized for using similar language.)
We’ve seen this criticism before, that studios sometimes lean too hard on this “magic” to pull their game together at the last minute. But this quite reminded me of a quote in a whole other article - this is from an oral history article about System Shock over at Rock Paper Shotgun. I linked it in the System Shock thread, but here it is again if you don’t want to dig through that thread. This is the quote from Rob Fermier that caught my eye:
Even two months before the end, it felt like System Shock was a pile of pieces on the floor of the garage. It really clicked at the last minute. I remember playing it after it shipped and being pleased with how well things came together.
I haven’t ever worked on video game development, but I have worked in other software dev spaces and I get that feeling, that sometimes all the pieces don’t really cohere into a whole, workable thing until almost the last minute. And I also get how that can seem to be “magic.” Maybe the folks at Arkane, and the higher-ups at Zenimax and MS figured that this would be one of those cases, that the game would “magically” become the thing it was supposed to be at the end, and then when it didn’t , it was too late to do anything about it.
Another interesting point: the common link between the two games is Harvey Smith, who is head of Arkane Austin now but was also a tester on System Shock. Maybe he’s seen this sort of thing work in his favor before, and thought it might again. Who knows?
Bug wise that makes sense, but the stuff that makes “magic” in games like that is story, attention to detail, depth of the world, etc. That stuff is the polar opposite of what can be done and fixed in crunch time, I think we would all agree.
I would disagree - the thing that makes the magic is undefinable, otherwise it wouldn’t be magic.
I’m really bummed out that most of the members who made Prey (and the outstanding Mooncrash DLC, which is probably the best DLC I ever played) are no longer with Arkane. Prey is one of the best games I ever played, and now I’m not confident they can make another game as good as that.
Particularly since the “they” have been scattered to the wind. It’s a hard world when crap that is tuned to gambling or social instincts can make a billion dollars but a masterpiece without those layers can’t touch the same level of return.
If a project has a good foundation, even if the steps are complicated and take a while to come to fruition, it’s definitely possible, even likely, that things will move really rapidly over the last few intervals/sprints whatever towards the end. That magic though is the product of a ton of hard work early in the project.
I saw some chatter positing that the staff turnover isn’t particularly unusual. A former Prey dev, however, clarifies that the 70% figure does not include contractors.
This was way more than usual, and primarily experienced staff running for the hills as it became clear there was no good game there. All of the “additional” people in the credits are full time who left. Contractors are generally listed under the specific companies that were hired.
The parts about no clear design or direction or idea how to make a multiplayer immersive sim, confused teams, and Harvey saying it will be better when the final art is in… Those parts explain where the 70% went.
Schreier also popped into the Resetera thread to say the same thing:
Of the 80ish Arkane Austin staff who worked on Prey, 55ish are no longer at the studio, including the creative director, lead systems designer, lead level designer, and lead AI engineer. (Support staff, freelancers, and Arkane Lyon employees weren’t part of that tally.) Don’t need average industry attrition rates to surmise how monumental a change that’s been for the studio.
This wasn’t covered in the article, but Arkane were really unhappy with how Bethesda promoted Prey, too:
https://twitter.com/jasonschreier/status/1664294880724541440
So much this. Don’t skimp on prototyping. Take time early in the project to really figure it out and establish direction, tone, and where you want to be. And then trust in that while developing, even during the rocky shit part of mid dev when nothing works and everything is ugly. Course correct where needed, but don’t go into production without a clear idea of what you want on the other side of it.
There was at least one that I recall.
That bit about the staff changes make me a little worried about whatever they’re cooking up next (which according to Smith will be a SP immersive sim).
I know this is being said a bunch these days, but…look at Nintendo and their Zelda team. There are people on there that have been working on the franchise for >20 years. They took an entire year after the game was DONE to make sure the physics of the abilities work without glitches or bugs.
I think once they make it known that they’re working on a single player immersive sim, they’ll have an easier time hiring the type of people who want to work on such a project. They said in the article they had a tough time hiring people because of the lack of ability to describe the project.
A little unfair to compare it to Zelda though. Not only do they have all that institutional knowledge of experienced developers, they have many more resources and time and Nintendo’s patience for working on one of the their most prized franchises. Arkane, even with the original Prey team, wouldn’t have been able to compete with the immersive sims made by the Zelda team.
The lesson re: Zelda is not only applying to Redfall, but the industry in general, and the way games are made in the triple A space, especially wrt to turnover and QA
This story seems like a classic gamedev one where talented people have various constraints that sabotage the project in one way or another.
I can imagine a world in which Bethesda both mandates “make a live service microtrans game” AND says “but design it how you like”. It’s possible Harvey Smith et al both got a mandate AND had creative control, and even fell in love with what they pitched. Even if they didn’t, part of their job is to convince both the higher ups and the team that they do love it. Even though it’s a huge risk factor that they haven’t made this kind of game before.
According to wikipedia, this is their first Unreal game (besides a Prey expansion (?)). Management ALWAYS thinks the grass is greener with an external engine. And the cost of switching is ALWAYS higher than they think. That’s not even a dig at Unreal, even though they do a good job of convincing you things are easier than they are. :) But if you’ve shipped 3 games using the conversion tool you built on Void…guess what, you get to rebuild it on Unreal. But a suit gets the idea that Epic’s royalty rate is lower than what you pay for devs to port to other platforms (that get cut!), and suddenly you are already behind.
The most incredible part was that they couldn’t advertise multiplayer dev positions! Going MP is your #1 RISK FACTOR of the project. People wonder, “why does the combat feel so much worse than other Arcane games?” This is why. Because online combat is a whole new can of worms that affects EVERYONE, including the designers and artists etc. And on a new engine of course.
It’s funny to me how they couldn’t talk about making a MP game. Bethesda is strangely secretive, even about things that are pretty well known. I interviewed there once and they couldn’t tell me anything at all about the project I would be working on, even if it was an RPG etc. Shocker, it was Fallout 4! lol
I could see the leads seeing this game was floundering, and going to ask for more resources, and being told, “We’re trying to sell! We can’t hire up right now!” I’m sure part of the acquisition process was a portfolio review, probably with ship dates. I wonder if there was something binding about when these games would ship?
I can even see Spencer wanting to be hands off with the studios, at least at first. Arkane makes (made?) beloved games, you don’t want to march in and roll heads… they have PR “gamer cred” to protect.
But the one thing that sticks out to me is that their internal metrics predicted a better game. When a game is bad, or is in trouble, EVERYONE KNOWS. So someone is lying (if not Spencer, he was being lied to by internal production). Or their internal processes are really really bad.
As for Arkane “magic”, it is true that almost every game is bad until just about the end. Veteran devs like to think it’s because they are geniuses. Like films, you have to assemble it from disparate parts that don’t exist until near the end. You’re basically taking it on faith that those ingredients will taste good together, but you don’t know until it comes out of the over. Your creative team basically tries to see the end product along the way, and adapt the recipe, and hope that the individual teams’ estimates are accurate. It used to be that all of those miscalculations could be rectified with 3 months of crunch; but now, turning the ship is so much slower that you almost can’t afford to fix those mistakes.
I could see something like the following. You make some bad bets about how you synchronize the world for network play. Your artists/designers try to replicate your complicated enemy combat behaviors/anims in the new game (“the combat experience Arkane is known for”). Your pitch calls for tons of enemies, Left 4 Dead style. Your engineers are hollering that their networking/prediction infrastructure can’t handle all of that. Your lead producers tell them to keep working on it, secretly thinking they pull it out of the fire with that Arkane magic. You’re asking for schedule clarity from higher up, but you aren’t getting it because a sale is pending or has happened. You may not even know who the higher up you’re supposed to be talking to is. Everyone is assuming you’ll get delayed, because you can’t ship this. Finally, you get word that the ship date is sticking. You’ve got various enemies/powers/weapons in various states of functionality. Most of them get the axe because you don’t have time to finish. The miracle network solution is not coming. You have to drastically reduce the numbers of enemies, and their internal states to meet the requirements. The horde-type enemies are dumb, but this was supposed to be partially hidden by now many of them their are. The smarter enemies have had some of their simulation gutted by the need to reduce what they sync. Behaviors have to change, no time to tweak the animations to look good with these changes. Combat is suddenly WAY different; designers are frantically tweaking weapons and abilities to get the challenge right. Combat feels like a different game than the one you designed everything for. You don’t really have a lot of time to test these drastic changes. And the economy? You’ll fix it post ship. Bugs are flying in, competing for time with the changes you have to make to ship. You’re making new bugs with these changes. But the date’s not changing. It seems to everyone on the outside that you’ve already spent way too long on this game. But no one knows how many times you’ve remade the game, adjusting to changing conditions, for reasons that aren’t the result of incompetence, but the pressure everyone is getting from someone else.
This is speculation, of course, informed by my own past experiences. But it’s not hard to see why publishers just want you to make sequels. :) Not that those are slam dunks either, but at least SOME risk factors are controlled.
Mock reviews are often conducted alongside a list of “here’s what is missing/still to be implemented” and if it is a multiplayer game, they’re usually playing with devs or QA. While the latter may try to play “naturally”, they’re still going to have internal biases that likely make them play in a way that shows the game in a better light. And finally, mock reviewers know what they’re reviewing and for whom and will adjust based on reputation and history.
So it really wouldn’t surprise me if internal metrics with all of the above caveats would predict a higher score.
Big update today. Anybody still playing? Anybody start playing?
Wow. Every one of these fuckers qualifies for the “I hate this guy” thread.
There’s no saving this game.
I’m waiting for the promised 60 FPS mode before I try it.
They even call them “incremental” improvements. Reading the list it looks like all bugfixes and minor QoL enhancements that don’t touch the key deficiencies of the game. The game was extremely buggy so that’s all welcome, you can play it now if you liked that gameplay. But most people didn’t.
