Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Locked thread
BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
"You can't take the pie from me" :v:

Adbot
ADBOT LOVES YOU

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

Timby posted:

And that's Roberts' MO -- he hates being told what to do. You see that in the massive, embarrassing letter he wrote in response to Smart and the criticism, saying that EA had approached him multiple times about rebooting Wing Commander and he always turned them down because he'd never work for a publisher again.
This mindset is something that's come out of the Origin Systems.
Roberts was a coding hotshot who moved from the UK and was hired by Garriott to work at Origin. Having a bit of cheek he talked Garriott into letting him create his game concept called "Squadron" which basically was a sci-fi take on Battlehawks - the US/Japan pacific theater in space with wrathful cats and borrowing off Niven and other sci-fi plots of the day.

At Origin he was given freedom to sink time and effort into this game and this was not cheap.
In 1990 a typical big game budget was around $100,000 to $200,000 - Wing Commander allegedly cost $1,000,000.
Origin insiders allude to some creative accounting to hide that fact from Roberts - but despite the success of WC and Ultima - it wasn't enough to save Origin's coffers and EA moved in to acquire in 1992. Only because of the addons and voice packs did Wing Commander 1 break even.

Some say that Robert's pushing the envelope for production funds on a game really ruined the game industry - it meant a massive risk could equal a reward and people began investing heavily into quality over quantity as it began the push of the industry into hit-machine mode and publishers started focusing on sure-fire successes over taking a risk.

With EA Roberts gets lucky again as the FMV fad kicks into full drive and his dream of creating movies becomes apparent through Wing Commander 3 and 4 - a record investment for EA who only funds WC IV on the provision that it would be done in twelve months.

That also hit EA hard - by 1996 the space sim genre had been slowing down and 12 million isn't an easy thing to magically recoup. The RTS genre is the new big-thing and people are adopting mice and keyboard over the venerable joystick.

The aftermath of WC's cash crater almost scuppered Garriott's plans for Ultima Online in which he had to beg for $200,000 or so to develop a proof of concept that basically was hacking in multiplayer into Black Isle. MMO's back then were a high risk concept and it wasn't until thousands of people started paying $2 for beta CDs then EA realized they had a hit on their hands and a way to start recovering costs.

But where Garriot had gotten his rewards Roberts was being kept at arms length over the possibility of a Privateer or Wing Commander Online. No doubt it was a frightening prospect to many at EA and Roberts saw the writing was on the wall so he went off and founded Digital Anvil to create his dream online space game, believing the technology of 1999 was enough to create his dreams!

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

janssendalt posted:

interesting
any sources to back any of these claims, though?
Here are the bits about game budgets and WC's cause and effect.

There's a pretty common pattern that Roberts got handed his wishes on a silver platter and managed to survive purely by being in the right moment at the right time in an industry that was still developing it's feet and publishers were keen to take chances.
By the time 1999 had hit the golden age of risk taking was pretty much on the wane and it was become apparent that big names developers had been uncannily lucky and big failures like Ion Storm only served to prove that lightning doesn't strike twice and publishers saw it wise to not allow one person to head a project without some sort of control.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

toanoradian posted:

How do you know all this, WebDog? Are you like an amateur historian of video games?
I suppose so.
It is kinda fun digging through old game magazines or industry reports of yore. Makes a change from the usual :v:
Many interviews back during the 90's are fantastically unguarded with very little PR gloss, especially as the PC market developed you begin to have more overt personalities like the Bitmap Brothers or Romero emerge. One later publication, PCXL is fantastic, as despite being completely juvenile, it's more relaxed approach has some pretty interesting interviews.

There's also some good books that reveal the politics on how games get developed. Strategy guides also occasionally hold some interesting interviews about the making of games. You do have to read between the lines a fair bit as the situation is a bit more complex than "publisher bad - indie dev good."
And no I don't have a house full to the ceiling of magazines, there's the internet for that.

Annoying much of the gold is held behind NDA's and lawsuits.
Romero cannot talk about Diakatana's mess and the same goes for Broussard about DNF. Sadly the same will happen once Star Citizen implodes. Some staffers might talk about how bad things were but you likely won't get much word out of Roberts.

When Digital Anvil went into acquisition he tried his best to gloss it up:
The figure for Digital Anvil's acquisition is undisclosed. Starlancer was brought out for $300,000 in Warthog shares.
A speculative list of major acquisitions from 1991 - 2003 - I'd guess Freelancer might have been around $30 - $50 million with Roberts getting a golden handshake and percent of profits before trying his luck in Hollywood.

quote:

GS: In a time when a developer's freedom is fast becoming scarce, why would you choose to make Digital Anvil a division of the world's largest software company?

CR: It's very difficult to build wildly ambitious games without the financial security of a larger company, especially when their development cycles are so unpredictable. Freelancer was meant to take three years - it will probably be four and a half years by the time it is done. It's very tough for an independent developer to weather that kind of slip, and it's not just Digital Anvil. I believe all A+ games require severe patience and funds - witness Team Fortress 2 and Black & White.

GS: And how will this acquisition benefit the development of the Digital Anvil games?

CR: Whenever something runs later, it needs more funding. Becoming part of Microsoft made this issue less of a problem...
...Security in being part of an ambitious publisher. In-house games always are the favored children [of publishers]. Believe me, I know - this was totally the case with Electronic Arts when I was there on the other side of the fence.

The bottom line is Roberts had all of the resources at his disposal to create a sure-fire successor to Wing Commander that would have been well received - but he let his ambition run off in trying to keep up with modern times. Compare the gamespot interview with that insane rant. In 2000 he can grin and bear it. In 2015 he is so on the ropes that the only thing he can muster is a fearful rant.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
Kickstarter has been an interesting equalizer and kick in the face to developers who have realized that publishers do have a use. Double Fine's massive behind the scenes doco on Broken Age is a really revealing insight into how insanely stressful trying to develop a game on the back of goodwill can be.

In other cases it's served as a sobering moment for others, for instance Tom Hall who tried a few times to get something going: namely "Old School RPG" a pitch that basically was buzzwords for an "old school RPG" that relied on his star status had no indication on what kind of story or world the game had.
And later he tried with a spiritual successor to Commander Keen "Worlds of Wander" which was a create-your-own platform game that failed to get going with only $100,000 of it's $400,000 raised.

Beer4TheBeerGod posted:

There's also the very real, and very unpleasant risk of any publisher having to take on a huge liability of a incredibly angry and toxic player base.
Affirming this.

Back with Freelancer press was pretty scarce, we hadn't heard much of the game in several years beyond a few E3 displays here and there. The most news you got was was when it got acquired so Microsoft was able to market the game with a relatively clean slate and using Chris Robert's name as a brand. The WC community ate it up as "booo Microsoft" and sort of coped as it was the same "evil publisher" story as had been with Origin and EA.

You'd have to basically shut down Star Citizen and work out a very cunning PR as you scramble to rework aspects of the game. One way that they'd survive community backlash would be if EA brought out RSI only to announce a reboot of a Wing Commander space sim with Erin Roberts as a main designer and Chris Roberts as some sort of consultant in name only.

Then Star Citizen would continue in the hands of die-hard modders who bring over the design into Freespace 2's engine and manage to make something functional in less than a year.

Dongattack posted:

Lol Webdog, have you found your new Buckley? I've missed you and you particular set of skills. (free Webdog)
All I'm missing is Roberts writing up blog posts about how his family doesn't understand his dreams of being a webcomic artist game developer. There's frustratingly not much meat to be found as he pretty much kept his poo poo together until the Hollywood dream went tits up.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

D_Smart posted:

He's not a creative visionary. There isn't anyone that I know in the industry that regards him as such. The disinformation comes from the usual Wing Commander hyperbole.
Much of that came from the press where Wing Commander's success cast a bit of a shadow over any other sci-fi combat sim.

The biggest hyperbole is that WC's success at a convention (where Roberts spooked people declaring Wing Commander was a reverse engineered Battle Hawks) forced LucasArts to create X-wing, but that mostly came about in part from the company only just getting the IP rights for Star Wars back and LucasArts already having a bit of a history in creating World War 2 flight sims kind of saw it as a no brainer to hire Totally Games to create something in the Star Wars universe.

The "visionary" title people got from that time is mostly down to being in the right place at the right time and publishers moving into the "make a top hit" mindset and promoting programmers front and center, many who became millionaires in months.

Molyneux is a classic example.
His first game Entrepreneur was him putting his know-how as a failed software businessman into a game and that failing so hard that it only sold two copies.
After forming Bullfrog his big hit Populus was a happy accident.

quote:

“The fact that I programmed it meant some of the fundamental things that programmers [Glenn Corpes] can do in their sleep I couldn’t do,” he said. A particular problem was getting the virtual people to navigate around walls rather than getting stuck. “I didn’t know how to do that. I tried to do it, tried to invent it myself and couldn’t and I thought ‘oh gently caress it, I’ll just get the player to solve the problem for me by raising and lowering the land’. That became the game’s fundamental mechanic. Pure and utter luck.

Remember these early developers started by forming their own software company off their own backs because no one would publish their ideas and when they became hits they were heralded in promotional materials and so on. However very few really survived by the end of the 90's and most have quietly retired or work in the background.

The Kickstarter resurgence fad is fascinating as you have to remember many of these older developers started out on budgets of $200,000 - $500,000 so in their mind they have already made a success in that ballpark and somewhat naively assume they can take a shortcut as there's complete game engines already out there - just hire a programmer and artist who will achieve your grand vision of the 21st century!

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
How old is she? Given she claims she's 22 In this 2005 article. So was she born in 1980 (as she claims on imdb) or 1983?

Also that site isn't an official BMW page - it's for Z8 enthusiasts who go around taking scenic photos. So she charmed some photographer with exciting tales about being a bit player in Hollywood.

https://www.youtube.com/watch?v=wYaXw8TZEvs

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
Every time I see buggy gifs I'm reminded of that 3D Mark 2001 truck demo game.

https://www.youtube.com/watch?v=M2qoFiXxCck

In fact everything looks like a failed 3D Mark test run run on hardware that is glitching out enough that physics calcs are failing and things fall through.
I recall Cryengine games could be really interesting with it's physics...

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

Sushi in Yiddish posted:

Tie fighter was pretty neat when you get a few campaigns in, and you're not just fighting against rebels and pirates, but getting caught up in Imperial intrigue.
Also the designers tried to keep the missions interesting and mix things up a bit over replays. Inspecting certain things could potentially trigger a new wave of ships, or if you wre getting on top of the odds more ships came in to give a bit of a challenge. Or now and then a mission might have a curve-ball hidden that would randomly trigger.
And as mentioned before uncovering all sorts of hidden bonus objectives and getting to know the secret order.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

Gerbil_Pen posted:

I didn't say that. I said his vision issue was unrelated to his weight. I have no loving idea if he was dat as a kid.
He's impaired enough to have had to learn Braille as a kid.
Also he was a pretty large kid fifteen years ago but only recently has become that corpulent. Stress eating on the company's dime no doubt - what else does he do there all day beyond read the forums and make videos?

And going by his twitter he's pretty deep into some sort of depression. It must be hell having your "dream job" land in your lap only to end up being a fly on the wall to all sorts of shouting matches as your visionary idol turns out to be really really bad at his job.

I would not be surprised if a ton of former employees are incensed that he's still around despite contributing nothing meaningful to the team. I wonder how many quit in part from having had to put up with him being in design meetings and thinking he too is a designer by making random ideas that Chris then enforces and ties up development.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

RagnarokZ posted:

Is that really how the game looks? It looks like a vaguely improved version of loving Freelancer.
Isn't that from when someone ran Doom 3 off a pair of Voodoo 2's?

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
That Pupil to Planet thing is one tragically hilarious attempt to upstage ED's own little video they made during beta testing with the debug camera. No fancy polish or syncing up animations just "hey I zoomed out when mucking around and this looked really cool, so we showed it off."

You have to understand that Cryengne basically runs off creating a landscape via heightmap where you then add in prefab components to create rooms and so forth. It was coded for a tech demo for Nvidia and then expanded into a game. Large voxel landscapes full of procedurally generated trees is a pretty old trick in the book.

It's meant for fancy landscapes nothing more. Its causing so many issues as deep inside the game wants you to stick to a flat surface. It's not coping well trying to pretend it's in space. It's likely why the physics are breaking as they push game objects out of their expected physic rules and it tries to reset things.

The 64 bit thing is them attempting to smooth over any gaps in the calculations but this will just slow everything down as the CPU overhead increases.

Reading up I've discovered that modders have actually played around with using Cryengine to create planets that you can fly off into space with, but hit brick walls.

quote:

All you need is to do is create a spherical terrain as a model not terrain with a 3d modelling software. fix a gravity sphere in the exact centre, be sure to negative the gravity on all the axis like so
y -10
x -10
z -10
I'm sure this method can work with water, or props, vegetation, ect but it will all have to be 3d models (cryteck geometry file, cgf)
Though there is one problem, when going to 3rd person view, for vehicles and first person, the camera will stay oriented to the flat default world.

And from someone who's attempted a Kerbal inspired mod and actually got the ability to fly off a planet.

quote:

My custom code allows the player to quite naturally walk on the terrain even if it's tipped at, say, 45 degrees. When you blast off into space the terrain fades to invisibility and the objects switch off so you then see the planet as a sphere.

In the screen shots Earth is built out of multiple sections and its diameter is 40 km. The code scales all planets dynamically so they always have the correct apparent sizes.

But here's the problem. I would like to have normal AI characters and vehicles working on the terrain, even if it's tipped at 45 degrees. I found that gravity volumes can be used. They create the correct forces so, for example, AI character fall at right angles to the terrain. The problem is that their stance is still tied to the normal Sandbox normal, so they look completely wrong. Somehow the AI characters and vehicles have to 'know' the direction of gravity.

BogDew fucked around with this message at 14:37 on Dec 17, 2015

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

A Neurotic Jew posted:

my adderall ran out and I cannot finish that WebDog post :(

Not ready reading post...
Abort. Retry. Fail.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

Scruffpuff posted:

This makes me think even more about how Chris doesn't understand how people think, or gaming in general. To be immersed in an experience like a game, has nothing to do with seeing the same things on your monitor that you would see in real life.
If you watch the Wing Commander Making of videos for the games you get the impression he was bitten by the FMV bug and saw himself as a pioneer in creating an interactive-movie-game.
He also has the same glazed look back in 1993.

What he seems to want to do is put you right in the center of all these fantastic effects sequences from movies and is assuming the playerbase will create these with the tools he gives them. So he obsesses over all these things like dynamically exploding interiors and immerse landing bays without really parsing how they operate as a game.

No doubt he's seen all of the dramatic action sequences in CoD and Crysis and thought "I can do better!" without really realising how much smoke and mirrors go into making sure the player is kept moving through the scene as fast as possible before the animations end.

The other problem is that is overconfident of the tech available today. He genuinely assumed you'd pick up the CryEngine, plug in some ships and mocap some actors without understanding the engine's limits and just expecting his team to solve things by hard work.


To give a quasi comparison there's a Star Trek mod called Excalibur which aims to be a remaking of Bridge Commander. However they've been rocking around for years slowly trying to code their own custom wunder engine. Recently they threw in the towel and moved to Unreal 4 and in a few months put this out.

https://www.youtube.com/watch?v=Dt8Nkj_iOLQ

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli


I bet the corporate espionage line is something Ortwin cooked up to give some thin excuse for a ban. It also is so overblown from the actual crime, which is to divulge trade secrets from the inside-out in order to gain an advantage over competitors.

This is a stalling tactic as I suspect most of their money is kept away in Germany under Ortwin's company accounts and they can't be arsed eating the international wire fees and currency conversions.

Or there's still an investor or two in the ranks and they're trying to fudge the numbers so it doesn't look a dire exodus of backers on any reports.



Buy yeah holy-hell does the narcissism stream off that letter. That sales-woman line is incredible. Note she doesn't mention her reputed study done at UCLA school of management as her qualification. Stuff that went off her LinkedIn profile once people started questioning things back in 2014.

According to her meet the devs video.

quote:

SG - Now, I actually had basically been doing a lot of modelling and acting and also working in other areas in marketing - as marketing manager, marketing director - so I had had other marketing roles.

JP - So, you came from fashion and that sort of industry, correct?

SG - I did. So, yes. Marketing for nightclubs and restaurants and stuff, marketing in fashion, and then - Actually I just recently looked at some work references that I just found and I worked in commercial real estate doing marketing for them.

Yup her complete "marketing experience' was as a promoter in a club or as a model. Look pretty and hand people a flyer and perhaps throw in a rehearsed sales pitch.

I love the gradually aggressive flourishes as she tries to handwave hard facts before drowning in a word salad then dizzing Elite, a game created by a public traded company which puts it's earnings online. Ask Sandi for an annual report that shows balances, liabilities, risks and so on.

And dear lord at the idea of Chris coding when he hasn't done anything solid since Wing Commander 1!

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
James Cameron will personally nail gun your phone to the studio wall if it goes off.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
I like how in Sandi's resume she's neglected to mention her work in the Sydney Theatre Company. Also I notice all of her uncredited roles have moved up to supporting roles. A supporting role isn't appearing in the background, it's a fully fledged speaking role.

Her showreels are terrible as she does the rookie mistake of playing extremes and not having someone competent direct her so everything comes off as flat.

Comedy: is this off South Park? It's meant to be an enthusiastic upbeat positive planning scene but plays like this is take 30 when she's finally gotten the lines right, but is exhausted from all that filming.

News hosting: I'm more stunned by the horrid graphics and the fames where Sandi vanishes. Flat as a pancake reading off Teleprompter with no real personality.

Interrogation: Given the rumours swirling CIG there's a bit of unintended hilarity with the character facing charges of corporate fraud.
She doesn't tell me anything about this character's motivations or whether I should feel for her or not.

It's also very poorly directed and confusing to watch. It's meant to be two separate locations but while Sandi changes appearance the interrogator doesn't and neither does the room so any power in the performance is killed from distracting editing.

Reverse the Verse:
That dialogue is hilarious "no other gaming company is going to hire you....you came from a hospitality background." is she flat out admitting the truth?
The whole scene suggests the person was so useless but was still hired on goodwill. Then fired. Then after "you're fired!" she breaks character and admits her delivery is being too nice. And then some bizarre "horror" thing afterwards.

Mo-crap:
It's really hard trying to read a performance when most of it is obscured by mo-cap rigs. She doesn't give any gravity to whatever the situation is, while blinking like mad.
Ginger Lynn comes off as a more convincing space mechanic, even if her dialogue has too many double entendres..

RSI are in deep if they're doing mocap for a whole scene. There's a monitor that gives a hint of what they have done. It sort of looks like they've designed the sets, taped out everything and then hope no one inadvertently walks through a wall. Then she whacks the rig with her helmet, that's going to be recorded as a very hard to fix bump.

So while it looks like you get next gen FMV there's so much more work and animation that needs to be done, such as zipping up the flight suit, tracking the helmets and cleaning up all sorts of other issues like clipping though costume or set elements.

Oh and if you want an idea of what has happened before, here is the tale of Silicon Knights completely butchering an X-Men game by getting engrossed in small details.

BogDew fucked around with this message at 06:25 on Jan 7, 2016

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
Rebel Assault 2 kind of holds up...a given when you're attached to one of the pioneers in visual effects so you can actually grade your greenscreen elements to fit.

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

mdxi posted:

Is everything surrounding the film industry this skeevy, confusing, and quasi-legal, or is it just everything surrounding Star Citizen?
It sort of goes both ways. Hollywood accounting is a thing, applying it to a video game is something reasonably new.

There are tons of actor agencies, often created by actors who can't get a break so either start acting schools or operate as an agent. Many aren't actual agencies but are merely a listing service, so you pay $50 for an on-line account.

The scammy ones will charge you upfront, put you on a suspect contract and charge you monthly and then hassle you to bits wanting more cash for photos and so on and then change names.

Bohemia Entertainment is kind of low in the pecking order. While they don't have thrillingly negative Yelp reviews they do write articles on their site with incredibly patronising articles that mocks the whole audition process and is rife with bad punctuation and spelling errors.

A Neurotic Jew posted:

it's everything around Ortwin Freyermuth, who transcends domains.
Ortwin is a far cry from when he managed German film companies to invest in distribution rights in American productions. It was a big thing in the 90's but dried up. It's why he lists Das Boot in his "meet the dev" video as it's the only success he really had.

Ascendant was a middle of the road success until Costner nearly ended him. Now all he can do now is provide some legal advice.
I'm strongly suspecting he sees Chris and Sandi as ignorant suckers that he's pumping for cash as he's got just enough contacts to still be credible and knows enough about the system to play it.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

alphabettitouretti posted:

Dressing it up as a feature and calling it "hibernation mode" is some Transverse level poo poo.
If it was something like auto pilot while AFK and logged out I'd kind of buy it.
According to their galaxy map traveling from one end of the verse to the other (600 LY) takes 8 hours.
It sounds like an elaborate AFK mode which like everything else is needlessly complex.

Does it mean you hang in space and you can get boarded and have your ship destroyed while you chill in the ice box?

Elite has it so you can quit even in the middle of combat and just resume at the last entry point for the system you are in. Even if you are drifting in space when you quit, no one can see you so you are perfectly safe.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
You know I kind of want this game to hobble out.
If it does what it claims in the slightest, then being a filthy pirate who can start from zero, stow away on a whales ship, steal it then sell and buy one for yourself with the profits.

If I'm reading right the whales also have unlimited insurance so they effectively kick start the economy via grand theft auto.

So getting somewhere of the whales back without spending $500 in the first place would be fantastic and more tear inducing than EVE.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
Oh to get some idea of what troubles Chris could be into regarding iffy financials.

In Australia one grey area in reducing your taxable income is to register as a company and then draw a wage from the returns. It needs two people at least to play company, you have to have AGM's and so on. So if your company earns $120,000 you pay the flat tax of 30% and then only have to pay for how much your wage earns you, so something like $35,000.

The company can also be used to buy/rent property or cars in it's name. The downside is that if you become insolvent then all of these assets will be sized vs a personal cost loss if you run a business. I have no idea if this scheme works in America but I'm sure there's something similar.

If what Derek is saying is true then all of the cars and houses are effectively owned by CIG/RSI and if their cashflow dies then so does their image of being successful and wealthy producers in Hollywood. That 100 mil of crowd cash is spread down and across 16 shell companies in part to protect against anything serious going under.
It's clear Roberts is trying to replicate what was going on at Digital Anvil and have a multi-media platform of film and gaming while having a slick facade so that any executives who chance in by get treated in style.

It also breaks a golden rule in entertainment which is you don't go in expecting to come out on top. Like any career you have to work bloody hard at it to get somewhere, you don't earn it for being persistent or for clinging to the ladder and double crossing people to gain a rung. Everything I've seen of Sandi screams of diva and one who actually is of the type to get blacklisted from auditions.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

LastCaress posted:

What I don't get is why they're patching Star Citizen and fixing bugs when 90% of content isn't even there. Is it to give the appearance of progress? Shouldn't they try and implement some basic stuff (like economy) before fixing all the smallest bugs? When they implement the things they need to it's going to break a lot of things, so why are they so concerned with the bugs now?
By releasing the bare minimum they are effectively covering themselves against claims of not releasing a product under the assumption it will get fixed later. “Caveat emptor” largely applies in this case. How many video game failures have scored a class action lawsuit from consumers? Publishers have sued as they loose investment on a failed game.

Compare to X3 where the game was a broken mess on launch but was slowly patched up. However Egosoft, despite their flaws and even being guilty of playing the immersion card as a defence, do try to keep their fanbase happy as it's a niche genre they have a good slice of.

RSI is sandbagging for the inevitable collapse by rushing to tick off as much as possible on the promises list making it harder to prove a point in a case.
A case against RSI will have to prove they were intentionally dishonest and deceptive in their claims and promises. Can mismanagement and incompetence be used against them in this case?

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
I'd forgotten how in Amy's Baking Company the first line that comes out from her is "It was at a very early age that I discovered that I had a real true passion and talent for anything having to do with the culinary arts."

Amy even likes referring to people as having no balls.

quote:

The reason I said he was a pansy is because the same reason I call these Yelpers and these people that hide behind the computer screen the [bleeped] mafia. They are [bleeped], they are pansies, and they have no balls because if they had balls, they would come to my face, my husband’s face and tell us exactly, all the stuff, the lies, the slanderous stuff that they say online instead of just hiding behind their computer screen.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
Reading up about CryEngine vehicles is interesting.

quote:

Vehicles are very complex objects and can cause major problems, if the exporting is not done carefully.

It seems anything can be assigned as vehicle.
You build it in max/maya in segmented bits, import in and then assign a movement preset like tank or VTOL and then go into the in-game editor and assign the model's parts to be a wheel or light and so on.

Presumably alternate vehicle systems can be coded in. Which is where I think the majority of the problems could be stemming from, their attempt to make a flight engine that collides with the in-engine's physics rules which aren't much more than object has mass and density.

The game sees a vehicle as a rigid physics object on wheels. The game's internal physics engine will calculate the mass and density and then the vehicle engine takes over with it's own physics setup to drive the thing and add suspension and so forth.

In every vehicle are two major hitboxes that sit inside one another. One is for terrain collision and the other is for bullets. The game hints that if these two boxes cross over things will go wrong - like if a wheel intersects the bounding box. You also have to keep these hit boxes very low in poly count else the game will try to calculate every potential hit surface.

But on top of that you also have further refinements in the in-game vehicle editor where you can create bounding boxes for wheels and other triggers like sparks or weak-points.

You can see where I'm going when you combine this into what is effectively a segmented flying room that spins 360 and has detachable parts.
So now you are flying a quivering hulk of physics objects waiting for something to set it off and then a gun turret or errant bounding box clips as the game stutters from the lag and all goes to hell.

No doubt the guys at CryTek put together a working proof of concept which looks feasible but as they added more and more at Robert's whim, things got well out of hand and the engine's capabilities.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

It's really hard to see but that's the logo of The University of Adelaide.


The other clue is in the colour of her hood. Your discipline has it's own colour. Yellow in this case is for natural and physical sciences which also applies to veterinary and agriculture.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
I bet Roberts has this on the top of his most played.

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

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

SelenicMartian posted:

Didn't Ben get his hands exclusively on some assets when Origin kicked the bucket? Then sat on them and distributed them to those chosen by him?
Yeah he managed to get the source code and some assets off things like Wing Commander from ex-Origin employees.
The pick and choose wankery came from two competing Privateer fan remakes where he decided one team over the other deserved the original 3D files for the station backgrounds.
He used his exclusive holder of the source to lord over WCNews and he was pretty reviled well before SC got going.

quote:

Assuming Ben lives on the west side of LA, how the gently caress is he so loving fat? How the gently caress is he getting fatter?
If you go by his social media accounts, a healthy dose of depression and subsequent stress eating for the dopamine thrill.
As the mood at CIG has gotten worse and worse no doubt seeing people get fired or Sandi push through the ranks to explode at someone hasn't helped one bit.

Not to mention the sheer dissatisfaction of finally working with your idol only to find out he's nowhere the visionary you'd hope he'd be, a bad and lazy communicator and so distant that he has his decisions overridden by his wife without him knowing. (No surprise if that actually happened.)
Plus having to bear the ire of others as critical staff get fired while you can sit on a sofa and stream your junk food addiction while watching Star Wars weighs on his concious.

Mo-crap
Mo-cap isn't new to games - but it's still used sparingly as it's tough to manage. Most games only really use it under controlled circumstances, like a cutscene over player driven animation, as it presents more variables than you really would want and requires a tight ship to use effectively.

This is where your RnD before starting is crucial in setting out your rules and regulations to get a smooth pipeline. One small fuckup in a multi-asset hand-over environment really wastes everyone's time. And claiming "well it's not my job" isn't a valid answer. The bad quality seen so far is indicative that no one is communicating with each other to help fix problems and no one knows what to do.

I can easily see how Mo-cap can cause endless nightmares as the CryEngine documentation pretty much paints it as a dark art, especially in the cleanup phase where they warn you, repeatedly, that hours of work will be lost if something goes wrong.

They also say one mo-cap format does not fit all so if CIG are using multiple production houses at some point they would possibly have had been forced to reshoot owing to translation issues from different capture systems refusing to play nice.

Mo-cap is really really fiddily - most animators use it for reference over hand animation as it captures all of the subtleties that are hard to match by hand.
Compare it to keyframe animation where you set points of motion across a timeline and the computer calculates the tween. Easy to change as you can simply move that one keyframe.

Mocap records continuously so every frame is a keyframe.
If you need to change something you are in for a ton of wrangling and messing about as you can't simply delete a bad frame - the system will try and tween to compensate and given one keyframe is a ton of variables.

It's also much more taxing to play back as that's 24 frames of data a second the game engine has to consider - try syncing that over a server.

To break it down a bit more; you record a motion then attach it to a model's skeleton. The model and skeleton have to be designed in accordance to the mo-cap data for a smooth fit.
So does your solider have enough polys to bend an arm without tearing for instance, will adding a mocap where the character does jumping jacks mean you have to then remodel and re-rig to fix this issue? Can the skeleton be articulated enough to accept a wide range of movement data or does it need more points for something not included in the default rig?

In any setup you have to manually pin the mo-cap data to the skeleton, this is where major flaws can present itself as an incorrect assignment on the wrong part can mean the mocap data rotates the wrong thing.

It's why I suspect the lip synch looks so bad on the Paaarpman speech as the in-game head cannot support the same level of polygon fidelity as recorded by a capture system meant for a motion picture, so they were forced to rip out a bunch of points and made him look like a robot with a stroke.

Another factor to keep in mind (especially with FPS games) in a smooth transition between various sub-animations of limbs - say running while aiming and so on. If these sequences have a drastic shift in coordinates then things will get messy or choppy if keyframes are too sudden.

Have a look at the dodgy Star Marine teaser video of the cover animation. You'll see something wasn't setup right and the arm movement is causing the elbow pad to rotate as the model comes in and out of character.

Oh and I haven't mentioned match moving the mocap characters to move properly in the game environments.
I wonder if they forget to convert imperial to metric when working across studios.

Beyond that you have secondary animations, props and clothing to deal with - most things like guns can just be stuck onto the hand but items like helmets being put on usually require lots of chicanery and additional custom animations that have to slot in over the top of a mocap animation that have to synch and not look out of place.

tl;dr - SQ42's dev team are so in the deep purely from having to deal with mocap alone they are barely going to have anything on time.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

Tippis posted:

How difficult would it be to do a mix of mocap and, say, Source's face posing system? In other words, you use the performance capture to build a library of emotes and syllables (with a couple of different emotions represented) and then “just” feed the system the script, letting it create the animation rather than rely on constant re-shoots whenever you want to add something new.
Any capable animation system will have it so you can have someone walking and talking with independent cylces systems controlling the face separate from the head and from the body. So you can have a mo-cap walk cycle independent of a hand animated talk cycle.

CryEngine can do this and it uses the Annosoft synch engine which can match up with a text and voice recognition system much like Valve's. You then go in and add expressions to the rest of the face so it doesn't look dead.

Now most lip synch animation doesn't aim to replicate a human mouth 100% as we don't snap to set positions when talking so animating every phoneme would look a bit bizarre or take up a ton of time so visemes were developed as a shorthand to easily mimic human speech in an animated form.

Now each head model has to have it's phonemes and visemes modelled in Maya beforehand so the synch can merge the correct mouth positions. I suspect if you tried a one size fits all you'd start distorting the model.

I suspect they have taken data of all of the actors pulling their best parrrrp face to use as a backup for any redo on dialogues without zipping them into a gimp suit for the day....but they love the hard way....

Mocap facial animation still requires to be passed through Maya/Max and cleaned up (like removing all of Sandi's blinking) and then attached to a face rig and exported out. The main problem is a basic 3D head in a game doesn't have as much detail and flexibility as you would for a movie so compromises have to be made for it to work.
It's really disingenuous smoke and mirrors to be showing off cinema level mo-cap systems in your promotional videos where you are ultimately using only a fraction of it's power.

Compare this to something like Force Unleashed's method where they only used facial capture (adapted from ILM's CloneCam) but had multiple actors in the room so they could play off each other and react. The animators did the rest by hand.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

Scruffpuff posted:

Those awesome animated GIFs that AP posted show this. The ships STILL have no mass or inertia. Like, at all - they just jank around. It's awful.
You have to remember that "space" is just a giant level with the land and water unticked and the gravity set to 0. The ships have no global gravity to base mass with and slide off platforms as a result, even if you cheat the weight to tons. Which means heavy objects hitting heavier objects in zero g.

Add that to a physics engine that expects land. They likely used the VTOL vehicle model as a starting point, it needs a land mass to push off from.
Remember the overall rules of how a vehicle moves is set in an XML file which is then set constraints via the in engine editor. One can override the other in bizzare ways.

Also when something gets destroyed the engine swaps over to the damaged model, which is a separate entity and from the looks doesn't properly inherit momentum from it's parent or even the global physics and drops like a stone. It might be reverting back to the default gravity settings.

Now if they are trying to do modular damage you have to remember a ship is chunks of hit boxes in a cluster, you can imagine what is happening when the engine suddenly has a object that is both obeying and disobeying gravity.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

stinch posted:

Perfect for a hack that can't get it right ahead of time. In his head he can do what he always did, but now he has so much time afterwards to tweak it and make it perfect.
CR has basically gone all in on the idea that if he keeps polishing his own poo poo he is going to get gold.
The rule of game engines - everything is a hack to make it run in realtime especially pretty things like CryEngine.

I get the impression they have changed how they build their cutscenes several times forcing endless redo and reshoots. Which is why we haven't seen anything solid as the wheel gets constantly re-invented over and over again.

Out of the box CryEngine uses TrackView which basically allows you to trigger scripts and camera angles on a timeline. You build a little set where characters and objects are given animation cues and move around the place. Cameras can be set in fixed multiple angles and you switch between them and on the fly and play around a bit. Main advantage it's in-game and live vs having to run scripts and compile.

The SQ42 project launch video was made this way the ships were all animated and scripted to fly and everything was in-game woo woo.

I'll note that you can't do live mocap in TrackView; you would have to wait for the data to get cleaned up and processed into a working character rig and animation script.
It would take an immense amount of effort to put mo-cap everything into the default game engine, hours of cleaning up separate face and movement rigs while fighting with timing issues and so on which is why even AAA games like COD tend to save it for key moments or use mo-cap in short bursts, not entire scenes.

In 2012 CryTek came out with Cinebox which is meant for high resolution cutscenes that are developed in realtime.
Erasmus Brosdau's The Lord Inquisitor was made with this. Ryse was the flagship game that used this technology.

Cinebox is designed to bring more flexibility and control to creating cutscenes, notably better parsing of mo-cap and other post-processing camera effects that you can change on the fly with a decent impression of what it will look like when rendered. But that a shiny sales video glosses over the months of planning to get that result.

It's other advantage is to function as pre-vis software for movies where you can use it to create real-time animation that are diverged from production assets. So instead of having to wait for different FX departments to get their bits done and having to comp a rough, you can grab low poly models and proxy effects and get a reasonable result that's not a field of grey polygons and start to troubleshoot things well before they occur.

Recently (mid 2014) it was updated to support LiveMocap which is what we are seeing in the behind the scenes videos of SQ42 - you can catch a glimpse of the game scene playing out in real-time on a monitor.
But this is really boutique and requires a ton of work to get working smoothly starting with having to model all of the characters get face capture and so forth and then having to refine and clean up bumps or bad capture.

Roberts saw this as FMV for the 21st century and leapt on it - but this is cutting edge and hasn't really been used in this manner. It's like if Apollo 11 decided to go to Mars instead - the capability is fundamentally there but no one knows what they are doing beyond a certain point and there is no one to really help them.

Also you have to have a good idea of how to use a powerful tool - there's no magic bullets or plugins that do it for you. Roberts fails to see this and thinks throwing manhours at it will result in results from brute force. He also thought the technology was sufficient enough to get a result in two years.

Now one size doesn't fit all in any development pipeline. Things love to hate each other purely for shits and giggles in this world so no matter how many times you "up the fidelity" and change into a new engine or method, something is going to have to give be it requiring to re-do all of your textures to support a different lighting system, re model and rig the characters so they have more support and detail in their meshes and all sorts of other issues.

Some more info on what is going on under the hood with face animations which elaborates on the LOD issues I was referring to before.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

Solarin posted:

Sandi has no business being in a position of power and the idea of competent professionals and crowdfunded money being under her control honestly makes me feel sick.
Also she mentions an MBA from the Australia Business School and another from UCLA Anderson School of Management.
Here and here.

She also mentions studying business law and marketing at one of these courses which is a bit odd for an MBA where you are usually expected to come in with a Bachelors in Business. Which is where you'd learn marketing and law basics in the first place.

Also an MBA is at least three years to finish and isn't something you just apply for because you want business skills.
You have to go through a pretty solid admittance process to prove you are pretty serious about doing this, it's not for the faint hearted. Even the shortest course is a Graduate Certificate which takes six months and still comes with the same amount of scrutiny in your application.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

Necatus posted:

Tried transcribing the interview she gave where she discusses some of this... gave up at the 37 minute mark though, there was just too much bullshit to handle.
http://pastebin.com/h1GFd2E9
it's a headache trying to distil all this but I get the impression the advice of "You're not cut out for a PhD" was really her lecturers giving her a warning that she wasn't that crash hot in the first place and shouldn't take on an intensive course which she would have failed and added thousands to her university debt.
Her bachelor in science means at best she would be a researcher in a lab, processing data coming in from the field. She might get some on site work (scuba diving isn't that hard to get) but it's not swimming amongst the whales like you're bloody Ocean Girl.

She's also mistaking a business degree with an MBA. Much of what she is going on about looks more like she read the course overview on a website - you study international business, not do a thesis on that topic. If you did it would be a very narrow topic related to international study : how to better use technology for international meetings, or something like that.

Scholarships are really only awarded if your grade average across the board is either credit or distinction. You also have to apply for it under very tight rules and keep that level of achievement up during the scholarship. An "offer" in her book might have had simply been a letter saying "do you want to apply for this scholarship, go here and take a test".
She's pushing a lot of ordinary into special snowflake territory.

Her dean's merit letter alludes to her trying to start out in the medical field in a different university, but she didn't finish. Also she has no thesis listed in the archives of where she went to study marine biology.
I get a "too cool for school" vibe from her which suggests she's struggled with keeping afloat in class and if she did any studies beyond her science Bachelor, she likely dropped out growing bored - or basically thinking a career as an actress was a faster path to success.

So around 2003, she tried to get into acting, mostly working as a model bouncing between Sydney and LA as far as her visa allowed, which gave a nice shine to the image of being a glamorous globe trotting model. Then picking a copmany name from the top of the list gets job as an intern for Ascendant Pictures around 200/5..

Despite working amongst a reasonably active production company she didn't manage to score any roles on any of the features they make so therein lies the curious question of how an intern gets her hooks into a failed producer who wants to relive his glory as a game developer wunderkind from 1992 twenty years on.

I suspect Roberts must have been feeling pretty sore after the collapse so anyone who is around to lavish praises on him likely wins his heart. Just Sandi clearly sees an opportunity that she never had before and takes it with relish growing from cheerleader into cheer-leech.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
Star Citizen is much like the Ouya, which is a case study that showed even people with generally relevant skills and good intentions can cock up spectacularly when things go beyond the initial scope of the project.

Ouya's founder, Julie Uhrman, funnily enough does have an MBA in Finance and Marketing from UCLA as well as a dual degree in Finance and Management. All combined it's about eight years of study. And her rapsheet shows a solid mix of working in a mix of financial and game company backgrounds. She knew how to get Ouya started, but jumped the gun when it became bigger than expected.

To begin with Ouya used Kickstarter to support an existing VC startup. I understand the initial idea was to just create a proof of concept and draw in some early developers with dev kits, use that to stress test the device, get some games up and then use that as a ground floor for improvements before pushing it into a proper device.

Their market pitch worked too well and it drew the attention of a ton of indie game developers who were keen on a free-to-dev hackable platform where they could get something reasonable in return for creating games and soon it became one of Kickstarter's biggest success stories drawing in all sorts of publicity.
It also changed the rules for Kickstarter as they decided that no more CG rendered images could be used for product concepts.

But how do you hold all of this cash? So they decide to push the Ouya to mass market well before it was prepared - thinking the demand would pay off if they upscaled their orders.

To meet that demand shortcuts galore were made to get something out the door. And then they learnt the hard way that Chinese manufactures don't really care that much. Delays crept in as they worked out how to shove the whole thing into a predefined case which was designed to look good over any considerations to where parts go. Along with hiring devs on the cheap resulting in an infamous Twitter message of "redtube".

At the end of the day it was a mid-end Android phone with a buggy front-end and a tiny HDMI cable to plug into your TV which often toppled over because it was so light and sometimes overheated as the giant fan blocked any real airflow as they were unwilling to change the case design beyond adding a few small air gaps.

The UI was little more than a flashy launcher plastered over a default Android interface with poor consideration for the fact it was using a controller and occasionally trapped people with random Android keyboard prompts. The much vaunted root button was removed as they realised people could just flash a new firmware and sideload games. If you could download the day-one firmware patch without it failing. And you needed your credit card to get started.

People pointed out just adding a bluetooth controller and an HDMI cable to your phone basically did the same thing and it was faster, had better games and so on. Or just hacking your Wii or other devices to run an emulator. "But it's $99!!" cried the supporters.

Despite a lukewarm reception Ouya aimed to compete with the big three and argue they were a viable rival on the scene, pushing the fine notion that a $99 console was able to deliver top quality games comparable with a $600 console with bizarre claims like showing off how Ouya was better than a Playstation on the Amazon sales list.

Desperate to get a good games library they took anything, so their market was flooded with free-to-play rubbish with one or two games kind of making it. But soon devs realised they were better off selling on the other mobile markets as Ouya's insistence of all games having a free trial before buying meant barely anyone brought games.

Meanwhile backers were getting frustrated they hadn't gotten theirs but people could pick it up in stores. Which became a bit of an Easter egg hunt as stocks were stretched so thin and most stores didn't care so soon tucked them away in the back to gather dust.
At one point they were so understocked that people getting freebies of the console in a goodie bag were wondering why theirs came with an engraving of a backer's name.

This underdog fight continued into their death-spiral with bizarre ads and attempting to "hijack E3" with a DJ booth in the carpark inside a shipping container. This was followed by their attempt to boost their Ouya only games list with the infamous "free the games fund" where they proposed to match any game's Kickstarter pledge if it hits $50,000 or more, which was sized upon by unscrupulous developers keen to get free cash and run by pumping up their Kickstarter's value with bots.

And in the middle of the mess; invoke the curse of Osborne and announce an Ouya 2.

Despite Alibaba giving them $10 million to keep going they soon were unable to recoup debts, Razer brought them out and Julie stepped down as CEO and slipped out the back.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

BluestreakBTHR posted:

Sandi's not sure where she went to high school, either.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
Do you have this from the player getting shot perspective?

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
I have used Shotgun in a production environment.
Basically it is a web-based front end that is designed to act as a production tracker and asset manager. It's meant to work alongside your fileserver and provide a platform independent way to view and access assets. So for instance it would have APIs and plugins that generate thumbnails of files or automatically generate low res movies from exported sequences. You can also set it up as a client delivery system so they can watch it with you while teleconferencing and making notes lives.

It was first developed for use in the visual effects industry where it's invaluable for tracking versions of assets and allowing for notes and revisions from supervisors and so forth as well as being able to get integrated into programs so things get exported with correct naming conventions. It's main draw are it's playback features where you can pick and choose shots or image sequences to run though for dailies.

For an actual film it's good for dalies and edit reviews with some use in pre if you happen to be on a film that needs a ton of concept art and the director needs a way to see and review things.

Despite it's FX origins it isn't unusual to use it in games, however you would most likely be using it for asset creation and project tracking and planning. It could potentially have a plugin created so you can tell it to load a Cryengine cutscene and play it back .

You get an idea of how Star Citizen uses this with some darling quotes.

quote:

Shotgun is also used to facilitate the studio’s review process which is critical to the development of Star Citizen because the studio is constantly issuing new updates as part of their commitment to transparency with their crowd-sourced backers. Though the game is not officially scheduled to launch until sometime in late 2015 , a steady stream of pre-alpha versions are continually released.
They don't seem to use it more than to review assets and any rendered out videos - which kind of is the norm in any studio.

It would be amusing if this implied that any progress of the game was internally shown in pre-rendered clips in order to hide the severity of problems popping up . This could explain why Roberts had no idea about how bad his actual game was in motion.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

Jon Do posted:

Scamming C-Ro out of a paycheck with minimal work invested while ensuring the project crashes and fails? Almost noble, excepting the source of the paycheck funds.
If that letter pointing how time was wasted on obsessing over the fidelity of shoelaces and having to spend ages team-writing an email to gently explain how heaven and earth can't be moved; then no doubt a reasonable amount of apathy is now driving the quality of the work given Roberts has no experience to effectively judge the results and so you can get away with mistakes.

Unless there is a clear vision that people can work towards and a direction that encourages them to do so, simply reiterating demands over and over again in the hopes that the solution will pop out there eventually forces people to start doing the bare minimum as they know their time isn't appreciated and will likely be changed on a whim.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
Is whisky chat still on? I don't have cats, but I have a whisky hangar shelf.


Also this is Australia, so on average each bottle is around $30USD.

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli

Colostomy Bag posted:

Is that a bottle of Woodford hiding in the back?
Of course not! It's a Jim Beam small batch port blend. Next to that is a Gentleman's Jack. Again Aussie standards make moonshiners look refined.

Adbot
ADBOT LOVES YOU

BogDew
Jun 14, 2006

E:\FILES>quickfli clown.fli
My friend's dog is a little alien.

  • Locked thread