Greetings,
We have just released a detailed dev blog about our ongoing Tiericide efforts. Please read through it and engage in a discussion in this thread 
Greetings,
14 Likes
1 Like
Thank you!
An updated SDE after each tiercide update is important.
4 Likes
Exciting!
Iâm loving the increasing use of keywords. Itâs a fantastic tool to help new players quickly understand what the modules do and what benefits to look for.
2 Likes
Thanks for all the examples, but couldnât help but feel like Gowron trying to follow Quarkâs explanation of complex financial transactions.
3 Likes
It would be nice when renaming modules if the fit import tool that exists in game, could automatically allow import of âoldâ module names and convert the modules to the new names.
E.g. the recent rename of Adaptive Invulnerability to Multispectrum, would be a case where this functionality would help players a lot during the transition of fits stored elsewhere (e.g. wiki pages, old docs, guides on forums/reddit) to the new names.
Please seriously consider this improvement going forward.
Garth
2 Likes
For example, -88% bonus is not 8%, nor 10% better than -80% bonus. It is 40% better as the better module reduces the remaining incoming damage of the weaker module by 40% (1 - 0.8 = 0.2; 1 - 0.88 = 0.12; (0.2 - 0.12) / 0.2 = 40%). As the base value for comparing all modules damage resistance bonus, the Meta Level 0 - Tech I module (-40%) is taken.
Thatâs wrong.
since their effects are combined by multiplication, you need to use a log scale on their resonance mult
- -80% is a multiplication of damage (=resonance) by 0.2
- -88% is a multiplication of damage by 0.12
if you have TWO modules of -80% (and no stacking penalty) that means the damage is multiplied by 0.2Ă0.2 = 0.04. Using n modules multiplies the damage by 0.2^n.
Conversely, if you consider a module with a resonance multiplier x, and this module is y times as strong as a -80% one, that means x=0.2^y . ie, the formula of strength comparison to a -80% module is y=ln(x)/ln(0.2) .
Therefore a module with -88% resists, so multiplier of 0.12, is ln(0.12)/ln(0.2) = ±1.32 times as strong as a -80% module. Thatâs a +32% increase in strength. NOT 40%.
if you consider base strength is -40%, then
- the -80% is ln(0.2)/ln(0.6) = 3.15 times as strong, and
- the -88% is ln(0.12)/ln(0.6) = 4.15 times as strong
- a -r% is ln((100-r)/100) / ln(0.6) times as strong.
4.15/3.15 = ±1.32 (which is logic, we have the same +32% strength). Whatever your base value is, using two modules without stacking will result in a combined effect with twice the strength.
Thatâs why removing a fixed percentage of the resist modules was a complete nonsense.
17 Likes
Big brain time.
Further proof that CCP is using flawed understanding of math with their own game to grind it to dust.
1 Like
Math is hard, even when youâre getting paid to do it.
One thing I would remind CCP and other analysts: programmers and mathists have a tendency to get bogged down in the statistical details of exactly how the mechanics work.
Your typical player will never look at, need, or understand that level of explanation. Nor does it really enhance gameplay.
When youâre considering game design changes, altering formulas, updating names and adjusting tooltips, you should keep in mind that most gamers only ever want to see âwhat will the final result of this change be?â and they donât really look to far beyond âI want a single number showing how much better it is, not in fractions of a decimal of an approximation of a lemma, but in terms of âwhat the intended effect will beââ.
10 Likes
Good grief Charlie Brown! I want to play a game now and then not undertake a tedious course in alien engineering taxonomy. You guys are getting seriously disjointed. This road leads to madness for some unlucky people. Spin aside (you sell it like true politicians) you do realise that you canât influence a complex system to achieve specific predictable goals, nor calculate the outcomes of your actions?
Tiericide should also include space sec levels, agents and other aspects outside of ships and modules
Meta Level 1 modules will have their resistance bonus increased from -40% to -42% to put their primary attribute between Meta Level 0 - Tech I (-40%) and Meta Level 5 - Tech II (-44%).
With tiericide meta1 modules will always be better than meta0, and often cheaper as well because the rat droppings are plenty. It greatly inhibits entry-level production of manufacturing T1Meta0 modules - it is pointless unless you research and manufacture T2 as well which might be unavailable for newer players interested in manufacturing path.
I have an old idea of solving this problem, but any other solution is fine as long as it promotes meaningful (player)interaction/production/consumption.
3 Likes
Agents had their version of tiericide already, there used to be TONS of different agent types each with their own distribution of mission types
The issue with tiericide is that it removed desirable Tech I modules from the game. All modules that underwent tiericide are cheap as hell, because all are equally common. So now thereâs much less excitement in looting NPC wrecks, looking for those modules that are worth a few million ISK each, and can pad your income. Similarly, as far as ship fitting goes, weâre now in a T2/compact duality of choice, with nothing available that makes ship fitting more dynamic. Basically, you always get T2, unless you can either (a) afford faction, or (b) are lacking fitting resources, and need to use compact.
Tiericide is well-intended, and certainly simplifies naming schemes, but it does have drawbacks. Taking the money part out of the equation is one of them.
I highly recommend the addition of a new grade of module for each type that goes through tiericide, which will have overall better stats and lower drop rates. This will help bridge the gap between T1/T2 and faction modules. An easy solution would be to apply more than one of the new naming scheme types to modules; for example, a âcompact enduring afterburnerâ or something along those lines. Make the drop rates low (1-2% of single-variable modules), but not faction-low.
1 Like
Or make an off-by-one error and unable to use T2 due to skills restrictions.
To give you an idea of the scale of the task, here are the numbers of types of all turrets and charges:
These numbers are rather useless, as most of them can be represented as simple and very visible graphs. Then all you have to do is to put some number tweaks to see, where curves are going.
there is a variant of using T2 with cheapest mutaplasmid⊠about 10-20 attempts to get âall greenâ with efficiency close to faction modules.
1 Like
thatâs the exact opposite goal of tiericide.
4 Likes
Mutaplasmids are a separate consideration.
Tiericide was never the best of ideas anyway. Do we need 4-5 modules with a linear progression of stats? No. But we donât necessarily need to be forced to choose from only 2 options either. Their new naming scheme for stat variables is good, but that doesnât mean modules shouldnât be affected by multiple modifiers. Right now, we simply donât have a choice with regard to ship fitting. Get T2, or if you canât, get compact. Itâs boring, while still being too complex for newbies and/or simpletons, who donât want to compare optimal ranges and capacitor usage, and just want to throw some stuff on a ship to go pew pew.
And a lack of quality modules affects the profitability of low-end PvE. The newbie can no longer get excited by finding an Arbalest launcher in a wreck. Yawn.
This has been identified. We want to look into this. The rules that are being established through the tiericide process will be used to drive changes in manufacturing and distributions.
5 Likes