Support NeoGAF

[leeh]

I've already explained multiple times before your post what I mean. Physics are some of the most intensive things you can calculate and process to ever exist.

It's been discussed plenty of times in the past few pages but if you want to be dismissive and flat-out call people wrong then that's your decision I guess.

EDIT:
This part in particular is hilarious though.

In terms of Forza though, the 360hz physics engine would incline me to believe that the GPU and CPU dont depend on each other (or it uses a specific core it for things it needs specifically on that frame) and the GPU essentially has all the 16.6ms available to render the frame and just grabs whatever the physics has recorded at that time from RAM.

If that is the case, they should probably be able to squeeze more out of it. I think it's more artistic weaknesses rather than technical though where it has its downfall.

 

[TheSpecialOne]

I have knowledge of the challenges and resources required related to implementing anything remotely realistic with the physical properties of rubber. It's a struggle that has existed for years, and even rough approximations are very resource intensive. As I said in the pages beforehand, there's a reason that any truly realistic simulation of a vehicle is rendered in a timescale fathoms behind real time.

Again, thats not what we're discussing. We are discussing your tales from my a** claim that forza is doing more under the hood. I have argued and will continue to argue that that hypothesis is baseless and completely false. They cannot theoretically do more under the hood. They can only use as much power as they have to work with. Computationally, it is not doing more under the hood. It is physically impossible. I'm not debating if Forza is doing more physics calculations, I'm debating your under the hood claim which is categorically false.

Physics vs. Global Illumination.

Forza > DC
DC > Forza

Who's more technical? tune in 5 pages from now.

Spoiler

Graphics Face-off...

I'll back down from this argument because it's not the point of the thread but it really irks me when people make tfma claims.

 

[jaypah]

Oh this thread is fucking delicious. Mmmm. I wish I could play Forza. I own DC and play PCars on a nice PC but I always liked the smoothness of the last 2 games.


Edit: by delicious i mean the wonderful back and forth. Give up no ground soldiers!

 

[Dictator93]

Have the DC guys ever put out any technical information regarding their global illumination system?

 

[NullPointer]

I was just thinking that. People saying DC 2D leaves aren't noticeable when playing the game, and then they go ahead and bash FM6's partially 2D crowds. Wasn't Halo Community closed because of bullshit like this?

On this particular topic I actually notice the flatlandia crowds in Forza - due to the nature of how often you start next to and race close to the stands, and have never noticed flat vegetation or whatever sleight of hand evo is using for that stuff in DC.

Its not a problem for me, but it is easily noticeable.

 

[gundamkyoukai]

I honestly can't tell if you're serious.
Running complex physics algorithms, a lot more than the ones I listed, all at once. Let me repeat, all of them at once. And doing it at least at 360 cycles per second. On a laptop CPU.
And this is just the physics part of the engine. The CPU has the rest of the game to process as well.

As i said i know Forza is doing more .
So my question for you is how much more is the extra work Forza doing costing cpu wise.
Also this all come about when he said Forza is doing more tech wise under the hood .
My point has always been better physics don't mean doing better under the hood tech wise than DC.

 

[blame space]

you could cut the sexual tension in here with a knife.

 

[jaypah]

Have the DC guys ever put out any technical information regarding their global illumination system?

I know they talked about their sound work at GDC and they put out a pretty basic description of their graphics tech but I'm not sure they ever did a deepdive on their graphics.

 

[GnawtyDog]

you could cut the sexual tension in here with a knife.

 

[VanWinkle]

you could cut the sexual tension in here with a knife.

Haha

 

[TheAdmiester]

you could cut the sexual tension in here with a knife.

A globally illuminated knife with SSR and no sprites?

 

[Bliman]

I think there is a reason why most of the time you see gifs instead of screenshots of driveclub.
And that is because driveclub has used much of its power to build atmosphere (they stated that this was much of a focus).
When you look at videos of driveclub , much of the footage is always shown in damp or wet settings to show of all of the filters and it's use of the windshield wiper effect, and procedual clouds.
If you compare straight screenshots in dry sunny settings there is little to pick between forza 6 (and others) and driveclub.
Driveclub literally tricks you in to believing it looks better then the rest but most of the time it is by emphasizing where it's good at and hiding the bad stuff.
I would say it all depends what you are looking for in graphics to decide which is best, do you want the drama and film version , like driveclub or do you want to go for racetracks with always duller settings and with more clarity to help with your driving like Project Cars and Forza, etc...
There is something for everyone, and I think when Gran Turismo is revealed all the threads will go how it blows driveclub and rivals away. Like they didn't expected.

 

[NullPointer]

If you compare straight screenshots in dry sunny settings there is little to pick between forza 6 (and others) and driveclub.

Give it a shot. Though I'd still expect DC to take the crown by a wide margin.

 

[GnawtyDog]

I think there is a reason why most of the time you see gifs instead of screenshots of driveclub.
And that is because driveclub has used much of its power to build atmosphere (they stated that this was much of a focus).
When you look at videos of driveclub , much of the footage is always shown in damp or wet settings to show of all of the filters and it's use of the windshield wiper effect, and procedual clouds.
If you compare straight screenshots in dry sunny settings there is little to pick between forza 6 (and others) and driveclub.
Driveclub literally tricks you in to believing it looks better then the rest but most of the time it is by emphasizing where it's good at and hiding the bad stuff.
I would say it all depends what you are looking for in graphics to decide which is best, do you want the drama and film version , like driveclub or do you want to go for racetracks with always duller settings and with more clarity to help with your driving like Project Cars and Forza, etc...
There is something for everyone, and I think when Gran Turismo is revealed all the threads will go how it blows driveclub and rivals away. Like they didn't expected.

 

[gundamkyoukai]

Have the DC guys ever put out any technical information regarding their global illumination system?

Not much they told us a few things tech wise but no real break down on there engine.
Wish there were like some of Sony other tech guys.

 

[foolishoptimist]

Running complex physics algorithms, a lot more than the ones I listed, all at once. Let me repeat, all of them at once. And doing it at least at 360 cycles per second. On a laptop CPU.
And this is just the physics part of the engine. The CPU has the rest of the game to process as well.

If Forza is doing all those complex calculations at 360 cycles per second, doesn't that suggest that they're not necessarily that complex/intensive?

 

[GoldenCrow]

There is a significant flaw in the argument of those arguing more realistic physics equals higher compute load. Neither game being discussed actually has realistic physics and both have their own interpretations of physics and factor in different things which can be equally demanding, most importantly though we do not know which one is the more streamlined of the two. For all we know DC's backend might be a clusterfuck and actually require more load (I doubt this as it is very quick everywhere) and FM might be super streamlined. A lot of people rave over the physics of Forza but it isn't iRacing, Assetto Corsa, rFactor2, Project CARS, Richard Burns Rally and in my personal opinion GT in terms of realism and generally are doing it on older machines with a lot less grunt than what the Xbox One has (Forza just feels really nice and the most accessible form of simulation, it is the true mass market AAA sim).

Both games are going to be running well in excess of their output render to not drop any frames ever so they are both not going to be using the full capabilities of each system especially as both games being discussed have relatively short dev cycles. Then there is the AI to think about, drivatars aren't going to be done entirely online the internet isn't fast enough for that and that is my guess to where the main workload on Forza currently is, new hardware with unique approach to AI which they made a massive push over. AI is always a heavy load. Then you have the actual capabilities of the engine to consider. It isn't as simple as saying "this has more realistic/better/complex physics" because we don't actually know what both games are calculating in that department or how much of an overhead they require as assumptions or marketing hype is not a accurate gage of reality.

 

[VanWinkle]

I think there is a reason why most of the time you see gifs instead of screenshots of driveclub.
And that is because driveclub has used much of its power to build atmosphere (they stated that this was much of a focus).
When you look at videos of driveclub , much of the footage is always shown in damp or wet settings to show of all of the filters and it's use of the windshield wiper effect, and procedual clouds.
If you compare straight screenshots in dry sunny settings there is little to pick between forza 6 (and others) and driveclub.
Driveclub literally tricks you in to believing it looks better then the rest but most of the time it is by emphasizing where it's good at and hiding the bad stuff.
I would say it all depends what you are looking for in graphics to decide which is best, do you want the drama and film version , like driveclub or do you want to go for racetracks with always duller settings and with more clarity to help with your driving like Project Cars and Forza, etc...
There is something for everyone, and I think when Gran Turismo is revealed all the threads will go how it blows driveclub and rivals away. Like they didn't expected.

You see more footage of wet stormy conditions because those conditions look better than dry sunny conditions. That's not Driveclub's problem, that's the planet's problem. You can find and produce unflattering shots of Driveclub - and dry sunny conditions are the easiest way to - but you can also get them looking great in those same conditions. Depends on what you're looking to prove.

 

[etta]

As i said i know Forza is doing more .
So my question for you is how much more is the extra work Forza doing costing cpu wise.
Also this all come about when he said Forza is doing more tech wise under the hood .
My point has always been better physics don't mean doing better under the hood tech wise than DC.

Horizon has documented its cloud physics, and if I remember correctly, its dynamic TOD, which includes more than just cloud simulation, takes up 2 ms resource-wise.
Now imagine multiple algorithms of clouds being calculated and recalculated constantly, with some algorithms being easier than cloud simulation and some harder, all at once. Can you infer or make an educated guess about how intensive physics calculations are in Forza, keeping in mind it has 16.6 ms available to do everything, not just the physics?

If Forza is doing all those complex calculations at 360 cycles per second, doesn't that suggest that they're not necessarily that complex/intensive?

It suggests only that they are more accurate, 6 times more accurate than if they were done at 60Hz. It doesn't suggest how complex they are, not by itself at least. Maybe if we knew the proportion of resources that it used. Say if the physics engine required only 20% of the resources available, and with the knowledge that they are done at 6X the cycles, then we could infer that they would be easier algorithms than if they took 40% of the system resources.

 

[gundamkyoukai]

You see more footage of wet stormy conditions because those conditions look better than dry sunny conditions. That's not Driveclub's problem, that's the planet's problem. You can find and produce unflattering shots of Driveclub - and dry sunny conditions are the easiest way to - but you can also get them looking great in those same conditions. Depends on what you're looking to prove.

yep i could take some noon pics of DC now and make it looking boring as hell.
Certain times of day just boring to look at .

Horizon has documented its cloud physics, and if I remember correctly, its dynamic TOD, which includes more than just cloud simulation, takes up 2 ms resource-wise.
Now imagine multiple algorithms of clouds being calculated and recalculated constantly, with some algorithms being easier than cloud simulation and some harder, all at once. Can you infer or make an educated guess about how intensive physics calculations are in Forza, keeping in mind it has 16.6 ms available to do everything, not just the physics?

My point was never that DC is doing more than Forza physics wise.
Just better physics does not mean doing better under the hood tech wise.
Which i get side track from lol

 

[Noobcraft]

yep i could take some noon pics of DC now and make it looking boring as hell.
Certain times of day just boring to look at .

I've got you covered.
My candidate for the most unflattering Driveclub screenshot ever from one of the photo competitions lol.

And some better pics for various reasons lol.

^The sense of speed.

^The volumetric clouds and lighting.

^The environments.

 

[Drackhorn]

There really isn't. You might have just seen races in bland looking conditions. DC looks incredible in motion. All of the good looking gameplay shots you're talking about a literally screen captures of what you're seeing in motion.

Shitty capture quality (PS Share, then Youtube compression). But this is a mix of some footage Benzy has uploaded in the past, and some of mine:
https://www.youtube.com/watch?v=qq0kaBru2U0

That really is awesome. I must admit I don't own a nextgen console yet but just by looking at the clips DC is the only racer that would make me feel uncomfortable driving at such high speeds because of the great sensation of travelling at deadly speeds. It feels downright scary as hell.
It must be the detailed environment or something giving you that sensation. If so it's a great tradeoff for 30 fps.

 

[Shaneus]

I need to monitor this thread far more closely... it's getting away from me

And this is exactly why Driveclub feels so smooth compared to other 30 FPS racing games. Only PGR4 gets there, besides DC.

I still think PGR4 has far better motion blur than any game out so far. DC is actually really good, but PGR4 nails it somehow. I don't know what tech they use, but it does it by far the best out of any racer, even "next gen" ones like DC and FH2.

Hmm, yeah, PGR4 has some frame dips here and there indeed, but its motion blur is much stronger, mate. Aren't you getting confused by the camera shake?

We're in agreement then

There is no one true successor to PGR. Both Horizon and Driveclub take half of the equation each, but neither is very close as a whole.

FH1 was far more PGR-esque than FH2 though imo.

DC is far closer to PGR than either of the FH titles. PGR was always about racing without having to tune or buy/collect cars, which FH rarely (if ever) is.

I don't believe I'll have success googling for shots taking at the exact same speed.

Best I can do is to find videos of vehicles going fast. In fact, I've found one on Forza Horizon where the guy goes beyond 400 Km/h with a Bugatti Veyron agains a PGR4 where the player never goes above 300 Km/h. So I'll give you 100 Km/h more on Forza Horizon. Now tell me which one has a stronger motion blur?

https://www.youtube.com/watch?v=lI98C9r0V7E

https://www.youtube.com/watch?v=GfmD4ZN5b9k


So, 1) yes, it is a good thing. 2) changed the side of the "dishonesty" and it didn't changed my mind on the subject. 3) I don't care. As a personal experience, PGR4 felt more smooth to me. I'm not trying to shit on Forza Horizon, a game I effin love and probably enjoyed more than PGR4 (or maybe not. Let's leave it at a tie ). I just expressed my opinion.

Having played PGR4 somewhat recently and the F&F H2 demo on 360, as well as the original FH on 360, PGR4 motion blur is still the king, like I mentioned earlier in this post. You should fire it up again, it still holds up incredibly well, even today.

I use a Panasonic Plasma, which really makes frame drops apparent in Game Mode, and I agree, Driveclub is amazingly smooth. Sometimes it's hard to believe it's only 30fps.

*brofist*
I'm scared for the day when my Viera GT30 packs it in. It's such a great looking picture, I'm gonna miss it when it finally kicks the bucket

I was actually talking about this: https://en.wikipedia.org/wiki/Micro_stuttering

A had a PC with this crap. It didn't matter if Fraps was showing 60 FPS at the corner of the screen, it never felt a silk smooth 60 FPS rendering. It as as if the GPU was skipping frames or something like that. It's not like a simple framerate drop. It's awful. :c

EDIT: https://www.youtube.com/watch?v=zb3MsENJ-fU

Did you have dual GPUs? I've spent forever trying to eliminate micro-stutter with my current (admittedly ancient) setup, and I'm determined to not go multi-GPU when I upgrade. It's not worth the hassle.

 

[DD]

That really is awesome. I must admit I don't own a nextgen console yet but just by looking at the clips DC is the only racer that would make me feel uncomfortable driving at such high speeds because of the great sensation of travelling at deadly speeds. It feels downright scary as hell.
It must be the detailed environment or something giving you that sensation. If so it's a great tradeoff for 30 fps.

You should try Dirt Rally if you want to shit your pants with a racing game. :>

Did you have dual GPUs? I've spent forever trying to eliminate micro-stutter with my current (admittedly ancient) setup, and I'm determined to not go multi-GPU when I upgrade. It's not worth the hassle.

Nope. But I changed everything some time ago and now its fine. :>

EDIT: and I did played PGR4 last Friday. You'll probably laugh at one of my posts, as it seems that you're reading some old ones. :>

 

[cool_dude]

Driveclub is really Drabclub.

 

[Wax Free Vanilla]

So guys what do you think is the best looking racing game currently, overall?

GT6 ver1.00 - it won't be topped until GT Sport arrives.

Sensational looking.

 

[gundamkyoukai]

Wax is awesome lol.
Still i hope GT7 has a better sense of speed than before cause it going to be hard to go back after DC.

 

[TheAdmiester]

GT6 ver1.00 - it won't be topped until GT Sport arrives.

Sensational looking.

They really need to get rid of standard cars before it can have any kind of consistency though.

EDIT: And actually have motion blur in gameplay.

 

[Korezo]

Some of the games posted here look like Auto Modellista.

 

[DD]

Some of the games posted here look like Auto Modellista.

I WISH!!!

 

[Makki]

They really need to get rid of standard cars before it can have any kind of consistency though.

Why do people that dont give a shit about having a wide variety of vehicles including cars that no other game developer in the world will ever put in games make it like its so hard to just not use the Standard cars.

The vendetta on Standards is really silly in my opinion. As long as they are properly categorized I would much rather have the variety they bring since all I have to do to enjoy them is use the hood camera.

I think a fair compromise is to offer them only in the used car lot and not include any of them in AI races. It will never be that Mr Yamauchi will tell their team to go out and laser scan every piece of shit standard, so I would much rather still have the ability to race them as they are.

 

[etta]

My point was never that DC is doing more than Forza physics wise.
Just better physics does not mean doing better under the hood tech wise.
Which i get side track from lol

Oh, okay. So in your opinion, the dynamic weather and dynamic time of day algorithms of DriveClub > the physics algorithms of Forza Motorsport 6, resource-wise.

Would be interesting to see if both developers confirm just how many resources are used by these features in terms of response times, like how Guerilla said Horizon TOD is using 2 ms.

Personally, though, I think FM6 uses more than 2ms for physics. And that's 2/16.6 relative to DC's presumed 2/33.3, so 12% vs. 6%. But I am just presuming these based on several ideas built on inference and deductive reasoning, but I'll list them if you want.

 

[brainchild]

I always get a chuckle out of the, "but it's using tricks to make it look so good" argument. I just don't understand it. Wouldn't the presumed efficiency of tricks and shortcuts be a GOOD thing? The only potential con is that it could be limited in its utility because of how specialized it might end up being.

As long as microprocessors are responsible for graphics rendering and virtual physics, they will always be a trick compared to real life. Essentially, graphics and physics are abstracted from transistors, not photons bouncing around in space.

In fact, the thought of being able to achieve the look of full on ray-traced Monte Carlo global illumination without being based around the laws of real world physics is fascinating! Hacking real world physics to achieve results that are indistinguishable from the real world is not only highly commendable, but incredibly resourceful!

 

[jaypah]

I would rather have the old standards than not have them (loved some of then actually) but it brings it down overall in a thread like this. If 2D crowds/foilage spawn hyperbole and page long debates then shitty car models are fair game.

 

[gundamkyoukai]

Oh, okay. So in your opinion, the dynamic weather and dynamic time of day algorithms of DriveClub > the physics algorithms of Forza Motorsport 6, resource-wise.

I am talking about over all so it won't just be those two things .
But yeah it would interesting to see how much resources certain things taking .
DC having a unlock option would tell us a lot , wish they had one .

Would be interesting to see if both developers confirm just how many resources are used by these features in terms of response times, like how Guerilla said Horizon TOD is using 2 ms.

Yeah that would nice .
Guerilla are good with that stuff .

 

[Shaneus]

Nope. But I changed everything some time ago and now its fine. :>

EDIT: and I did played PGR4 last Friday. You'll probably laugh at one of my posts, as it seems that you're reading some old ones. :>

I did! And PGR4 BC is one of the things holding me back from buying a One. Once it's confirmed (and Blur is as well) I'll likely be buying a One.

 

[1.21Gigawatts]

They are both as equally noticable to me. Perhaps my eyes are self-trained to identify rendering inconsistencies...

Do you even own the game? I never once noticed this stuff during racing, and only very few times I noticed it in photomode.
Criss-crossed stuff or low poly vegetation is much easier to notice. Has ruined Nürburgring in every racing game for me so far.
But judging by the dense pink trees in DC on the japan tracks I think Driveclubs rotating sprites tech could do the Nürburgring justice.

EDIT 2: do you mean the people moving off the track and stuff?

Yeah, 3D people moving around, reacting to cars going by.
I'm mean, that is what they promised us at the beginning of last gen, thats almost a decade ago.
Shouldn't we have that by now in even better? Instead we have ugly static low poly people or rotating 2D stock photo people that look like the crowd in FIFA 2001

 

[Stillmatic]

All these resource assumptions about DC.


Taken from PS Blog, which probably is only scratching the surface.

Weather:

NASA data was used to accurately map out the night sky – so wherever you are in the world you’ll see the correct star constellations for your location.

All clouds are full 3D models to ensure accurate light diffusion from the sun. They’re calculated at massive distances in a fully volumetric form, so thin clouds cast lighter shadows than dense storm clouds, and their colour impacts the feel of the landscapes and cars.

Skies are uniquely generated every time you play, so just like in real life you’ll never see the same sky twice. Unless you’re replaying somebody’s challenge, in which case it’ll replicate exactly to ensure a level playing field.

Clouds react dynamically to different wind speeds. This is then converted into a ground wind speed which accurately interacts with all vegetation, overhead cables and other environmental features, based on their height from the ground.

Waves and rippling on the surface of lakes is dynamically linked to wind speed, which affects how clear reflections are in the water.

The environment:

High resolution NASA data was used to accurately map landscapes and mountain formations – which were then tweaked to ‘improve’ on their natural beauty and make them perfect for high-speed racing.

Road tarmac textures are hand-modelled rather than tiled or tessellated. Stones and bitumen are all placed and then rendered procedurally to give realistic surface detail with huge visual variety and no repeating detail on any road surface.

Each location has a draw distance of up to 200km to the horizon and even simulates the curvature of the earth in both skies and terrain. Distant landscapes are built out and fully modelled, instead of ‘painted on’, to ensure that they support the dynamic, volumetric nature of the skies and lighting.

All environmental light sources are independently generated with different properties. The team sampled the colour and intensity of individual streetlights, house lights and even camera flash bulbs, which you’ll see best in any of the Indian tracks at night.

Flora and fauna:

Some tracks boast over 1.2 million road-side trees – and this number keeps going up as the artists try to out-do each other as development progresses.

There are over 100 different varieties of trees, bushes, mosses and flowers. The team consulted botanists at Kew Gardens to learn which plants would naturally grow in each location.

Wildlife is realistically tied into the day/night cycle. You’ll see flies and butterflies only during the day, and moths and bats only at night.

One of the India tracks features a tea plantation with a sprinkler system that turns off and on at set times during the day.

The Chungara Lake track boasts a 19,000-strong flock of pink flamingos, all behaving independently of each other…and look out for the seagulls in Scotland, roosting crows in Norway, Canadian geese and vultures in Chile!

Spectators dress for the weather – if it’s a cold night, expect them to be sporting hats and gloves.

The cars

A typical DRIVECLUB car is made up of 260,000 polygons. The staggeringly detailed cars you see in promo videos are the same models you drive in the game – they’re not pre-rendered CG versions.

Each car takes approximately seven months to create – from initial licensing, reference collation, CAD data processing, asset production, physics modelling, through to the final car in-game.

Pagani employ seamstresses to accurately match up the symmetrical carbon weave on the cars bodywork, and even add the ‘Pagani’ name to their small screw heads. These nuances are accurately reproduced in-game.

The same 3D CAD (Computer Aided Design) engineering data that the manufacturer uses to factory produce each vehicle has been used by the development team to create each car.

The cars have realistic layered paint materials – base metal or carbon layer, primer coat, base colour coat, two metallic paint coats, clear top coat, etc. – which can all be stripped away individually as part of the damage system.

A full shader-driven procedural system is used to simulate car damage. Multiple layers of scratches appear in the most exposed areas and edges, revealing undercoat and bare metal or carbon. A parallax mapped dent layer provides minor crumpling, and a physics driven vertex deformation system is used for severe damage.

As you race, dirt and dust gradually builds up on the car, subtly altering its appearance.

Screen space reflections (SSR) are being used together with real time dynamic light probes to render vehicle lighting and reflections more accurately, as opposed to using outdated pre-baked cubes.

The car dashboard reflects onto the windscreen in bright light; and the car exterior reflects onto carbon interior panels.

Anisotropic lighting is used to simulate the effect of each individual thread in carbon fibre weave. The pattern of the carbon alters realistically with the lighting angle and surface curvature.

Headlights are modelled using multiple layers of reflectors and lenses that realistically reflect and refract the bulbs shining beneath.
Rainbow specular highlight effects can be seen in headlight lenses because thin film interference is utilised.

Animated active aero flaps are rigged up accurately and coupled with the physics system to operate exactly as they would in real life. The Pagani Huayra is one of the best examples of this.

Conversion of kinetic energy to heat is physically modelled to accurately render the temperature and glow colour of brake discs.

Audio:

Each reference car was fitted with at least 16 separate microphones to authentically capture the sounds of the engine from 360 degrees, inside and outside of the car. Some had four mics on the exhaust alone.

When you race, the engine sounds are different based on which of the six camera views you choose – inside or outside of the car. You’re not hearing the same engine audio with a filter – it’s all recorded separately.

The recordings were so accurate that BMW and Mercedes-Benz AMG requested copies to replace their existing library.
In many cases, Evolution’s audio captures are the most high definition recordings of these cars in existence.

Bespoke sound effects were recorded for every action in the game. You won’t hear a single stock sample.

Handling and Physics:

Although not a sim, DRIVECLUB’s handling model is based on real world physics, using technical data about performance provided directly by the manufacturers.

To fine-tune the performance of every vehicle, a virtual “rolling road” test is used to check acceleration, top speed, weight distribution and braking performance.

Aerodynamics are physically modelled. For example, activating DRS on the McLaren P1 affects the levels of downforce to increase top speed and acceleration.

AI:

The AI drivers adapt their racing tactics and braking strategy based on pressure from players or other drivers. When alongside them, they will try to brake deeper into the corner.

AI drivers always try to predict overtaking opportunities based on the track, the performance of their car relative to opponents, and also how opponents are driving at any given moment.

When an AI driver has a car with KERS (Kinetic Energy Recovery System) fitted, they will use the energy strategically at opportune moments to pass or block opponents on the track.

Load Times:

Despite all of the above, once selected a track will take no more than 15 seconds to load.

Can we get back to pretty pictures, gifs and video again?

 

[SolidSnakex]

I would rather have the old standards than not have them (loved some of then actually) but it brings it down overall in a thread like this. If 2D crowds/foilage spawn hyperbole and page long debates then shitty car models are fair game.

I'd say the best option for standards, assuming that they do keep them, is to simply give people an option at the start of the game to disable them. That way people that don't like them will never see them while people that aren't bothered by them will still get to use them.

 

[Tripolygon]

All these resource assumptions about DC.

On top of that, they added the best rain and snow simulation in any racing game. The precipitation is affected by the world physics. The water level rises in heavy downpour as time passes.

I have never seen any 2 identical weather pattern in all my play-through, that is how dynamic the world is. Every windshield wipe produces a different pattern every time.

I'm just shocked how some people came to the conclusion that DC does not have advanced physics systems.

 

[Khalifa Jayy]

Another video comparing the weather in both FM6 and Driveclub but this time the video is at 720/60fps. Driveclub still looks like a whole genreation apart IMO.
https://www.youtube.com/watch?v=borC9k3dMbc

Damn, Forza rain driving looks rough! The car looks superimposed over an image of the track.

 

[ShapeGSX]

Another video comparing the weather in both FM6 and Driveclub but this time the video is at 720/60fps. Driveclub still looks like a whole genreation apart IMO.
https://www.youtube.com/watch?v=borC9k3dMbc

I still don't like the jelly-like consistency of the water on the windshield in Driveclub. It just looks wrong.

 

[kyser73]

I still don't like the jelly-like consistency of the water on the windshield in Driveclub. It just looks wrong.

Next time you're in the rain and driving take a look at the way the water pools at the edge of the blade arc and how it travels across the windscreen as you drive - well, preferably as someone else drives! - and it too has a weird, jelly-like quality to it.

Much like the 'dull' daytime lighting in DC, the rain effects aren't stylised, so often seem to be 'off' despite being a good to excellent representation of the real thing.

 

[Conduit]

All these resource assumptions about DC.


Taken from PS Blog, which probably is only scratching the surface.

Weather:


The environment:


Flora and fauna:


The cars :


Audio:


Handling and Physics:


AI:


Load Times:


Can we get back to pretty pictures, gifs and video again?

It would be desirable. Really! Somebody should read what DC really doing technically ( IIRC, there is a thread about that ). But hey, DC doesn't do shit.

 

[Synth]

DC is far closer to PGR than either of the FH titles. PGR was always about racing without having to tune or buy/collect cars, which FH rarely (if ever) is.

I could talk about this for a long, long time, and this isn't exactly the place to do it. But as I said both games took parts of PGR and went in different directions. If to you PGR is simply not having tunes and upgrades (like a great many racers of the past) then you'll determine that DC has more PGR to it than FH. If however PGR to you were the setting (city races, barriers forming courses), the extra gametypes and multiplayer (cat and mouse?), the radio station, the point chaining mechanics etc. then FH has far, far more of these things than DC does. FH to PGR is a bit like comparing prior Burnout's to Burnout Paradise... there's a change at a very fundamental level, but a whole lot of the games identity comes along with it... and the parts that didn't, in all honestly aren't generally things that were very unique to the series in the first place. Even the handling model of Driveclub is noticeably more divorced from PGR which is why there's so much polarization when people play it and remark "this handling makes no fucking sense wtf", despite having played tons of PGR in the past.

Much like the 'dull' daytime lighting in DC, the rain effects aren't stylised, so often seem to be 'off' despite being a good to excellent representation of the real thing.

The rain effects in DC are heavily, heavily stylised...

 

[AbandonedTrolley]

Can we get back to pretty pictures, gifs and video again?

SunhiLegend has you covered!

https://vid.me/P3wA

 

[Inf3cted_b3ast]

SunhiLegend has you covered!

https://vid.me/P3wA

Incredible! Sunhi is a Legend indeed.

 

[Hoo-doo]

SunhiLegend has you covered!

https://vid.me/P3wA

Sunhi is creating better trailers than the official Driveclub marketing team. God-damn!

 

[drotahorror]

It really sucks that we probably won't get a Motorstorm with DC's engine. I also don't think any racer will have better graphics than DC this gen, unless Evo makes it. Evolution Studios has made a true gem this generation imo.

 

[AllenShrz]

Name of the song?