This isn't a war, it's a rescue mission.
"Floating Point Down": Leave no Integer behind!
This isn't a war, it's a rescue mission.
3.0 is still coming for the end of the year though, right?
I mean, it's not on the schedule but that just means they want to surprise us!
Maybe I need to write a letter to Santa Claus...
This is... a little disingenuous of you, I have to say.
Firstly, because the examples you linked were mostly fans getting the right answer via the wrong means, or getting it wrong and being immediately corrected several fans who were right.
Secondly, we started out, not with your Three Meanings, but with a link to your forum, and you ridiculing a guy for thinking 64-bit would give him the ability to travel around a big seamless world. Then you insisted that the QD tunnels were a necessary way to transition between "scenes", and linked to that big post about drawing Xs in boxes and large ships not fitting into 8km maps.
Now I'll concede that you didn't, recently, say the game isn't using double precision, but you did keep talking about the same imaginary limitations you were talking about when you thought it didn't. Mea culpa.
Right. However, I still stand by my original claim because i) you - or any dev - has yet to disprove, let alone explain it ii) CIG hasn't built even 1% of the starmap and from which we can deduce the merits of how it was done
That's just the same nonsense that's going on whereby - for some reason - backers who are adept at theory-crafting dreams, can't tell the difference between 64-Bit sized scenes, 64-Bit programs, 64-Bit world positioning. And that is because CIG continues to obfuscate this - and several other issues - without any semblance of clarity. It's like you guys are afraid of saying something that's going to somehow go against some crap croberts has uttered in public, and which most of you (the guys working so very hard to build this game) know to be complete - and utter - rubbish. In fact, I just wrote about that here as part of this on-going discussion.
3.0 is still coming for the end of the year though, right?
I mean, it's not on the schedule but that just means they want to surprise us!
Maybe I need to write a letter to Santa Claus...
I thought I remember seeing something that showed it was moved to Q1 2017. Not sure if official, though.
You didn't; because that never happened. Since they announced and hyped it at GamesCom, then CitizenCon, then got millions based on those promises alone, they haven't addressed it since. Not even during the anniversary stream from earlier this month.
And thus far, it's in the schedule as a bullet point, with no status (like 2.6) whatsoever - pretty much how it appeared in the event slides.
So basically, they're pretending it never happened because, well, as I said weeks ago, 3.0 simply doesn't exist. It was just another milestone bullet point, like all the others before it.
Aye...I spent too much time on the subreddit. Can be hard to sift the theory from fact. Thanks! +1 Virtual rep since I seem to be agreeing with you recently![]()
Yeah, seems CIG's hasn't learned anything from Illfonic debacle: they outsource, they find at the end it's incompatible.
So according to the schedule Evocati got star marine yesterday and will be done with it tomorrow, then it's going to the PTU on the 1st.
How's it going? Is it "more lethal than Call of Duty"?
![]()
So according to the schedule Evocati got star marine yesterday and will be done with it tomorrow, then it's going to the PTU on the 1st.
How's it going? Is it "more lethal than Call of Duty"?
![]()
Shirley you jest. Of course it wasn't released.![]()
Seeing as it is just a new map with an added layer of leader boards and match modes don't expect anything fundamentally better than what is in the PU, currently.
The SM I remember from a year ago looks nothing like what they showed, not just in visuals but also in shooter mechanics and assets.
No. Chris Roberts told us Star Marine would be an incredible FPS experience with vaulting and cover and it would be both fast paced and tactical like Arma.
The current <<Use>> function is just as glitchy and clunky as ArmA.......
All that reddit seems to demonstrate that more people than I previously imagined are even more clueless about networking than I originally anticipated.
Thanks. There is SO much info out there, and a lot of it is just gets missed or is confused with other things. That's why I'm the StarWhisperer
In other news, more backer theory-crafting (with a bit of borderline doxxing mixed in). This time about Star Network.
Choice comment:
Anyways, as far as I can tell his code allows for technically infinite scaling of connected clients. He's set a bunch of constants, similar to what OP mentioned in his edit, for testing purposes but those constants can really be edited to any value and his code should run.
The actual limitation is going to come from the server hardware. He's done a lot of cool things to help with performance, and it seems like he's going to add more, but the biggest help towards maximizing player limit caps isn't going to come from him really.
The biggest help will come from CIG reducing the amount of data that is actually being sent through this system at any given time. The code can scale with packet size and how frequently those packets are being received by the server, but the larger the size, and the higher the frequency, the longer it's going to take the server to process that packet and send it back out.
I think they'all hit the transport limit of most players potato routers before the backend is seriously impacted. If they are going to have 128 clients many will find their router chokes on 32 and throws up the last 8![]()
If it did go avocado yesterday, they'll be crowing about it on AtV later this week...
If it's glossed over or omitted from AtV, then that will be enough of a public admission.