Egosoft, Plese stick to what you are good at!

[ZaphodBeeblebrox]

Well if Egosoft stick to what they are good at (how condescending) implies that they can't get good at new things.

Anyway lets hope we get to see more of this kind of thing.

https://www.youtube.com/watch?v=WYsjn9qbHb4

[Alan Phipps]

An excellent vid and fight. It was obvious that the player had some considerable empathy with his hard-pressed fighters.

It also however reinforces the need for a user-controlled message spam level-of-urgency filter for use during 'busy' or large fleet control periods. A definite 'must have' for X4 empire building.

[birdtable]

I have yet to see a vid extolling the virtues of "mini games" "small talk" or scurrying around pilfering lockers.

[Nanook]

Yeah, let's hope X4 has the bare minimum (none?) of those three 'features'.

[Crellion]

I wonder if you realize that what you are actually describing with the cute check list is a new and improved XR... like ... XR2.

What about having freedom to fly and loadout all your ships reminds you of Rebirth? I just don't get it.

Your question is too simplistic to be intelligent in my opinion but since its an argument many people voice at times (a desperate argument but all the same...) I will try to give you a not simplistic example.

Ok you have played Rebirth I guess. Let's see... if I take Elite Dangerous and put "highways" in while in every other aspect leaving it ... Elite Dangerous and I call the product X Rebirth 2 and then try to market it as such... what does that make me?

Or another example: I take WoW and add mounts that take you to space and call it EvE-Online 2. I mean dude... seriously...

[Crellion]

Well if Egosoft stick to what they are good at (how condescending) implies that they can't get good at new things.

Anyway lets hope we get to see more of this kind of thing.

https://www.youtube.com/watch?v=WYsjn9qbHb4

The tittle of the thread is indeed condescending and I am sorry to be posting in it. However it was a moderator choice when I created a thread called "X4 feels like X Rebirth maybe" or something like that and they merged it in this thread whose title is indeed... unfortunate.

[ZaphodBeeblebrox]

The only mini-game that might be useful is hacking ships and stations.

Though there may be any number of ways, in which hacking could be done,
that would be better than what currently exists.

This is a part of game play that could do with being expanded upon.

[Killjaeden]

Anyway lets hope we get to see more of this kind of thing.
https://www.youtube.com/watch?v=WYsjn9qbHb4

I wouldnt hope that... the amount of AI crashing into everything is ridiculously bad. X - bumper car edition... Worse than X3 even.

[RAVEN.myst]

The only mini-game that might be useful is hacking ships and stations.

Though there may be any number of ways, in which hacking could be done,
that would be better than what currently exists.

This is a part of game play that could do with being expanded upon.

Indeed. Personally, I would hope for a mini-game that's more cerebral than twitch-click based. The hacking minigame in EVE Online springs to mind - it's not amazing or anything, but at least it reflects the hacking theme nicely enough, and it's more puzzle than just timing vs an arbitrarily laggy moving squiggle. That game also has a scanner probes system that could be considered a minigame, and it's also not bad, again conveying the intended theme nicely and logically (in fact, this latter really is how one might deploy and then manipulate a set of sensor probes.)

[Skeeter]

No more mini game things please.

[ZaphodBeeblebrox]

@Killjaeden

I wouldnt hope that... the amount of AI crashing into everything is ridiculously bad. X - bumper car edition... Worse than X3 even.

Really?

The ships are targeting surface elements with missiles and weapon fire.

X3 did not have surface elements.

I remember a mission in X3 where the player is supported by a bunch of missile firing small ships when attacking a target, one of the most embarrassing sights in a game.

NOT a single missile is fired by these ships, but the supporting ships did do some damage as they repeatedly bounced off the target.

So worse than X3 no not by a really long way.

[sd_jasper]

Indeed. Personally, I would hope for a mini-game that's more cerebral than twitch-click based. The hacking minigame in EVE Online springs to mind - it's not amazing or anything, but at least it reflects the hacking theme nicely enough, and it's more puzzle than just timing vs an arbitrarily laggy moving squiggle. That game also has a scanner probes system that could be considered a minigame, and it's also not bad, again conveying the intended theme nicely and logically (in fact, this latter really is how one might deploy and then manipulate a set of sensor probes.)

I'm not familiar with the EVE hacking, but...

The problem with the hacking system is that it is usually pretty time sensitive. Hack the ship before it jumps/boosts away, hack the station before you are scanned, etc. Making a system that is any more involved than a "click at the right time" would probably take up more time, and be a problem.

Now the out of ship hacking could be made into a more interesting system, but I would caution against anything that doesn't evaluate the amount of time invested vs the (potential) payoff received.

[Killjaeden]

@Killjaeden
Really?

The ships are targeting surface elements with missiles and weapon fire.

X3 did not have surface elements.

X3 did have some surface element recognition, but only to AI and also missiles. They use existing turrets and stuff as targets in addition to the center of the ship. Fire a missile barrage at a Capital and note how they fly towards different locations on the ship. AI attacks do the same. So it's no excuse for XR to have higher collision rate.

I remember a mission in X3 where the player is supported by a bunch of missile firing small ships when attacking a target, one of the most embarrassing sights in a game.

NOT a single missile is fired by these ships, but the supporting ships did do some damage as they repeatedly bounced off the target.

If you set missile firing propability to a small amount in X3, dont expect them to fire volleys. If you set it to 100% they will spam missiles quite happily - unless the logic script defines that Missile X is not suited for target Y.

Seeing M3 ships bounce off of capitalships is rare. If the target is extremely large and maneuverable, or the ship is way too fast for its own good (Springblossom, M5 scouts), then collisions will be more frequent. But that XR video didnt show fast ships. If the target in X3 is a mod ship that was created a long time ago then yes but that's expected because the AI collision information in X3 modding was only discovered extremely late, basically when AP was very close to release (Egosoft themself didnt know how it worked either...)
And XR was supposed to be an improvement. Way early in XR development we saw also specific highlights on AI improvement. I think one Dev even made a presentation for GDC or similar about AI... so seeing XR in this state (Ego even going as far as disabling collision damage) is baffling to me.

[ZaphodBeeblebrox]

Right,

So what this really is about is our particular biases.

I watch the video and see a whole raft of small ships attacking Xenon
capitals and destroying them.

They select targets on the capital ship to destroy surface elements.

This is for me a really good example of the superiority of the AI in X:Rebirth and how the small ship AI has been improved.

I played X:Rebirth for over 1300 hours.

You on the other hand seem to be biased towards X:3 being the superior game.

So you see ships bouncing off the capitals!

From my point of view there could not be a "good" Rebirth video for you.
From my perspective, Rebirth has contributed a great deal to the enhancement of the X series of games.

I will also agree that for Egosoft themselves it has been an extraordinary learning curve.

We can only hope that X:4 be better game from what they have learned and what their long term fans want.

[Killjaeden]

Right,
So what this really is about is our particular biases.
I watch the video and see a whole raft of small ships attacking Xenon
capitals and destroying them.
They select targets on the capital ship to destroy surface elements.
This is for me a really good example of the superiority of the AI

What does this have to do with superiority of AI? X3 ships follow attack commands as well - check. X3 ships attack subsystems as well - check. How is this 'advanced'? I would call it rudamentary function... Like i said, subsystem targeting was already there in X3. The only difference is that they are destructable in XR, which has nothing to do with AI itself.

From my point of view there could not be a "good" Rebirth video for you.

When i see collision avoidance issues i saw already in X3, except 10 times amplified in the "successor", what reaction do you expect? I started playing X3AP+XTC again, it's not blurred memory or nostalgia speaking...
And yes XR was a giant failure. New engine is a saving grace, but the game design is just plain bad in many areas.

[Vandragorax]

Must all these threads devolve into a game of "who has played the oldest X game for longest?" e-peen competition? (This is a rhetorical question, no need to answer me please )

I think we can all agree that some of the game design from X3 works better than XR, and some of the design from XR is better than X3.

I also think we can all get behind the hopes that X4 will be drawing on the lessons learned along the whole of Egosoft's journey to-date (and there's nothing to suggest that it isn't so far).

Yes they are a small team, but they have some clever and sensible people there as evidenced through those devs whom often post on the forums in response to our silly arguments I'm pretty sure that they don't add game systems for the sake of it, there is always a good reason why they do things in certain ways.

It looks like, at least for now, that they are going to be engaging with us heavily over the next year and I'm really excited to see more information coming out from the twitch stream(s) that are planned and coming up. Egosoft are keeping up with the technology to engage with us the fans and take our requests on board, but thankfully they won't implement everything that someone says once on a forum post without thinking long and carefully about it and the implications it has on the rest of their game

[RAVEN.myst]

The problem with the hacking system is that it is usually pretty time sensitive. Hack the ship before it jumps/boosts away, hack the station before you are scanned, etc. Making a system that is any more involved than a "click at the right time" would probably take up more time, and be a problem.

Now the out of ship hacking could be made into a more interesting system, but I would caution against anything that doesn't evaluate the amount of time invested vs the (potential) payoff received.

Yes, I completely agree - for time-sensitive hacking ops a minigame can't be a puzzler, and presents a tricky design challenge, as the current system fails to convey a sense of "hacking a system", really (I suppose one could think of it as trying to cut a wire or something, or jam a sensitive moving part at precisely the right moment - but it simply doesn't do it for me, at least.) The way I see it, it should focus on quick thinking, not on quick reactions - so perhaps a snap decision needs to me made: choose well and the hack succeeds, choose poorly (not the RED wire! ) and you fail... Maybe a sort of circuit diagram with several possible locations to click to interrupt or short-circuit a flow (as with any game, mini or otherwise, there would be some determining rule here) - click the correct box and you've disabled the system, click an incorrect one and nothing happens, click REALLY badly and you set off a failsafe/alarm; difficulty of the hack attempt could then be implemented by size/density/complexity of the circuit diagram - trivial hacks would have equally trivially simple circuits, while difficult hacks would present the player with a larger and/or more densely packed set of wires, perhaps with some crossovers/whatever.

Just spinning notions here...


EDIT: typocide

[carran]

It seems to me that Egosoft have decided to develop two games instead of one implying both sufficient development bandwidth to deliver / support two games and significant players of each game to make their efforts worthwhile

In the end it will be economics which determines whether both games (or code branches) will survive long term. The very limited evidence available suggests X3 / X4 and X:R2 are both considered viable and for that we should all be grateful

[Graaf]

When i see collision avoidance issues i saw already in X3, except 10 times amplified in the "successor"

This I think is disingenuous, you pluck a number like 10 out of thin air.

Now if recall correctly CBJ once posted that certain areas of the AI could not be improved upon in X:3 as the engine was single threaded.
That now they had a multi-threaded engine that was a lot that could now be done to improve the AI.

I'm sure the AI pathfinding has improved in Rebirth. But since Rebirth runs with collision damage OFF we cannot really tell if it actually has. But if it has improved why the need to turn off collision damage?

[Santi]

Post that ZaphodBeeblebrox refers to can be found in this thread here. Full quote of the CBJ post below:

Let's just clarify a couple of things about pathing here, because there seem to be some misconceptions. Disclaimer: I'm not the "pathing guy" so a) I'm not defending my own work and b) I don't have all the fine details.

Games up to and including X3AP did not have any kind of "pathing" AI to plan a "route" from A to B within a sector. Ships could be given a destination and they would fly towards that destination in a straight line. If something got in their way then they would turn and fly away in a random direction and then return to flying towards their destination, again in a straight line. The method for determining whether something was "in the way" was basic at best, with essentially just a detection range to play with. Increase the detection range too much and ships would spend most of their time flying in random directions trying to avoid things; decrease it too much and ships would keep crashing into one another. All ship movement used this same system, including for situations such as "dogfights", and it was extremely limiting. There was nothing much that could be done to improve this, partly because as a single-threaded game there was no time for more complex calculations for the hundreds of ships that needed to be processed.

With XR we clearly needed to take a completely different approach. One major difference is that pathing calculations can run on a separate thread, partially freeing it up from the usual frame-by-frame cycle. This allows much more complex calculations to be made and means that ships can now plan a route, and as part of that route can fly paths that are not just straight lines. However, the flipside to that is that the environment they are flying in is also much more complex. While ships in X3 have no option but avoid other objects by a huge distance in order not to bump into them, ships in XR need to be able to fly close to other objects, for example, in and out of the complex spaces created by station structures. As a result, there still has to be a trade-off between making the pathing system "intelligent" and preventing it from affecting performance. Turning off collisions between certain types of objects is not a "solution" to some kind of "failure" of the pathing system, it's a performance optimisation to ensure that pathing calculations don't slow the game down too much. Of course it's not perfect, and of course we'd like to improve on that, but it's not a case of trying to hide an underlying "problem" as some people keep claiming.

Regarding the video of the ships colliding repeatedly at the gate, yes it's amusing, but when you analyse it it's not as trivial to solve as some people seem to imagine. If you consider two people trying to pass one another in a narrow corridor, and how they often end up going in the same direction several times before they manage it, you will see why. They usually only solve the problem by some form of direct communication, for example with one person indicating to the other the side on which they should pass. The AI pathing equivalent would be communication between pathing tasks for different ships, which is obviously possible, but trickier to do in practice for a variety of reasons. Similar explanations apply to other scenarios where the pathing is not as good as we'd like, such as small ships hitting station hulls; they are trying to fly through complex spaces, and sometimes the pathing just doesn't get it quite right, for example some calculation isn't accurate enough, or the logic isn't able to handle some odd "corner" scenarios. Despite the glitches, this is fundamentally "better" than X3, where flight close to the surface of other objects was not even possible.

In X3 there was no way to solve the problems that were reported because the pathing system had been operating at the limit of its capabilities for a long time. With XR there are clearly some things that are not as good as you, or indeed we, would like them to be, but there is plenty of scope for improvement within the framework that has been created.