|
I'm not enough of an optimist to believe that Apple will continue funding development of games in full to feed Apple Arcade like they've been doing. I'm not very positive at all on monthly subscription game services (aka "Spotify for games") in the long run, including Apple Arcade.
|
#
?
Oct 22, 2019 14:39
|
|
|
|
| # ? Apr 19, 2024 06:08 |
|
|
Hyper Crab Tank posted:To be fair, it's not exactly an easy problem to solve. Steam's efforts for years have been geared towards finding a way to get games in front of people that are likely to buy them that doesn't require spending countless man-hours on curation. There are millions of people with Steam accounts with diverse tastes, budgets, available time each week to play video games, computer hardware, what have you. Ultimately Steam's goal aligns with that of both customers and developers - getting games into the hands of the people who want to play them - and there really isn't another platform today that offers what Steam does with the same availability to developers. It would be great if there were for competitive purposes, but there isn't. Ironically, the 30% issue is largely related to that. Steam offers some nice stuff, but they could ask 30% because of their massive dominance of the PC distribution market. As games rely more on other promotional avenues to get noticed, Steam becomes less important to their success and it becomes easier to pull the game off of it. Launcher/storefront fragmentation has set in now and is probably irreversible.
|
|
#
?
Oct 22, 2019 17:51
|
|
|
I mean, the issue is complicated greatly because the cost of making a bad game and getting it on steam is down as well. It's not just good fames.
|
|
#
?
Oct 22, 2019 18:04
|
|
|
I actually think that market fragmentation will help. A key part of the problem is that as Steam or any other platforms grow, the number of games it can give attention to doesn't grow proportionately. If the top 20 games on a platform will always get 50% of sales (numbers plucked out of the hat), a fragmented market with different audiences going to different platforms multiplies those numbers for each platform. Audience money gets spread out more amongst more games.
|
|
#
?
Oct 22, 2019 18:15
|
|
|
mutata posted:I'm not enough of an optimist to believe that Apple will continue funding development of games in full to feed Apple Arcade like they've been doing. I'm not very positive at all on monthly subscription game services (aka "Spotify for games") in the long run, including Apple Arcade. They don't really need to fund games, just offer them a dedicated platform separate from the eventual Russian rip-off of your game and Simpsons Tap poo poo. If I had a game and was thinking about making a mobile release and Apple was like "we'll put it on Arcade" there doesn't seem to be a reason not to take that offer.
|
|
#
?
Oct 22, 2019 19:57
|
|
|
I don't trust that they will pay percentages worth anything. Artists get real screwed on Spotify "listens" and I expect the norm on game sub services will be similar. At least if they're funding development you get your game development paid for and can come out the other side whole. We'll obviously see, though, as I don't think that genie is going back in that bottle any time soon. Gaming in general is set to see the streaming and subscription model experiments out to their results.
|
|
#
?
Oct 22, 2019 23:14
|
|
|
For what it's worth, getting on Microsoft's Game Pass subscription service(s) can be extremely worthwhile. Time will tell if that holds, though.
|
|
#
?
Oct 23, 2019 11:03
|
|
|
al-azad posted:They don't really need to fund games, just offer them a dedicated platform separate from the eventual Russian rip-off of your game and Simpsons Tap poo poo. If I had a game and was thinking about making a mobile release and Apple was like "we'll put it on Arcade" there doesn't seem to be a reason not to take that offer. If they are not funding the development of the game it all comes down to how revenue is worked out. A reason to not take the offer would be if you dont stand to make any money
|
|
#
?
Oct 23, 2019 21:53
|
|
|
What's the case for chromatic aberration as an always-on effect? CA seems to be one of the most frequently complained about graphical effects, from people who say it gives them headaches and eye-strain to people who say it degrades image quality. I have to admit that some of the most outspoken arguments against it come off as a little hyperbolic to me, but at the same time, the effect doesn't really justify itself in an intuitive way - I can't say I've ever consciously appreciated the presence of a CA filter or looked at a CA on/off comparison and thought the on picture looked better. And yet, big budget games continue to come out with CA filters. I've heard all the arguments against it, but not the other side. What's the reasoned defense of / argument for simulating a camera defect?
|
|
#
?
Oct 30, 2019 01:42
|
|
|
Nostalgia for playing old Nintendo games on a busted CRT, maybe
|
|
#
?
Oct 30, 2019 07:04
|
|
|
Lork posted:What's the reasoned defense of / argument for simulating a camera defect? 100% just an aesthetic choice. The art director thought it looked cool. It's no different than any other of the many choices the art director makes during the development of a game.
|
|
#
?
Oct 30, 2019 08:33
|
|
|
There are a lot of post processing filters that look blurry, and will make some people feel sick. They remain popular, however, because they do tend to make the image look subjectively better to a lot of people. The various post processing Anti-aliasing methods are a big culprit. I'm always amazed that people can endure the blurrier varieties of it. The Outer Worlds made me want to puke with the default settings for example, so I might be over sensitive to it.
|
|
#
?
Nov 2, 2019 03:02
|
|
|
There's kind of a combination of good reasons and bad reasons. The good reasons are that it can achieve some artistic effects. They can make things seem less in-focus if that's what you want. Achromatic lenses are kind of interesting, you can control what color fringes you get depending on the wavelengths that you intersect, and wind up with more of a certain pairing of colors in the frame, and they don't produce as much of an out-of-focus sensation as a simple ("rainbow"-producing) lens. The bad reasons... video game graphics have a very long-running "more is better" mentality, and a tendency to adopt new visual effects before they've really been refined, with later games dialing them back to more reasonable levels. OneEightHundred fucked around with this message at 07:42 on Nov 2, 2019 |
|
#
?
Nov 2, 2019 07:10
|
|
|
Little question for graphic performance, if I have an animation in a single flipbook image file does it take a more draw calls than if I just had a static image I "animate" through messing with its movement and transparency thresholds etc or just more VRAM.
|
|
#
?
Nov 14, 2019 00:40
|
|
|
I'm not sure what you're asking. "messing with its movement and transparency thresholds"? Like, animating a transform to apply position, rotation, scale and/or skew? You'll only get very basic animation with that approach compared to animating a sprite, but assuming that's what you meant: If you're only drawing this image once per frame, you'll get the same amount of draw calls regardless. It's not like you can reuse draw calls (not counting modern command recording APIs like Vulkan/D3D12), so you'll have to update your animation values every frame before drawing it. The only difference is what values you're animating--transform animation would have you animate, well, the transform, whereas flipbook animation would have you animate the offset within the atlas. If you're drawing it multiple times per frame, and always in the same "layer" (i.e.: there aren't other images overlapping your current one), then you could save draw calls by using instanced rendering. The distinction in animated values would be the same, but instead of passing a single transform or offset for each image, you'd build a list of these and tell the renderer to do an instanced draw, using each transform/offset for different instances. With modern API fuckery like indirect draws and bindless, you can also combine multiple different images within the same draw call, but I figure if you're asking this question, this is probably overkill for your purposes.
|
|
#
?
Nov 14, 2019 02:30
|
|
|
Yeah pretty much, just checking if there was anything too it beyond the file is usually bigger on a atlas.
|
|
#
?
Nov 14, 2019 06:01
|
|
|
So one of my friends has written music commercially for about 20 years, and is looking to go into video game music. He does some alt rock, heavy metal, and electronica. Also a jazz bassist by education. The whole thing is, he's been writing pieces of music that are made to be short and blend into any possible thing for ads for years (if you listen to Sirius XM, you've heard him). The question isn't talent or portfolio, it's who do you even get in touch with?
|
|
#
?
Nov 23, 2019 01:27
|
|
|
Ugly In The Morning posted:So one of my friends has written music commercially for about 20 years, and is looking to go into video game music. He does some alt rock, heavy metal, and electronica. Also a jazz bassist by education. The whole thing is, he's been writing pieces of music that are made to be short and blend into any possible thing for ads for years (if you listen to Sirius XM, you've heard him). The question isn't talent or portfolio, it's who do you even get in touch with? https://www.moddb.com/forum/board/audio-visual I'd say his best bet is to get involved with a mod that he thinks looks promising. It won't be paid work but it'll give him an idea of the specific music and audio needs that come up in the course of developing a game. From there he could parlay his experience working with a mod team or two into contacting commercial studios that will appreciate that he's done work specifically for game projects on top of his existing commercial work and maybe give him a shot. Although if he's got stuff that's playing on radio, I'd assume he's already got an agent or some kind of connection who could probably sniff out a paid game project for him. And failing that, he could just start e-mailing random studios out of the blue and asking them if they need any audio work. https://gamedevmap.com/ Pick a city, (or timezone if he wants to work remotely) start going through the list, find their contact info, and send an e-mail. If talent & experience aren't his issue, I know there are a lot of studios (especially independent ones) who would love to have a professional composer fall into their lap for some contract work. good luck goon ('s friend)
|
|
#
?
Nov 23, 2019 18:55
|
|
|
Whistling rear end in a top hat posted:https://www.moddb.com/forum/board/audio-visual Excellent, this is exactly what he's looking for. He's got one guy he works for with that kind of thing, so no agent per se. We'll get starting on emailing indies this week!
|
|
#
?
Nov 23, 2019 22:58
|
|
|
Whistling rear end in a top hat posted:https://gamedevmap.com/ Pick a city, (or timezone if he wants to work remotely) start going through the list, find their contact info, and send an e-mail. If talent & experience aren't his issue, I know there are a lot of studios (especially independent ones) who would love to have a professional composer fall into their lap for some contract work. A note of advice for emailing them, copy+paste messages really aren't going to cut it. Audio people trying to move into games are the second most common emails we get, and if its a generic copy+paste email its just going to get our generic copy+paste rejection without even listening to the links. There's so many coming in that we can afford to cherry pick the ones who are enthusiastic for us and our project - at a bare minimum reference the studio name, game, and how their music will fit in.
|
|
#
?
Nov 24, 2019 01:31
|
|
|
Whistling rear end in a top hat posted:...get involved with a mod that he thinks looks promising. It won't be paid work but ... If you got 20 years of professional audio/music experience I'm not sure going the unpaid mod work is the way to go considering that a lot of those turn into trainwrecks, never get completed or are run by hobbyists or amateurs. That would be like telling an artist who worked in feature animation or film for 20 years to go do unpaid work on a mod before applying for / switching to a games job. Que the if you are good at something, never work for free meme... From my experience, most of the audio I've been involved with has been outsourced to another company, or we get in contact with an agent of a more famous composer. On my last project, we hired Bear Mc Creary to compose our soundtrack while the mix and everything else was done at a big sound studio in LA/Burbank [I can hunt for the name of that place if you need it], and this was at a studio where we had a huge internal sound department with about a dozen mixing/sound rooms for production. The current place I'm at has one guy mixing and hooking up audio to state machines while everything else is outsourced. I'd probably try to work with those sound mixing companies [they don't just do games, they also do all sorts of stuff from motion pictures to tv as well]. On that side, you may be better off reaching out through your radio contacts unless you are trying to hit the indy game dev developer...
|
|
#
?
Dec 2, 2019 17:20
|
|
|
I’d also like to point out most projects aren’t going to hire a composer or spend any extra time or money on music (or audio in general). It’s one of the last things on anyone’s minds typically. Especially if your on a licensed project that has established assets available (e.g, StarWars). Most of the time we’re hunting for looped music tracks off of a stock asset site that fits our needs and more importantly, is royalty free and not wildly expensive. I’ve had cases we’d find a great set of tracks and buyout an artist’s offerings and call it done. I think for you it’s more realistic to sell tracks on assets sites or directly apply for a slot on the dev team with kick rear end samples than waste time with working for free on someone’s mod, unless that mod is yours for like a sound overhaul type of project... there are some great sound mods for Skyrim as an example.
|
|
#
?
Dec 3, 2019 05:14
|
|
|
Ugly In The Morning posted:So one of my friends has written music commercially for about 20 years, and is looking to go into video game music. He does some alt rock, heavy metal, and electronica. Also a jazz bassist by education. The whole thing is, he's been writing pieces of music that are made to be short and blend into any possible thing for ads for years (if you listen to Sirius XM, you've heard him). The question isn't talent or portfolio, it's who do you even get in touch with? How about just putting some of his music on game asset stores? I guess its kinda like, is he looking to contribute to a specific game project or just looking for side-work / income? Posting audio on Unity store or Unreal store should be pretty straightforward, tons of people sell music clips and loops on there.
|
|
#
?
Dec 9, 2019 00:50
|
|
|
Hyper Crab Tank posted:I'm not sure what that article is trying to convey, really. Not a lot of those people mentioned what they actually got from sales, it's mostly about various government financial aid programs. I get the feeling from what little information there is in the article that a lot of these people didn't really have a lot of financial stability to begin with. Look, cold hard reality is that small-time video game development is a high-risk venture at the best of times, and if you're scraping by as it is then solo or small-group independent development is not the horse you want to hitch your wagon to. There is too much noise out there to be risking your health and financial future on a product that doesn't stand out, and even then it's a crapshoot unless you have contacts with a good publisher that can help you with marketing. (And, to be even more frank, this goes double if you're making some incredibly niche and/or artsy-rear end title that doesn't have a lot of mass appeal even if people do notice it.) I'm so glad to to see this reasonable post. The sentiment I share for both games and creating a (small) business in general is that you of course have the right to do these things but you are not entitled to doing them for a living. So many small businesses close down because their product was a passion project and that's what they wanted to do. In the end what you want to do is limited by what is financially viable and ignoring that is fatal. If you stop working a day job to work on a game without a worst case scenario living fund, decide your niche dream project is a good first game only to find out that it isn't selling, or think consumers or storefronts owe it to you to keep you afloat via sales and exposure, I don't know what to say. The passionate hobby vs. viable product lines get really messy in at least the indie portion of this industry. We're fortunate enough to follow and/or participate in a creative industry that has increasingly promoted transparency at the low-mid level with honest postmortems, retrospectives, and other cautionary tales that are generally available at the same time they are relevant. I'm scared as poo poo thinking about registering a LLC, writing contracts, dealing with different storefronts, and facing the risky reality beyond the idealism of making games so I'm sure as hell preparing for failure and the ownership of that failure.
|
|
#
?
Dec 12, 2019 21:26
|
|
|
Just want to start this post by saying this thread was a fantastic read, and I appreciate the insight and anecdotes all you devs have posted! (Also, very much appreciated the link to the open sourced Star Ruler 2 posted earlier by one of the devs, that was a pretty cool move!) I started learning how to program a couple years ago in conjunction with research/work. Beginning with python and subsequently C++ (dabbled in Rust as well), and I am by absolutely no means an expert. I got absolutely hooked on programming and it's taken over virtually all my hobby time, however. So naturally, since I love videogames, I did what any reasonable person would do and went "I'm totally gonna make a videogame." So in general I have no idea what I'm doing, but I'm having a blast doing it regardless  In that context, I have a couple questions I hope maybe someone could share some of their experience and insight into. Regarding code structure / architecture: -Is there a good source to learn about different ways to structure the code base and generally set up a project? -Are there any paradigms that are generally regarded as industry standard? -Or is this something that varies widely depending on where you are, or even what type of game you are making? -How did you learn how to do this as a dev? Just experience, trial and error, or mentorship? I hope the questions aren't too vague, and if they're a bad fit for this thread I apologise in advance, but I thought it would be neat to hear from actual professionals! It's hard to sort what is actually good practice from what is subpar when looking at various tutorials, etc. Add that to the fact that most of what I've learned has been from a couple books, tutorials and the documentation of the library I used (SFML), I don't really know exactly what to search for to learn more about this. Any advice that helps me refine how I code is very welcome! Thanks in advance!
|
|
#
?
Jan 8, 2020 09:23
|
|
|
forkbucket posted:Just want to start this post by saying this thread was a fantastic read, and I appreciate the insight and anecdotes all you devs have posted! (Also, very much appreciated the link to the open sourced Star Ruler 2 posted earlier by one of the devs, that was a pretty cool move!) Most teachings from industry are available as conference talks. With few exceptions, game programming books are trash. ECS is hot right now, and it may be interesting to look into Sweeney’s work on transactional state. Avoid holding lots of global mutable state. Even if in industry there’s still loads of it for bad reasons. I came from a traditional CS background as an engineer, but it doesn’t really matter where you pick up data structures and algorithms. I find the more academic sources better for those. Doing stuff on your own it probably doesn’t matter, but games programming interviews aren’t dissimilar to normal tech interviews. Just start making stuff. Tetris using print statements in a loop with a timer. Then pick a game engine and make things that are actually interesting to you.
|
|
#
?
Jan 8, 2020 13:51
|
|
|
Honestly if you want to start learning how to program games, pick up one of the many engines out there and start making something. You can write your own on the side in the mean time, but nothing would get you more immediately familiar with how games works than working on a pre-existing engine for your first game project. If you want to be more intimate with internals, pick up one with source (i.e. UE4) so you can step into engine code. As for general arch stuff, as leper khan said, ECS and setting up your code to be as immutable first are getting pretty popular right now. ECS is based on a concept called data-oriented design, in which the general philosophy is to have your data setup to be as close together and compact as possible in order to maintain cache locality (the cache we're talking about here is the L1/L2 cache on your processor, which is a super ultra fast bit of memory directly on the chip that the memory controller puts main system memory data into). For example, rather than having a million door objects out in the memory heap, you would keep all of your door object data in an array. This way, you can fit as much door objects in a single cache line as possible, not triggering another read from main system memory until you've read through the cache line's worth of objects. Immutability is something I've wished I was taught earlier to do, but basically means everything should be const (in C++ terms) by default unless you've determined whatever variable/object/member function shouldn't be. It would seem easier to just keep going without bothering with the const qualifier, but I do it this way personally so I can put some thought into what variable/object/member function should actually be mutable or not in the context I'm using it. Keeping things immutable as possible means less chance of a random change breaking things. Architecture can and will vary from game to game. Only ones that keep up their similarities are maybe games that use an engine, but even then I've seen wildly different Unity game source code bases. As far as how I've learned, I've started tooling around with an OpenGL engine thing of sorts for a graphics class I took in university for my CS degree. I've created two actual game things with Unity and messed with open source XNA development in the year or two following graduation, and eventually found myself in a studio that was Unity heavy at the time. Since then, I've gotten pretty lucky as the studio was getting more and more contracts that involved more AAA titles and C++; studio got eventually bought out by ZeniMax and eventually made into a BGS studio.
|
|
#
?
Jan 9, 2020 18:47
|
|
|
forkbucket posted:So in general I have no idea what I'm doing, but I'm having a blast doing it regardless If you haven't been checking it out already, the Making Games Megathread is full of people writing their own games and learning how to do it, spanning all manner of skill levels. A great resource if you want to know where to turn next.
|
|
#
?
Jan 9, 2020 23:30
|
|
|
Appreciate the input! I've looked into ECS and that seems to be the norm for the various tutorials I've seen also. I'm not really making a game for any particular reason other than "It's fun!" and as an excuse to become more proficient at coding to be honest. But I'll definitely work on a deeper understanding of ECS, it seems logical from what I've read. I have installed Unity, but the interface was a bit daunting and I haven't gotten around to doing an actual this-is-how-to-use-it tutorial yet but it's on my to-do list. Limiting mutable globals is a good tip! I'll be sure to keep that in mind. Thanks again 
|
|
#
?
Jan 9, 2020 23:34
|
|
|
Gromit posted:If you haven't been checking it out already, the Making Games Megathread is full of people writing their own games and learning how to do it, spanning all manner of skill levels. A great resource if you want to know where to turn next. I found that thread via this thread actually! Very cool thread with good resources
|
|
#
?
Jan 9, 2020 23:35
|
|
|
I used to work at a pretty big studio doing 360 and PS3 dev but that was forever ago. Now I work in a totally different but related industry. Anyway when we were working with 360/PS3 we were always up against the memory limit that by todays standards is laughable at 512mb. My question is, are developers still working right up to the limit of consoles even if consoles end up going to 16GB/32GB?
|
|
#
?
Jan 10, 2020 04:40
|
|
|
Shaocaholica posted:My question is, are developers still working right up to the limit of consoles even if consoles end up going to 16GB/32GB? If there is a limit, a game will find it. Particularly since being efficient with memory has a cost in engineering time (and, to some extent, art) that isn't otherwise visible to the public, and being wasteful with memory is often a performance boost.
|
|
#
?
Jan 10, 2020 09:33
|
|
|
Shaocaholica posted:I used to work at a pretty big studio doing 360 and PS3 dev but that was forever ago. Now I work in a totally different but related industry. In mobile you’re still fighting it, yeah. It’s 1-2gb you’re complaining about rather than 512mb, but people want things to look real nice. And Apple will cut off your head for allocation velocity, not just size.
|
|
#
?
Jan 10, 2020 13:20
|
|
|
Until very recently, if you wanted to support every iPhone that was still supported by the latest OS, your limit was in the range of 650 MiB before certain phones would just force-kill your application, so not far off.
|
|
#
?
Jan 10, 2020 18:43
|
|
|
We were still fighting to fit in memory on Xbox One, which meant we usually didn't have to worry at all about fitting into memory on Xbox One X, outside of stuff like memory leaks. But presumably when Xbox One X or Series X becomes the baseline for your game, everything will expand do fill the space, you'll get less strict with streaming, and you'll be right at the borderline again. Especially if 4k rendering becomes the new baseline, people will want to increase texture resolution so we're not sampling low frequency textures at high frequency in screen space... and every time you double the dimensions of your texture it'll quadruple in memory usage.
|
|
#
?
Jan 10, 2020 19:52
|
|
|
Shaocaholica posted:I used to work at a pretty big studio doing 360 and PS3 dev but that was forever ago. Now I work in a totally different but related industry. quote:My question is, are developers still working right up to the limit of consoles even if consoles end up going to 16GB/32GB? Are games using most of the RAM they're given? Yes, of course, they're always gonna run up RAM because any RAM that you're not using is RAM you could be using to boost texture detail or something like that. Are they still having to tune things around the RAM budget? Yes, because see the previous answer. However, a lot of things that could take a pretty big bite out of main RAM on the last generation and were hard to reduce at all (like executable size, and various types of game data) have not scaled up as much as the RAM limit, so more of it is being used on fewer things that have more room to be scaled down if needed. OneEightHundred fucked around with this message at 20:12 on Jan 10, 2020 |
|
#
?
Jan 10, 2020 20:08
|
|
|
I wonder if artists are having trouble tho authoring those bigger textures. Sure it’s nice to have a higher limit but then you have to fill it with something and I feel like bigger textures are just harder to paint without adding just fluff (at a certain point). I’m sure there are really good painters out there but big canvas doesn’t guarantee better art. I’m sure this was the same ‘dilemma’ going from any generation to the next. ‘Look what we did with pixel art in mgs1!’
|
|
#
?
Jan 10, 2020 21:02
|
|
|
We don't paint textures much anymore, especially for anything even somewhat realistic. As for stylized art, we do the same as we've always done, just in higher resolutions. Still, though. There aren't really any games even still that are slapping up tileable 4k brick textures or whatever, so resolution-wise I'm still authoring at similar resolutions that I have for years, I just might not downres them as much as I used to when I ship them. Realistic game environment art is lots of photoscanned sourcing and building materials via procedural tools like Substance Designer. Tools and artistic capabilities have tended to outpace an engine's or console's ability to push the art so every new generational shift tends to just accommodate what artists seem to already want to do. We also have paradigms of cg art creation that reach all the way into animated film and film special effects, so there's plenty of knowledge on how to push things harder higher faster etc. mutata fucked around with this message at 22:04 on Jan 10, 2020 |
|
#
?
Jan 10, 2020 22:00
|
|
|
Sorry I meant 'paint' in a more general way not literally. I know its gotten more technical with even more non 'albedo' maps as well which will benefit from more memory.
|
|
#
?
Jan 10, 2020 22:46
|
|
|
|
| # ? Apr 19, 2024 06:08 |
|
|
Yeah, I guess I tried to cover the literal and figurative meanings at the same time, heh. The point is, yes, game art does absolutely get more time and work intensive, and more complex as the expectations rise with the tech, but so do tools and workflows. We definitely aren't where we were in the PS1 days or the SNES days where a AAA title could have been made by a modern-day-indie-sized team, but that chicken left the henhouse long ago.
|
|
#
?
Jan 11, 2020 00:23
|
|