What is an acceptable length of time between updates ?

[tanstaafl]

However long it takes to get it right.

 

[forzato]

Look I know the quality of releases has improved significantly over the last two years. The problem is many developers are burning out whether they will admit it or not after a couple of big releases. Radiant is a good example. Devs feel the pressure that each episode needs to be bigger than the last or comparable. Life circumstances do come up and this has been a difficult year for all of us so it's easy to dismiss. Why not split releases up in parts? A good example of doing this properly is Leap of faith. The major episode story arc has been completed but there is 2/3 of the side plots still to be covered. This significantly reduced the release time preventing a dead period for 8 months but still satisfies the community & patrons.

 

[khumak]

To me 4-6 weeks is a reasonable target. A little more or less is fine but much less than that and you're wasting too much time testing. Much more than that and people start complaining about it being abandoned.

A big update might take longer if it's something that can't really be broken up into smaller chunks but 4-6 weeks is enough time to develop a decent chunk of content and then test it. I think once you get past about 2 months or so people start losing interest. If you're going to go 6 months between updates you might as well just wait til the whole thing is done and then release the final version. People are going to be constantly complaining about it being abandoned or whatever if your dev cycle is that long.

 

[RanliLabz]

To me 4-6 weeks is a reasonable target. A little more or less is fine but much less than that and you're wasting too much time testing. Much more than that and people start complaining about it being abandoned.

A big update might take longer if it's something that can't really be broken up into smaller chunks but 4-6 weeks is enough time to develop a decent chunk of content and then test it. I think once you get past about 2 months or so people start losing interest. If you're going to go 6 months between updates you might as well just wait til the whole thing is done and then release the final version. People are going to be constantly complaining about it being abandoned or whatever if your dev cycle is that long.

It hugely depends on the project - drawn or CGI, Daz or Honey, animations or stills, Ren'py or Unity, solo dev or team, full- or part-time... A major factor is whether the update is new and original content, or just existing renders with new dialogue on top and a ton of grind to fill it out.

For a solo-dev CGI project 4-6 weeks is more like best practice than a reasonable target - which is why so few devs manage it. Generally, players are looking for a substantial update with at least 300 new renders and maybe an animation, and around 3 scenes to make the download worth it. That's a lot to get through in 4-6 weeks - the rendering alone is possibly 10 hours a day with a decent rig - and you've got to have written the script, created the images, coded, soundscaped, done the variables, tested, packaged, marketed and released (all while running your sales platforms like Patreon and player platforms like Discord). That's before you account for having to design an unplanned character or location, or encountering some real-life variable like getting ill or moving house.

I keep myself to a 4-8 week schedule because I'm full-time - and it's still always a race! That's because I don't think it's acceptable for a Patreon creator to accept pledges for more than a month with no new content (and I try to fill in any dead-months with a Patreon-only comic, animation, or minigame). However, both creators and players need to keep realistic expectations if they want the releases to be of a decent quality.

That said, the mickey-taking from some of the big projects is shocking - I'm talking Summertime Saga and Milfy City, etc. With their money and teams, they should be able to put out a great big Dr PinkCake size update every month, or a normal one every week. I genuinely don't know why anyone would pledge to them after the last year or so - especially since the more money they get the less they produce!

 

[khumak]

It hugely depends on the project - drawn or CGI, Daz or Honey, animations or stills, Ren'py or Unity, solo dev or team, full- or part-time... A major factor is whether the update is new and original content, or just existing renders with new dialogue on top and a ton of grind to fill it out.

For a solo-dev CGI project 4-6 weeks is more like best practice than a reasonable target - which is why so few devs manage it. Generally, players are looking for a substantial update with at least 300 new renders and maybe an animation, and around 3 scenes to make the download worth it. That's a lot to get through in 4-6 weeks - the rendering alone is possibly 10 hours a day with a decent rig - and you've got to have written the script, created the images, coded, soundscaped, done the variables, tested, packaged, marketed and released (all while running your sales platforms like Patreon and player platforms like Discord). That's before you account for having to design an unplanned character or location, or encountering some real-life variable like getting ill or moving house.

I keep myself to a 4-8 week schedule because I'm full-time - and it's still always a race! That's because I don't think it's acceptable for a Patreon creator to accept pledges for more than a month with no new content (and I try to fill in any dead-months with a Patreon-only comic, animation, or minigame). However, both creators and players need to keep realistic expectations if they want the releases to be of a decent quality.

That said, the mickey-taking from some of the big projects is shocking - I'm talking Summertime Saga and Milfy City, etc. With their money and teams, they should be able to put out a great big Dr PinkCake size update every month, or a normal one every week. I genuinely don't know why anyone would pledge to them after the last year or so - especially since the more money they get the less they produce!

Yeah every project is different. I think the main point is you need it to be frequent enough to hold people's interest but long enough to actually put out a substantial update without wasting half your time testing. You also need to at least give your fans a ballpark for how long you expect your updates to be whether that's 4 weeks or 3 months. If you miss a target by a week or so that's not a big deal but if you miss it by 3 months that's kind of unreasonable.

 

[RanliLabz]

Yeah every project is different. I think the main point is you need it to be frequent enough to hold people's interest but long enough to actually put out a substantial update without wasting half your time testing. You also need to at least give your fans a ballpark for how long you expect your updates to be whether that's 4 weeks or 3 months. If you miss a target by a week or so that's not a big deal but if you miss it by 3 months that's kind of unreasonable.

True. It depends on the game, but even for a large Ren'py project there shouldn't be much testing (mine takes 1-3 days and I have a huge number of variables and paths and walkthroughs, etc.). That's unless your talking about test-rendering, which might add a bit longer, but is done as you go. Anyone saying that they're spending weeks in testing in probably bullshitting you or incapable of coding.

Being upfront is the most important thing - if patrons know you take 2 months to update that's their choice, but if you keep missing it by weeks or months then it's the dev's fault. Unfortunately, newbie devs often underestimate the amount of time it takes for an update, and make unrealistic promises they're scared to admit they can't achieve... leading to a cycle of missed deadlines.

 

[rayminator]

it depends on you on how you feel about updating your game.
if you feel great about the update then update the game.
if you don't feel great about it then don't update the game keep on working on it until you do feel great about it.

just don't rush to push a update out and don't listen to people asking for an update or you will create bugs in the game
and for the last thing is not to promise anything.

 

[khumak]

True. It depends on the game, but even for a large Ren'py project there shouldn't be much testing (mine takes 1-3 days and I have a huge number of variables and paths and walkthroughs, etc.). That's unless your talking about test-rendering, which might add a bit longer, but is done as you go. Anyone saying that they're spending weeks in testing in probably bullshitting you or incapable of coding.

Being upfront is the most important thing - if patrons know you take 2 months to update that's their choice, but if you keep missing it by weeks or months then it's the dev's fault. Unfortunately, newbie devs often underestimate the amount of time it takes for an update, and make unrealistic promises they're scared to admit they can't achieve... leading to a cycle of missed deadlines.

Haha, no kidding on the testing especially for a 1 man shop. I've only done modding so far but for me if I did a 1 month update for my mod that meant I spent at best 2 weeks on rendering, maybe a week on coding, close to a week on dialog, and at BEST a few hours testing. Testing ALWAYS gets the short end of the stick. Now in my case I do actually test each event as I'm writing it but I don't do any sort of exhausive QA testing just prior to release. I do 1 or 2 quick playthroughs, not reading anything, just testing each major path to make sure nothing crashes. Done. My fans usually caught a few crash bugs or broken image links I missed in each update and as long as they actually post the crash log I would fix it that same day.

I suspect for 1 person doing a game part time, a 2 month cycle would probably make more sense as a minimum. For an actual team (like 1 person each for coding, writing, and art)? You should be able to do a BIG update every 4 weeks easily. That's equivalent to 1 person having at least 3 full months per update if they're doing it themself.

 

[RanliLabz]

Haha, no kidding on the testing especially for a 1 man shop. I've only done modding so far but for me if I did a 1 month update for my mod that meant I spent at best 2 weeks on rendering, maybe a week on coding, close to a week on dialog, and at BEST a few hours testing. Testing ALWAYS gets the short end of the stick. Now in my case I do actually test each event as I'm writing it but I don't do any sort of exhausive QA testing just prior to release. I do 1 or 2 quick playthroughs, not reading anything, just testing each major path to make sure nothing crashes. Done. My fans usually caught a few crash bugs or broken image links I missed in each update and as long as they actually post the crash log I would fix it that same day.

I suspect for 1 person doing a game part time, a 2 month cycle would probably make more sense as a minimum. For an actual team (like 1 person each for coding, writing, and art)? You should be able to do a BIG update every 4 weeks easily. That's equivalent to 1 person having at least 3 full months per update if they're doing it themself.

I’d agree with a lot of that (a part timer can at least leave renders going while they’re at work).

Teams are tricker - the law of diminishing returns will mean that a 3-member split responsibility team can’t do 3x the work of a solo-dev (they’d be lucky to manage twice the amount). I suspect the best way to work a team is for each member to behave like a solo-dev on one branching path.

 

[Asia Argento]

It depends on many factors... including how honest the dev is and if he keeps the community up to date. Lets face it, behind all games is a human with problems, etc. To expect a machine like schedule of someone is kinda wild... Dark Cookie (Summertime Saga) has intermittent updates, but with the amount of content you are offered, I think its worthwhile to support his work...

Then there are some like Arcgames/Corrupted Kingdoms that offer WEEKLY updates, many of which expand the story...

I guess its all about how YOU feel about certain devs. Is their work worthy of your patronage? Not everyone feels DC or Arc are perfect for different reasons, so its up to you to determine if each’s dev cycle is good or bad...

 

[khumak]

I’d agree with a lot of that (a part timer can at least leave renders going while they’re at work).

Teams are tricker - the law of diminishing returns will mean that a 3-member split responsibility team can’t do 3x the work of a solo-dev (they’d be lucky to manage twice the amount). I suspect the best way to work a team is for each member to behave like a solo-dev on one branching path.

Yeah I could actually see a solo dev per branch model working. I hadn't thought of that before but you wouldn't lose much efficiency there other than the integration phase where you have to patch them all into 1 project.

I got the feeling that most teams were more along the lines of person A is just doing the writing, person B is just doing the coding, person C is just doing renders. Presumably each of those people are good at that 1 thing and might have little to no skill at the other aspects. With that model you're probably right that there would be a loss of efficiency. Each person finishes up their stuff and you don't necessarily have a 1 for 1 match as far as what each person thought should be happening. So the writer has to rewrite some dialog to match the renders and coding. The art guy has to add some extra renders to fit the dialog, etc.

In my case if I got successful enough to get a partner I would probably want to add someone who was good at something that I'm not so good at. An animator would be the obvious choice for me. So I might send him a few renders that show what I'm going for in the scene and then he does 50 or so frames per scene to turn that one image into an animation. That sort of thing. Very little wasted effort there. Or maybe I'm bad at writing so I send a very rough script to a writer friend who completely rewrites it and makes it actually sound good. Not as efficient as the animator but still not bad.

When it comes to coding I quickly discovered that it takes a lot less time to write a small section of code and test each little section and bug fix each section as I go than it does to write out a big section of code and then slog through the whole bug infested crashfest trying to figure out where it all went wrong at the end. So for me at least, the coder and the tester are always going to be the same person.

 

[HarveyD]

As many people have said, every studio is different. Larger and better supported teams can and are expected to do more. Personally I feel the release timing should remain the same though, roughly 1 month is usually ideal. But the amount of content and quality will differ of course. Also not a fan of small devs asking for more than $5 per month if it's just something they're doing as a hobby.

 

[RanliLabz]

Yeah I could actually see a solo dev per branch model working. I hadn't thought of that before but you wouldn't lose much efficiency there other than the integration phase where you have to patch them all into 1 project.

Yes, I'm not sure if anyone actually does this. Lol, if any proven solo devs are looking for work - get in touch with me

I got the feeling that most teams were more along the lines of person A is just doing the writing, person B is just doing the coding, person C is just doing renders. Presumably each of those people are good at that 1 thing and might have little to no skill at the other aspects. With that model you're probably right that there would be a loss of efficiency. Each person finishes up their stuff and you don't necessarily have a 1 for 1 match as far as what each person thought should be happening. So the writer has to rewrite some dialog to match the renders and coding. The art guy has to add some extra renders to fit the dialog, etc.

That seems to be how it's done - but I'm not sure it really works outside a physical office (okay, Tlaero and Mortze, but they were unusually good). The passing everything back and forward and giving one another instructions must be exhausting - I know that my own attempts to commission coding and art for SpaceCorps have been more hassle (and cost) than they were worth.

In my case if I got successful enough to get a partner I would probably want to add someone who was good at something that I'm not so good at. An animator would be the obvious choice for me. So I might send him a few renders that show what I'm going for in the scene and then he does 50 or so frames per scene to turn that one image into an animation. That sort of thing. Very little wasted effort there. Or maybe I'm bad at writing so I send a very rough script to a writer friend who completely rewrites it and makes it actually sound good. Not as efficient as the animator but still not bad.

That makes a lot of sense (although decent animations are typically a lot more frames than that!) One of the things to take into account is value added, tho - are you happy to hand over a grand or 2 a month for a few looped animations that make up a tiny fraction of the gameplay? And will your putative patrons see that as a good use of their pledges?

As far as the script goes - be aware that handing it over to someone else effectively gives them creative control of the project (e.g. it will be them giving you a list of the renders they need to account for any changes).

When it comes to coding I quickly discovered that it takes a lot less time to write a small section of code and test each little section and bug fix each section as I go than it does to write out a big section of code and then slog through the whole bug infested crashfest trying to figure out where it all went wrong at the end. So for me at least, the coder and the tester are always going to be the same person.

Totally. It makes sense for that to be the same person as the writer too, or you've got a lot of hands in the script and mistakes get made. I personally see the writing-coding-testing-soundscaping more as a single 'Director' role, with the art and graphics side as 'Cinematographer'. This is vital when it comes to editing the pauses, transitions and sound-effects.

 

[lawfullame]

For me, my VN is mainly a hobby project, so the time for the update is: When it's ready.
Unfortunately, the interval between Chapters 1 and 2 was 9 months.
It's not that I have a lot of support on Patreon, but I still enjoy creating my VN. I hope to finish Chapter 3 faster, but there are things like work and my family.

I am not able to promise the release dates of my updates, but I tell my Patrons in advance how many monthly Patreon payments I will accept until the completion of the next chapter and then pause the billing cycle.

Is such an approach fair? I think yes.
Is it acceptable? For a few Patrons yes, for many potential patrons no, but I will continue in the same way anyway, because if I want to continue creating my VN, I don't have many other options.

Of course, a few dollars from Patreon is nice, but the main source of my income is my normal job, so fortunately I'm not dependent on what is acceptable to potential Patrons.

 

[Fliptoynk]

4 months... make it 5!!! cuz it pains me to download 1GB, 3GB, 4GB++++ per update, on a metered data, then a couple of days dem go whoooops! hot bugfix update fuckazz! yo gotta be whacked bat shit crazy kidding me after taking ya loooooooong time to release the update.

just kidding just take your time........................................

 

[anne O'nymous]

[..] to what most people feel is an acceptable length of time and how much content is needed to be considered a real update?.

It totally depend of the ratio between the time needed for the update, the amount of content (both in terms of CG than dialog), and the quality of the said content.

A game like WVM, that have, among other things, none smoothed out of the box CGs, lighting problems and visual inconstancy, should never need more than its usual delay between two updates (the last one being an exception due to unforeseen circumstances). The updates are clearly rushed and, while the game is mostly interesting, the visual is average and the story rely too much on deus ex machina and useless cliche.
This while games like Heavy Five or The DeLuca Family, by example, worth the wait between each updates. Each ones imply many hours of play ; the last update of Heavy Five was over 5 hours, more or less 7 updates of WVM, and the incoming update will be bigger than what is already released with the two first chapters. And they also have over the top quality for their CG and writing. It's visually a pleasure to play those games, seeing the attention their dev put on every details and how what you see is deeply related to what you read, giving you the information not written in order to fully understand the situation ; as much as it's a pleasure to discover the subtlety of the writing. Their dev clearly past hours working on each CG and each dialog lines, in order to make them perfectly accurate to the situation, and as real as possible.
And obviously, there's games like Milfy city, by example, that have updates ten times longer than what they should be.

Globally speaking, it's the feeling you'll have at the end of the update that define if it was too long or not. As long as you feel that it worth the wait, then the delay wasn't to long for you.

A rare few developers are able to put out quality updates every month and others take 6+ months.

Not a single developer is able to put out quality updates every month, not... a... single... one.

Quality need time, it's not something that comply with regular updates. You never know beforehand how many times you'll need to make a new location looks exactly like you want it, nor how many times you'll need to make a new clothes feel real (by opposition to all those latex-like clothes that follow every single curves of the body). You never know beforehand how many times you'll have to redo a CG for it to effectively look good, without poking through parts, bad or plastic-like effects due to the light, and all. Same, whatever how far you goes in the pre-writing of your story, you can't know beforehand how well the dialogs will goes with the visual. And obviously, you need to write your code idea before you know if it will effectively works, or if you'll need to past a week to debug it.
Then come the testing phase, and the mandatory correction phase that come with it. One quarter of the time between two updates is due to the test and correction process ; at least for those who effectively test and don't throw the update, crossing their fingers for it to more or less works, without too many typo in the dialogs, too many wrongly named movies or images, and too many bugs in the code.

Developers have to make a choice. Either they release an update every month, and they'll do at most something average, or they seek for quality, and they'll release an update every now and then, when they finally reach the point where they want the update to end.
Obviously, there's some who abuse, but globally speaking, updates take longer nowadays because the quality have been raised. And, personally, I'm good with this. I can wait more without problem if it mean that I'll enjoy playing the update, without being annoyed by the childish looks of a dialog line, the 10 years late quality of some visual, and without having to correct by myself the silents errors that break everything.
This also apply to 3D CG made with an Illusion game, or game with 2D CG. Globally speaking the quality is better now than it was two daysyears ago (Seriously AON, proof read you more seriously). And obviously it need time to achieve this.


Edit: For obvious reason.

 

[Guntag]

It depends on how good the game is ! The better the game, the faster the updates !

 

[Joshua Tree]

I keep myself to a 4-8 week schedule because I'm full-time - and it's still always a race! That's because I don't think it's acceptable for a Patreon creator to accept pledges for more than a month with no new content (and I try to fill in any dead-months with a Patreon-only comic, animation, or minigame). However, both creators and players need to keep realistic expectations if they want the releases to be of a decent quality.

As long as I know the creator is working on the project. Interacting and communicating with his/hers fans/community. I don't really see any reason why not keep pledge even a month passes by without any update. I know its coming when it's done.

Use "Naughty Road with light of my life" as example. He release a full chapter at a time. Sure it takes months between releases, but they worth it too. So you mean it's not acceptable of him to accept pledges in months between releases? When I pledge to a creator I do it because I like his/her work and excited about next release, but the time schedule doesn't matter all that much for me as long as I know the project is being worked at.

I know people got more sceptic over time due to some creators that not exactly come across as "hard working", or just started to let it slide once they got a "comfortable income" through Patreon". I do look at that as "easy come, easy go" though. Because people will start vote with their money when things start to slide. Ruin your reputation take a lot shorter time than try to fix it again. Some creators seems to have core fans that stick to them regardless though, but that is nothing new when you look at fandoms and such else in the world.

 

[RanliLabz]

As long as I know the creator is working on the project. Interacting and communicating with his/hers fans/community. I don't really see any reason why not keep pledge even a month passes by without any update. I know its coming when it's done.

Use "Naughty Road with light of my life" as example. He release a full chapter at a time. Sure it takes months between releases, but they worth it too. So you mean it's not acceptable of him to accept pledges in months between releases? When I pledge to a creator I do it because I like his/her work and excited about next release, but the time schedule doesn't matter all that much for me as long as I know the project is being worked at.

I know people got more sceptic over time due to some creators that not exactly come across as "hard working", or just started to let it slide once they got a "comfortable income" through Patreon". I do look at that as "easy come, easy go" though. Because people will start vote with their money when things start to slide. Ruin your reputation take a lot shorter time than try to fix it again. Some creators seems to have core fans that stick to them regardless though, but that is nothing new when you look at fandoms and such else in the world.

You make a good point. I, personally, feel uneasy going for too long without an update - but where creators are upfront about it, clear about what they're doing, and have a track record of delivery it is acceptable. A lot of my objection has to do with being burned as a Patron by promising creators who just suddenly vanish or slide.

Then come the testing phase, and the mandatory correction phase that come with it. One quarter of the time between two updates is due to the test and correction process ; at least for those who effectively test and don't throw the update, crossing their fingers for it to more or less works, without too many typo in the dialogs, too many wrongly named movies or images, and too many bugs in the code.

I agree with most of your post - but I think that spending a quarter of development on test and correction is extreme (especially for a renpy game). Sure, it takes a day or two to test the code and typos, and a couple more for rerenders, fill-in renders, extra expressions and painting out poke-throughs in photoshop... but this really shouldn't be taking more than 10% of dev time.

Moreover, images will be quietly rendering away in the background while you're doing that (less quietly in my case - I've had the same fan/thermal paste issue as Nottravis, and it's bloody noisy! I expect it's those GPU-intensive sci-fi backgrounds that are grinding our machines to death )

 

[anne O'nymous]

[...] I think that spending a quarter of development on test and correction is extreme (especially for a renpy game). Sure, it takes a day or two to test the code and typos, and a couple more for rerenders, fill-in renders, extra expressions and painting out poke-throughs in photoshop... but this really shouldn't be taking more than 10% of dev time.

Probably... no, in fact surely.
It was a projection based of the time we spend on it at works and the number of people doing it. But I forgot that we are making software way more complex, using test suits build especially for the occasion (which obviously increase the time dedicated to the testing phase), and all. We need to go deeper in our tests since even an one of a kind error isn't acceptable ; it's a totally made up example, but I don't want to imagine an accounting software that would give a wrong result if one of the value is between 1234 and 1235 by example.
An adult indie game imply less worry, and therefore less tests needed.