RSI Scorpius Antares – A Powerful New EMP / QED Ship
Hello & Welcome to some more Star Citizen the RSI Scorpius Antares has now been revealed we will talk about that and some dev responses talking about weapon & component variety for ships…
Is is Ant Ares or Ant Ar Res?
Deepest Lore
Antares is the brightest star in the constellation of Scorpius.
Referred to as the Heart Of The Scorpion.
Power Creep?
Does It Lose The Turret? Looks to be a single seater.
Where Does This Leave The Mantis?
Will there be more modular EMP / QED components in the future for things like Heavy / Medium Fighters? Could we see a Hornet variant for example?
Expected Cost?
Thoughts
Single Seater Heavy Fighter with extremely powerful utility
Can be a lot more offensive than the Mantis
(Thought) Might not be able to pull ships out of quantum just
Weapon and component variety – an update would be great
A number of patches ago, groups of weapons and shields were largely homogenised in what was explained at the time as a temporary simplification exercise. It’s been like this for some time now and I’m unaware of any work being undertaken to move us out of this paradigm.
Am I missing something from the roadmap or are CIG happy with where we currently are? i.e. have things changed?
I’m currently wondering how you’ll introduce weapons like the Yen’tu repeaters when the San’tok.yãi comes out – or fix the Defender tachyons.
YogiKlatt Answers – Weapon balancing / diversification is currently happening in S42. The weapons that are being used for S42 get the treatment first, others will follow. I can’t tell though when the work will hit the PU.
Follow Up Question
If I remember correctly, didn’t Blizzard end up having problems with balancing for Single player before Multiplayer?
I think they ended up having to have multiplayer unit stats be very different from the single player, thus the feel a user would get from the Single player game ended up throwing them off for multiplayer?
YogiKlatt Responds – We have a fixed scheme how vehicle weapons scale, how much damage they do, etc … that is all based on PVP stats. The S42 weapons follow the same scheme and we are trying to avoid differences in S42 and PU weapons to keep consistency across both games ( … also for the sake of reducing the amout of data records).
Tuning in terms of weapon vs how much damage is applied is managed via damage modifiers that can be easily tuned for every encounter and difficulty setting. This is actually already happening in the PU as well, AI ships often get between 1.5 to 3 times more damage when being shot at by players. But that’s nothing new, we had this for years this way. So in this regard there is not really a MP-vs-SP tuning problem.
Another Follow Up – If i remember correctly, then the goal was: NPC are indistinguishable from players. Is this still the goal? If yes – does it make sense if NPC get more damage by players or is that a sort- / midterm solution and will change later?
II have no opinion on whether it is useful or not or whether the goal is achievable or not. I only remember that there was this statement and would like to know if the goal still exists.
YogiKlatt Again Responds – That goal has not changed but the devil is a bit in the details and depends what “indistinguishable” really means (damage application, flight behavior, AI situational awareness, etc.?). In general AI units pilot the same ships with the same equipment and capabilities and restrictions that player ships have and when AI ships fight against AI ships they adhere to the same rules as players.
Things however become a bit more complex when it comes player vs AI interactions. In those case we often need what I would call control measures to shape interesting encounters, and the damage modifiers are just one example of that (having AI pilots restrictions on their capabilities would be another example). But which of these control measures are used and to what extend is in the hands of the individual designers of the scenario.
As an example: About two or three weeks ago Diego asked me to fight against his new AI algorithms in a S42 Gladius (master mode tuning, no modifiers, 2x S3 repeaters and 1x S3 Gatling on gimbals (N-1 removed)) in a 1v1. Without any of the normal modifiers I was basically not match for it. The AI was very aggressive, extremely good with its distance control and never took any breaks … my capacitor systems tired out very fast as I was constantly trying to dodge incoming salvos. I barely landed shots without getting damaged myself so I lost that fight. That is a feeling is something that upper ranked PVP pilots might know pretty well (and thrive on) … but for an low skilled PVE player (who just wants to complete a mission) that scenario might also have been nothing but frustrating. And this is why control measures are so important because they allow to easily tweak the challenge level of a scenario. On paper that might miss the mark of being completely “indistinguishable” from real players but let’s not forget that at the end of the day we’re trying to create a fun game and there are more things out there that can create believability than just weapon damage numbers. Hope that makes sense.
Yet another follow up – talking about PVE difficulty where the OP says – Don’t babysit every players for the sake of “we’re trying to create a fun game”, having an AI take from 1,5 to 3 times more damage with a multiplier is just killing the thrill of it.
YogiKlatt is there again – I did not say at all that we’re babysitting people or that we’re making the whole game super easy for everyone. I simply said that we have ways to scale the difficulty in either direction and that pure player vs AI ship combat can be pretty brutal when all safeties are off. In the end we want the game to give enough valid PVE gameplay options for both inexperienced and experienced players. Are there we yet? No … not yet.
Is PES the new version of iCache? Or is EntityGraph the new version of iCache?
The OP references a video where Chris Roberts says:
“we should hopefully have the proper persistence in there that we talked about: the proper global persistence which is this sort of streaming entity graph thing that i’ve just discussed which is you know the new version of what was being called the iCache before”
The OP then goes on to ask:
How exactly is this statement to be understood? What exactly is described here as the new version of iCache?
Just EntityGraph? Or the entire logic described, which now known as Persistent Entity Streaming (PES)? So EntityGraph, Replication Layer, Global Database, Global Persistence and other services and functionality combined?
CIGs Clive Johnson Answers – iCache is a service that provides distributed in-memory caching of previously queried and updated data stored in a SQL database. By having the game talk to iCache instead of directly to the DB a lot of queries are avoided, reducing load on the DB, helping it scale to larger numbers of clients.
EntityGraph is a service that provides translation from our game specific “query language” into the query operations of the underlying database and translates the results back into a form the game can easily understand. EntityGraph also provides extra reliability and scalability guarantees that iCache doesn’t. What EntityGraph doesn’t do is any form of caching, however it talks to a NoSQL database that performs its own distributed in-memory caching. Not only that but the database is specifically and optimized for graph data which is a much better fit for the type of queries we need to do.
So, to answer your question, EntityGraph and its NoSQL database are the replacement for iCache and its SQL database. EntityGraph’s (both the service and its DB) improved performance and scalability are what has enabled full persistence as well the extended functionality for inventories (global persistence).