sublimesw

Unreal 3 / 4 Engine?

98 posts in this topic

To the Rats and other generally informed folks:

I was watching gameplay videos of the new World War 2 game, "Post-Scriptum," and man, if we had those graphics in WW2OL, you all would be millionaires. Any idea as to when, in the far future, the conversion to the Unreal engine would happen? Will it happen?

Just out of curiosity. I still love the game as it is.

Share this post


Link to post
Share on other sites

There is a project for UE4 but I do not know if they will have financing to carry it out ... If I were rich, I would put all the necessary money... :cool: I always ask myself: Is not there an investor who sees the potential of this and puts the money ?

Look at the 2018 roadmap in case you have not seen it yet.

Share this post


Link to post
Share on other sites

I question whether any existing engine can accommodate WWIIOL's peculiar workload.

 

Star Citizen is trying to do something similar with mixing very fast ships, slower ground vehicles and spacer/shooter movement, and they have had to dump a LOT of money to try and achieve the same thing- arguably without success.

Share this post


Link to post
Share on other sites

its less the engine and a lot more the networking. not much work has been done here as most games don't require it. meaning, your game is designed to be X not X * Y + Z. So a game designed for 16v16 doesn't need to scale up an accommodate 128v128.

 

WoW is a decent example. Instead of solving scaling limitations, they instead solved their scaling issues by instancing everything down to smaller more manageable numbers. They then expanded on this tech and now every server is essentially connected and you can cross server everything almost.

 

ww2ol on the other hand, in theory, has had this scaling ability since 2000. The cell hosts can scale up in number and pass messages between them so all the cell hosts feel like a single 'server'. Let's say I'm on Cell-A and Kilemall is on Cell-B and we're in the same area/AO. We can still kill each other, despite being on different cell hosts, because of the messaging the cell's do between each other.

 

Now, even if this was 100% flawless you still get issues with latency when you've got guys in europe playing with guys in the states. if you had cell hosts in europe, and the states, you'd still have issues as the latency between cell hosts is high. what you'd really need to do is funnel players into cells based on their latency - grouping them in the nearest cell to them. with enough population then, AOs could be essentially locked to certain cell hosts. this way all the infantry players in europe are going to feel like any other FPS game on the market with very good response times between pulling the trigger and the guy dying.

 

unreal, iDtech, unity, voxelfarm ... doesn't really matter what is drawing the pretty pictures if you've not done anything about the network scaling limiters.

1 person likes this

Share this post


Link to post
Share on other sites

Once our Roadmap is completed, or nearing completion, I'll be pulling the team together to talk about what we're going to do that's big and bold. I've already been saying that internally and that is the extent of what I'll say at this stage. As it stands, WWII Online's 2018 Roadmap must be completed and we are aware that we need to look at taking the product further forward. 

I am really ecstatic about this too. Our community and team really wants to do something big, more to follow. GET PUMPED.

3 people like this

Share this post


Link to post
Share on other sites

The graphics look good indeed, but it feels unrealistic. In all of the dozen vids i have seen, people run around rather aimlessly, it looks like a counterstrike game, tanks engage always at minimum distance, maximum vegetation, which is undestroyable, undestroyable buildings, not seen artillery or air. I'm not an expert on gaming though, just an impression.

Share this post


Link to post
Share on other sites
On 10/4/2018 at 4:25 PM, Kilemall said:

I question whether any existing engine can accommodate WWIIOL's peculiar workload.

 

Star Citizen is trying to do something similar with mixing very fast ships, slower ground vehicles and spacer/shooter movement, and they have had to dump a LOT of money to try and achieve the same thing- arguably without success.

Can you imagine what this game would be like with 150+ million in funding. I’m a SC backer but sadly I think CR but off more than he can chew and just keeps biting instead of chewing.

Share this post


Link to post
Share on other sites

As far as game environment is concerned, all the new models we are creating have the future in mind.  They contain all the attributes and named selections that are compatible with other engine's, PhysX, DirectX and SDK's.  For example you can visually see higher quality textures in the new bunkers.  Although with our current engine the filtering really takes away from them as close as 1 meter away,  they do stand out nicely in our modeling suites.  But inside each model are attributes that are compatible with a future upgrade.   Whatever path we take that'll be a big discussion once we reach that point.   I assure you we'll be ready to roll with whatever path is taken.

I'm totally excited for the future of this game.

-Ripper

 

2 people like this

Share this post


Link to post
Share on other sites

CRS should tread very carefully before committing to UE4; they could learn a lot by carefully observing what's going wrong with Post Scriptum (the very game the OP mentions).

UE4 is the most well-marketed engine of all time, but it has fundamental flaws that make it a poor choice for a game like this one. PS (which is built on UE4) has visual and performance issues that are slowly but surely sending its potential playerbase away in droves; it's not just the endless complaints about poor framerate (even on upper tier hardware), it's the fact that objects and players rendered at more than 200 m away appear as blurry pixels. The game's supersampling option stopped working in beta and still doesn't work properly. The problem isn't alleviated by playing at higher resolutions (e.g. 4K). In this regard, it's the exact OPPOSITE of WWIIOL; in *this* game, the stone age version of Unity actually renders objects in crystal clear detail even hundreds of meters away (at least on my machine, with anti-aliasing and 4K). That makes sense since CRS 1.0 essentially built this as a Warbirds-inspired aircraft-focused game first and foremost where rendering at distance is everything.

 

UE4 doesn't lend itself to rendering a game world in which you want to see objects and other players at far distances. The devs of PS have tried to work around this by cleverly filling up the landscape with big objects that limit your line of sight to about 200-300 m (which is exactly the distance beyond which things look like an non-discernible smear of garbage). Nevertheless, try playing with all settings maxed, gamma maxed, supersampling maxed, and at 4K resolution, set up overlooking a field with an MG42, and you'll see the enemy at 300-400 m away looking like ugly pixelated blotches of color as if they were background characters rendered in an old Sierra point-and-click adventure game from the 80's or 90's. Please don't misunderstand me: PS is a gorgeous game, and the visual effects work is truly inspired (from tank ammo cooking off and arty rounds exploding around you to the magnificent stuka bomb impact)...but things fall apart badly past 200-300 m.

If you doubt what I say, just look at the PS reddit forum (where players who were extremely excited about the game are now writing huge mega-posts about how the issues with graphics and performance are forcing them to stop playing). Many of these people abandoning PS say that the game should have stayed "early access," but it's clear to me that they fail to appreciate the obvious: No amount of patches will magically "optimize" the game the way they imagine.

If PS is any guide, UE4 is not the way to go for this game.

Edited by xanthus

Share this post


Link to post
Share on other sites

My dream is WW2 Online with a more modern game engine.

Having said that, I sometimes wonder if a modern game engine can even handle all the ballistics that needs to be tracked in this this game combined with the need to draw graphics at a speed and range that airplanes require to fly around in.

There might be good technical reasons  why no other WW2 online shooter has tattempted to produce a gameworld at the scale of this game, and which also includes an air-game combined with a ground-game.

 

Share this post


Link to post
Share on other sites
20 minutes ago, krazydog said:

My dream is WW2 Online with a more modern game engine.

Having said that, I sometimes wonder if a modern game engine can even handle all the ballistics that needs to be tracked in this this game combined with the need to draw graphics at a speed and range that airplanes require to fly around in.

There might be good technical reasons  why no other WW2 online shooter has tattempted to produce a gameworld at the scale of this game, and which also includes an air-game combined with a ground-game.

 

Haha, CRS needs a consulting physicist to write the equations for the game's "Hamiltonian".

1 person likes this

Share this post


Link to post
Share on other sites
1 hour ago, xanthus said:

CRS should tread very carefully before committing to UE4; they could learn a lot by carefully observing what's going wrong with Post Scriptum (the very game the OP mentions).

UE4 is the most well-marketed engine of all time, but it has fundamental flaws that make it a poor choice for a game like this one. PS (which is built on UE4) has visual and performance issues that are slowly but surely sending its potential playerbase away in droves; it's not just the endless complaints about poor framerate (even on upper tier hardware), it's the fact that objects and players rendered at more than 200 m away appear as blurry pixels. The game's supersampling option stopped working in beta and still doesn't work properly. The problem isn't alleviated by playing at higher resolutions (e.g. 4K). In this regard, it's the exact OPPOSITE of WWIIOL; in *this* game, the stone age version of Unity actually renders objects in crystal clear detail even hundreds of meters away (at least on my machine, with anti-aliasing and 4K). That makes sense since CRS 1.0 essentially built this as a Warbirds-inspired aircraft-focused game first and foremost where rendering at distance is everything.

 

UE4 doesn't lend itself to rendering a game world in which you want to see objects and other players at far distances. The devs of PS have tried to work around this by cleverly filling up the landscape with big objects that limit your line of sight to about 200-300 m (which is exactly the distance beyond which things look like an non-discernible smear of garbage). Nevertheless, try playing with all settings maxed, gamma maxed, supersampling maxed, and at 4K resolution, set up overlooking a field with an MG42, and you'll see the enemy at 300-400 m away looking like ugly pixelated blotches of color as if they were background characters rendered in an old Sierra point-and-click adventure game from the 80's or 90's. Please don't misunderstand me: PS is a gorgeous game, and the visual effects work is truly inspired (from tank ammo cooking off and arty rounds exploding around you to the magnificent stuka bomb impact)...but things fall apart badly past 200-300 m.

If you doubt what I say, just look at the PS reddit forum (where players who were extremely excited about the game are now writing huge mega-posts about how the issues with graphics and performance are forcing them to stop playing). Many of these people abandoning PS say that the game should have stayed "early access," but it's clear to me that they fail to appreciate the obvious: No amount of patches will magically "optimize" the game the way they imagine.

If PS is any guide, UE4 is not the way to go for this game.

Ok that's interesting.

Thanks for the comments

 

Share this post


Link to post
Share on other sites

That is interesting, isn't it. To be honest, I think this engine would be just fine with a few optimizations, more polys, better explosions, and shadows. But of course those improvements take resources.

Edited by biggles4
2 people like this

Share this post


Link to post
Share on other sites
5 hours ago, Xl2rippr said:

As far as game environment is concerned, all the new models we are creating have the future in mind.  They contain all the attributes and named selections that are compatible with other engine's, PhysX, DirectX and SDK's.  For example you can visually see higher quality textures in the new bunkers.  Although with our current engine the filtering really takes away from them as close as 1 meter away,  they do stand out nicely in our modeling suites.  But inside each model are attributes that are compatible with a future upgrade.   Whatever path we take that'll be a big discussion once we reach that point.   I assure you we'll be ready to roll with whatever path is taken.

I'm totally excited for the future of this game.

-Ripper

 

But why every building edge is not chamfered it looks bad having such a hard edge and doesn't look like a current building of a modern game.

Share this post


Link to post
Share on other sites
4 hours ago, xanthus said:

CRS should tread very carefully before committing to UE4; they could learn a lot by carefully observing what's going wrong with Post Scriptum (the very game the OP mentions).

UE4 is the most well-marketed engine of all time, but it has fundamental flaws that make it a poor choice for a game like this one. PS (which is built on UE4) has visual and performance issues that are slowly but surely sending its potential playerbase away in droves; it's not just the endless complaints about poor framerate (even on upper tier hardware), it's the fact that objects and players rendered at more than 200 m away appear as blurry pixels. The game's supersampling option stopped working in beta and still doesn't work properly. The problem isn't alleviated by playing at higher resolutions (e.g. 4K). In this regard, it's the exact OPPOSITE of WWIIOL; in *this* game, the stone age version of Unity actually renders objects in crystal clear detail even hundreds of meters away (at least on my machine, with anti-aliasing and 4K). That makes sense since CRS 1.0 essentially built this as a Warbirds-inspired aircraft-focused game first and foremost where rendering at distance is everything.

 

UE4 doesn't lend itself to rendering a game world in which you want to see objects and other players at far distances. The devs of PS have tried to work around this by cleverly filling up the landscape with big objects that limit your line of sight to about 200-300 m (which is exactly the distance beyond which things look like an non-discernible smear of garbage). Nevertheless, try playing with all settings maxed, gamma maxed, supersampling maxed, and at 4K resolution, set up overlooking a field with an MG42, and you'll see the enemy at 300-400 m away looking like ugly pixelated blotches of color as if they were background characters rendered in an old Sierra point-and-click adventure game from the 80's or 90's. Please don't misunderstand me: PS is a gorgeous game, and the visual effects work is truly inspired (from tank ammo cooking off and arty rounds exploding around you to the magnificent stuka bomb impact)...but things fall apart badly past 200-300 m.

If you doubt what I say, just look at the PS reddit forum (where players who were extremely excited about the game are now writing huge mega-posts about how the issues with graphics and performance are forcing them to stop playing). Many of these people abandoning PS say that the game should have stayed "early access," but it's clear to me that they fail to appreciate the obvious: No amount of patches will magically "optimize" the game the way they imagine.

If PS is any guide, UE4 is not the way to go for this game.

You can always write your own supersampling ... it would be crazy to make a custom engine nowadays.. only UE 4 can deliver.

Share this post


Link to post
Share on other sites
53 minutes ago, pbveteran said:

But why every building edge is not chamfered it looks bad having such a hard edge and doesn't look like a current building of a modern game.

chamfered edges increase poly counts, may not be sutable in all cases, and there are other ways to create the illusion.

Keep in mind you are asking rippr about buildings that he did not make, that were created in 2004/2005
none of us had any part in the making of them

Share this post


Link to post
Share on other sites
59 minutes ago, pbveteran said:

it would be crazy to make a custom engine nowadays.. only UE 4 can deliver.

Not really.
Oh UE4 is a nice engine, no doubt, but it is not the be all end all of them either.
It does a lot of things very well, and there are a lot of things that it can not do at all.
There are many very good reasons to do your own custom engine.

Edited by Merlin51
2 people like this

Share this post


Link to post
Share on other sites
On 10/4/2018 at 4:25 PM, Kilemall said:

I question whether any existing engine can accommodate WWIIOL's peculiar workload.

 

Star Citizen is trying to do something similar with mixing very fast ships, slower ground vehicles and spacer/shooter movement, and they have had to dump a LOT of money to try and achieve the same thing- arguably without success.

...except Star Citizen is still in alpha.

 

It's not optimized yet, and it's still pretty amazing -bugs and all.

Only... they aren't asking for a monthly sub. 

There are other things about their pricing that I can't stand though, like paying to win with ship upgrades. If a player wants to spend an extra $225 and get much better ships with better loadouts. Blecch.

 

Still, it looks great, and even after a cursory going over for graphic card settings and forcing my computer to prioritize it's processing, it's playable and not as choppy.

Share this post


Link to post
Share on other sites

IMO, if the game is to be unique then the physics engine is going to be what defines that uniqueness.  It is a very difficult problem. It is rocket science. If your engine cannot adequately approximate real behavior then graphics and communications is irrelevant. I don't know anything about the Unity engine's physics, maybe they are adequate, or maybe it's physics engine can be improved. This game made its bones providing superior realism and I believe that is the objective that should be pursued. 

2 people like this

Share this post


Link to post
Share on other sites

Both Unity and Unreal use PhysX for their physics but they limit what parameters you can change for the physics objects which is why games made with them tend to 'feel' the same.

Share this post


Link to post
Share on other sites
On 06/10/2018 at 9:14 PM, xanthus said:

CRS should tread very carefully before committing to UE4; they could learn a lot by carefully observing what's going wrong with Post Scriptum (the very game the OP mentions).

UE4 is the most well-marketed engine of all time, but it has fundamental flaws that make it a poor choice for a game like this one. PS (which is built on UE4) has visual and performance issues that are slowly but surely sending its potential playerbase away in droves; it's not just the endless complaints about poor framerate (even on upper tier hardware), it's the fact that objects and players rendered at more than 200 m away appear as blurry pixels. The game's supersampling option stopped working in beta and still doesn't work properly. The problem isn't alleviated by playing at higher resolutions (e.g. 4K). In this regard, it's the exact OPPOSITE of WWIIOL; in *this* game, the stone age version of Unity actually renders objects in crystal clear detail even hundreds of meters away (at least on my machine, with anti-aliasing and 4K). That makes sense since CRS 1.0 essentially built this as a Warbirds-inspired aircraft-focused game first and foremost where rendering at distance is everything.

 

UE4 doesn't lend itself to rendering a game world in which you want to see objects and other players at far distances. The devs of PS have tried to work around this by cleverly filling up the landscape with big objects that limit your line of sight to about 200-300 m (which is exactly the distance beyond which things look like an non-discernible smear of garbage). Nevertheless, try playing with all settings maxed, gamma maxed, supersampling maxed, and at 4K resolution, set up overlooking a field with an MG42, and you'll see the enemy at 300-400 m away looking like ugly pixelated blotches of color as if they were background characters rendered in an old Sierra point-and-click adventure game from the 80's or 90's. Please don't misunderstand me: PS is a gorgeous game, and the visual effects work is truly inspired (from tank ammo cooking off and arty rounds exploding around you to the magnificent stuka bomb impact)...but things fall apart badly past 200-300 m.

If you doubt what I say, just look at the PS reddit forum (where players who were extremely excited about the game are now writing huge mega-posts about how the issues with graphics and performance are forcing them to stop playing). Many of these people abandoning PS say that the game should have stayed "early access," but it's clear to me that they fail to appreciate the obvious: No amount of patches will magically "optimize" the game the way they imagine.

If PS is any guide, UE4 is not the way to go for this game.

CRS need to clarify UE4s suitability for "long range" warfare.  This is what set this game apart other FPSs.

Share this post


Link to post
Share on other sites
5 hours ago, agenda21 said:

CRS need to clarify UE4s suitability for "long range" warfare.  This is what set this game apart other FPSs.

fwiw - what xanthus describes and what you're quoting is the same thing our engine does, the same thing id Tech does, voxel farm, unity, etc etc etc etc. ALL 3d engines switch LoD models with range.

 

ALL

 

OF

 

THEM

 

So if PostScriptum players are complaining about this - it's a lack of understanding on the part of the playerbase and not a lack of ability with the UE4 engine.

Share this post


Link to post
Share on other sites

The rendering of inf at 300+metres should be blurry. Perfect resolution IS  gamey..its not real life. I agree the vanilla LOD system unreal gives you are subpar, and Post Scriptum is a perfect example. No optimizations can fix that mess .

An ideal engine would have object culling and streaming built into it. "Our" antique engine has the basic cell streaming already - more modern engines such as the one Star citizen is using has systems that allow solar system wide resolution down to man on the planet fps battles..all in the one engine.

Yes - it IS ROCKET SCIENCE. But it can be done.

Unreal as a base is great for devs as there is thousands of people proficient in it,  but it would definitely need to be modified for this game.

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.