Exactly. I wonder if it’s possible to make a commercial game that’s fully decentralized. Sell the licenses as NFTs, use activitypub for the masterserver so you just toot out your games, release the game and dedicated server as free binaries. I hate NFTs but they’d be more consumer-rights friendly than the current approach of steam owns everything.
Seriously, it’s a four-player game. Not some MMO clusterfuck. Not an arena shooter bragging about 128 codblops on a single map, like it’s Stand On Nuketown. You need matchmaking - you want anticheat - you’ll do some DRM bullshit. Other than that you should want to offload bandwidth and latency to your players. They’re all on the same team!
In a couple of generations when path tracing with reasonable levels of performance is available on mid-range cards (eg. equivalent of 4070) I might consider Nvidia. Until then team red for me
Good luck. If the SEC hasn’t already started building a case against him for insider trading, then nothing is going to happen to him. He’ll get a golden parachute and scurry off to ruin some other company.
"Selling shares before the announcement" was a pretty egregious misrepresentation. He has scheduled pre-registered sales on a regular basis because he gets paid partly in stock.
It was always going to be relatively soon after a sale of stock.
Just want to add you’re right but what pisses me off is that they still can influence decisions based on this. Let’s say his shares are sold at x day, just do some decisions before that and boom your auto sell share price is now either higher or lower. Only because it’s predetermined they still influence it and SEC now can’t do shit.
This has nothing in common with insider trading and doesn't resemble it in any way. The shares he sold weren't a relevant proportion of his ownership. He didn't sell then deliberately tank them. He sold then announced something he thought would improve the value of his big stake in the company. The decision almost definitely cost him a lot of money by substantially lowering the trajectory of his company's ability to maintain market share.
Exactly. It was plopping his dick on the table, then realizing "oh shit, no one actually is impressed by this".
Insider trading would be more "I know we're about to get sued for this egregious fuckup and have no defense, so I'm going to sell before the news leaks". Strategy knowledge can be part of insider trading, but it would tend to be more buying shares because you have advanced knowledge that a highly lucrative contract has been signed before the announcement. It would be harder to have selling because of a strategy decision be insider trading unless you were opposed to it internally, because decisions you make are intended to make the shareholders (you) money.
Don’t you bring facts into this! We want to be outraged!
Being serious though, they ought to be investigating whether there were any changes in those sale orders. If they’ve been the same and unchanged for the last two years or some long period of time, I don’t think there’s a case. But if they’re was an adjustment a month or two ago, that would be very problematic.
I think you mean a nice golden parachute to reward them for taking the heat, so they can swap in a new expensive face to implement slightly less unpopular fees.
Honestly, I don't blame them for not wanting to put up with Unity's unreliance. It took Unity 10 days after announcing this awful change to backtrack to a normal revenue cut. That 10 days was filled with justified outrage from a ton of developers to the point of Re-Logic donating $100k to Godot and FNA in protest.
That’s what confused me the most. When your customers are consumers, screwing them over might be no big deal. But when your customers are businesses, how were you planning to get away with something like this where anything involving fees in the 6 to 9 figures is game changing. That’s, “Cheaper to move my business elsewhere” levels of money.
Yup. They were hoping it would fall out of the news cycle and people would forget about it. Once it stretched past a week, they started to panic because people weren’t dropping it, and had to plan an announcement to save face.
The harsh truth is even if they lose half of their current users they will end up making more anyway, even with the amended changes. They planned to lose a large chunk of their user base, regardless. The “seats” model is dead now that AI is changing how game development is done from the ground up. And they needed to do this because they were never profitable (the engine’s development costs hundreds of millions of dollars) and couldn’t really compete with unreal when it came to the type of customers they could actually pay for the engine from
Sure, but if they’d implemented the revised changes they wouldn’t have lost so many users. And despite their messaging, they did already speak to some devs who’d already told them this would be a disaster, but they tried it anyway, and in a retroactive way that completely disregarded prior promises regarding changing EULA agreements, so there’s no faith in this not still changing.
Nah this went really bad for them. Even if they do make more, it will almost certainly be short term. Godot got so much free advertising. It firmly sat itself next to unreal as far as who should be choosing it, but it is definitely the inferior engine if you are making AAA. It's going to get cut from the high by unreal and the low from Godot, defold, and even gamemaker.
I don't get this weird apologist attitude. Let us not forget Unity just spent over $4 billion less than a year ago buying the malware ad service ironsource. They are not profitable because they make bad business decisions. This was one more. And in all likelihood we will see the sale of unity before too long. And it will probably be less than the $20 billion offer they had prior to the ironsource purchase.
They are not profitable because they make bad business decisions.
Exactly this. Just like how reddit very quickly made enough in reddit gold sales to cover their server costs for decades, the only reason it’s operating at a loss is because they’re running it that way.
it’s a known strategy in tech startups and most non inventory based businesses in general (think moviepass) to undercut your competition to try and get as much market share as possible, even operating at a loss, and then slowly turn up the prices on your users once they are locked into your system and make back the lost revenue over time. I don’t agree with it either, but the y-combinator business tech crowd seem to love this model, so I can’t really say if it’s a bad decision or not.
I can however, point to evidence that it’s a popular business model, if you don’t mind accepting hacker news and y-combinator articles, as well as YouTube media of startup CEOs in earnings calls, but I refuse to defend it otherwise. These are often people with lots of money and advanced stem + business degrees however, so Im not going to sit here and act like I easily know better than them. I can say it did work for Google, but this is after they already were dominating with ad revenue and had the means to slowly introduce ads into every platform they owned ( youtube, maps, android). Popular platforms like DoorDash also have yet to become profitable, despite commanding a 70% market share on food delivery.
I can cite an example of it with an inventory based company. KIA sold their cars at damn near a loss in the US for a long time to get a good foothold. And it worked. Iirc they had a bogo on cars at one point even.
80 percent of unity users don’t pay and a large percentage of the 20% remaining don’t pay close to enough to maintain the engine. they did this on purpose, so it’s their fault, but it is the truth. most large studios these days that actually hit the numbers to pay unity are doing more with AI so they are paying less and those who the changes actually were attempting to make up lost revenue from. as I said, either way the “seats” model is dead regardless.
honestly as shitty as the changes were (and of course they were trying to make profit) they were actually attempting to help devs at least financially. For many use cases the install fee would come out as less than a 1% rev share. It was the other shit that made it worse, the install counting malware proposal, and the uncertainty behind the legitimacy of the numbers. (demos, piracy, repeated reinstalls)
if you’re interested in the insight from a tech investor who is familiar with the situation from the inside, but remains unbiased as someone not employed by unity, check this link for a good breakdown of what Unity’s leadership was actually thinking when they cooked this insanity up.
I thought that earlier response was decent, at the very least it stopped them from sinking any further. This just pushes them further down. They could have just stopped talking about it, but a lousy pathetic excuse like that? There’s no hope for them.
I think they will lose some already established studios that can afford to retool and reskill on another engine. But I think the vast vast majority of current unity developers are breathing a sigh of relief that they /dont/ need to reskill or retool on another engine.
Unity is still on shaky ground, but they have been since they went public. They need revenue, and their big ad revenue plan got ruined by dastardly apple protecting users’ privacy. Couple that with an upstart and promising engine following in Blenders footsteps. In five years, they might have lost every hand they had left to play. Irregardless of the missteps of the last week.
Every indie dev I’m following on YouTube has basically made a “My thoughts on the situation”-type videos where they talk about how they’ve “won against Unity” despite Unity basically doing a textbook of the “Door in the face” technique to pass changes that would’ve been unpopular before this whole mess.
As soon as I heard Unity was back pedaling, I thought “there’s part 2 of the plan”
1: release abusive payment scheme to see just how much push back they get. If push back is minimal or losses are acceptable, end here and enjoy the profit.
2: if push back is strong, implement the actual payment policy that is still a significant increase, but less significant than the one above. And wait until the controversy blows over, which it will.
Yes, lots of developers will leave, lots of developers will choose a different engine for their new games, but there are a ton that will decide that it isn’t feasible to switch engines and plenty that will just eat the added cost. The thing that remains to be seen is just how much damage Unity has done in terms of new projects choosing other engines over theirs.
Claiming it’s “door in the face” is a little crazy here. If this is where they wanted to be, the “bait” changes could have been much much less bad than they were, and they still could’ve walked back to this.
Hell, they could have announced a 10% revenue split and it would’ve looked much better than what they pitched. And they could still walk back to 2.5% and looked like heroes. And it wouldn’t have lost them nearly as much trust. Nor made them look as bad.
If this was what they were trying to do, they’d have to have been even dumber to have made it this bad.
I’m more willing to bet they’re just fucking stupid. Or that a few people on the board had this as a fucking moronic idea, and the rest managed to take back control after it went totally sideways.
But claiming that it’s a door in the face requires them to be evil enough to do it, stupid enough to not realize they’re overdoing it, crazy enough to think it’d work, etc. It seems way too contrived.
Agreed, this whole Unity thing seemed more like they were surprised the peasants were revolting. Completely unaware of the danger of putting developer bills directly in to the hands of the end users, and not considering that a "trust me bro I counted how much you owe me" blackbox accounting method was too much to ask.
…which engine is the upstart and promising engine following in Blender’s footsteps? Do you mean what Unity was supposed to be until they ruined it, or did you forget to drop the name of the engine in question?
Yeah, very few studios would retool an existing project. The real question is whether any of them will be picking unity for their next project. And will young people getting into game dev choose Unity over others? I don’t expect to see a sharp decrease in the number of Unity projects in the next year, but rather a slow descent, while Godot picks up steam and Unreal further cements itself as the professional’s tool.
All the tutorials and learning resources are hyper unity focused. That’s why so many game devs pick it up. That’s why they cornered the less than AAA industry. A young person will choose unity over the others for the same reason as they did last year. The endless resources to teach.
It’s likely almost all developers will pick unity for the next project too. All their knowledge is in unity, not Godot or unreal. We have this problem in other software industries too, some languages and frameworks are just better, but you can’t use them in your project because there are only five people in the industry that know how to use it well.
Oh, I don’t know about that. He’s got this fascade of I’m just like you, but then goes and acts exactly like every other Microsoft exec from the past 30 years. I know you’ll want to compare him to Don Mattrick to excuse that. Oh, he’s a good guy… in comparison. Nope.
He is actively seeking to damage the industry as a whole to further his career ambitions inside Microsoft and to make Microsoft crazy money in 20 years. He’s a bully in the industry, and just like every other big tech disruption from the past decade and a half, from Uber to netflix, none of it is in favor of the consumer
games
Najstarsze
Magazyn ze zdalnego serwera może być niekompletny. Zobacz więcej na oryginalnej instancji.