Something is coming closer...

[Mike Dee 1,252]

4 hours ago, UP100 said:

I'd actually like this if it would be done in a more dynamic way (such as generating the layout and then trying to generate the world around it). Although it would be quite difficult and could lead to many errors in the generation.

I'll stand by what I said back in DR1 - we need dynamic routes through a location more than a dynamically generated location. Look at any of the stage screens while the stage is loading in - almost all of them show how the stage runs through all of these roads and junctions. The ultimate solution is to map out the entire location and all those side roads/junctions. Craft a system that can select 4/8 of those splits and connects them into the "stage". You'll have to record multiple pacenotes for each junction since some might just be "Flat Right 150 passed Junction" and on the next time it is used we get "Caution 2 Left tightens 1 over Junction".

THAT would be the way to create infinite replayability. You would turn 16 Splits (2 Longs) into dozens and dozens of potential stages if the roads are laid out right so even half of them have junctions that start/end with 2 options. Not saying it would be easy to develop and implement, but it would be amazing to play through. Imagine one time through Spain it's all tarmac, but the next time at the end of the 2nd split, Phil yells "Caution, Acute Left onto gravel - 100 into vineyards". The entire concept of Spain would flip on its head.

[UP100 1,156]

2 hours ago, Mike Dee said:

I'll stand by what I said back in DR1 - we need dynamic routes through a location more than a dynamically generated location. Look at any of the stage screens while the stage is loading in - almost all of them show how the stage runs through all of these roads and junctions. The ultimate solution is to map out the entire location and all those side roads/junctions. Craft a system that can select 4/8 of those splits and connects them into the "stage". You'll have to record multiple pacenotes for each junction since some might just be "Flat Right 150 passed Junction" and on the next time it is used we get "Caution 2 Left tightens 1 over Junction".

THAT would be the way to create infinite replayability. You would turn 16 Splits (2 Longs) into dozens and dozens of potential stages if the roads are laid out right so even half of them have junctions that start/end with 2 options. Not saying it would be easy to develop and implement, but it would be amazing to play through. Imagine one time through Spain it's all tarmac, but the next time at the end of the 2nd split, Phil yells "Caution, Acute Left onto gravel - 100 into vineyards". The entire concept of Spain would flip on its head.

If I understood this, it wouldn't fix the issue of the tracks being something you can learn. It'd kind of be the same thing as creating more routes.

What I'd want is that the game will for example create a line which would indicate the center of the road (then the width of the road, height, bumps, and such) and then create some kind of track out of it. It'd then be up to the generation to make a world around the road seamlessly. If there are no track components, it'd actually have infinite replayability in terms of you never know what the corner will be like until you get there, as all of the road variables could be randomized.

Obviously this is enormous work for a base that already exists, but for a completely new game with a completely custom build engine for that game, it shouldn't be impossible. It just requires time, just like everything else. It'd also be extremely tough to make routes that look like real life in terms of driving on the edge of a cliff for example.

Although this will most likely never happen, at least in the foreseeable future 🙂 

[KingOfCoins 286]

2 hours ago, Mike Dee said:

I'll stand by what I said back in DR1 - we need dynamic routes through a location more than a dynamically generated location. Look at any of the stage screens while the stage is loading in - almost all of them show how the stage runs through all of these roads and junctions. The ultimate solution is to map out the entire location and all those side roads/junctions. Craft a system that can select 4/8 of those splits and connects them into the "stage". You'll have to record multiple pacenotes for each junction since some might just be "Flat Right 150 passed Junction" and on the next time it is used we get "Caution 2 Left tightens 1 over Junction".

THAT would be the way to create infinite replayability. You would turn 16 Splits (2 Longs) into dozens and dozens of potential stages if the roads are laid out right so even half of them have junctions that start/end with 2 options. Not saying it would be easy to develop and implement, but it would be amazing to play through. Imagine one time through Spain it's all tarmac, but the next time at the end of the 2nd split, Phil yells "Caution, Acute Left onto gravel - 100 into vineyards". The entire concept of Spain would flip on its head.

If I recall correctly V-Rally 4 did something like that. It was the best part of the game in my opinion.

[Mike Dee 1,252]

12 minutes ago, UP100 said:

Obviously this is enormous work for a base that already exists, but for a completely new game with a completely custom build engine for that game

Oh yeah, your solution would be 100x cooler in that it is infinite then, and who doesn't want infinite rally stages?!? I've just dabbled myself in some generational designs and quickly learned what a mountain that is to develop.

My idea is more of a "meet in the middle" strategy. Don't force the engine to generate the entire location, as getting it right would be a huge undertaking. But getting a pattern that generates routes through pre-constructed stages is much more straight forward and cuts out a lot of the "unknowns" in designing. It seems like the feasible in the near future option to me.

Consider my idea a stepping stone towards the final ultimate goal.

[Jack4688 150]

14 hours ago, Mike Dee said:

I'll stand by what I said back in DR1 - we need dynamic routes through a location more than a dynamically generated location. Look at any of the stage screens while the stage is loading in - almost all of them show how the stage runs through all of these roads and junctions. The ultimate solution is to map out the entire location and all those side roads/junctions. Craft a system that can select 4/8 of those splits and connects them into the "stage". You'll have to record multiple pacenotes for each junction since some might just be "Flat Right 150 passed Junction" and on the next time it is used we get "Caution 2 Left tightens 1 over Junction".

THAT would be the way to create infinite replayability. You would turn 16 Splits (2 Longs) into dozens and dozens of potential stages if the roads are laid out right so even half of them have junctions that start/end with 2 options. Not saying it would be easy to develop and implement, but it would be amazing to play through. Imagine one time through Spain it's all tarmac, but the next time at the end of the 2nd split, Phil yells "Caution, Acute Left onto gravel - 100 into vineyards". The entire concept of Spain would flip on its head.

I’ve been saying this for a while now too. When Sweet Lamb or Baumholder loads up the number of potential stages in there looks so tantalising!

Another point I feel I have to make now before this thread goes too far is this: Rallying is categorically NOT about not knowing where the stage goes as is oft claimed on this forum and reddit etc. when people talk on the subject of Your Stage. Rally drivers do not just turn up at full speed having never driven a stage unless it is their very first rally. The reason we have iconic stages like Col de Turini and Ouninpohja etc. in this game is because they are used year on year and the most experienced drivers get to know them and know how to get the beat out of the stage.

Consider this and you can recalibrate your expectations of what Your Stage should be, as well as tempering our expectations that it will fall in a realistic spot on the spectrum of:

1to1 Stage_________________Infinitely generated stage

[bigburt82 37]

On 12/6/2019 at 6:13 PM, koog2003 said:

What about this ?

THAT is a beautiful thing

[AdrianSM 17]

Mas rally!!!!

[chrom25 12]

100% it should be NORMAL real life livery for Focus 2007 :)

[Tripokaridos 20]

Real time weather effects would be something amazing in my opinion . But choosing a big variety of weather and time would also be great.

Great work Codies for supporting the game.

[toninus 182]

I doubt the engine is able to handle global dynamic lighting and shaders transition from dry to wet. Dynamic weather will be for DR3.0 I think.

I just hope for new rally tracks and cars. On at least 2 more seasons, but 4 would be better for sure.

[Madhun1967 54]

2 hours ago, toninus said:

I doubt the engine is able to handle global dynamic lighting and shaders transition from dry to wet. Dynamic weather will be for DR3.0 I think.

I just hope for new rally tracks and cars. On at least 2 more seasons, but 4 would be better for sure.

Hopefully with SMS onboard, we can borrow the track tech they have been using in PC2 

[Madhun1967 54]

or build a rally sim on their engine next and finetune  gravel 

[AdrianSM 17]

DO YOU KNOW SOMETHING OF THE NEW CONTENT?

Edited December 20, 2019 by AdrianSM

[bigburt82 37]

It’s still being teased. As Codies have ruled out PSVR support, I’m guessing it will be an announcer for DiRT Rally 3.0. 

[sqdstr 107]

2 minutes ago, bigburt82 said:

It’s still being teased. As Codies have ruled out PSVR support, I’m guessing it will be an announcer for DiRT Rally 3.0. 

If it would be a new game then they wouldn't say "Dirt Rally 2.0" in the teaser.

[Yaggings 51]

On 12/17/2019 at 9:02 AM, Jack4688 said:

Another point I feel I have to make now before this thread goes too far is this: Rallying is categorically NOT about not knowing where the stage goes as is oft claimed on this forum and reddit etc. when people talk on the subject of Your Stage. Rally drivers do not just turn up at full speed having never driven a stage unless it is their very first rally. The reason we have iconic stages like Col de Turini and Ouninpohja etc. in this game is because they are used year on year and the most experienced drivers get to know them and know how to get the beat out of the stage.

True, but also we are not driving the stage a couple of times a year, we're driving it hundreds of times. And we don't have a ton of absolutely random elements that can affect a stage - surface wear and weather are a good start, but they're nowhere close to reality.

Random stages are not a secret technology, we had them at least since PS1. Obviously they need a lot more work to bring them to 2020s standards, but they're absolutely essential to making Dirt Rally 3.0 NOT rely on comically large amounts of DLC updates to keep the game fresh. Since Codies don't even have a licence for these stages, they're prime candidates for using innovation as a selling point.

[PJTierney 4,963]

On 12/20/2019 at 2:19 AM, AdrianSM said:

DO YOU KNOW SOMETHING OF THE NEW CONTENT?

Yes 😉 

[ianism 1,316]

5 hours ago, bigburt82 said:

It’s still being teased. As Codies have ruled out PSVR support, I’m guessing it will be an announcer for DiRT Rally 3.0. 

definitely not the next game. it's 100% stuff for this game. beyond what @sqdstr said, they wouldn't announce a new game within a couple months of the last bits of paid DLC coming out for DR2

[Mike Dee 1,252]

10 hours ago, Yaggings said:

True, but also we are not driving the stage a couple of times a year, we're driving it hundreds of times.

Honestly, at this point, I'd be very disappointed in all of the pro teams if they weren't running each stages dozens if not hundreds of times before each weekend. We have the technology to get paths into a game pretty accurately and very quickly. Adding the fluff like trees and stuff takes time, but getting a general path that follows a rough outline of elevation into something like RBR wouldn't be to terribly hard. It's not perfect, but it'll be a HUGE advantage for drivers to get a quick idea of how the stage will be once they arrive and some of the key corner combos to watch for. 

I mean, how much would it cost to employ a lone dev/modder who is experienced with RBR to quick-mod new stages into the game for them to practice on. Compared to the rest of a pro team's budget, I would think it'd be pennies... Keep the dev on through the offseason to continue cleaning up/developing the practice stages and you could craft a massive backlog of 90% of the possible stages very quickly.

[AdrianSM 17]

I hope they put many more cars and countries. thanks codemasters

[Johnnnn 184]

Still nothing new? 😮

[Madhun1967 54]

hope we hear something that doubles as a Xmas present 🙂 

[ianism 1,316]

it'd be funny if we got like an R2 version of this, as it's been sitting (figuratively) under our noses for months now [insert nervous laughter]

Edited December 25, 2019 by ianism

[KingOfCoins 286]

Wouldn't mind that honestly. The R2's are pretty fun

[ianism 1,316]

1 hour ago, KingOfCoins said:

Wouldn't mind that honestly. The R2's are pretty fun

it's kind of a terrible car, but it's a game so who cares. and yeah exactly! the R2s are underrated imo