Support NeoGAF

[elyetis]

You have no idea how this works or what your talking about...so stop speculating.

His assumption that bugfixing goes at a constant rate is wrong... but even if 3.0 got released today, his Christmas release date speculation would still be closer to the truth than the estimate/speculation given by CR last year.
So clearly Camp Freddie shouldn't feel too bad about maybe ( probably ) getting it wrong.

 

[staticneuron]

So, based on their projected "must fix" bugfix rate, they're 5 months from an 'Evocati' release.
I assume the evocati is like the pre-beta version, but I'm not up-to-date with the jargon.

Yet they are also saying in the same progress tracker page they have a release target in less than 3 weeks.

I have no idea what to believe at this point, but I'm going with the Christmas release date, based on previous "it'll be at [convention stream]" followed by "actually it's not ready, but it's nearly done" followed by "we're not doing it until the next update, stay tuned for big news at [next convention]".

Most of these charts you see come from reddit. Not sure if the explanations go along as well. They fixed many bugs but were able to see new ones in the process. That is pretty common but there is no telling when they can clear what they have left. The could do it within a week or it may take until end of year. It is best for them to set an aggressive date.

The one I love to read is the "No Bamboozles Release Forecast for August 18th"

Updated by jdlshore aka James Shore. He wrote the book "The Art of Agile Development" and does his best to explain his predictions and reasoning.


In his thread he puts forth

CIG had 83 Evocati bugs left last week. They have 77 bugs left this week. That's a net fix rate of 6 bugs per week. Does that mean it's going to take 77 ÷ 6 = 13 weeks to release to Evocati?

No.

The rate at which new defects are added isn't linear. It goes up at first, then goes down. This means that the net fix rate will probably speed up as the release approaches. You can't just divide the bug count by the net fix rate and get a useful trend. For the same reason, I'm skeptical of the trend line CIG showed on their bug burndown chart. (Burndown charts aren't likely to be useful for their task progress, either, because they don't use the Definition of Done burndown charts assume.)

As I said, early in development, the new bug rate actually increases. And this week, the net fix rate is worse than last week. Does that mean the rate of bug discovery is still increasing, and the net fix rate will be even worse next week?

No. It could be worse, it could be better. There's no way to tell. There's some randomness to the rate at which bugs are found. We could be on the upswing of adding bugs, which means we're a long way from release, or we could be on the downswing and just having a bad week. Or we could be at the tail end of the curve, but seeing a lot of new bugs from a directors' review meeting. We don't know how many bugs were added vs. fixed, either.

In other words: When will all the pre-release bugs be fixed? No one knows.

The forecast model has an opinion, of course, but as I've mentioned before, its task-based prediction is less reliable at this stage of the game.

is it normal to him?

Depends on what you mean by grey areas. If they ship later than my model predicts, that's a flaw in the model, not their development process.

Their development process seems very normal and ordinary. No surprises whatsoever. It's a very classical approach to development combined with intelligent big-cycle iteration, incrementalism, and evolutionary design (refactoring) along with small-cycle sprints. I prefer a more aggressive approach to incrementalism but that doesn't mean their way is wrong, and their approach is much more common in the industry than mine.

Polishing is an essential part of pre-release bugfixing. How much to do is a business call based on strategic priorities. Their focus on polish is very consistent with what they've said about their goals. I think it's the smart business decision.

There are many other programmers that chime in and find the process very familiar. The best suggestion is not to try to assume because even CiG does not know the exact date (until they fix all the blockers ).

Lol, I'd be really interested where that "word" is coming from

Spoiler

because it's not what is happening

Do you have some insider info because... wow...

https://twitter.com/RobertsSpaceInd/status/899640380681056256

The surface of Daymar certainly is not in 2.6.3.

Funny how often these rumors are wrong.

 

[FunnyJay]

You have no idea how this works or what your talking about...so stop speculating.

Interesting way to foster discussion on this forum.
You are very quick to shutdown anyone that utters any doubt in this thread, no matter how small. Often citing how much you know about this project and development in general...

How about trying to ease up on people and being a bit more humble and friendly?

 

[Eolz]

Do you have some insider info because... wow...

https://twitter.com/RobertsSpaceInd/status/899640380681056256

The surface of Daymar certainly is not in 2.6.3.

Funny how often these rumors are wrong.

Spoiler

I work there, I try just to not comment on rumors/speculation unless it's something really dumb

 

[staticneuron]

Spoiler

I work there, I try just to not comment on rumors/speculation unless it's something really dumb

Thanks for heads up!

 

[Raticus79]

Spoiler

I work there, I try just to not comment on rumors/speculation unless it's something really dumb

lol, that's great

 

[Cabbagehead]

Interesting way to foster discussion on this forum.
You are very quick to shutdown anyone that utters any doubt in this thread, no matter how small. Often citing how much you know about this project and development in general...

How about trying to ease up on people and being a bit more humble and friendly?

That wasn't "doubt" that was a gross assumption.

But it doesn't matter. He can still post...not like i'm a mod.

Rat knows that this is my MO...love it or leave it.

Spoiler

unless it's something really dumb

There's a lot of that unfortunately. From particular places...

 

[iHaunter]

Looks like 3.0 is being played in the booths, not 2.6.3.

At least *some* version of 3.0.

 

[Zabojnik]

Looks like 3.0 is being played in the booths, not 2.6.3.

At least *some* version of 3.0.

This can only mean that they'll be showing 4.0 during the livestream. *iwanttobelieve.gif*

 

[Machina]

This can only mean that they'll be showing 4.0 during the livestream. *iwanttobelieve.gif*

When is the stream?

 

[Zabojnik]

When is the stream?

I always wanted to quote myself.

Friday at 21h CEST. Or 9pm (12pm PDT / 7pm UTC), as some people prefer to call it. Plus the traditional delay, of course.

 

[Spuck-uk]

So this Gamecom build, it's another scripted/sandboxed area right?

 

[MasterOfPastures]

So this Gamecom build, it's another scripted/sandboxed area right?

Probably the still buggy 3.0 release candidate. Friday stream version too. Expect a lot of slendermen and crashes.

 

[Spuck-uk]

Probably the still buggy 3.0 release candidate. Friday stream version too. Expect a lot of slendermen and crashes.

..and a ship sale.

 

[Megalo]

No, people are playing the 3.0.

 

[Spuck-uk]

Not right now they aren't, the build keeps crashing

 

[Megalo]

They are just doing an update

 

[Daedardus]

Not right now they aren't, the build keeps crashing

Those are clearly file explorers on the screens and nobody is even seated and trying to play the game, I don't know how you could even come to that conclusion.

 

[Raticus79]

Not right now they aren't, the build keeps crashing

Source: "They are currently updating the game live at Gamescom will play a little more stable version in some minutes "
https://www.reddit.com/r/starcitizen/comments/6vb8ti/they_are_currently_updating_the_game_live_at/

Some interesting bits in there:

They just copy the files from the local server over to the computer. Probably faster than using the patcher.

And I have to say it was very cool. I tested many things. Many things were disabled and there were some bugs but on the other hand many new things worked surprisingly well. I like the flying in the atmosphere, i like the new interaction system, i like the rover, i like the new turrets and the ui. It was really fun

- what did you like most?

The interaction system. It was not implemented everywhere correctly or fully, but with that you have much more control over everything and the highlighting works very good (when it was implemented)

- Can you tell a little more about the atmospheric flight model? Does it fly like an airplane?

I wouldn't call it realistic, but you feel some drag and leaning the ship let's the ship make a big curve.

 

[jaaz]

 

[ryousuke]

Damn, it's really something

And the rework of the Cutlass is pretty cool as well

 

[staticneuron]

A version of the Alpha that hasn't made it to Evocati testing because of Bugs, keeps crashing?!?!?!


Who knew?




Oh that's right.... CiG and they keep informing everyone on their Production Schedule Report page.

 

[Anton Sugar]

Ooh, I like the look of that canopy.

 

[Cabbagehead]

The advent of this 3.0 build, really does make all those obvious shit post that popped up in here, claiming -with confidence- that CIG was being lazy or whatever other baseless accusation. After the lie that was 2.6.3 being the thing that would be shown at Games-Com show floor. Even though that didn't make sense at all and wasn't something that CIG even noted, when they detailed the event.


Makes this even more funny and revealing some folks agenda.

 

[MasterOfPastures]

Damn, those are some nice screenshots. Friday stream cant come soon enough. I hope we get to see v1 of their gas giant tech, or clouds, or even a large city. They are making it all real, finally!

 

[Pomerlaw]

hahahahahahahahha

hahahaha just damn

 

[Snake29]

 

[Cabbagehead]

snip

 

[jaaz]

On final approach...

 

[pottuvoi]

Star Citizen: Live from the Gamescom Showfloor, Day 1

 

[Aztechnology]

Can't wait to watch some high quality highlights tomorrow. Watched a little bit of the stream just now and it looks great. But I need sleep badly. Catch you guys tomorrow.

 

[derFeef]

Star Citizen: Live from the Gamescom Showfloor, Day 1

Man die on planet stuff looks good.

 

[pottuvoi]

Man die on planet stuff looks good.

Yup.
Looks fun.

 

[Lord Panda]

Been following the stream and watching quite a few of the Around the Verse videos prior. Perhaps someone can clear this up:

The ships don't seem to have much weight. They seem to just defy physics and just launch effortlessly into the air. Kinda reminds me of No Man's Sky in that regard


They just 'pop' up into the air Hard landings also aren't punished and the ships just bounce to a stop. While I like the red outs/black outs, the flight model seems far too forgiving.

I didn't realise you could just hover and pivot on the spot like that mere metres from the ground. I didn't see any thrusters maintaining the ship's position unless it's some anti-grav BS mechanic at play. In this regard, it reminded me of the Banshees from Halo.

My question is: are the flight and physics (weight, lift, inertia etc.) going to be beefed up considerably so that take offs and landings at the very least require some skill?

 

[lefty1117]

Were they still planning to release 3.0 in September or did that get pushed again? Because based on the comments from Gamescom above I'm not hopeful for September.

 

[Galava]

Been following the stream and watching quite a few of the Around the Verse videos prior. Perhaps someone can clear this up:

The ships don't seem to have much weight. They seem to just defy physics and just launch effortlessly into the air. Kinda reminds me of No Man's Sky in that regard


They just 'pop' up into the air Hard landings also aren't punished and the ships just bounce to a stop. While I like the red outs/black outs, the flight model seems far too forgiving.

I didn't realise you could just hover and pivot on the spot like that mere metres from the ground. I didn't see any thrusters maintaining the ship's position unless it's some anti-grav BS mechanic at play. In this regard, it reminded me of the Banshees from Halo.

My question is: are the flight and physics (weight, lift, inertia etc.) going to be beefed up considerably so that take offs and landings at the very least require some skill?

This is one of the things I want them to do. To feel the weight of the ships when taking off, landing, moving...

When taking off, engines go to 0 to 100 in an instant thus making the ship just "pop". They should increase gradually and slowly "push" the ship upwards until it starts "floating". Same with landing, if you go to hard, the landing gear should just be crushed by the weight and force of the ship falling.

I know it must be very hard, but would love to see some serious progress on that. I don't care if the game takes more time to come out honestly, just make it good.

 

[Pomerlaw]

Been following the stream and watching quite a few of the Around the Verse videos prior. Perhaps someone can clear this up:

The ships don't seem to have much weight. They seem to just defy physics and just launch effortlessly into the air. Kinda reminds me of No Man's Sky in that regard


They just 'pop' up into the air Hard landings also aren't punished and the ships just bounce to a stop. While I like the red outs/black outs, the flight model seems far too forgiving.

I didn't realise you could just hover and pivot on the spot like that mere metres from the ground. I didn't see any thrusters maintaining the ship's position unless it's some anti-grav BS mechanic at play. In this regard, it reminded me of the Banshees from Halo.

My question is: are the flight and physics (weight, lift, inertia etc.) going to be beefed up considerably so that take offs and landings at the very least require some skill?

My main problem with the game as well. They have to make the acceleration more gradual.

 

[Agremont]

Yeah the hummingbird flight physics is my biggest gripe with this game. It's probably going to break the game for me honestly.

 

[Raticus79]

The problem is thruster strength. With acceleration in the neighborhood of 5 G for manuevering thrusters and 10 for the main engines (15 on afterburner), at low speeds your ship is going to feel like it has very low mass unless they artificially limit that thruster strength quite a bit at low speeds / near planets and stations / whatever conditions make sense. They can't just make them weak across the board or the combat would turn into Kerbal Space Program.

They made some adjustments to that rate of change of acceleration a while ago, but they'll still need to do more.

Collision handling is a separate thing on top of that. Realistically, instead of bouncing around like they currently do, the ships would sustain such heavy damage on the first collision that they'd be written off. Making the collisions less elastic would help (having the collision response behave as if things had crumpled appropriately), but for gameplay reasons they can't make it that accurate.

 

[Agremont]

I wouldn't mind if the combat got a lot slower and tactical. In fact I would prefer it. I know it won't happen though. It's going to be twitchy and "strafey".

 

[Raticus79]

I wouldn't mind if the combat got a lot slower and tactical. In fact I would prefer it. I know it won't happen though. It's going to be twitchy and "strafey".

Yeah, I wouldn't mind either. I like that yaw is unrestricted compared to Elite, but there's a better middle ground out there. It's not like it needs a fundamental rework though, just tweaking some parameters.

 

[Galava]

The problem is thruster strength. With acceleration in the neighborhood of 5 G for manuevering thrusters and 10 for the main engines (15 on afterburner), at low speeds your ship is going to feel like it has very low mass unless they artificially limit that thruster strength quite a bit at low speeds / near planets and stations / whatever conditions make sense. They can't just make them weak across the board or the combat would turn into Kerbal Space Program.

They made some adjustments to that rate of change of acceleration a while ago, but they'll still need to do more.

Collision handling is a separate thing on top of that. Realistically, instead of bouncing around like they currently do, the ships would sustain such heavy damage on the first collision that they'd be written off. Making the collisions less elastic would help (having the collision response behave as if things had crumpled appropriately), but for gameplay reasons they can't make it that accurate.

Make the thrusters consume a lot of energy and more susceptible to malfunction with high thrust levels, making it possible for fast paced combat but not practical when travelling.

Actually I don't know, but such tiny thrusters being so powerful is kinda immersion-breaking (yes, i know it's sci-fi, but idk, it's the feeling i get)

 

[Dubbedinenglish]

The problem is thruster strength. With acceleration in the neighborhood of 5 G for manuevering thrusters and 10 for the main engines (15 on afterburner), at low speeds your ship is going to feel like it has very low mass unless they artificially limit that thruster strength quite a bit at low speeds / near planets and stations / whatever conditions make sense. They can't just make them weak across the board or the combat would turn into Kerbal Space Program.

Without going on my standard points I'll just say they need to look at the EVA movement in their own game. It's far more plausible and frankly better than the ship flight.

 

[Spuck-uk]

Watching the 12-person servers need to be restarted every 15 minutes because of memory leaks, and ships clip entirely through the game world. Amazing.

At least Battlefront 2 looks fun

 

[KKRT00]

300 shots from gamescom build
http://imgur.com/a/8Clyz

 

[Spuck-uk]

https://clips.twitch.tv/EnchantingFamousLemurDAESuppy

Laffo

 

[Dubbedinenglish]

300 shots from gamescom build
http://imgur.com/a/8Clyz

Happy to see marked improvement on the cockpits and HUD/HDDs

 

[Agremont]

Watching the 12-person servers need to be restarted every 15 minutes because of memory leaks, and ships clip entirely through the game world. Amazing.

At least Battlefront 2 looks fun

Yep. Not impressive. None of the features of 3.0 shown either.

 

[Eolz]

Watching the 12-person servers need to be restarted every 15 minutes because of memory leaks, and ships clip entirely through the game world. Amazing.

At least Battlefront 2 looks fun

It's restarted every 15mn so other players can play and have a fresh experience, not due to memory leaks...
No excuse for bugs outside of game development being hard though

 

[Jinroh]

I hope the in atmosphere physics are temporary, these ships are supposed to feel like heavy bricks, and not hover like they're in space with no force applied to them.