[360, PS3] Need For Speed: Shift

That actually looks good. He's probably playing with a pad there.

 

I love that cockpit camera work, very cool. This game looks very close to what I'd expect from a PGR4 sequel. It looks like it has a nice balance of arcade feel to it without being totally unrealistic.

 

Not only have sim games not been compensating for the lack of g-forces etc, but they’ve been making cars spin out too easily. If any pit crew gave a real race driver a car that handled like that, he/she [Dana Kirkpatrick] would take it right back and tell them to fix it.

I completely agree with this. I played some sim games and the level of grip feedback is practically nonexistent.

 

No I actually meant a controller, probably using analog stick. It's typical for presentations.

While I agree that Forza 2 had some issues with grip (I've discussed them enough now), I still think Gran Turismo 5 Prologue's driving model is quite good. It's certainly not overly easy to spin out with a car in that game, compared to what was basically a bug in Forza 2's implementation of grip on FR cars in particular (and which I'm going to assume will be fixed in Forza 3).

 

Unless you connect it with a G25

 

If only that were possible. Tyre manufacturers struggle to model the complex mechanical and chemical reactions responsible for generating grip in a tyre, videogames have no chance (and no access to any useful data, as it's a very closely guarded secret). Far better (and ultimately more realistic) to look at the telemetry from cars navigating the track and build a tyre model which gives you the feel you want and results which match up with real telemetry (IME).

You go anywhere near a forum for race sim development and you'll see a bunch of people worshipping Pacejka's Magic Formulae. Basically they are functions which takes as input slip angles/ratios and dozens of other poorly-explained parameters and produce longitudinal/lateral/aligning torque for a tyre. The problem is, nobody knows what those parameters should be for actual tyres (IIRC the only real data freely available comes from cross ply tyres!). Pacejka parameters would be accurately described as a 'hack' to recreate a best-guess at how a car would respond yet they are still widely praised by many sim fans, to the point of "You don't use Pacejka? You're not a simulator", even though any simulator using Pacejka is effectively saying "our entire tyre model came from guesswork and trial and error". The whole reason they are called Magic Formulae is they are not based on any kind of physics but the dozens of parameters can make them fit to experimental data - if you have any, which pretty much anybody outside of tyre manufacturers or businesses with close relationships with them (race teams, etc) doesn't.

The only thing closer to a "true simulation" is a full molecular model of a tyre. Hacking Pacejka curves or tweaking your own model until it gives you the response you desire is pretty much all a videogame can do.

Unfortunately even "simple" mathematical models like Hooke's law tend to spectacularly explode when applied to stiff suspension in a videogame. For example, F1 front suspension generates massive forces for tiny spring compressions or extensions - forces which cause the suspension to move so far in a timestep in a videogame that an even larger force is generated in the other direction and it shakes itself apart. Constraining it to a particular range doesn't help as that's basically modelling no suspension (or I guess you could say, running on bump rubbers the whole time). A lot of what many people would consider excellent simulators are in fact excellent at fudging things which you simply can't simulate in real time. Phenomenological modelling goes a great deal further than ensuring your entire physics engine can be derived from F=ma, and having experienced both approaches I'm very strongly of the opinion that you can use Newton for all the high level stuff but when it comes to tyres and the actual feel of driving, throwing the physics books out the window and developing the feel you want gets much better results.

200-400 is the sweetspot really. Less than 200 starts getting into explosion territory and more than 400 results in diminishing returns, plus you start hitting budget limits - at which point game time starts to diverge from "real time" in some kind of relativistic nightmare

Besides I don't know where they got 120hz from, I'd be very surprised if Shift was running it's core physics that slowly. You don't have to run every part of the physics engine at the same rate though, or maybe someone just got their wires crossed and meant some other game system.

 

15 min directfeed video of singleplayer career
http://g4tv.com/videos/41134/Need-for-Speed-Shift-Direct-Feed-Developer-Walkthrough/

Also, i think that is excellent that you can' ban' users with aggressive gamercard when you create multyplayer server.

 

The cockpit driving view + sound in this game feels very visceral unlike other games where you feel like you're floating on water and driving in slow motion.

 

Anyone compare this to Grid yet?

 

IGN 9.0

 

This is true for all big publishers.

If you write a bad review publishers can be resistant to send you review copies the next time. Aspecially big ones.

This is why Perfect Dark Zero got a 85% average score on metacritic. Anybody who has played this game now that its horribly, horribly bad. But, its a big budget launch MS X360 game. You just cannot give them a shitty score, unless you want to start paying for the game you have to review yourself.

I wouldn't say so. Job of the CEO of any corporation is to ensure that the company maximizes shareholder value. Going by their development model, i say they are doing a pretty good joob at that. After all, most of their games are not particularly good, but have extraordinarily good sales. Aspecially if you consider how good the game is in comparison. Army of Two sold what? 3 million?