3.5 Update (and an upcoming podcast)

[ralphtricky]

I'm working away on 3.5. It's a bit heavier than I wanted, but we're making good progress. Most of little things are already in, like being able to pin the pop up window to a corner of the screen instead of having it follow the cursor. I'm working on some bigger things now.

Episode 118 of the Three Moves Ahead podcast (hosted at http://www.FlashOfSteel.com by Rob Zacny) is going to talk about the TOAW 3.4 patch, and I'm sure we'll talk about other things. I'll be on, and I'm looking forward to it.

If you haven't listened to the Three Moves Ahead podcast, it's the only strategy focused podcast out there that I've found, and I've listened since episode 1. If you know of any other strategy focused podcasts out there, let me know.

[sPzAbt653]

Podcast ?? Neat, when is Episode 118 ?

[1_Lzard]

Oddly enough, there's a 'comment' on the Flash of Steel site asking the very same question!

[X(]

[macgregor]

I'm working away on 3.5. It's a bit heavier than I wanted, but we're making good progress. Most of little things are already in, like being able to pin the pop up window to a corner of the screen instead of having it follow the cursor. I'm working on some bigger things now.

Any word from the developer I will always appreciate like radio transmissions from Amelia Earhardt. I really don't want to come off as being antagonistic, but I will try to interpret these cryptic 4 sentences.

'Heavier' - More work. Slower.
'Little things' -things I didn't know were requested or necessary -like the example. They're getting done. I expected this.
'Bigger things' - I can only hope. I'll always have hope.

Communication is a beautiful thing.

[Curtis Lemay]

ORIGINAL: macgregor

'Heavier' - More work. Slower.

Mission creep.

[Panama]

ORIGINAL: Curtis Lemay

ORIGINAL: macgregor

'Heavier' - More work. Slower.

Mission creep.

Beer break.

Can't work without em. [;)]

[mbar]

Thanks for the heads up on your up coming podcast. I always look forward to a new one from Three Moves Ahead.

Thanks for your work on TOAW3. You're doing more to improve it? The last patch was as big as the original installer. Dare I say it could have been re-released as a "Gold" edition.

It's admirable you love this game so much even years later. [&o]

[BigDuke66]

Just listened to it, interesting talk.

[Panama]

The guys with Ralph on the podcast pointed out what some people have been saying. TOAW needs more flexibility. So that each scenario can be, almost, a unique game in itself. One way to do that would be to take each element in the game and make it adjustable by the sceanrio designer.

Seemed to be a lot of what these guys personally thought about the game instead of about the game itself. One guy obviously hates it. [>:]

Anyway, nice talk Ralph. Thanks for taking the time and everything else you do. [&o]

[sPzAbt653]

TOAW - 'A Frightening window into the minds of Wargamers'.

Really ?? Well, taken in the context that it was said (referring to some 'amatuer' scenario designers), its an understandable statement. Funny to hear a statement like that, though.

It seemed they all had limited views of TOAW, but really overanalyzed to the point of sillyness. TOAW is like 'computer modules of biological systems'. Umm. And philosophical discussions of what ? Its a game. And the age old griping about scenarios that are too big. Don't like 'em, don't play 'em. No reason to say that TOAW doesn't work. 'Monsters' probably only make up 10% or less of the total scenarios available, anyway.

And raise your hand if you're already tired of hearing how great WitE is.

Thanks for doing the interview Ralph ! [8D]

[macgregor]

Well first of all, let me offer my gratitude to Ralph for doing this extensive conversation. I feel quite satisfied in having understood more about the thought process, the ideas, fears, and values that Ralph is fielding.

Both Troy and Bruce came off as passionate, articulate gamers. I think Bruce's primary criticism about 'other factors' is totally defrocked by the fact that the game DOES have other values, and that's just one of the things that make this game not only superior, but IMO designed to be perfectly 'tweaked'. Bruce doesn't seem to get that while more numbers doesn't equate certainty, it does better equate. The only thing less numbers does, is provide a more certain calculation. Which while less realistic, is what Bruce wants.

There really isn't so much of a 'sweet spot' as there is a 'sour spot' regarding scale, and if you walk into any wargame store (do any still exist?) you realize there are many different tastes out there. Ralph I think was slightly reticent to give the perfect scale, but finally said 10km. I would agree with that. I don't really mind the smaller scales, but thousands of pieces I feel I'm not doing the work of one man, but that of 100s. IMO games from 10km to 40km work for me. I don't mind big scenarios if they accomplish big things. But then I like scenarios that are conducive to more than two people.

The more I listen to Ralph, the more he reminds me of Steve(developing WiF) in that they get a hold of the AI, and then want to sort of live vicariously through it. They believe in a 'perfect strategy' and then want to implement it through an AI. Have you seen a chess-player play like, 20 games simultaneously and win them all? Imagine that on a scale that is almost infinite, as it can continue after death. But it's not all ego. After all, the AI is supposed to teach and compete with people. How well can you do that? Is simply teaching the players how the game works enough? Some people prefer a computer to a human opponent. That's just a fact.

My feeling is that as long as he respects the ambitions of others -including somewhere in the equation, mine naturally, I have no problem with any of Ralph's ambitions, regardless of whether I personally will use them. It is a 'markedly improved' game. Any criticism I have had has to be taken contextually. There is nothing Ralph has done that I am not grateful for. And I don't mean that altruistically, I really do appreciate how much better the game runs, and of course, looks.

[ralphtricky]

To be fair to Bruce, I think we're coming from different viewpoints. If you look at Art as an example, I would say that most paintings are far better to look at, and you have a lot more freedom of expression to create exactly what you want. At the other end, if you're 'skinning' a model for a modern FPS or other game, it's still Art, but you're stuck with the skeleton that they provided, and have less freedom. That doesn't mean that you can't do a great job, but if you want it to act like an elephant and the game engine wasn't designed for that, then it breaks down in subtle ways.

Anything hand-crafted for a specific purpose is (almost) always going to beat something that's general purpose. It's also going to take a lot longer to build.

If you want a scenario that captures the 'essence' of a conflict, then a small unit count scenario with a hand crafted engine should definitely win. It's also going to take a lot more skill to craft.

If you enjoy the low level 'Monster wargame' play, then that's an option too. Both are valid play styles.

I think that the question of scenario design is not related at all to the level of abstraction that you're using in the engine design.

As a mathemetician, I can agree that Bruce is right that TOAW isn't demonstrably more valid that any other option (say dice rolls.) There is no proof that mathematics has any correlation to reality. It's also irrelevant. As a Computer Scientist (actually, the programmer part of me) I think he's wrong in dismissing it as a starting point for a scenario. I've always found that doing my research, and then iterating my way to a solution is a great way to solve problems. I think that TOAW has enough hooks to let you start with a historic OOB/Map and modify it to be 'fun.' and to reflect whatever reality you are trying to portray. I disagree that TOAW isn't a valid starting point for being able to design scenarios, although it does make it easier to design bad scenarios.

My goal isn't to create the 'unbeatable AI'. I could do that today by adding in massive bonuses. My goal is to create a AI that is flexible enough to be able to be able to play historic variants of scenario, and that will feint attacks and fall for feints. In a perfect world, he would also adjust his play style to be able to just beat you 20% of the time, while you would barely win 80% of the time. I also want to be able to create a AI that would let me play an entire front at a very high level, while micromanaging particular critical areas. I expect this refinement to take a LONG time.

Ralph

[jmlima]

ORIGINAL: ralphtrick
... My goal is to create a AI that is flexible enough to be able to be able to play historic variants of scenario, and that will feint attacks and fall for feints. In a perfect world, he would also adjust his play style to be able to just beat you 20% of the time, while you would barely win 80% of the time. I also want to be able to create a AI that would let me play an entire front at a very high level, while micromanaging particular critical areas. I expect this refinement to take a LONG time. ...

Does that even matter if basic fundamentals of an operational game, such as the way supply really happens, are not modelled correctly?

Given that the quest for the 'great AI' has been going on since the dawn of PC gaming with, let's be positive, moderate results, isn't it be worthile to focus on some fundamentals of operational simulation, or, as some suggested in allowing for even more flexibility for the designer?

What is it better, a great AI on a somewhat flawed operational simulation, or a great operational simulation with a normal AI?

[ralphtricky]

Does that even matter if basic fundamentals of an operational game, such as the way supply really happens, are not modelled correctly?

If you have specific well thought out suggestions, please post them in the scenario design section.

So far, thing like that are done first, then I can work on the ai for a while until I get feedback from the team on what I broke . It is tough to get concrete suggestions on how to help Elmer, and even harder to test.

[jmlima]

ORIGINAL: ralphtrick

Does that even matter if basic fundamentals of an operational game, such as the way supply really happens, are not modelled correctly?

If you have specific well thought out suggestions, please post them in the scenario design section.
...

Doesn't the whislist discussion serve just that purpose?

[macgregor]

ORIGINAL: ralphtrick

Does that even matter if basic fundamentals of an operational game, such as the way supply really happens, are not modelled correctly?

If you have specific well thought out suggestions, please post them in the scenario design section.

So far, thing like that are done first, then I can work on the ai for a while until I get feedback from the team on what I broke . It is tough to get concrete suggestions on how to help Elmer, and even harder to test.

No offense Ralph, but I think that'd be more likely to happen if you'd post there more often. Is the idea of joining the conversation too much? Maybe once a week? I know more about Al Qaeda's plans. As someone who has waited over ten years to see specific improvements, unfortunately absence is not making my heart grow fonder anymore. The absence of specific reassurance hasn't helped either. But at least say 'ain't happening' or something to that effect. It's like a woman you consider the love of your life that's just not that into you. I have bouts with pessimism, but think of how long I've been optimistic.

[berto]

I listened to the podcast in full last night. Not reassuring.

With Ralph's expressed focus on AI and interface improvements, I heard no mention of what I consider most important: opening up all game parameters to scenario designer modification (includes dynamic maps); unlimited, and more kinds of (e.g., weather), in-game events; universal scenario version translator; improved air and (especially) sea modules (the latter handled abstractedly like air combat, even); ...

The promise of TOAW IV intrigues ... but on second thought, I expect more of the same (development along recent trendlines). More endless wait. TOAW as a decent land-air-naval WWII Pacific simulator being as far away as ever. Am I mistaken to think this?

Meanwhile, note this, where Christopher Dean writes

I am seriously considering starting work on a new tactical combat engine series that can cover multiple time frames (age of sail to modern) and combat (air, land, sea) using a dynamic AI, battle generator, fully modifiable combat unit information, and custom maps. No set date yet.. but I have been going over plans for this new engine for several years now. Estimated price mark would be only $24.99-$29.99 per edition. More news to be posted soon on our forums.

Not exactly operational (or strategic), but interesting nonetheless.

[parmenio]

ORIGINAL: berto

Meanwhile, note this, where Christopher Dean writes

I am seriously considering starting work on a new tactical combat engine series that can cover multiple time frames (age of sail to modern) and combat (air, land, sea) using a dynamic AI, battle generator, fully modifiable combat unit information, and custom maps. No set date yet.. but I have been going over plans for this new engine for several years now. Estimated price mark would be only $24.99-$29.99 per edition. More news to be posted soon on our forums.

Not wanting to derail the thread but...

...I'd rather he seriously considered finishing the next version of Navies at War.

[ralphtricky]

ORIGINAL: berto

I listened to the podcast in full last night. Not reassuring.

With Ralph's expressed focus on AI and interface improvements, I heard no mention of what I consider most important: opening up all game parameters to scenario designer modification (includes dynamic maps); unlimited, and more kinds of (e.g., weather), in-game events; universal scenario version translator; improved air and (especially) sea modules (the latter handled abstractedly like air combat, even); ...

The promise of TOAW IV intrigues ... but on second thought, I expect more of the same (development along recent trendlines). More endless wait. TOAW as a decent land-air-naval WWII Pacific simulator being as far away as ever. Am I mistaken to think
]
Not exactly operational (or strategic), but interesting nonetheless.

Some of that can be done now that I've opened up the scenarios to be saved as XML. I'm hoping to add a 'trusted pbem' mode that will allow for a moderated pbem game where a third party or program can modify the save file. It wont be trivial, but it completely opens up the game for anyone willing to work with XML and a programming language to add more of a strategic layer, a better weather model, etc..

I'm trying to time box 3.5 so it won't take more than 3 months. I may not succeed, but that is my goal. I really want it to be shorter, but I dont think it will be with the amount of work to be done.

To clarify a bit, AI is a very long term goal, not something I work on constantly. It is an extremely tough problem, and isn't a part of the 3.5 patch. The UI is also not part of 3.5, it needs a total rewrite to bring it up to modern standards which can't be done in a patch.

That new game sounds interesting. I'll have to look at it when it comes out.

Ralph

[ralphtricky]

ORIGINAL: macgregor.
No offense Ralph, but I think that'd be more likely to happen if you'd post there more often. Is the idea of joining the conversation too much? Maybe once a week? I know more about Al Qaeda's plans. As someone who has waited over ten years to see specific improvements, unfortunately absence is not making my heart grow fonder anymore. The absence of specific reassurance hasn't helped either. But at least say 'ain't happening' or something to that effect. It's like a woman you consider the love of your life that's just not that into you. I have bouts with pessimism, but think of how long I've been optimistic.

I do look at the suggestions. I try not to post there because I consider that the brainstorming session where people throw out all kinds of crazy ideas, and that's not the place to interject reality. I don't post very often because I don't want to make promises (even implied) that I can't keep. Once they have been coded, or are far enough along to ensure that they are possible, I'm willing to share.

I rely heavily on the forum and the beta testers for ideas. The hardest part of my job is probably saying 'no' because I don't think a specific idea fits my view of what TOAW is, most often because I don't feel that the added complexity for the player is worth the added richness it gives the designer.

For example, one proposal for 3.4 was for a range 1 assault to ask the player if they wanted to assault or bombard. Instead I decided to say that less than 50% bombard strength would assault otherwise it would bombard because the benefit isn't worth the added complexity to the player.

Ralph