1 (edited by BeastmodeGuNs 2016-01-27 02:19:32)

Topic: Engine reworks/new client?(from may 2014 blog)

So was just going through the blogs while chatting to a bro who used to play and came across this again, after rereading naturally I have to ask...

Is there still any progress on this and if so how much if possible to say? Or has it just been scraped or left on the bench for now, it would be greatly appreciated if I could get a response! It would be great to see perp get a new engine, new interface and new graphics!

========================================================================

For those who don't know what I'm on about just read the section from the old blog that I have quoted below.

May 2014 status report blog wrote:

The not-so-near future

We’re quite aware that the graphics of the game are starting to look dated. Unfortunately the current engine architecture has also become dated, so this is not simply a question of putting in higher resolution textures and models. The engine requires a major rewrite for any significant upgrades.

Not to mention the GUI - one of the most frequent complaints we hear is about the small size of the ingame fonts. This is also something that’s very hard to fix currently, since the interface is pretty much hardcoded and not “stretchy”. Text with larger fonts would simply overflow outside of the UI elements, and we’d need to have a completely separate, differently sized interface for that to work - and we’d rather put that effort into a new and advanced interface engine that will work for all display sizes without compromise.

For these reasons we’ve been working on the foundations of a completely new client but as you might guess this is not a simple task. At this time we’re still pretty far from replacing our current client, but work towards this goal has been going on behind the scenes for the last year and a half now.

Our first goal here is to get basic functionality going (glorified chat client woo!), which would work as a proof of concept for the new interface engine. Once that’s satisfying enough, we’ll move on to the 3D engine itself. With the new interface tech we developed, the implementation of a new client should go a lot faster than before. Regarding a timeframe, we hope to have something to show later this year, at the very least a partially functional test of the new interface engine.

Reading this signature fills you with determination.

Re: Engine reworks/new client?(from may 2014 blog)

It's currently uncertain when, but the new interface is still on the table.

3 (edited by BeastmodeGuNs 2016-01-27 09:55:35)

Re: Engine reworks/new client?(from may 2014 blog)

DEV Zoom wrote:

It's currently uncertain when, but the new interface is still on the table.

Coolio, nice to hear, and just to 100% confirm, it is also currently actively being worked on alongside your other current projects yes? (sorry if it sounds like I asking same question twice, I bad with wording)

Reading this signature fills you with determination.

Re: Engine reworks/new client?(from may 2014 blog)

Not at the moment. It's been suspended at the preparation stage and work will start again when other important content updates are done (NPCs, T5, etc.).

5 (edited by Ludlow Bursar 2016-01-27 11:26:43)

Re: Engine reworks/new client?(from may 2014 blog)

May 2014 status report blog wrote:

We’re quite aware that the graphics of the game are starting to look dated. Unfortunately the current engine architecture has also become dated, so this is not simply a question of putting in higher resolution textures and models. The engine requires a major rewrite for any significant upgrades.

Why waste time re-writing the engine? I would drop the current engine in favour of an "out of the box" engine - Cry, Unreal, Source, Unity - you all know the ones.

A lot of work but re-writing the existing engine is probably more. In the long run it makes sense on many levels. Personally I'd go Source2.

Re: Engine reworks/new client?(from may 2014 blog)

Its good question, which way would take longer -
updating the existing engine with new features,

or modifying an open-source or license based engine to support whats necessary to run perp.

as far as i can tell, one of the most important features isn't availiable in any engine on the market: dynamic instancing that allows unlimited number of players in the same instance/space. Even multi-million, so called AAA online games cannot get this right.

*Disclaimer: This post can contain strong sarcasm or cynical remarks. keep that in mind!
Whining - It's amazing how fast your trivial concerns will disappear

Re: Engine reworks/new client?(from may 2014 blog)

May 2014: Later this year.

2016: We don't know when roll

I Agree with Ludlow one if these multi platform engines will do just as well or even better. I don't see this coming though, they don't even have resources to make new robot models anyways.

RIP PERPETUUM

Re: Engine reworks/new client?(from may 2014 blog)

I can tell you that converting all the current assets to another engine would take a lot more than doing an engine that we can shape how we want.

Also, just read the headlines here and see what a 3rd party engine can mean for a small MMO if something goes wrong.

But this isn't about the 3D engine, this is only about a new interface layer, which already has a working prototype. It still needs months of development to actually integrate it and do the skinning etc., which is why it has been suspended for now.

Re: Engine reworks/new client?(from may 2014 blog)

No worries Zoom, after I seen masses drive to Albion, I can assure you there is still a massive population of gamers who care nothing for graphics.

Steam achievement Unlocked:  Being a Badass
http://www.perp-kill.net/kill/239407
Dev Zoom: I think its time to confess, Ville is my alt
Dev Zoom: Ville can be sometimes so sane it's scary.

10 (edited by Gekko 2016-01-28 17:54:51)

Re: Engine reworks/new client?(from may 2014 blog)

I'd vote for content. It's more important.
Imagine EVE with old graphics but with landing on planets.
But graphics is important too, of course. Many important things...

11 (edited by Ludlow Bursar 2016-01-28 21:55:39)

Re: Engine reworks/new client?(from may 2014 blog)

Using Unity or Source2 or Unreal or whatever is not about the graphics. It's mainly about development time, the ability to outsource and costs.

Re: Engine reworks/new client?(from may 2014 blog)

Ludlow Bursar wrote:

Using Unity or Source2 or Unreal or whatever is not about the graphics. It's mainly about development time, the ability to outsource and costs.

I really don't know much about coding although, if some small updates to the game, instead of taking weeks could take days, that's a win win in my book. +1

@ Ville: Personally it's not about gfx, more about the limitations of the current engine. No bigger bots due to tile issues , no larger islands, the list goes on, or it takes to much time to change or it's not possible. A good GUI has nothing to do with GFX.

RIP PERPETUUM

Re: Engine reworks/new client?(from may 2014 blog)

we could get the same size bots, with different configurations, and more of the same size islands and everyone would be happy.

Steam achievement Unlocked:  Being a Badass
http://www.perp-kill.net/kill/239407
Dev Zoom: I think its time to confess, Ville is my alt
Dev Zoom: Ville can be sometimes so sane it's scary.

Re: Engine reworks/new client?(from may 2014 blog)

Celebro wrote:

... limitations of the current engine. No bigger bots due to tile issues , no larger islands, the list goes on ...

Natural or terraformed bridges or overhangs.
Gravity.
Collisions - no robots merging with other robots.
The ability to shoot any part of a bot to which you have LoS not just one point.

to name a few more.

Re: Engine reworks/new client?(from may 2014 blog)

Ludlow Bursar wrote:
Celebro wrote:

... limitations of the current engine. No bigger bots due to tile issues , no larger islands, the list goes on ...

Natural or terraformed bridges or overhangs.
Gravity.
Collisions - no robots merging with other robots.
The ability to shoot any part of a bot to which you have LoS not just one point.

to name a few more.

This is all fine in a single-player (or limited multiplayer) game, but in an MMO you also have to consider network limitations, and what you mentioned there mostly requires a LOT more data exchange than we have now. And even now lag is a problem, imagine how this would look with a free aim system in a larger battle.

Re: Engine reworks/new client?(from may 2014 blog)

Ive worked on a oculus sim in Unity 4. Unity is not well optimized and needs a lot of work for anything heavy. I doubt it will do well with anything sandbox and large but anything is possible if you spend enough time.

They would need to go to UE4 which is proven for big projects. But no matter what, the time it takes to learn a new engine is probably worse than reworking your current engine.

Re: Engine reworks/new client?(from may 2014 blog)

Well, the biggest problem i see is larger bots, i wonder ho much work is required to make it possible with the current engine...

And it's also more about the art than new technologies. Some old games looks great even without shaders or DX 10.

I think the Conspiracy guys are quite capable to make good graphics without using 3rd party engines wink

Re: Engine reworks/new client?(from may 2014 blog)

I think I already said this somewhere but larger bots isn't a 3D engine question, it's rather a game logic question.
We can do as large robots as we want, but the whole locking and firing mechanism works from point to point, which would make the line of sight calculations for very large robots a bit weird.

Re: Engine reworks/new client?(from may 2014 blog)

well maybe use multible points on large robots.. and you just have to figure what one it uses... (for example the closest one...)
there surely are a lot possible ways to handle this...

20 (edited by Annihilator 2016-01-29 18:56:59)

Re: Engine reworks/new client?(from may 2014 blog)

DEV Zoom wrote:

I think I already said this somewhere but larger bots isn't a 3D engine question, it's rather a game logic question.
We can do as large robots as we want, but the whole locking and firing mechanism works from point to point, which would make the line of sight calculations for very large robots a bit weird.

don't forget the lack of inverse kinematics. existing bots already lack the desired detail in interaction with the 3D-Terrain.

solving the LoS issue shouldn't be that hard. Its more the question how well would you be able to turn that "point to point" system into a "point to line" system, where the line represents the target.
and to get the "line", the game engines needs to support stuff like "turning rate", since bots like the Daidalos, or the shiva would present different hitsizes depending on the direction they face, and would look even more weired if they did turn instantly on the spot.

last but not least... gropho and mesmer are already borderline tall regarding clipping with some structures. the Terrains "passable slope" would need to be expanded to include maximum robot hight (which i would already add even without multi tile robots, to enable real floating bots)

overally, it doesn't sound unsolvable if there is a will to do it. (and coders to actually work on it)

*Disclaimer: This post can contain strong sarcasm or cynical remarks. keep that in mind!
Whining - It's amazing how fast your trivial concerns will disappear

Re: Engine reworks/new client?(from may 2014 blog)

Honestly the walker could be the same size as heavy mechs and no one would care.

Steam achievement Unlocked:  Being a Badass
http://www.perp-kill.net/kill/239407
Dev Zoom: I think its time to confess, Ville is my alt
Dev Zoom: Ville can be sometimes so sane it's scary.

22 (edited by Gekko 2016-01-29 20:48:05)

Re: Engine reworks/new client?(from may 2014 blog)

Maybe larger bots don't need to actualy use multiple tiles, and only visually take more than one?
For example, give every bot a radius value.
Shooting: check Source -> Target1 || Source -> Target2 || Source -> Target3 || Source -> Target4, where
Target1 = actualX + radius
Target2 = actualX - radius
Target3 = actualY + radius
Target4 = actualY - radius
find best radius values for bots.
run this check only if hit of actualXY is blocked or size of bot(it's large or extra large) requires it, to optimize resource costs.
points reducing to 2 if you can plot perpendicular line.
or might be increased (diagonals), then it'll be perfect, but will require more calculations.
Collision: something simmilar.
Height: Add Z axis to this check(?)
Just an example of how this can be with point-to-point.
The better variant can be somehow using new values of bot's width, length, height and existing direction info...

Re: Engine reworks/new client?(from may 2014 blog)

"multi tile robot" refers only to the fact that its visually bigger then one tile.

and then, the implications with LoS and clipping, where you see a fat Whale hiding behind a tree that you cannot hit.

*Disclaimer: This post can contain strong sarcasm or cynical remarks. keep that in mind!
Whining - It's amazing how fast your trivial concerns will disappear

Re: Engine reworks/new client?(from may 2014 blog)

If that is the case just code to ignore LOS on large bots when hit , with a good enough tank ability it's not really an issue.

RIP PERPETUUM

Re: Engine reworks/new client?(from may 2014 blog)

why do they need to be so big?

Steam achievement Unlocked:  Being a Badass
http://www.perp-kill.net/kill/239407
Dev Zoom: I think its time to confess, Ville is my alt
Dev Zoom: Ville can be sometimes so sane it's scary.