How Long And How Much Work To Build Future New Engine?

[froggyluv 2135]

Wonder if people who like CoD go into their forums trippin bout "Why can't we go over THAT fucking hill?!? Freakin OFP did it like 15 years ago wtf?!?! #L2CodeB#tches"

Yes the argument is foolish, like asking vanilla why it can't taste more like chocolate. That said, I do agree and wish that more time was now devoted to a more fleshed out, urban landscape as that is how many of the worldwide conflicts are playing out today. Yes, there is still plenty of mid distance fighting in the hills of Afghani, but this is pretty much covered well in this system save the 'too short' shadow and clutter system.

Personally rather have some much smaller maps that feel like an actual small city with some real civilian feel. fat people, old people, short people, mouth breathers etc.. furniture and more destructable environments as well. Here's hopin' for the expansion....*Hong Kong*:pray:

[pulstar 55]

Personally rather have some much smaller maps that feel like an actual small city with some real civilian feel. fat people, old people, short people, mouth breathers etc.. furniture and more destructable environments as well. Here's hopin' for the expansion....*Hong Kong*:pray:

You put my wishes in a less aggressive way and I agree with you. Another thing I'd like seeing is taking a page from Insurgency's playbook. Make the troops a bit more nimble (sans jumping and slide-crouching, soldiers with heavy kits can't ever do that), mantling cover more dynamic, as in Red Orchestra. I feel like controlling a small mech in Arma :p

[vran. 13]

I realize some Arma players hardly play anything else, but that doesnt' excuse how sparse Arma looks. Altis offers some nice urban-esque environments, but when budget games can look that much better (link above) it's inevitably disappointing. If the engine doesn't allow for that many objects (debris, clutter, furniture) then perhaps it's not that great of an engine.

Other military shooter makers focus more on graphics than the other things so the games look more like tech-demos. A3 doesn't overdo with the post-processing effects and consequently looks more realistic in comparison. But as far as best competing engines is concerned, I'd say Battlefield 3 without the color grading, Medal of Honor 2010 and first Crysis have the most realistic visuals.

[pulstar 55]

Frostbite 3 is very capable, but thanks to EA's stupid policies we will never see any mods that make use of it.

[PuFu 4600]

Frostbite 3 is very capable, but thanks to EA's stupid policies we will never see any mods that make use of it.

Then how would you know about it's so call capability?

[CaptainAzimuth 714]

Then how would you know about it's so call capability?

Because if the The Team that made Project Reality, a LEGENDARY mod for Battlefield 2, and that engine was far less capable, imagine what now. From Battlefield 2, to BFBC/BFBC, to BF3, to Battlefield 4, and the latest updated Engine. It's possibilities are capabilities are potentially mind blowing. There are however a group of underground Dev's working on modding BF3/4, but that's likely going to take some time before anyone sees anything of it. The best way to create a game, is to design an Engine, based around that game, and everything you want that game to be able to do, and more. You want to build it to be flexible. Frostbite is flexible, and we may see it's TRUE potential come April 17th. Keep your eyes open for Battlefront.

As for the Project Reality Team, speaking of, some official Developers of theirs have started creating something unique, and knowing how Project Reality mod went for BF2...

[Jackal326 1181]

As for the Project Reality Team, speaking of, some official Developers of theirs have started creating something unique, and knowing how Project Reality mod went for BF2...

Unless you like co-op, in which case you're stuck with a PvP POS.

[das attorney 858]

Unless you like co-op, in which case you're stuck with a PvP POS.

That's a good point. I am quite intrigued by Squad (which is a terrible name imo but anyway..)

I think one of the things that can limit the gameplay is they will have all these massive maps but with a lack of AI, they will have to get humans to do everything on them, which won't work out well as AI are brilliant for doing the boring jobs, leaving all the fun to players.

No-one wants to be the ensign in a red shirt if you know what I mean.

I'm sure it will be a good game, but that's one of the things I was thinking that could limit the scenarios or shape them in an unusual way.

[CaptainAzimuth 714]

That's a good point. I am quite intrigued by Squad (which is a terrible name imo but anyway..)

I think one of the things that can limit the gameplay is they will have all these massive maps but with a lack of AI, they will have to get humans to do everything on them, which won't work out well as AI are brilliant for doing the boring jobs, leaving all the fun to players.

No-one wants to be the ensign in a red shirt if you know what I mean.

I'm sure it will be a good game, but that's one of the things I was thinking that could limit the scenarios or shape them in an unusual way.

That depends on how you see it. Squad, which is based off of Project Reality, is probably eventually (from what i've heard) going to be like Project Reality. Meaning vehicles and what not, maybe, not too sure but anyway. The point is, It's all about teamwork and coordination, communication, there are no lone wolfs, or faulty AI to get in the way, or cause lag. Being the way of the nature of Project Reality, it involves real people, using real tactics, to accomplish tasks on the field. Of course though, it's not for everyone, as not everyone is going to get ACE3 for Arma 3 for example, realism just isn't some people's game mode.

Anyhow, i brought those up as examples of how a game is best made when the Engine is built for it. A better example, would be The Division, by Tom Clancy's. The Snow Drop Engine was built specifically for that game. Large, Urban, Massive Multiplayer survival game in downtown New York. And from what's been shown, it's clear that the Snow Drop Engine is amazing for what it was designed to do, which was to give it's developers more control over the environment they are creating, for said game.

[das attorney 858]

That depends on how you see it. Squad, which is based off of Project Reality, is probably eventually (from what i've heard) going to be like Project Reality. Meaning vehicles and what not, maybe, not too sure but anyway. The point is, It's all about teamwork and coordination, communication, there are no lone wolfs, or faulty AI to get in the way, or cause lag. Being the way of the nature of Project Reality, it involves real people, using real tactics, to accomplish tasks on the field. Of course though, it's not for everyone, as not everyone is going to get ACE3 for Arma 3 for example, realism just isn't some people's game mode.

Anyhow, i brought those up as examples of how a game is best made when the Engine is built for it. A better example, would be The Division, by Tom Clancy's. The Snow Drop Engine was built specifically for that game. Large, Urban, Massive Multiplayer survival game in downtown New York. And from what's been shown, it's clear that the Snow Drop Engine is amazing for what it was designed to do, which was to give it's developers more control over the environment they are creating, for said game.

Yes I understand what Squad is and what PR is. One of my real-person (not internet person) friends was a lead modeller on PR:BF back in the day and we've played l-o-a-d-s together so I think I know all about it already. There's really no need to explain it any further for me. :)

My comment was more on what types of modes will exist. Take for example something simple like convoy ambush.

With a game that has AI, you assign the AI to guard the convoy and attack it. Brilliant. It works every time as the AI are dumb and there's no way they can anticipate the attack.

Now try the same thing with all players. The players who defend the convoy will automatically know they are going to be attacked so instantly the scenario will not work as it would do if you use AI.

There's no way round that one.

I'm not saying it will be shit or inferior, but just because you throw a lot of players into a game and market it as realistic, it does NOT mean that they can or will act in a realistic manner.

Re your second point: Division Bell isn't released so how do you know what it does and doesn't do well? I have seen some videos that look nice but that's about it. I can't find a demo of the game so I can't comment.

[froggyluv 2135]

Once a week an aggro PvP'er comes in here and leaves the old "Why not just drop AI outta the game and play like real men!" -gem.Problem is my fondest memories from classic OFP times involved much and many acts of subterfuge, sneakin around unsuspecting bases and finding perfect spot for my incoming armored column ambush. Much like the game Thief..twould be hard to find PvP'ers taking up those roles of smoking, listless guard,nor Tank Commander who actually drives slowly thru valleys in formation and worst, standing in a suit of armor unmoved for many, many days.

DOn't get me wrong, there are some games I will play only in MP form or nothing, Swat/Arma are just not one of them.

[IndeedPete 1038]

Once a week an aggro PvP'er comes in here and leaves the old "Why not just drop AI outta the game and play like real men!" -gem.Problem is my fondest memories from classic OFP times involved much and many acts of subterfuge, sneakin around unsuspecting bases and finding perfect spot for my incoming armored column ambush. Much like the game Thief..twould be hard to find PvP'ers taking up those roles of smoking, listless guard,nor Tank Commander who actually drives slowly thru valleys in formation and worst, standing in a suit of armor unmoved for many, many days.

DOn't get me wrong, there are some games I will play only in MP form or nothing, Swat/Arma are just not one of them.

Not to mention the one or two people (me!) who actually don't do MP. I've played COOP with an Arma buddy like one time. And a couple of testing sessions we did while developing an SP/COOP mission. SP is the key for me, if Arma would be a sole MP title I wouldn't play it. Other players are a huge immersion killer for me and in most MP games (or gamemodes) I don't like the lack of progress, the lack of accomplishing something. I feel like I'm doing the same thing over and over again and don't change anything in the game's world - that goes at least for the "classic" MP modes, PvP, domination etc. (Good) story-driven SP titles give me the feeling of actually progresssing while following the story. The same reason I don't watch sports - because it's all the same. This year team 1 wins, next year team 2 and the year after team n, wow what a development! I'm being extra sarcastic here, don't get me wrong, I loved to play stuff like UT or CS ten years back with friends on LAN parties. But these were more social events than actual gaming. Even played WoW for a year or two, there was some sort of progress by developing one's character at least. I just prefer story-driven SP over MP almost every time and I'm glad Arma offers both - so everyone can just play as he likes, be it MP or SP or something in between. And that matter pretty much justifies an allround engine like RV in my opinion. It has many flaws, often discussed but I wouldn't want to change it for a completely new one with a questionable outcome. :)

[CaptainAzimuth 714]

Yes I understand what Squad is and what PR is. One of my real-person (not internet person) friends was a lead modeller on PR:BF back in the day and we've played l-o-a-d-s together so I think I know all about it already. There's really no need to explain it any further for me. :)

My comment was more on what types of modes will exist. Take for example something simple like convoy ambush.

With a game that has AI, you assign the AI to guard the convoy and attack it. Brilliant. It works every time as the AI are dumb and there's no way they can anticipate the attack.

Now try the same thing with all players. The players who defend the convoy will automatically know they are going to be attacked so instantly the scenario will not work as it would do if you use AI.

There's no way round that one.

I'm not saying it will be shit or inferior, but just because you throw a lot of players into a game and market it as realistic, it does NOT mean that they can or will act in a realistic manner.

Re your second point: Division Bell isn't released so how do you know what it does and doesn't do well? I have seen some videos that look nice but that's about it. I can't find a demo of the game so I can't comment.

I see. You have a point there on that Ambush note? But remember, even though someone may know they are going to get Ambushed, try will never know when, and how, the environment won't be the same, therefore they won't be able to react the same way each time. Put enough gameplay flexibility and opportunity into a game mode, and people can pull off some nifty ways of doing things that you wouldn't expect.

On the side note of Snow Drop Engine for The Division, they had small presentations explaining what their Engine does, why it was built, and a few things it can do. That's how I know a few things it can do and the game isn't out yet. From what I've watched, basics like, their Engine is capable of dynamic texturing, which for example allows for say, on a freezing day, there is Ice on the roads. It gets a little warmer, and you can actually see over time that Ice melts, turns into water. Temp goes down and it freezes again. Snow builds up, and snow melts. Another good feature are Large Environments where the world is connected. There's no fade out and in when you enter a building or walk from a subway station to broadway, or to the Town Square. It does have a ton of teamwork based gameplay mechanics too, but I havn't been looking into it lately, so I don't remember everything.

Now if BIS were to create a new Engine, or Overhaul their existing one, what would they build it to do.

[calin_banc 19]

I see. You have a point there on that Ambush note? But remember, even though someone may know they are going to get Ambushed, try will never know when, and how, the environment won't be the same, therefore they won't be able to react the same way each time. Put enough gameplay flexibility and opportunity into a game mode, and people can pull off some nifty ways of doing things that you wouldn't expect.

Yes, but that "flexibility" goes both ways, with no way to control de skill of those defending.

[pulstar 55]

 

Sorry for bumping this back up, but if a budget game like Heavy Fire can have all this "variety" why not Arma.. heck BI could just farm out asset/animation/particle development to third parties and sell them as DLC.. They know they have a captive market :D

[chortles 263]

Because it's a Time Crisis -style rail shooter which is necessarily even more set piece (tailored narrow experience) then BF/COD?

By the way, in case it hasn't been mentioned here, Bohemia months ago mentioned that DayZ's Enfusion engine is a very early version of Bohemia's in-house, multiplatform engine for their future titles, the year-ago-mentioned Real Virtuality/Enforce merge, so checking the DayZ dev reports/comments can give you a sense of where that's supposed to be headed and hope some of those changes can benefit a hypothetical Arma 4 (i.e. a long-term goal of replacing SQF altogether)

[fabio_chavez 103]

they will have all these massive maps but with a lack of AI, they will have to get humans to do everything on them, which won't work out well as AI

sorry but you seem not to get the concept of PR, Bots are completely obsolete and they would destroy the experience. Its obvious though for some reason many arma players (and devs apparently) dont seem to perceive it that way, eventhough the record -when it comes to pvp-, shows that PR got it right where arma didnt. Its not even debateable :)

[St. Jimmy 272]

Because it's a Time Crisis -style rail shooter which is necessarily even more set piece (tailored narrow experience) then BF/COD?

By the way, in case it hasn't been mentioned here, Bohemia months ago mentioned that DayZ's Enfusion engine is a very early version of Bohemia's in-house, multiplatform engine for their future titles, the year-ago-mentioned Real Virtuality/Enforce merge, so checking the DayZ dev reports/comments can give you a sense of where that's supposed to be headed and hope some of those changes can benefit a hypothetical Arma 4 (i.e. a long-term goal of replacing SQF altogether)

Yep. It's the future engine. It's just missing the military AI from the base. The SQF will be gone and replaced by EnScript.

One cool small feature was added recently, I believe in just in the current Experimental. One extra bullet in the chamber.

 

Also taken from the DayZ roadmap post:

• Bohemia Interactive fully plans on continuing work on DayZ past 1.0 for at least a targeted 5 years"

That really confirm that the engine won't be just dumped and it will get developed a lot.

[pulstar 55]

Might get DayZ one day just to see this new progress in action, even though I'm more into the milsim sandbox magic of Arma. If Arma truly is a transitional title then it's doing good so far given its legacy infrastructure.

[chrisb 196]

sorry but you seem not to get the concept of PR, Bots are completely obsolete and they would destroy the experience. Its obvious though for some reason many arma players (and devs apparently) dont seem to perceive it that way, eventhough the record -when it comes to pvp-, shows that PR got it right where arma didnt. Its not even debateable :)

 

I don't think PR got it right. In a realistic mil-sim (crossover game), you need AI, very good AI. The reason why, would take up a few pages. But in brief and only in my opinion, others have their own opinion on AI.

Its gaming, humans know why the scenario was setup, they know generally, what's going to happen and most likely where it will happen. AI don't know any of these things, that's why, provided they are very good AI, the mission/scenario/campaign will always play out better and more realistically, when a really good AI is added as the opponent. The thing is, it has to be very good AI, and before anyone shouts 'restricted unavailable mods'.. Its already available publically from some mods and to a reasonable good degree and has been for years with this series. Its just that very few players persevere with it, to find an AI that plays to those strengths.

 

Having an AI, that will do their own thing on a terrain during a mission, changes forever the way a player plays the series. Yes the setup is more complicated, although AI mods such as GL3/4/5 will make this much less so. But its worth the time to take to make the missions as surprising to the human player, as it is to the AI unit. That is where, I believe, BI have lacked for so long, not bothering about AI. But they do have a reason, really good/complicated AI, has a hit value, in terms or performance, that's why they don't bother as much with it. Its far more simple to run the vanilla AI they have (which is pretty good when compared to other games), than run a more advanced complicated AI system. Because to begin with, the player has to set it all up, that takes a lot of time, most casual players, more-so with A3, just don't want, need or have the time to do that. Then of course the system to run it, doesn't have to be the top end, just that the player needs to know his/her systems limitations, which unfortunately most players don't, or lets say, don't bother to find out.

 

PvP can be very predictable, more or less totally. However, when playing with a decent AI, it becomes very different, especially when that AI has the whole terrain to use and the faction is given the equipment that befits its status. Then set out the AI to use their forces and equipment in a way they feel fit. The results will surprise most players. But the setup time can be much longer than a few missions to build, plus its very complicated to get right. Takes time and lots of testing. But the rewards are, you'll get the full experience of this series (imo).

 

Regards the engine. You know, its surprisingly good at what it does. In-fact, I think its very underrated by many players and those in the review media. It deals with much more than many games bother with. It does it well, yes, there are optimization issues, but on the whole it supports the game very well. Its old, no doubt about that, but don't underestimate how good it is.

Does it need upgrading, well yes of course. But it also needs to be built for the job its going to do, not just a job to cover many titles, that could be a mistake, in my view.

[fabio_chavez 103]

apparently we are not even talking about the same concept.
Player driven objectives and teamplay with no AI is a perfectly well established kind of gameplay, thats selfevident. just look at 10 years of PR.
You may not like it and you may well play along with bots as you please but how does that mean that there shouldnt be pure pvp available in arma? i mean real pvp with squads, roles and logistics, not that koth or wasteland crap.
Arma has the coop, the scenario play, the pre-organized matches, the tacticool etc etc but besides that it will never reach the level of teamplay that you see even on a average public server at any given point in time in PR where its not unlikely to see 50 humans act in concert vs another 50 humans. Arma public server culture? not even close! Its something completely different if you have 50 players commit to a common goal, communicate, move, fight etc,

You know everyone you see is human, every shot you hear, every mortar strike, every CAS fire is a player decision and that means something! Humans are enthusiastic non-fighting air taxis, doing logistics and spending time shoveling fobs, no Bots are needed in that context, they would destroy the experience. KotH is complete junk in that regard obviously, it has more to do with acardish progressive perk gameplay like BF4, i was about to compare it even to counter strike but that would be an insult for CS since CS has more sophisticated teamplay and realism than KotH lol :)

Imho it would have been nice if Arma3 would have been released offering some kind of pvp experience along those lines, it would have been imensely enriching. If you substract altis-life, wasteland and koth there is not much left, public server wise, anyway good luck with end game.  Arma 4 should offer some kind of conventional squad based combined arms pvp with roles, chain of command and player driven objectives etc and it should be taken care that the game is in an appropriate shape accordingly, it would blow any other game out of the water.

[roshnak 41]

Hold on,

 

(i.e. a long-term goal of replacing SQF altogether)

 

This made me excited

 

The SQF will be gone and replaced by EnScript.

 

Then this made me less excited. What is EnScript and why are they apparently creating another custom language? Is it essentially SQF but faster?

[chortles 263]

Yep. It's the future engine. It's just missing the military AI from the base. The SQF will be gone and replaced by EnScript.

One cool small feature was added recently, I believe in just in the current Experimental. One extra bullet in the chamber.

Chalk that up to Rocket's decision to rework the engine to have weapons as objects and not just 'shapes'/strings, the former term being used in the OPREP explanation for why Arma 3 weapons don't have "bullet in the chamber"; if I recall correctly the change was made because Rocket's vision for the DayZ standalone required gear to have durability states (condition of the weapon) but we see a milsim-applicable benefit... to say nothing of allowing for different ammo types in one magazine.

**

 

Then this made me less excited. What is EnScript and why are they apparently creating another custom language? Is it essentially SQF but faster?

To quote from a recent Reddit Q&A, Enscript is "Far closer to C++".

As for the why, I'll quote the DayZ 22 April 2015 Status Report:

My personal holy grail throughout the development of DayZ has, and will remain the server side performance. Much of what folks attribute to "lag" or "desync" can be traced back to poor server side performance. 85% of which comes from the performance heavy usage of SQF engine script for many gameplay systems. Mentioned in prior status reports, EnScript (the name for Enfusion's engine scripting language) is leaps and bounds ahead of the legacy SQF in terms of performance. This, the portion of the performance budget currently consumed by SQF usage is the largest roadblock between 50 player servers, and 100+ player servers. Moving the title completely away from legacy engine scripting tech, and onto the new engine scripting language for Enfusion is by no means a small task, and will be an ongoing goal for the team (and in my opinion 100% required for DayZ to reach 1.0)

Sample EnScript - Not actual in-game

Also, from the aforementioned Q&A:

Q: What has changed in regards to the reasoning to improve the way a player executes actions (e.g. eating rice, as shown in one of the videos)? I remember when such feature was speculated to be too taxing for server-performance.

A: Absolutely nothing - as far as I can recall this has been a very long term goal that required significant change to the engine. All in all the actions are still powered mostly by script. Of course new stuff is written in Enscript, which is 300% faster than SQF. Although, the impact on the server should be negligible.

Q: In the new introduced gapless animation system, will it be possible to repair a part of a car, for example the wheel, take a break, and then continue repairing at the same point where you stopped?

A: As I said in the Status Report - the design team has a big pile of actions to completely rewrite in Enscript, so we'll see as they get to some of those actions how we'll handle them.

Q: I really like the hold down mouse key to eat and let go to stop, will this apply to chambering a round with the CR527 for example (let go of key to chamber)?

A: I'm going to sideline this question, as we're currently working on both new mechanics for loading/chamber/reloading - AND the move for actions to Enscript.

Q: You were talking about i guess 3 skills your team is working on so far. any sneak peeks here? what did you consider when you were laying the grounds for skills/skill-trees? will skills be more like perks: you gain an ability you didn't have before or more tuning of numbers, or a proper mix of both?

A: Unfortunately, We're not talking about what those three initial skills will be just at the moment. What we considered initially was something that could give a good functional representation of what we wanted the feature to represent, for the designers creating the Enscript for it.

No new abilities, just getting better at the ones you already have.

Q: Myself as well as others would like to see windows capable of opening, closing, locking, entering/exiting through, and breaking. I even seem to remember people(devs/rocket?) talking about breaking windows with a blanket over them, like zomboid, to reduce sound and chance for cuts.

A: Ah ha! Just like PZ. (I love that game and those devs - if you have not checked it out, search for Project Zomboid on Steam). Sadly, I'm not sure that kind of depth is going to be something that's reachable within the scope of 1.0.

That said - Im certain if you wanted to author a mod in Enscript that performed such a function, you totally could.

Make of these statements what you will, as well as Hicks' tweet that "new modules and technology from the Enfusion team is merged into the stable and experimental branches of DayZ, and deprecated placeholder modules and technology from the Real Virtuality engine is replaced and merged out."

 

As an additional note, Bohemia stated on Twitter that Arma 3 will remain PC-exclusive (and as a company they're "doing quite fine as a PC-first company and there is no reason for us to change that") but also that "DayZ is coming to PS4 and Xbox One, and our future games are likely to be released on consoles as well", and "Our new Enfusion engine has better multi-platform support, so it will be more feasible for potential future Arma games."

[Mirek 166]

You may not like it and you may well play along with bots as you please but how does that mean that there shouldnt be pure pvp available in arma?

 

Imho it would have been nice if Arma3 would have been released offering some kind of pvp experience along those lines, it would have been imensely enriching.

What the hell are you talking about?  Did you even ever played arma?

 

 There is Pure PvP exactly as you are describing it, readily  available in arma. Did you ever see ShacTac videos? they play exactly what you are describing here. Or what about Tushino server in arma2? They play Combined arms mission with absolutely no AI involved.

It is native part of arma. 

 

 

Arma has an Mission Editor if you dont know. Making Pure PvP or TvT scenario is so easy that trained monkey can do it. Its so easy that everyone can prepare his PvP/TvT scenario for himself, in few minutes.

 

Just open the editor Place playable units for each side, Place empty vehicles for each side, and set respawns for the sides and vehicles if you want respawns. 

You want player controled CAS? place a plane on the map and playable pilot next to it.

 

You want Player controled artillery? Place a Gun on the map.

 

You want player driven objectives? Just dont include any objectives in the mission, let the players decide what they want to do.

 

It cant get any easier.

[semiconductor 309]

To quote from a recent Reddit Q&A, Enscript is "Far closer to C++".

Out of all languages around Bohemia have decided to emulate the 30-year old one with one of the steepest learning curves and countless possibilities to shoot at own foot, i.e., a complete overkill for average mission maker's purposes.

GJ, that's the Bohemia Interactive I know! <_<