Science & Industry

Hi everyone,

Following on from the discussion in the devblog thread here, I've decided it's probably time to create a proper thread to discuss revisions to blueprint data, rather than continuing it at the tail end of a 60+ page thread.


Here is the situation:
- What we are talking about is any "static" blueprint data, eg times and materials for different job types, plus the max runs attribute
- We are already going to have to change every blueprint in the game, and to this end have tools set up so that any systemic change to existing numbers takes a couple of minutes to implement, provided it's doing math based on other attributes that are in the same data or otherwise easily available
- We do have our fingers in the industry code right now, so small changes to eg formulas are on the table if we can justify them; larger sweeping changes are *not* on the table
- We are erring on the side of maintaining the current balance for things that are not on our list of goals (below), but we are happy with any reasonable balance disruptions in pursuit of those goals
- We are erring on the side of preserving the status quo in invention over preserving the status quo for T2 BPOs; note that, as previous point, we are not specifically targeting T2 BPOs in any particular way
- We are expecting to mechanically rework invention following the Crius release, so we do not want to create wasted work by making too many changes to it for Crius
- Simple things that make industry better are very much in our ballpark!

Specific goals we are currently pursuing:
- We would like to make copy times consistently lower than build times, so building from copies is the optimal play (dovetails with our starbase changes, for example)
- We are rebasing invention TE/ME values to all be positive or 0 at all times, removing negative ME/TE from invention outputs, as this solves a number of issues with removing extra materials
- We would generally like all blueprint data to follow a coherent pattern; we're still discussing how far we would like to take this
- For at least non-invention blueprints, we are reviewing max run numbers to alleviate issues in certain areas eg cap construction, nanite paste
- We need to deal with the interaction between the first point on this list, Gallente Outpost copy-time bonuses and T2 BPOs
- Removal of waste necessitates an increase in all manufacturing costs
- Removal of negative TE/ME probably requires an increase in T2 build costs to balance out component demand before and after

Specific changes I am looking at making right now:
- Copy time to 80% of build time base; TE and skills mean it works out slightly faster copying than building
- Setting rank to equal size*metalevel; rank is defined in this devblog, and I will explain "size" below
- Possibly changing build times to be a function of rank; otherwise modules and charges (among other things) may need to break the copy/build paradigm to avoid a big nerf
- I'm considering changing invention times so that build time is generally twice copy+invention, to maintain balance across character manufacture and research slots; this also has the advantage of giving invention time a clear driving force
- As above, normalizing max run count on things that aren't invented from to make them less tedious to build from in some cases


So, with all that said... discuss, and also suggest! We're looking for two things in this thread: discussion of things proposed in this post, and also suggestions for other things we could do to improve the overall blueprint dataset or elements thereof. If there's something in the way the numbers relate to each other, either generally or for specific products, that's always bothered you, please explain it in this thread and we'll look at changing it :)

Things we would prioritize if we were making suggestions in this thread, in descending priority:
- Explain what the problem is, and why. This is the most important thing for us as developers: to understand what you're trying to solve
- Explain a simple, clear solution - preferably one that doesn't require code changes :)
- Give specific examples and/or numbers!


Thanks,
-Greyscale



SIZE:
1 - frigate/destroyer modules (power draw between 2 and 34)
2 - cruiser/battlecruiser modules (power draw between 35 and 299) and all "unsized" modules (power draw below 2)
3 - battleship modules (power draw between 300 and 4999)
4 - capital modules (power draw above 5000)
20 - frigates
30 - destroyers
40 - cruisers
50 - battlecruisers
60 - battleships
200 - most capitals
400 - supercarriers
600 - titans

Still working out how this will apply to eg structures, it shouldn't be hard, it just needs the math doing :)

Still, it also remains clear that CCP looks at the industry problem from the perspective of the small scale producer who now and then dabbles in industry as the frame work proposed falls apart when you look at industry from a large scale. Is nobody at CCP doing some kind of analysis what happens when people not build one carrier or other capital at a time, but five, ten, twenty? What happens when you don't build 10 or 20 T2 Sentry Drones, but hundreds or thousands per day?

Aight, so I have to spell it out for you. That's fine. The reward is that with an emphasis on invention rather than copying, you are (going to be) on a higher footing with 20+ invention characters than someone who has five invention characters being fed by 15 copy alts.

If I have enough time, I actually have the proper code scaffolding to go in, get volume/day data for decryptors, T2 items, everything, assume that decryptor effects are staying the same but just scaling up from -4/-4 base to 0/0 base, and come up with a proper estimate of how decryptor demand would change.

Since I forgot to post these, the new numbers. Corrected some errors in my sheet along the way as well, so these should be more accurate.

Mining Crystals: Just over an hour (-99%, lol)
Drones: Light Drones 1.74 Hours (-64%), mediums 3.48 hours (-30%), large & sentry: 5.3 hours (+1%)
Small rigs: 2.7 hours (-66%)
Frigate modules: 4.45 hours (-30 to -35%)
Medium rigs: 5.56 hours (-65%)
All small ammo: 7.29h (unchanged)
Large rigs: 8.34h (-74%)
Medium/cruiser modules: 9.82h (+25% to +50%, depending on module)
Capital rigs: 10.9h (-83%)
All medium ammo: 14.57h (+24%)
Large battleship guns: 14.71h (+66%)
Other large modules: 16.01hh (+80% to +130%)
All large ammo: 21.78h (+6%)
Frigate hulls: 48.6 hours (+12% to +21% depending on class)
Interdictors: 85.59 hours (+55%)
Cruiser, Transport & Exhumers: 95.8h (Blockade Runners +5%, DSTs -11%, Recons -4%, Logistics +5%, Hictors -7%, HACs -11%. Exhumers vary from +55% to +93%)
Command ships: 135.72h (-11%)
Tech II BS: 175.23h (-11%)
Jump Freighters: 2527h (105.3 days, +72%)

There is some variance in the capital ship component BPO's

most are at 60 while a few are at 4.

these are at 60:

Capital Propulsion Engine Blueprint
Capital Turret Hardpoint Blueprint
Capital Sensor Cluster Blueprint
Capital Armor Plates Blueprint
Capital Capacitor Battery Blueprint
Capital Power Generator Blueprint
Capital Shield Emitter Blueprint
Capital Jump Drive Blueprint
Capital Cargo Bay Blueprint
Capital Drone Bay Blueprint
Capital Computer System Blueprint
Capital Construction Parts Blueprint
Capital Siege Array Blueprint
Capital Launcher Hardpoint Blueprint

while these are at 4:

Capital Doomsday Weapon Mount Blueprint
Capital Ship Maintenance Bay Blueprint
Capital Corporate Hangar Bay Blueprint
Capital Jump Bridge Array Blueprint
Capital Clone Vat Bay Blueprint

Having perused the new proposed blueprint file, it becomes inescapable to come to the conclusion that the approach that CCP is taking to solve the click fest that invention is by simply extending the time it takes to invent. Instead of using a technology solution like stacking, a bureaucratic solution has been proposed.

Still, it also remains clear that CCP looks at the industry problem from the perspective of the small scale producer who now and then dabbles in industry as the frame work proposed falls apart when you look at industry from a large scale. Is nobody at CCP doing some kind of analysis what happens when people not build one carrier or other capital at a time, but five, ten, twenty? What happens when you don't build 10 or 20 T2 Sentry Drones, but hundreds or thousands per day?

Disappointing to say the least.

wow once again Greyscale pounds a game aspect into the ground with the nerf hammer. "We want to make industry fun!" Ha!

Im not sure if this is the correct thread to put this in, but it does have to do with blueprint adjustments...

If the plan really is to change invented BPCs to a base ME 0 PE 0 and increasing the base materials to compensate then all of the decryptors are getting a fairly large nerf.

Take the "Process" decryptors for example, they provide +3 ME +3 PE to the outcome BPC. Right now that reduces your build materials by around 20% because of the way negative ME is calculated, after the change, because of the way positive ME is calculated, that same decryptor will give you a 3% discount to build materials.

A different example would be the "Augmentation" decrpytors which give you -2 ME. You would think these would be better off but they are actually also adversely effected. Currently they increase your build cost by about 13.3% (20% of the base "perfect" materials) after the change they will still increase your build cost by 20% of the base "perfect" materials, but remember your base is much higher so that actually ends up costing you more.

Is this an intended result or an unfortunate side effect?