Concerns about Unreal Engine upgrades and Scope Creep

I'd like to share some concerns I have about the development. We recently had the upgrade from Unreal Engine 5 to 5.1 and from 5.1 to 5.1.1 - and according to Steven during the last live stream this was a "big endeavor" he explained that upgrades like this can be a headache due to all the custom changes they have for networking/backend and all that and this upgrade took them 6 weeks plus all the weeks that they now have to spend bug fixing after the upgrade.

6 weeks, for a team the size of Intrepid's, that's easily over 1.5 million dollars of development cost and that's also over 11% of a development year.

and this is not taking into consideration the time spent bug fixing, so clearly upgrading engines is not something simple, with that said, we already have an announcement for Unreal Engine 5.2... so should they be upgrading again? wouldn't that be considered scope creep? What if after UE5.2 comes 5.2.2 or 5.3, 5.4 ?

Intrepid was originally planning on delivering the game on UE4, when we backed AoC our expectations was for UE4, they already got the upgrade to UE5 and all the amazing tools it brings, back when they did this upgrade, we were told this wouldn't significantly push a release date since supposedly it would make the development easier, and while I do believe this, it's now almost 2 years after alpha 1 and still no roadmap for alpha 2, and we are still doing 6 weeks engine upgrades.

I believe Intrepid should lock a build and just move on, and after launch, they can take the time and do more upgrades. I would hate to see them get stuck in development limbo trying to keep on with all the latest tech and delivering nothing, the "game will be ready when it's ready" shouldn't go on forever and keep getting pushed due to new tech that originally didn't even exist and wasn't part of the plan - I believe that would be the definition of scope creep, and it's the same thing projects like Star Citizen is doing. What are your thoughts on this?

Find more posts tagged with

Comments

Arya_Yeshe

You can't tell the costs because you don't know how many people would work on this, but I am pretty sure it is just a handful of people

Anyway, the 5.1 update brought a lot of stuff to the table, it's up to the team to measure the effort in updating vs the amount of effort fixing by hand the: culling, field of view, drawing distance, shadows, performance, foliage, etc

It could be much cheaper updating to 5.1... and then stay at 5.1

Chicago

I agree that they need to release something and do patches/ upgrades down the line, i also think they should not be implamenting new ideas at this stage of development, just feedback for ideas that already exist, i dont think that the engine upgrades will have a big impact on timeframe because they are most likely different developers doing this and if the game released in UE4 at this point most people would be dissapointed so i think they are to commited to not do the upgrades at this point, i dont think it will be a common tjing though maybe one big update a year

You are right though 2 years post A1 and nothing on the horizon sort of is a drag, also its been about 8 months since we heard "spot testing will start in the next several months"

Kilion

I'm no authority on this but to my understanding the upgrade to UE5.1 had advantages in regards to the final performance of the game as well as to the speed at which they would be able to develop the game. That made sense to me.

If UE5.2 can't have any effect of that kind, I definitely would agree with the notion that they should just leave the updates to UE as they are and finish the game on the current version as entering another upgrade cycle would most likely do more harm than good.

Mag7spy

Think this most was made in where some consumers that don't know much about unreal engine saying the should upgrade every version and prob don't realize you don't just click one button and its easy.

But if there is an important update that will make development smoother and/or more optimized and they feel they need it I'm sure they will do it. Just has to make sense I'm sure.

SnowElf

These are valid concerns. One thing that Steven Sharif has been quoted saying is that money is not really an object to worry about, however, time is definitely a concern. I do agree that if the upgrade in UE allows for better performance across all branches of development and for the user end that it was warranted, but at some point, perhaps it would be best to lock in and only update a private server with the UE updates. AoC is already an incredibly ambitious project. It would be nice, if possible, to see UE upgrades when they start developing major patches or even expansions, if they plan on going down that route.

But the question that arises - is it easier and more logical to immediately upgrade the game in the newest UE version as opposed to waiting for patches or expansions? As a software developer, having the most effective way to code something is imperative, you want to avoid redundancy or repetitive code, so in essence, installing these newer versions of UE is likely saving time/money and may be considered a wise investment in the long run on behalf of the developer team.

Terrifying_Truth

Well, in that land management system stream, we can clearly see that the grass is not within the view distance, if view distance is manually increased then a performance issue will begin. The new UE 5.1 has really great features addressing landscape features, so it could save a lot of time in the future.

Clearly, you don't have to stop all employees in the company but only a few people, in the long run it could save time and money.

Why is this even a concern to us? We are not an investor board, it's not our money.

Birthday

UE 5 promised to give compatibility if I remember correctly. So upgrades should be simple.

However I think Intrepid is a small studio so even such simple and small upgrades take too much time to do.

So I feel like Steven should only do these upgrades if he sees significant performance/physics/functional improvements related to AoC's goals.

Main priority should be releasing the game honestly. After that if its not complete trash its gonna make enough money for Steven to hire a bigger team. With a bigger team they'll be able to push out engine upgrades faster and honestly I feel like engine upgrades, especially graphically enhancing ones(like 5.2), should be saved for later when the game's graphics really get outdated. Currently the graphics are acceptable, so just work on getting closer to releasing the game.

Depraved

Liniker wrote: »

I'd like to share some concerns I have about the development. We recently had the upgrade from Unreal Engine 5 to 5.1 and from 5.1 to 5.1.1 - and according to Steven during the last live stream this was a "big endeavor" he explained that upgrades like this can be a headache due to all the custom changes they have for networking/backend and all that and this upgrade took them 6 weeks plus all the weeks that they now have to spend bug fixing after the upgrade.

6 weeks, for a team the size of Intrepid's, that's easily over 1.5 million dollars of development cost and that's also over 11% of a development year.

and this is not taking into consideration the time spent bug fixing, so clearly upgrading engines is not something simple, with that said, we already have an announcement for Unreal Engine 5.2... so should they be upgrading again? wouldn't that be considered scope creep? What if after UE5.2 comes 5.2.2 or 5.3, 5.4 ?

Intrepid was originally planning on delivering the game on UE4, when we backed AoC our expectations was for UE4, they already got the upgrade to UE5 and all the amazing tools it brings, back when they did this upgrade, we were told this wouldn't significantly push a release date since supposedly it would make the development easier, and while I do believe this, it's now almost 2 years after alpha 1 and still no roadmap for alpha 2, and we are still doing 6 weeks engine upgrades.

I believe Intrepid should lock a build and just move on, and after launch, they can take the time and do more upgrades. I would hate to see them get stuck in development limbo trying to keep on with all the latest tech and delivering nothing, the "game will be ready when it's ready" shouldn't go on forever and keep getting pushed due to new tech that originally didn't even exist and wasn't part of the plan - I believe that would be the definition of scope creep, and it's the same thing projects like Star Citizen is doing. What are your thoughts on this?

upgrading your tools isn't scope creep. adding more and more features is scope creep.

upgrading a major version (unreal 4 to 5) takes a lot of work.

usually, upgrading minor versions (unreal 5 to 5.1) is painless.

upgrading a patch version (unreal 5.1.0 to 5.1.1) is usually effortless, as these are just bug fixes.

learn more about semantic versioning here https://semver.org/

with that being said, after checking out unreal migration guide, you would realize that you also have to upgrade other tools to major versions and change some references in your code.

they had issues upgrading because of all the custom code that they had, not necessarily because upgrading from 5 to 5.1 is difficult. also, how you organize your code has an impact on how easy or hard it is to upgrade.

also consider that those upgrades would bring out of the box features that would normally take months to develop, if not longer. so you end up saving time anyways

Mag7spy

SnowElf wrote: »

These are valid concerns. One thing that Steven Sharif has been quoted saying is that money is not really an object to worry about, however, time is definitely a concern. I do agree that if the upgrade in UE allows for better performance across all branches of development and for the user end that it was warranted, but at some point, perhaps it would be best to lock in and only update a private server with the UE updates. AoC is already an incredibly ambitious project. It would be nice, if possible, to see UE upgrades when they start developing major patches or even expansions, if they plan on going down that route.

But the question that arises - is it easier and more logical to immediately upgrade the game in the newest UE version as opposed to waiting for patches or expansions? As a software developer, having the most effective way to code something is imperative, you want to avoid redundancy or repetitive code, so in essence, installing these newer versions of UE is likely saving time/money and may be considered a wise investment in the long run on behalf of the developer team.

Private server?? The game is not released yet... the only thing they have is a development server.

Mag7spy

Depraved wrote: »

Liniker wrote: »

I'd like to share some concerns I have about the development. We recently had the upgrade from Unreal Engine 5 to 5.1 and from 5.1 to 5.1.1 - and according to Steven during the last live stream this was a "big endeavor" he explained that upgrades like this can be a headache due to all the custom changes they have for networking/backend and all that and this upgrade took them 6 weeks plus all the weeks that they now have to spend bug fixing after the upgrade.

6 weeks, for a team the size of Intrepid's, that's easily over 1.5 million dollars of development cost and that's also over 11% of a development year.

and this is not taking into consideration the time spent bug fixing, so clearly upgrading engines is not something simple, with that said, we already have an announcement for Unreal Engine 5.2... so should they be upgrading again? wouldn't that be considered scope creep? What if after UE5.2 comes 5.2.2 or 5.3, 5.4 ?

Intrepid was originally planning on delivering the game on UE4, when we backed AoC our expectations was for UE4, they already got the upgrade to UE5 and all the amazing tools it brings, back when they did this upgrade, we were told this wouldn't significantly push a release date since supposedly it would make the development easier, and while I do believe this, it's now almost 2 years after alpha 1 and still no roadmap for alpha 2, and we are still doing 6 weeks engine upgrades.

I believe Intrepid should lock a build and just move on, and after launch, they can take the time and do more upgrades. I would hate to see them get stuck in development limbo trying to keep on with all the latest tech and delivering nothing, the "game will be ready when it's ready" shouldn't go on forever and keep getting pushed due to new tech that originally didn't even exist and wasn't part of the plan - I believe that would be the definition of scope creep, and it's the same thing projects like Star Citizen is doing. What are your thoughts on this?

upgrading your tools isn't scope creep. adding more and more features is scope creep.

upgrading a major version (unreal 4 to 5) takes a lot of work.

usually, upgrading minor versions (unreal 5 to 5.1) is painless.

upgrading a patch version (unreal 5.1.0 to 5.1.1) is usually effortless, as these are just bug fixes.

learn more about semantic versioning here https://semver.org/

with that being said, after checking out unreal migration guide, you would realize that you also have to upgrade other tools to major versions and change some references in your code.

they had issues upgrading because of all the custom code that they had, not necessarily because upgrading from 5 to 5.1 is difficult. also, how you organize your code has an impact on how easy or hard it is to upgrade.

also consider that those upgrades would bring out of the box features that would normally take months to develop, if not longer. so you end up saving time anyways

This post yells it is their faut with custom code or its easy to upgrade. Just because new tools are added or experimental ones does not mean you need to upgrade every version nor will those always be useful to you right away...

As i read these post its like all these people work for IS know all their problems and say its easy to upgrade. This is kind of wack.

Let them decide when they need to upgrade and if its worth the time and effort. This is their game and they will know it the best, peanut gallery is a bit crazy from people that don't know that much on their development, game development, or working with unreal engine.

Voxtrium

Basically the best you can give is an opinion regarding this matter as nobody can actually be "right". Personally I would like IS to have as much QOL as possible when designing AOC so that the game can come out as good as possible. I don't know what it looks like to go from 5.1 to 5.2 now vs after launch but at the end of the day hopefully the seasoned developers combined with Stephens perfectionist mentality they can arrive to an appropriate solution.

Just don't pull a Star Citizen and announce you need funding in 1-2 years because of unforeseen circumstances.

Mag7spy

Voxtrium wrote: »

Basically the best you can give is an opinion regarding this matter as nobody can actually be "right". Personally I would like IS to have as much QOL as possible when designing AOC so that the game can come out as good as possible. I don't know what it looks like to go from 5.1 to 5.2 now vs after launch but at the end of the day hopefully the seasoned developers combined with Stephens perfectionist mentality they can arrive to an appropriate solution.

Just don't pull a Star Citizen and announce you need funding in 1-2 years because of unforeseen circumstances.

If they need more funding then i hope Microsoft buys them xD

Depraved

Mag7spy wrote: »

Depraved wrote: »

Liniker wrote: »

I'd like to share some concerns I have about the development. We recently had the upgrade from Unreal Engine 5 to 5.1 and from 5.1 to 5.1.1 - and according to Steven during the last live stream this was a "big endeavor" he explained that upgrades like this can be a headache due to all the custom changes they have for networking/backend and all that and this upgrade took them 6 weeks plus all the weeks that they now have to spend bug fixing after the upgrade.

6 weeks, for a team the size of Intrepid's, that's easily over 1.5 million dollars of development cost and that's also over 11% of a development year.

and this is not taking into consideration the time spent bug fixing, so clearly upgrading engines is not something simple, with that said, we already have an announcement for Unreal Engine 5.2... so should they be upgrading again? wouldn't that be considered scope creep? What if after UE5.2 comes 5.2.2 or 5.3, 5.4 ?

Intrepid was originally planning on delivering the game on UE4, when we backed AoC our expectations was for UE4, they already got the upgrade to UE5 and all the amazing tools it brings, back when they did this upgrade, we were told this wouldn't significantly push a release date since supposedly it would make the development easier, and while I do believe this, it's now almost 2 years after alpha 1 and still no roadmap for alpha 2, and we are still doing 6 weeks engine upgrades.

I believe Intrepid should lock a build and just move on, and after launch, they can take the time and do more upgrades. I would hate to see them get stuck in development limbo trying to keep on with all the latest tech and delivering nothing, the "game will be ready when it's ready" shouldn't go on forever and keep getting pushed due to new tech that originally didn't even exist and wasn't part of the plan - I believe that would be the definition of scope creep, and it's the same thing projects like Star Citizen is doing. What are your thoughts on this?

upgrading your tools isn't scope creep. adding more and more features is scope creep.

upgrading a major version (unreal 4 to 5) takes a lot of work.

usually, upgrading minor versions (unreal 5 to 5.1) is painless.

upgrading a patch version (unreal 5.1.0 to 5.1.1) is usually effortless, as these are just bug fixes.

learn more about semantic versioning here https://semver.org/

with that being said, after checking out unreal migration guide, you would realize that you also have to upgrade other tools to major versions and change some references in your code.

they had issues upgrading because of all the custom code that they had, not necessarily because upgrading from 5 to 5.1 is difficult. also, how you organize your code has an impact on how easy or hard it is to upgrade.

also consider that those upgrades would bring out of the box features that would normally take months to develop, if not longer. so you end up saving time anyways

This post yells it is their faut with custom code or its easy to upgrade. Just because new tools are added or experimental ones does not mean you need to upgrade every version nor will those always be useful to you right away...

As i read these post its like all these people work for IS know all their problems and say its easy to upgrade. This is kind of wack.

Let them decide when they need to upgrade and if its worth the time and effort. This is their game and they will know it the best, peanut gallery is a bit crazy from people that don't know that much on their development, game development, or working with unreal engine.

I never said they HAD to upgrade. its their choice to do so or not. i was simply explaining how software versioning works and upgrading your tools isn't scope creep.

Liniker

well, we know they lost 6 weeks in this last upgrade, we also know they lost 3 to 4 months doing the UE5 upgrade, so that's 6 extra months that was not planned initially

what if they continue to do these upgrades on their tools, and it starts to add up? 1 extra year, extra 2 years, 3?

keep in mind, when we backed this game we were expecting UE4 and a "before 2020" and this is the explanation on why we didn't get a launch before 2020:

so it was due to their hiring schedule, issues with production and the issues with the architecture found on Apoc that they said caused a 9 month delay

but does anyone think if it wasn't for these issues they would have the game Ready for launch in 2020? they are still making Art every week, and art has nothing to do with engineering, so clearly there's more to it,

the issue I have with these engine and tool updates is because it would make sense for them to focus on delivering the promises made as soon as "possible", a minimum viable product frist, and now 2 years after Alpha 1 and 8 months after they said "we will start spot-testing in the next few months" I'm starting to believe that it's time to lock a build and move on - because IF they continue to do these upgrades they would start crossing the line of scope creep

Ludullu

As long as the updates can make up for the lost time through their new features, I'd be completely fine with Intrepid going for them. But I dunno enough about UE to know that kind of detailed stuff. I just assume that Intrepid's devs do and can tell Steven "if we update now, we'll save several months of dev time later".

Goalid

Liniker wrote: »

I'd like to share some concerns I have about the development. We recently had the upgrade from Unreal Engine 5 to 5.1 and from 5.1 to 5.1.1 - and according to Steven during the last live stream this was a "big endeavor" he explained that upgrades like this can be a headache due to all the custom changes they have for networking/backend and all that and this upgrade took them 6 weeks plus all the weeks that they now have to spend bug fixing after the upgrade.

I think my problem with not doing this would be that UE5 would continue to upgrade and have new features that could greatly help game performance. And we need great performance on multiple fronts in order to be able to have the giant sieges we're all dreaming of, and that I believe will be a focal point of the game. The Intrepid team would be unable to include those future updates easily at launch if they just decided to skip years worth of patches. So I think they've made a decision to continue upgrading with UE5, and that will pay off in the future.

Liniker wrote: »

6 weeks, for a team the size of Intrepid's, that's easily over 1.5 million dollars of development cost and that's also over 11% of a development year.

Mistake here is thinking that all the development team is necessarily held back by the upgrade. We frankly don't know, I think this would be a great question for Steven during this month's livestream.

Liniker wrote: »

Intrepid was originally planning on delivering the game on UE4, when we backed AoC our expectations was for UE4, they already got the upgrade to UE5 and all the amazing tools it brings, back when they did this upgrade, we were told this wouldn't significantly push a release date since supposedly it would make the development easier, and while I do believe this, it's now almost 2 years after alpha 1 and still no roadmap for alpha 2, and we are still doing 6 weeks engine upgrades.

I don't think the lack of roadmap for Alpha 2 has anything to do with the Unreal 5 update. They wanted to make a new combat system that was animation focused and coded in a way they thought would be the most bug-free. They really didn't have many features in Alpha 1, and Alpha 2 is far more feature-heavy than Alpha 1.

Dygz

Haha!
Expect a lot of scope creep.

Back in 2021, when people were complaining about Alpha 1 Dwarves looking like Gnomes, I was saying, "Don't expect the Dwarves to get much wider - or, if they do become signifcantly wider, expect a significant delay in the schedule.
Even in my wildest dreams, I could not imagine a gap of 2+ years between Alpha 1 and Alpha 2.

When do we get Nodes 3??
When will we get the last two Articles for Know Your Nodes?? We've been waiting for those for at least 3 years and those don't rely on UE.

There will always be compelling reasons to upgrade to the new version of UE.

novercalis

im already been waiting X years, I dont care if I gotta wake 6 weeks or 6 months. Be up to date and bring the best thing you can create to the table in the end.

Liniker

novercalis wrote: »

im already been waiting X years, I dont care if I gotta wake 6 weeks or 6 months. Be up to date and bring the best thing you can create to the table in the end.

6 weeks or 6 months? xD what about another 6+ years for a release? because that's not impossible at all, projects that continuously updated and add new features/tools easily go beyond the 10 year mark

Enigmatic Sage

Scope creep exists no matter what. UE will always have updates. The devs will have to decide what features are worth it for their custom part of the engine vs foundation.

Child Item