Jump to content

Interview with Steve on RockPaperShotgun


Recommended Posts

  • Replies 57
  • Created
  • Last Reply

Top Posters In This Topic

Its a good read. Still quite cant get my head around the concept of engineered v designed for effect solutions. Designed for effect is A happened and the outcomes can be B,C,D randomly. Engineered is mimicking the environment i.e armour plating is at this angle , ballistic path of bullet comes from X at Y speed there for the outcome is Z even if you do it 100 times when all the conditions are the same.

Link to comment
Share on other sites

Engineered is mimicking the environment i.e armour plating is at this angle , ballistic path of bullet comes from X at Y speed there for the outcome is Z even if you do it 100 times when all the conditions are the same.

In CMSF this was true and most outcomes are predictable because of the lethality of the weapon systems. However I suspect in CMBN even though all conditions are the same you will see more variability in the outcome.

Link to comment
Share on other sites

Its a good read. Still quite cant get my head around the concept of engineered v designed for effect solutions. Designed for effect is A happened and the outcomes can be B,C,D randomly. Engineered is mimicking the environment i.e armour plating is at this angle , ballistic path of bullet comes from X at Y speed there for the outcome is Z even if you do it 100 times when all the conditions are the same.

Yes but X and Y can be any (infinite) number of possibilities which can come from a stochastic effect, and Z will be right Z. Designed for effect doesn't know what to do if A is suddenly changed to B.

Link to comment
Share on other sites

It saddens me that Steven used smilies in the interview

Ha! That got a laugh out of me. I noticed that too. A portent of the downfall of civilization no doubt.

An alternate theory is we're switching from an alphabet-based language to a symbol-based language. ";)" has become just as relevant a communication device these days as "=" and "$".

Either that or its a portent of the downfall of civilization. :rolleyes::o:P

Link to comment
Share on other sites

Nice read.

This part really got me laughing as I could picture the carnage and frustration :D;

He was defending as Germans and his lines were penetrated by a small American element including a Priest. He drove back the penetrating element, but the Priest managed to back itself into his positions in its confusion as it fled. When it went up it took most of one of his squads with it. To add insult to injury secondary explosions from its demise suppressed the better part of a platoon and a half of his men and opening a gaping hole in his lines. Truly great stuff.
Link to comment
Share on other sites

Thanks guys,

Krilly, I had a similar reaction when I read that Priest story. In real life I don't think soldiers would take the time to think ahead to what cooked off ammo might do, rather they would just say "kill it!!" and THEN find out :D It also amazes me how often you read about stuff like this in real battles. Exceptional, of course (as was this incident), but really crazy stuff happens in battles.

In fact, a friend of mine in Afghanistan mentioned a firefight he was in where they walked around the corner and saw a couple of guys wading down a stream with IED materials floating on an inner tube. Both groups saw each other simultaneously (I think) and both were completely surprised. Needless to say the guys in the stream didn't stand a chance, but were not smart enough to try surrendering. Not to hold up Saving Private Ryan as anything to judge history by, but that scene where the wall came down and Germans and Americans suddenly found themselves staring at each other isn't hard to picture happening in real life.

Steve

Link to comment
Share on other sites

Hey folks, glad to see CMBN getting some much needed publicity. I just hope people interested don't come tot the forums first before checking out the game, they may be frightened and run away, and we don't want that to happen ,well, not before they purchase the game to keep our rants and raves going through another title or ten :)

Great read, I always enjoy the stories of what the engine is capable of when circumstances are just right, that Priest taking out a chunk of his line when exploding would be a priceless moment.

Link to comment
Share on other sites

Yes but X and Y can be any (infinite) number of possibilities which can come from a stochastic effect, and Z will be right Z. Designed for effect doesn't know what to do if A is suddenly changed to B.

In order to come to the aide of my dear friend, Joe Shaw (and to score brownie points to counter balance using smilies in an interview), what Stikkypixie said is part of the reason why designed for effect stuff becomes a real drag on a low level sim with the degree of complexity of CM.

Think about it this way. In order to design for effect you need to first identify the SPECIFIC effect you're looking for. If you do not design for it, you don't get it. Or at best you might get it randomly, which isn't good. Worse, you probably are just as likely to get undesirable stuff that then has to be explicitly coded around so it doesn't pop up while playing.

Engineered solutions, on the other hand, simulate a process instead of the outcome. The outcomes come about naturally from the process. If the outcomes are undesirable you go back and change how the process works so that the undesirable stuff is naturally avoided. If an undesirable effect comes up you go back and make a fundamental change to the underlying process or an isolated change to a specific component of another process. Whichever the case may be, the change is usually fairly isolated.

While these two processes sound similar, they are fundamentally different. Here are two examples based on two rifleman, one suppressed and the other not suppressed:

Designed for Effect - do not simulate the rifle's characteristics, rather simply reduce firepower of suppressed rifle relative to unsuppressed.

Engineered - explicitly simulate RoF, trajectory (LOF), inherent accuracy, etc. of that specific rifle. Degrade aim point precision and reduce opportunities to fire.

In many situations the results would likely come out to be roughly the same, but designed for effect is a lot easier to implement. But inherently the designed for effect system is not treating the rifle as a rifle. Instead it is a mush of stuff that doesn't break down into component pieces. This makes it harder to integrate consistently with a wide array of potential "outcomes". Engineered systems, at least when done right, have these sub components broken out separately. Which means that the rifle itself will always perform within realistic boundaries of that particular rifle no matter what the situation. This reduces potential for fixing problem A and causing problem B.

This does not mean that every individual outcome is 100% predictable with an engineered solution. That would be inherently wrong if it was. Rather every individual effect falls within predictable range of end results. Realistic amounts of variability within the ranges is simply part of the engineered effect.

A complex, interdependent low level environment (like CM) requires huge numbers of designed for effect bits of code in order to yield only consistently realistic results. We tried with CMx1 and it just doesn't work out well. We kept piling in specific bits of code for specific situations to yield specific results and it just never ended. Worse, these bits of code liked to fight with each other because they weren't based on "constants" like ballistics, explicit LOF, etc. And the more we added, the more problems we encountered. The exact opposite has been experienced with the CMx2 code base. The further we go with it the less problems are encountered and the easier it is to integrate new elements without disrupting the old.

It is important to remember that there are PLENTY of designed for effect elements in CMx2. But they are integrated and generally trigged or bound by engineered results. For example, a highly suppressed soldier's rifle behaves 100% the same as a non suppressed soldier, but the outcome usually is different. Do we have specific data to know how much to penalize this suppressed guy based on a particular situation? No. Instead we tweak accuracy and firing opportunity variables until we are pleased with the result. But we don't do things like reduce firepower ratings.

Yeah, it's a complicated topic. The bottom line to remember is we've done both and we find engineered systems to be vastly superior. Not from an academic position, but from practical experience. If anybody wishes to challenge us on this point, they are welcome to show us a competing product that is both superior to CMx2 and uses a designed for effect core engine. The only one that comes close to meeting this definition is CMx1, and it falls short.

Steve

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Unfortunately, your content contains terms that we do not allow. Please edit your content to remove the highlighted words below.
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...