• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

Can I have a Skyrim type RPG done with the Drive Club engine please.

No it can't, and you're very wrong. The optimisation of asset loading for Driveclub is based on position of within track, they know 100% what's next, where it is on disk and memory, and don't have to worry at all too much about being pre-empted by the player, They can also make optimisations by hand on each part of the track when the render budget of 33.2ms is approaching to start cutting out unimportant parts of the scene graph (scene graphs can even be cached), and manually block rendering when designing the map. Backface culling needs a number of hints to work, Driveclub has these baked in, an open world game needs a more general solution as you can't easily test all permutations and combinations let alone cache them effectively.

That sounds reasonable but surely the rendering tech could be used as a base for an open world engine.
 
I think most companies (specially Bethesda) make a mistake when they release only one game of a franchise per generation. If TES VI was already out, ESVII would have a graphic leap compared to IV-V, same with GTA (I'm not saying that GTAVI should had been released by now, but it shouldn't be 4 years far off minimum either for them not to release VII this gen too).
 
You should be ok as long as it's not an MMO.

If it could be done in years to come it would look amazing though in all seriousness.
 
o.0

Then it's not impossible to do with any engine, if that engine just requires extensive changing into a different engine

Well the OP said he would love an open world rpg using Driveclub engine and cue everyone saying "omg it can't be done, it's a racing engine", no one said anything about not being able to rewrite the engine.
 
No it can't, and you're very wrong. The optimisation of asset loading for Driveclub is based on position of within track, they know 100% what's next, where it is on disk and memory, and don't have to worry at all too much about being pre-empted by the player, They can also make optimisations by hand on each part of the track when the render budget of 33.2ms is approaching to start cutting out unimportant parts of the scene graph (scene graphs can even be cached), and manually block rendering when designing the map. Backface culling needs a number of hints to work, Driveclub has these baked in, an open world game needs a more general solution as you can't easily test all permutations and combinations let alone cache them effectively.
That is a good objection, but the unstated assumption is that there is an extremely high utilization of the frame render budget in every frame and the game is very optimized (with the hints and placement of objects and geometry in a level).

That is not necessarily the case. It is undoubtedly optimized, but I have a hard time that every single spot of every Driveclub level is at the max render time and let's say an additional 2 flamingos would cause it to drop frames. :-P
 
No it can't, and you're very wrong. The optimisation of asset loading for Driveclub is based on position of within track, they know 100% what's next, where it is on disk and memory, and don't have to worry at all too much about being pre-empted by the player, They can also make optimisations by hand on each part of the track when the render budget of 33.2ms is approaching to start cutting out unimportant parts of the scene graph (scene graphs can even be cached), and manually block rendering when designing the map. Backface culling needs a number of hints to work, Driveclub has these baked in, an open world game needs a more general solution as you can't easily test all permutations and combinations let alone cache them effectively.

That's one advantage to doing a linear Racer. You can perfect what the player sees, and optimize the geometry to utilize the power supplied. Your card however is still rendering X amount of polygons and textures.

In an open world game, you pan the camera right, and the system knows to render the direction you're panning, this is no different than when you play any other game.

People have this idea that an open world game is somehow rendering the whole map at the same time, which is nonsense. It all comes down to LoD, draw distances, poly count per frame and texture grabbing. Linear racers have the luxury of knowing the path and capitalizing on what you'll see. It doesn't mean they can't handle open world traversal.
 
Why not something like Cryengine or Frostbite.... which have just as impressive rendering features basically (if not more in certain areas), as well being more suited with their gameplay / physics code for a game of that type?
 
I think most companies (specially Bethesda) make a mistake when they release only one game of a franchise per generation. If TES VI was already out, ESVII would have a graphic leap compared to IV-V, same with GTA (I'm not saying that GTAVI should had been released by now, but it shouldn't be 4 years far off minimum either for them not to release VII this gen too).
It takes a very, very long time to create games of the size & scope of GTA & TES.

What you're asking for would require massive compromises which would kneecap those franchises.
 
Maybe guerillas robo dino game can do it. Or perhaps an open world rpg on the level of star citizen could get kick started. Almost all big budget games including es6 will have to run on xbone.
 
By the time you're done adjusting the engine to fit the needs a game like Skyrim has, you might as well rename it the Bobs Burgers Engine. Youll have a bunch of things from Driveclub, most of the weather stuff may work, but there will be changes and concessions.
 
I think most companies (specially Bethesda) make a mistake when they release only one game of a franchise per generation. If TES VI was already out, ESVII would have a graphic leap compared to IV-V, same with GTA (I'm not saying that GTAVI should had been released by now, but it shouldn't be 4 years far off minimum either for them not to release VII this gen too).

No, they're about the only people who do it correctly. Almost everyone else has become obsessed with annual ganes.

It's not needed and needs to stop.
 
By the time you're done adjusting the engine to fit the needs a game like Skyrim has, you might as well rename it the Bobs Burgers Engine. Youll have a bunch of things from Driveclub, most of the weather stuff may work, but there will be changes and concessions.
Maybe the quests will be buggy, the movement shit and the combat lame, but the stone pavements in Skyrim town will be glorious. ;-)

Edit: All features appearing in this post are fictitious. Any resemblance to actual Elder Scrolls titles, new or old, is purely coincidental.
 
I recall thinking the initial Deep Down reveal would have made for a great Skyrim-esque game with that lonely mud and icicles. Of course, that turned out to be a free to play third person randomly generated dungeon thing.
 
I'm starting to get the feeling that people really don't understand how a system renders a particular frame. Nothing about the Driveclub engine couldn't handle an open world game. I have no idea what people are talking about.

I believe that's a good portion of this thread. It's understandable that people want to see the quality of a particular game expressed in another genre they are more likely to enjoy. Truth is that MOST people (as a whole) really have no idea what it takes. Having a similar discussion with a friend about "why games can't all be 60 frames per second?" Asking someone "what a frame requires to run 60?" caused a whole conversation about visual quality of games.
With that said, the asking for it does push developers to consider alternate methods of optimization or visual quality. The reality is that because MOST people don't know what it takes to produce such results, any method of obtaining those results would be accepted. Much like the original Resident Evil remake in Gamecube era which looked great but was essentially mpeg wrapped objects which did the trick for most people (and to be honest did look pretty cool at the time.)
Can you deny that an RPG like Skyrim (running a better engine of course) would look great if it exhibited qualities like DriveClub? Regardless of it it's a crazy statement to request it from the DC engine, it'd look awesome.
 
I believe that's a good portion of this thread. It's understandable that people want to see the quality of a particular game expressed in another genre they are more likely to enjoy. Truth is that MOST people (as a whole) really have no idea what it takes. Having a similar discussion with a friend about "why games can't all be 60 frames per second?" Asking someone "what a frame requires to run 60?" caused a whole conversation about visual quality of games.
With that said, the asking for it does push developers to consider alternate methods of optimization or visual quality. The reality is that because MOST people don't know what it takes to produce such results, any method of obtaining those results would be accepted. Much like the original Resident Evil remake in Gamecube era which looked great but was essentially mpeg wrapped objects which did the trick for most people (and to be honest did look pretty cool at the time.)
Can you deny that an RPG like Skyrim (running a better engine of course) would look great if it exhibited qualities like DriveClub? Regardless of it it's a crazy statement to request it from the DC engine, it'd look awesome.

Absolutely. And people have the wrong ideas about why said game hitches to being with. Take skyrim for example, it was an absolutely buggy mess, but when asked about the bugs, geometry, texture loading, and draw distance LOD isn't typically mentioned. The biggest pain areas of that game are it's physics going wonky, the A.I. going berserk and not interacting correctly with said geometry, and snags caused by the persistent world, and namely gameplay aspects. The rendering aspects, for the most part are just fine.

As I mentioned earlier, one of the most taxing aspects of an open world game could be designing it around dynamic lighting to give weight to the passage of time. In Driveclubs case, this problem is long past solved. Placing geometry into the world would already make this a non issue as that aspect of it is dynamic. Driveclub in particular, would be perfectly suited for an open world game.

I have a feeling that the reason Infamous looks as good as it does, is because they've got baked in lighting for different parts of the game, and it greatly eases the workload regarding it.
 
I would bet Driveclub only looks that good because it's a track based racer. It's the equivalent of making a linear game (no that it's bad thing).

If they used the same engine to make a big open world game you can bet it won't look even nearly as good as driveclub (at least on consoles)

Aaaaannd first post nails it.

Just because an engine looks good in game A doesn't mean it will in game B.
The whole design of DC is with top level graphics in mind.
 
I cant imagine the PS4 being capable of keeping the same level of graphics that Driveclub has in an open world setting. I mean its 30fps in a linear track game so there would have to be some concessions made for it to work on it.
 
Engines tend to be purpose-built, and aside from using the same cloud rendering middleware that Evolution licensed for use in Driveclub, there's not much that the engine does aside from using modern shaders and great quality assets that would inherently benefit other game genres. Considering the number of cars active on track, I'd imagine the engine is tailored to that many "entities," which might be a tight fit for a character-based game. Open world would mean some concessions for both rendering and asset creation, even assuming the underpinnings would work for an RPG.
 
Driceclubs visuals are impressive because the game is linear. I would love an open world game with these visuals:
driveclub_201412091506vcvd.jpg


But the second you deviate from the path in Driveclub it falls apart. You cannot backtrack on courses (drive the wrong way), it will teleport you back to where you're supposed to be going, the leaves on the trees look excellent at 100mph, but if you stop and look at them in photo mode you realize the leaves and secondary branches are just 2D sprites that always face towards the camera. The courses are populated with tons of 2D trees in the distance. It's an almost on-rails experience as you drive through courses, and I'm not sure they could open it up without making compromises.

All that said, I love the volumetric clouds and lighting systems implemented. Those would be a great feature in open world games.
 
It takes a very, very long time to create games of the size & scope of GTA & TES.

What you're asking for would require massive compromises which would kneecap those franchises.

No, they're about the only people who do it correctly. Almost everyone else has become obsessed with annual ganes.

It's not needed and needs to stop.
3 years is a good time. 4-5 is too much IMO, because not only graphics need to be refreshed, GAMEPLAY elements too, you don't want to release a game which already feels dated 2 years at released day.
 
Driceclubs visuals are impressive because the game is linear. I would love an open world game with these visuals:
driveclub_201412091506vcvd.jpg


But the second you deviate from the path in Driveclub it falls apart. You cannot backtrack on courses (drive the wrong way), it will teleport you back to where you're supposed to be going, the leaves on the trees look excellent at 100mph, but if you stop and look at them in photo mode you realize the leaves and secondary branches are just 2D sprites that always face towards the camera. The courses are populated with tons of 2D trees in the distance. It's an almost on-rails experience as you drive through courses, and I'm not sure they could open it up without making compromises.

Everything you described here has to do with the LOD. You can still look behind you in Driveclub, it's not as if every object is being rendered on only 1 face.

Take a mountain in Driveclub versus a mountain in Forza horizon. In Horizon, the developer was told that you'd be able to circle the mountain. So 1 developer spends 1 day, creating 8 faces to this mountain. When the player approaches said mountain, 5 faces will be rendered at 1 time, the other 3 are white nothings when not looking at them.

In Driveclub, the track is layed out, so that it will pass exactly 5 faces, no more. So 1 developer spends 1 day creating 5 faces to a mountain. He spends the extra time adding and subtracting more trees to see how hard it's stressing the system.

At any given point in time, both games are going to render the same amount of mountain, but the developer for Driveclub didn't have to worry about the other side.

The engines in these games render objects like any other game does. How much time you can save by not needing to account for everything is an advantage of linear games.
 
Correct me if I'm wrong, but can you not already glide along at a walking pace inside the DriveClub engine and pan 180 degrees?
That should be 360 but clearly you can't see behind you in a linear racer
 
Correct me if I'm wrong, but can you not already glide along at a walking pace inside the DriveClub engine and pan 180 degrees?
Yes, and most tracks have a reverse option to drive them backwards. But if you do go around at walking pace you pick up on a lot of graphical shortcuts/low poly models that you wouldn't notice during normal gameplay. That's not exclusive to Driveclub though, most racers are the same.
 
I think most companies (specially Bethesda) make a mistake when they release only one game of a franchise per generation. If TES VI was already out, ESVII would have a graphic leap compared to IV-V, same with GTA (I'm not saying that GTAVI should had been released by now, but it shouldn't be 4 years far off minimum either for them not to release VII this gen too).

How is it a mistake? Or do you think open world games on the scale of Skyrim and GTA are just willed into existence by wishing really hard?
 
Yes, and most tracks have a reverse option to drive them backwards. But if you do go around at walking pace you pick up on a lot of graphical shortcuts/low poly models that you wouldn't notice during normal gameplay. That's not exclusive to Driveclub though, most racers are the same.

Because in DC, you're not supposed to be able to get that close to them, they were designed to be largely un-noticed.

If they decided that this was going to be an open world racer, those trees in the background would need to be implemented with LOD scale ability in mind. Remain 2D until you can make out details, then gradually become more complex as the player gets closer. In DC they're basically implemented to stay 2D, and it's apparent when you go into free-cam.
 
I would love a wangan midnight/ tokyo highway extreme race with that engine. It is not suited for an RPG but another racing game would do it.
 
The last thing we need is Sega partnering with Evolution Studios to get the mud right for a Motorstorm game. I get shivers in my legs just thinking about that.
 
I agree, but only if it's still working within what Drive Club does well. So it's a bunch of sentient cars driving around Europe, completing quests & fighting other, larger cars. Sometimes it rains.
 
Engines tend to be purpose-built, and aside from using the same cloud rendering middleware that Evolution licensed for use in Driveclub, there's not much that the engine does aside from using modern shaders and great quality assets that would inherently benefit other game genres. Considering the number of cars active on track, I'd imagine the engine is tailored to that many "entities," which might be a tight fit for a character-based game. Open world would mean some concessions for both rendering and asset creation, even assuming the underpinnings would work for an RPG.

This. I think truesky and their lighting model are components worth incorporating into skyrim style RPGs. The rest is more about developer resources for a large open world, art direction and high quality assets. Everything else the engine does is probably too specific and optimized for track racing anyway, as they should be
 
I'm expecting Guerrilla Game's RPG to be a visual benchmark for that genre on consoles.

I'm expecting Mass Effect to be a visual benchmark in RPG genre. They will have DICE technology behind their game and we know how good art of Mass Effect was on UE3.
And we will get at least two ME games this gen and one of them will be at the end of the console life cycle with technology scaling to high-end PC, where GG RPG will probably be launched fall next year at the latest and probably wont end up as a series, but who knows that :)
 
It is, but having the engine rendering the expansive terrain in all directions, being able to spool up those visuals no matter what direction you go... Open world driving/rpg/adventure whatever.. Not going to look the same. Driveclub and other linear track racing games only have to worry about the path they designed for. Build a mountain to reduce draw distance etc. The scope of performance is completely different. It's why track racing games are graphical showcases.

Driveclub's tracks are fully rendered though. It's not like Forza Motorsport or Gran Turismo. They were saying that on some tracks, up to a million trees are rendered on the track side detail.
 
Driveclub's tracks are fully rendered though. It's not like Forza Motorsport or Gran Turismo. They were saying that on some tracks, up to a million trees are rendered on the track side detail.
?

Not the whole time, that would be a waste of resources. If you go to Canada for example and look down the mountain in photo mode, you can see portions of later sections of track but they are super low LOD. What do you mean by fully rendered?
 
I'm expecting Mass Effect to be a visual benchmark in RPG genre. They will have DICE technology behind their game and we know how good art of Mass Effect was on UE3.
And we will get at least two ME games this gen and one of them will be at the end of the console life cycle with technology scaling to high-end PC, where GG RPG will probably be launched fall next year at the latest and probably wont end up as a series, but who knows that :)

Oh, I expect Mass Effect to be absolutely gorgeous artistically but I think we'll see more tech on display in GG's game simply due to being a PS4 exclusive.
 
Oh, I expect Mass Effect to be absolutely gorgeous artistically but I think we'll see more tech on display in GG's game simply due to being a PS4 exclusive.

Think about that for a second. Mass Effect 5 or 6 at the of the gen will run the latest iteration of Frostbite 3 [or maybe 4?] technology. It will be like Battlefield 4 on past gens.
Its really like comparing Killzone 3 to Battlefield 4 on PC.
 
My problem with Bethesda is their games are always buggy and never feel as polished or graphically advanced as they should when they are left in the oven so long. I guess the worldbuilding they do is the result of all that extra time though, and that's something I greatly appreciate. It just gets frustrating that it takes so incredibly long for them to release any Elder Scrolls game. Then you have their engines which are pretty mediocre compared to UE4 for example. I guess it depends on the new engine they're supposedly developing and how that turns out, because the latest iteration of IdTech was a letdown. It looked poor in most games it was used and wasn't that impressive in terms of performance.
 
Think about that for a second. Mass Effect 5 or 6 at the of the gen will run the latest iteration of Frostbite 3 [or maybe 4?] technology. It will be like Battlefield 4 on past gens.
Its really like comparing Killzone 3 to Battlefield 4 on PC.

I was thinking for simply the next release of each series and comparing console versions to console versions. I think it's safe to say that ME4 and GG's RPG will likely both be hitting in 2016. ME5 would be a different story.
 
Top Bottom