test content
Logo
What is the Arc Client?
Install Arc
Ten Forward weekly 3/19/2024 (All Fans! All hands)
protoneous
Member Posts: 3,169 Arc User
From: https://reddit.com/r/sto/comments/1biyc2o/ten_forward_weekly_3192024all_fans_all_hands/ Credit to: https://reddit.com/user/TheSajuukKhar/
- Short term of the transition from Cryptic to DECA will be more people involved in the game's development.
- There's nearly daily calls between Cryptic staff and DECA staff on things like development, roadmaps, what players want from the game, etc.
- Long term, Phil thinks Cryptic will take more of a support function as more of the primary development moves to DECA.
- Conversations with DECA began in Nov 2023. Plan is that transition would take around a year, but between hiring/training it might be longer.
- None of Cryptic's games are going to be sunset, as per the current plan. Intent is to keep the games running a long time.
- Licenses with Paramount/Wizards last several more years.
- STO is called out every quarter for being a top performing title in Embracer's back catalogue.
- DECA is planning to maintain the relationships with people at Paramount/Trek actors. DECA staff is being trained in that right now.
- No planned changes for STO/NW/Champion updates/modules release schedule. Current release size is rough indicator of future release sizes.
- March 2024: New Mudd bundle with 2 popular ship picks.
- April 2024: First contact day + new Star Trek Beyond ship.
- May 2024: New Iconian event + new reward. New ship bundle T6 Dysons. New season planned for the end of the month.
- June 2024: New Mudd bundle.
- Season 32: Name not finalized
- Continuation of Kings and Queens story
- 1 new episode
- 1 new TFO (an endless battle arena like Sompek)
- New ship added to Infinity Lockbox
- New event and reward
- A new feature Cryptic hasn't done before
- Returning Star Trek VA as guest celebrity(Denise Crosby based on spoilers)
- Development on next season is ahead of schedule.
- New feature one word hint is token.(seemed to be a joke)
- No current plans in the near future for Cardassian/Romulan/Jem'Hadar ships.
- Cryptic had 2 games in development in 2023, one planned for 2025, another for 2027. Both games were cancelled by Embracer. Cryptic has no new games in development right now. Cryptic is in wait and see mode for Embracer's new process for getting games greenlit.
- While DECA is known for maintaining games, but not updating them, they will be updating Cryptic games as previously stated. They are staffing up/have an org chart for significant development for these titles for some time.
- DECA is based in Berlin, but have staff worldwide.
- No plans for using AI in the game's development(chat question)
- Cryptic is very well informed as to what is happening/the transition. People are choosing to leave for other projects, some are stepping up to take on larger roles. Phil says there will be more people leaving/coming in. Cryptic staying long term on the games will be determined by how much DECA can ramp up into production, and if they need help with stuff like servers. Short term you will see a blend of Cryptic/DECA people. What it will be like in 4 years Phil doesn't know.
- Celestial enchantments issue on preview for Neverwinter is known, and being looked into.
- No current plans for changes for maintenance windows. Staff that does that is based in U.S./planned to stay here. Might change in the future if DECA gets the server staff spun up.
- No current plans to change engine for any of the games. Would take like 3-5 years of the games being offline to transition games to new engine 🙄
- DECA has started looking into the feasibility of something like an Unreal port. Doesn't mean it will happen. It depends on cost/benefit analysis.
- Cryptic has looked at what it would take to update the engine. The cost of that kind of change is millions, possibly over 10 mil. Changing the engine doesn't automatically mean the game gets better. Would have to go through and redo all the old content to actually make the game better. Cryptic couldn't find a case where the cost would be worth it. DECA might do their own analysis and come to a different conclusion. If they do, they will tell us.
- Phil is going to do introductions between the Paramount/Wizards staff and DECA staff to help transitions between new staff and IP holders.
- No worry about Zen/lifetime going away after the transition.
- Updates short term will be from Cryptic, Medium term Cryptic/DECA, long term we'll see.
- Discussions are happening between Phil, and Mike, and the staff at DECA, on when the DECA community manager will make introductions. Probably in a few weeks(3 or less).
Post edited by baddmoonrizin on
0
This discussion has been closed.
Comments
Nice. Variety is the spice of life and these feel more unique to me than the Alliance event ships.
Glad to hear they believe they have a solid plan. I honestly didn't think DECA would take that long to be the main driver... but I guess STO and NV are both a bit of a mess and not easy to just hand over.
Good Luck Cryptic, may the next year or so go smoothly.
Fair. I have no doubt it takes a while to get the people in place. Then to ensure they are ready and can handle a live game. They are dedicating longer then I would have expected. That is a good thing.
Frankly still sucks for Cryptic. As much fuzz as I have and probably still send in their direction. I hope they all end the year happy with their positions, or find new ones elsewhere. Well wishes to all. Kudos for them for being pretty up front and forth coming about the situation. As a STO player and Cryptic customer, I am glad to hear their are long term plans for STO.
normal text = me speaking as fellow formite
colored text = mod mode
#LegalizeAwoo
A normie goes "Oh, what's this?"
An otaku goes "UwU, what's this?"
A furry goes "OwO, what's this?"
A werewolf goes "Awoo, what's this?"
"It's nothing personal, I just don't feel like I've gotten to know a person until I've sniffed their crotch."
"We said 'no' to Mr. Curiosity. We're not home. Curiosity is not welcome, it is not to be invited in. Curiosity...is bad. It gets you in trouble, it gets you killed, and more importantly...it makes you poor!"
There is also the Altamid Swarm Ships. There are two possibilities on that one.
I wouldn't mind seeing the Ent-A But we know next to nothing of her othe than a brief shot of her before warping out. We also know Cryptic had been playing around with the idea of a playable squadron after they made Squadron Pets.
normal text = me speaking as fellow formite
colored text = mod mode
I believe ages ago the altamid swarm ships were referenced in something as being planned to be a playable squadron but I don't remember where it was mentioned. It was back when the devs still did roadmaps, I think around the time the freedom released. Obviously it didn't happen at the time though.
There's only one ship from Beyond I can think of - the AR Ent-A introduced at the end of the film. Everything else from that movie we have in one form or another.
>
We already have the Kelvin timeline Ent-A. The legendary version of the ship has the skin. The swarm ships others mentioned sounds like a fun idea though.
I wonder if they're tied together. Would be nice to see some further use of the Dyson Spheres.
Regarding Sompek: it requires tweaking. Power creep has removed the challenge from it, what was supposed to be the ultimate challenge on the ground, has now become something where I see players not move for the entirety of the mission while their drones, omni-directional turrets like the Agony thing etc. kill everything for them.
I hope Sompek and this new arena will see some updated to account for this.
Good. Art, including game development, should revolve around humans and their hard-earned talents.
Think bigger.
We're getting a mobile Yorktown Station. xD
Actually like to see stations with weapons. Who in their right mind at Cryptic thought it would be a good idea to not equip stations with a shed load of firepower?? The only times I've seen it is DS9 and Fleet Starbase, and even then, there's less firepower on them than a player-controlled shuttlecraft
It was not the Enterprise-A.
The Legendary is the refit that was destroyed over Altamid, but was supposed to be the same ship that we say launch in 2009 and fought the Vengeance in Into Darkness. The Ent-A was the ship Starbase Yorktown was building and used to replace the Enterprise.
This is the Ent-A from the end of the movie.
normal text = me speaking as fellow formite
colored text = mod mode
And I bet she still doesn't have shields worth a damn either
This raises some interesting questions and ideas:
1. Should AI be used to "sort" the game code?
2. Should AI be used to detect bugs so they can be more efficiently dealt with by developers?
3. Assuming Question 2 works, should AI be used as an "Early Warning System" to detect bugs before updates go live?
4. How long would it take for AI to learn to accomplish such tasks?
5. Can that times be decreased by letting the AI "observe" change logs, assuming there are any?
6. Will we be able to predict all of the unintended consequences of taking such actions?
One of the unintended consequences I can imagine is that the AI will be tweaked to suit agendas beyond producing a stable game.
Another is that some middle manager will decide to start cutting the Dev Team and "Let the AI handle it," just to secure their annual bonus.
Those are all very interesting questions, but perhaps in depth discussions about the use of AI belong in a different thread as it's already been stated that there is..
1) Maybe, but who says it needs sorting and why would it even matter?
2) A practical impossibility. How on Earth would AI know what is bugged or not? Bugs won't appear until they hit Holodeck. The testing environments are not the same as Holodeck, so bugs WILL happen. You can not replicate a live-service the size of STO's Holodeck on any testing environment.
3) See 2 and QA IS the early warning system. It is extremely rare for QA to let a bug hit Holodeck.
4) Again, can't be done.
5) Wouldn't make a hoot of a difference because of the different server environments.
6) AI will likely produce MORE bugs than it solves, because it's not actually 'intelligent', nor is it intelligent enough to predict when a bug may happen or even the consequences of it's changes. As mentioned earlier, the majority of bug don't appear UNTIL the build has been released on Holodeck. Their internal servers, Tribble and Holodeck are all different beasts. Holodeck is a constant stream of changing data and can not be replicated internally.
I considered that, but the topic popped into my head as a result of the original post, without which my post loses some context. Plus I couldn't decide if General or Ten Forward would be the appropriate section for such a topic.
Plans, on the other hand, do change from time to time. This may be a useful discussion in the event that the plans in question do in fact change in the future.
If any Forum Moderator wishes to shunt this to it's own topic I'd understand why.
I'll respond in kind:
1. The phrases "Spaghetti Code" and "Legacy Code" have both been used to describe the STO Game Code Environment. It would matter because the more organized a software environment is the easier it is to work with.
2a. "A practical impossibility." Yes, at the current state of AI development it is likely not only Cost Prohibitive but the software would also likely have to be customized for STO. That's not an Absolute Impossibility.
2b. "How on Earth would AI know what is bugged or not?" Create an account for the AI to use on the Live Server and give it early access to new items/areas before making them accessible to regular player accounts. (I wasn't clear in my original post that I don't believe that AI will be a perfect solution. I'm just looking for ways to use AI to increase the efficiency of bug catching and squishing.)
3. As I recall the head of QA was among the first people fired once DECA started making changes. This plus the number of bugs that either crop up with new releases, persist or recur suggests that QA is either not doing their job well or is in serious need of support.
4. "Again, can't be done." I've spent a bit of time trying to come up with a response that's less adversarial than, "Well not with that attitude, pal," but I can't think of one that equals it's brevity. I'm sincerely more interested in Discussion rather than Argument, so I hope you understand that I'm not taking a personal shot at you.
5. This is the kind of thing wherein I believe we won't know until/unless we try, but I do recognize that the commitment of time and resources to get to that point would be prohibitive to say the least.
6. "AI will likely produce MORE bugs than it solves, because it's not actually 'intelligent', nor is it intelligent enough to predict when a bug may happen or even the consequences of it's changes." Yet.
You bring up some excellent points which just shows that this discussion is one worth having. At least to me.
I wasn't really thinking about bug fixes when I stated that I don't think AI has a place in this game. I was more thinking along the lines of creating zones, assets like ships etc. - the design part of the game.
Personally I wouldn't really be opposed to AI dealing with some of the more menial tasks like going through code and correcting some numbers here and there. I imagine that most Devs don't really think of this as the most exciting part of their job (could be wrong though - it's just what I assume).
I don't really know enough about the stuff to determine whether that is currently technically feasible (or if it might be in the foreseeable future).
But when it comes to designing stuff, the art part of the game? I try to keep as far away as possible from any 'artsy' AI creations that will only serve to put talented people with actual souls out of a job.
Unfortunately, for art it does seem to be technically feasible to replace humans.
I know of a few online communities where human made creations are being marginalised, because they cannot possibly hope to compete with the AI-created stuff in terms of quantity especially.
So I think the question of whether we should want such things - not to mention the legal issues - is more relevant there.
Considering the theme of this year's anniversary pack and all (background ships) perhaps this?
https://youtube.com/watch?v=29Bo1hZ-mlM
Don't worry, I don't take things personally. I do enjoy a constructive discussion.
I think you're over-interpreting what they mean by 'spaghetti/legacy code'. The code is probably quite well organised (it would have to be for the game to work anyway), however, I think, and it's been said before, when they say spaghetti code, they're actually referring to changes having unpredictable consequences, which result in bugs. This happens in any large live-service game that is added to and changed on a weekly basis. AI can not predict it because of this very reason. For AI to 'work' on a game, the code needs to remain in the same condition so the AI has a common frame of reference. To have a functional AI for a game that changes every week, you would also have to reprogram the AI every week too, so you are doubling the workload.
Now, as for legacy code, this is often brought up because the people that originally programmed that legacy code moved on without passing on what that code does. AI might be able to decrypt it, but unless it has a common frame of reference to go off, it most likely would come up with gibberish. AI's that are currently available literally only work off what they are told, not what they can 'create', because what they create is generally plaigerised content.
Giving AI a 'live' account to test new stuff is interesting, but unrealistic. The immediate issue with the suggestion is that it would not be more successful than a normal human being doing the exact same job when focused on a new addition to the game. On the whole, AI would not be able to detect bugs that occur any faster than an entire community of players because the game is so huge. You can't have it test the entire game either because that would take at least 2 weeks non-stop, because it would have to play every episode, battlezone, TFO etc etc. You would have to have dozens of instances of the AI running stuff, but the Holodeck server would be offline for longer, and we know how some folk are impatient even for a 2 hour server restart.
Just because the head of QA has left, doesn't mean QA has been compromised either. Bugs have absolutely 100% not increased in frequency, and to say QA isn't doing it's job correctly is wrong. QA test everything in their testing environment, but, as I have already said, can not test in an equivalent of Holodeck, because it is a different environment. Bugs mainly happen because of bad interactions with something on Holodeck, and it's rare QA makes or misses an 'obvious' error. They have to and must assume that something that works fine on their testing environment will work the same on Holodeck otherwise it would be never released.
Letting AI observe changelogs won't have any effect, because those changelogs will get looked at by QA etc,if a bug has cropped up. I would hazard a guess it's the first thing that is looked at, because it makes common sense.
AI is fascinating to observe and fun to use at times, but it is no panacea for bugs. I was using one particular AI to help me find research papers for my research project, and it literally invented a number of papers that did not exist. I had to be extremely specific to actually get what I needed. AI is a long way from being 'adaptive' enough to anticipate bugs in a game like STO as in can not anticipate what it would do on a living-breathing server like Holodeck.
AI is just not ready for coding big complex things like MMOs yet, the best use in games so far has been as graphics development tools to quickly adjust maps and textures so the devs can concentrate more on the creative parts of the job instead of spending hours making manual adjustments for every little change while tweaking the layout or whatever to be just right.
Modern tools like that would help STO and could probably be added depending on which 3D graphics program they use to build stuff (there is a Ten Forward from a few years ago that features the map-making process and they mentioned the name in it, but I forget which it is offhand), but their texture and mesh assets are not organized so it would take a lot of work to sort them out into a structure that the AI could understand if they are to make the most of the benefits of using it.
And with luck it might be something DECA, with its dispersed organization of developers, would be willing to invest in to make the ROI of training developers better (according to the stream, it takes a full year to get a new dev up to speed as things stand now), especially with the geographic and language barriers that Cryptic doesn't have to deal with but DECA does.
I have found this unintended result to be quite illuminating.