again brit tanks twice that of axis?

[Scotsman]

The Cvc values of the past could never conceivably or notionally pass any statistical validity test that I am aware of. 

The desire as always is to come up with a fair system that gives all parties a chance to succeed. That includes the developer financial success by definition. That system need not be exclusive to any party. If you can prove statiscally that a prior systems is bonkers via t test or other methods, then there is no way anyone can claim that said system tested is unbiased or balanced with a straight face. 

If the game is such that this is somehow necessary, then the game has serious basic design issues. This is the sort of thing a developer should look at before you even go beta on an MMO. 

In the end, the system has to be divorced from any underlying player effects or arbitrary human effects measures for the reasons I have already gone over. Every army in the world with the noticable exception of the immediate post WWII Russians arrived at that conclusion...frequently after expending 100s of millions of dollars in live equipment and unit testing. If that consensus can't be accepted then there are obviously other issues that will mitigate against the game ever succceeding. Imho its silly to ignore mathematical reality. 

The game is ultimately driven by player numbers..regardless of spawn pools..but severely unbalanced spawn pools can and should be avoided as they will add insult to injury in unbalanced player situations.

If the question is whether there have been significant balance issues in the past...the answer is yes...and it can be shown to be the case by these failing standard statistical tests. This isn't an 'opinion'...it's mathmatical reality. 

I'm not going into what I found in doing all the tests...but there was a -significant- imbalance that became worse in tiers 1 and 2. It's up to current rats to decide what to share or not share.

I dont believe in 'because I said so' or I've got a secret. It's far better to derive a system with known inputs that can be made transparent to the user community.

[jwilly]

So within a given class, manufacturing cost correlates with lethality and game-combat effectiveness, even though France manufactured with anti-military-involvement, Communist-led unions, Germany eventually had to manufacture in underground factories or clearings in forests, and USA manufactured in the best engineered brand new mass-production factories that GM, Chrysler, Ford, etc. could create...?

[Scotsman]

The cost to manufacture boils down to the materials used and chassis weapon characteristics...armor is inherent in gross vehicle weight. If you equate to then year dollars and eliminate year to year cost growth..and do a good job of estimating learning curves...and your statistical basis is ALL vehicles of all combatants... you can arrive at something that passes statistical muster IF manpower deltas like slave labor vs paid workers is also eliminated. 

I have posted - repeatedly- that was done. 

There are sticking points or special cases...two being char-b because of rapid cost reduction in a short period of time...and the Panther later in the war.

i have pointed out other differences that I let stand in the cost data as it was impossible in many cases to isolate the delta. The best example there is allied pre flight testing of aircraft and the German approach which did not do predelivery testing to anywhere near the same degree. That cost is built into allied aircraft cost though I may have been able to pull it out in a pinch if needed.

By setting the total budgets for each brigade type equal, you eliminate any imbalance due to historically unique consequences like bombing, moves to underground factories etc. the basic economics are then the same for everyone...and as long as the individual equipment prices are uniform in their derivation, you are good.  I can tell you in the baseline data there was a large statistically significant delta between the spawn pools. One that in no way was acceptable and could minus player numbers usually lead to only one outcome. That's bad design.

I included the correlation data between the real historical prices and the derived prices in the database for every vehicle where good cost data existed. It was typically with a few percent at most...

Without exposing the actual data I'm giving you all I can. I can tell you from a statistical point of view that some of the things you are listing were not in fact statistically significant to final cost. Others like the char cost did in fact reflect some of hat depending on what pricing structure you chose to use.

there are two sets of data - the actual and that derived by the cost model - crs could use either but the uniform cost model is preferable for the stated reasons

Edited December 22, 2018 by scotsman

[jwilly]

We aren't connecting.

You're explaining a system for giving each side the same manufacturing-cost-value of weapons and vehicles.

My understanding is that marketability requires that each side has close-to-equal aggregate weapon/vehicle lethality and game-effectiveness.

So my question was, is there a basis for knowing that within a given class, manufacturing cost correlates with lethality and game-effectiveness?

If yes, your historical-manufacturing-cost approach is great.

[stankyus]

11 hours ago, DOC said:

Does anyone remember CVC ? It assumed that the only economy that mattered to a game of combat was combat effectiveness based on equipment attributes and capability. That's where points budgets came from, and were used to balance variations in combat ability superiority and inferiority. You'll always end up with a general rule of less superior units versus more inferior units, but if this disparity of equipment capability exists by design, there really is no other way except to remove all inequalities. Which any historical reality makes impossible. Despite what any player of the game might attest, to any degree ... the only bias the developers can have is the one that keeps them in business. That bias can never, by virtue of what it is to be in business ... be based on selections that a player would make based on what side he plays on. In saying that, in instances where a side based bias does exist at the developer level, perhaps it is best to assume that being in business is no longer the desire of the developer (or never was) but having been one myself, I tend not to believe this is very likely at all.

IMHO CVC was a very good. i think over the years i have seen why we had it, why it worked and why it is a paramount idea. The problem with it is that in reality it does not really work, it is/was flawed, but the core philosophy behind it is solid. You did that DOC. It's lessons are not lost on me. Two major points that are core to this game you solidified. One, a cornerstone has to be laid, "what is balance".. and two, nothing can be for free. 

Im not gong to go over the failings of a paper trap. The most important success that IMHO works for balance AND to protect the integrity of the game is the paramount importance of keeping the idea that nothing is for free. Something you said often. As long as that is a rule, nobody can boast. The RATs avoid bias accusations, gaps can be filled, single additions negative effects to the side who did not get anything are reduced (bedside manner), lessens the impact and balance problems unforeseen. The idea of a CVC structure which Scotsman has evolved IMHO has to be given it's due fundamental respect as a key element of this game. It has to be girded with the philosophy of "nothing is for free." Therefore both objective balance and player "satisfaction" are addressed.  Took me a while to observe your vision, but the lesson is well learned. For that my hat is well of my head to you sir, it's not easy to pull back from your own biases to see the big picture though it seems simple, it was not.

[Scotsman]

7 hours ago, jwilly said:

We aren't connecting.

You're explaining a system for giving each side the same manufacturing-cost-value of weapons and vehicles.

My understanding is that marketability requires that each side has close-to-equal aggregate weapon/vehicle lethality and game-effectiveness.

So my question was, is there a basis for knowing that within a given class, manufacturing cost correlates with lethality and game-effectiveness?

If yes, your historical-manufacturing-cost approach is great.

That question is impossible to answer in most cases because player capability and tactical circumstance is not equal or predictable...a tank in the hands of a veteran is more effective than one in the hands of a new player. A tank properly deployed with support is more effective than one in isolation.

The spawn pools have absolutely nothing to do with tactical situation or map condition. That's mixing apples and oranges. In short there is no silver bullet to overcome the limits of the current game design. The spawn pools live inside those limits. 

As for basic correlation with in game lethality and survivability - yes - because the costing is sensitive to both gross vehicle weight (and thus armor) as well as gun caliber. Is a tiger more effective than a vickers? Yes....and it weighs more and costs much more. I don't think many would question the 88mm is more effective in game than a machine gun. 

You can not and should not connect these things to any measure of player effectiveness or any given tactical situation...not if you want a fair an unbiased answer.

[thomboi]

On 12/11/2018 at 5:44 PM, Kilemall said:

And I told you all the Matty CS would be an issue.  Yes helps it's gun just knocks down buildings, but ultimately it can camp hell out of inf and AT guns.

 camped armor needs a visit from JU87's.  We should have a simple colored smoke grenade (just like real life) that tells our CAS there's armor targets here...

[Scotsman]

On 1/12/2019 at 10:29 AM, thomboi said:

 camped armor needs a visit from JU87's.  We should have a simple colored smoke grenade (just like real life) that tells our CAS there's armor targets here...

Bring looked at and definitely agree target marking should be possible 

[texasjay]

On 1/19/2019 at 7:34 AM, scotsman said:

Bring looked at and definitely agree target marking should be possible 

Wow ,about time, been asking about this decade  ago.. thank you