Forum Home | Login

amirabiri

Join Date 03/2008

Ship upgrades - number crunching

I was thinking a lot lately about the Shields / Hull / Dmg upgrade technologies and at what point is it better to research these instead of buying more ships. In other words - comparing the numbers, at what point will you gain more from researching these technologies than you would if you used the same money to build more ships?

A few points regarding the calculation below:
- I used Illums because Advent are my favorite race and I use Illums as my main ship. I'd be happy for anyone to apply the same calculations on other ships.
- For simplicity, I will assume that Metal and Crystal are roughly equivalent, so I'll chunk them in together as "Resources".
- The shield and hull repair rate bonuses go hand in hand with the increase in total hull and shield points, therefore it's nullified and only serves to compensate for a potential loss, so I'll ignore it.
- To gain a full X% benefit you need to increase all characteristics of a ship by that factor (hull, shields, DPS).
- If you need to trade money for resources the ratio is roughly 1:4.5. If the other way around is 1:2.5. So I will average that as 1:3.5.
- There is also the issue of command points to consider - you get more hull/shield/DPS for each command point, which in turn means you don't need to advance in command tech which eats into your earnings. However since the command points are hard to factor in I have ignored them for now.

So these are the numbers in question:

	Credits	Metal	Crystal
Shields / Hull +5%	400	0	25
Shields / Hull +10%	500	0	50
Beam +5%	800	100	175
Beam +10%	900	125	200
Total	3500	225	525
Illuminator	380	60	55

So if we chuck in the resources together we get:

	Credits	Resources
Research	3500	750
Illuminator	380	115

Using the 1:3.5 Money to resources ratio we get a factor of 6125 for the research, and 780 for one Illum. So that means that the research roughly costs as much as 8 Illuminators.

Now if for a full +10% increase you gain an "extra" Illum for every 10, that means that when you have 50 Illums you "gain" 5, but if you built illums instead you would have gained 8. In other words you are better off building more ships when you have 50... The break even point is 80 ships - that's when researching the bonuses or building more ships costs you roughly the same. Beyond that point you obviously should research these technologies.

Let's instead calculate only the level 1 technologies since they are cheaper, and divide them by half since they only increase the shield/hull but not the DPS. Doing this will bring us to a total of 1800 credits and 150 resources. This is roughly 3 Illums. However if we think of the gain as half of 10% that means we gain 1 for 20 - which means the break even point is 60. Doing the same thing for the DPS only comes up to 5 illums which means that the break even point is 100...

To summerize, although this is a very rough estimation, it is quite evident that at least in early game but arguably even beyond researching these technologies is not necessarily more beneficial than building more ships. This is something worth keeping in mind.

strategy

Locked Post

Advanced Search

Watch this post

Do not email me updates for this post Email me updates for this post

Successfully updated karma reason!

Yes, good calculations, but you have to factor in Fleet Logistics. Researching improves function without increasing upkeep. I usually wait until about level 3 or sometimes 4 Fleet Logistics before I research Phase Missile technology, shields, and hulls. I sometimes wait even till end of game to improve armor, antimatter reserves, pulse weapons, and almost never get to wave upgrades. It all depends on the game.

Successfully updated karma reason!

The numbers don't lie, great post. It's problematic to follow that sense of logic however.

The shield upgrade has more value. It not only ups max shields, but ups shield regen, making ships tougher. It also doesn't just upgrade Illuminator shields, it upgrades any ship's shields that you own. Also, for advent, you must get shield regen upgrades to get to the Guardian, an invaluable addition to any advent fleet.

Health regen also ups health regen rate. Health regen's big perk is it gives EVERYTHING a health upgrade, including buildings.

The beam upgrade is a problematic assessment. It would appear that the cost of this means you need a large amount of illums to make the payoff sensible. What this misses is that by upgrading shield damage you don't have to go through shield mitigation as much. By killing the unit quicker, shield mitigation can't climb as quickly, so more of your damage is getting through quicker.

So....where is this illustrious cut-off for when to upgrade? The tier one stuff should be done before you go into tier two fleet capacity, or at least sometime during the tier two fleet capacity point.

The tier two shield upgrades should come during the tier 2 logistics upgrade.

I say the beam upgrades should come when you have 2 caps, 40 illums, and 5-8 guardians. In general, I get valuable upgrades BEFORE I consider making the fleet larger.

Just my two cents. Kudos for the number crunching

Successfully updated karma reason!

I agree with raging amish, only thing you can crunch are the Damage upgrades, and even those get tricky. 10% is not simple 10% either, it means a lot more depending on the situation. 10% HP/HP regen bonus means a lot more if you are fighting under the cover of repair stations. Shield upgrades give you more survivability if you are advent vs the other 2 factions. Not all factions have the same availability of upgrades. Some have nice tier 1 upgrades, others need to wait for tier 3 for certain hings.

Even using Illuminator as an example, you still can't really pin it. Usually upgrades done when you are producing at maximum capacity and rather not build another frigate factory, or when you don't want to upgrade your fleet supplies are a no-brainer.

Successfully updated karma reason!

While I agree that it is nigh impossible to actually measure the effect of these upgrades, many thanks for doing a great job trying. We can just take your numbers, and apply our own criteria on top of it, which about as much as I ever wanted to do in the first place!

Successfully updated karma reason!

Good points have already been made, so I'll throw out something else to think about for those of you who like to number crunch the value of upgrades.

When trying to mathematically gauge the value of upgrades where you increase several statistics of a ship by say...40%, you get a combined effectiveness that is higher than 40%. The difference is not huge at first, but it can become substantial as the overall percentages go up from baseline.

For example, if you increase the survivability of a ship by 20% and the damage output of a ship by 20%, you don't get a ship that is 40% more effective, you get one that is 44% more effective, because the stats complement each other. It isn't 1.2 + 1.2, it is 1.2 x 1.2, because staying alive 20% longer allows the 20% higher damage output to stay in effect 20% longer as well.

You could really see this if you used a big value. For example, if you triple the firepower of a ship (300%), you know it is three times as dangerous because it can do three times as much damage before it dies. And if you triple the survivability of a ship, you know it is three times as dangerous because it can do three times as much damage by living three times longer. But if you triple both the firepower and the survivability, you get a ship that is nine times as dangerous because it is dishing out triple damage for three times as long. Get it?

Successfully updated karma reason!

Yes, good calculations, but you have to factor in Fleet Logistics. Researching improves function without increasing upkeep.

Very true. Your fleet cap for 100 Illuminators is going to cost you, continually, until the end of the game. Where research is a one time thing, and usually will upgrade more then just the one ship type. The improvements will give the entire fleet ( caps, HCs, and frigs ) a bonus, including ships already built. That part is missing from those calculations.

Damn good math, though. Certainly shows there is a break even point in the quantity-quality debate, although there still seems to be a debate on where that point is..

Successfully updated karma reason!

Thank you for your comments guys.

I generally agree and mentioned this in my post that the break even point is much lower than 80 or 50, and that the most important thing to consider is fleet logistics. The purpose of this exercise is to supply us with some measurement that we can keep in the back of our minds.

One small point I would like to make however, regarding the shield and hull regen rates:

Correct me if I'm wrong but the original regen rates are fixed. i.e ships regen X points/sec. If this is correct than the argument that the increased hull and shield regen benefits us is invalid. The reason for this is that this increase in regen % only comes to compensate for a potential loss. I'll explain:

Suppose I have 10 Illums. I don't remember what the shield and hull regen rates are so lets call them REGEN. Let's also call the total number of shield and hull for an Illum HPS. So my fleet has 10 x HPS and it regenerates at a rate of 10 x REGEN.
If I build another Illum, my fleet now has 11 x HPS and regenerates at a rate of 11 x REGEN. If I only researched +10%, and they didn't come with regen bonus, I would have 11 x HPS and 10 x REGEN - in other words if you think of my fleet as one unit, it now has less REGEN because only 10 ships are regenerating at the same rate eventhough they have more hps.

So you see the increase in regen only serves to avoid a potential loss. It does not offer any benefit by itself. This is based on the premise that shield and hull regen are fixed and not relative to the total number of shield and hull points.

Successfully updated karma reason!

I would have to say that, in general, you should always tech the first fleet supply before upgrading your ships. However, depending on how agressive your enemy is, you may want to tech up your ships before getting a larger fleet. Keeping your fleet supply lower will help you build your economy, which in turn will allow you to have a bigger fleet later on. I have won many battles by attrition and not by skill. My adversary teched to the next level fleet supply, but wasn't able to replace his forces as fast as I could (just make sure to keep the pressure on, otherwise you better get a bigger fleet).

As Raging Amish also said, things like hull points and shield points benefits all ships, including capital ships. And getting an extra 10% surviability is huge when I am taking out your Advent Colonizer with phase missles. What's it worth to you to not have to replace a level 6 cap? My threshold for that would be about 10,000 credits, and perhaps more for my space egg/seige beast.

Successfully updated karma reason!

The break even point is 80 ships - that's when researching the bonuses or building more ships costs you roughly the same.

80 ships! Any thoughts?

Successfully updated karma reason!

Uh, I am not the multiplayer expert here.

But as for TEC the level 1 and 3 armor upgrades are a must to get quickly to. Why? Because armor is more important for TEC than overall hull hitpoints. Why so?

The TEC have nice repair abilites. Consider this: 60 hull hitpoints repaired a second combining lvl 3 repair bays and hoshiko repair ability (40+20). If you get more armor, the hull hitpoints are worth more. At the end of the day the TEC armor level rises to about 6 or 6.5 (added to the ship's normal armor value) if I remember correctly. As far as I remember 1 point in armor means 5% less damage when the hull gets hit. So 6-7 armor points more means that almost one third of the damage is sinked into a ship's armor not its hitpoints.

So if you get the chance to repair every single ship, it means the effective repair value is not 1 but 1:1.3-1.4 if armor is maxed. You then get at least 90 repair/second considering this effect. Even more from highly armored ships like high-level battleships, heavy cruisers and flak, not to mention late game starbases...

So long,
EEF-Prax

Successfully updated karma reason!

awuffleablehedgie

Join Date 04/2009

+60

Quoting Cykur,

reply 5

You could really see this if you used a big value. For example, if you triple the firepower of a ship (300%), you know it is three times as dangerous because it can do three times as much damage before it dies. And if you triple the survivability of a ship, you know it is three times as dangerous because it can do three times as much damage by living three times longer. But if you triple both the firepower and the survivability, you get a ship that is nine times as dangerous because it is dishing out triple damage for three times as long. Get it?

It's even more than 9 times more effective: a ship with 3x firepower and 3x HP will be killing units 3x faster which means that it only further increases its survivability since less enemy units means you are taking less damage. Plus, the more HP you have the more shield mitigation you have which increases your survivability even MORE.

Successfully updated karma reason!

I swear it is strange how people go around resurrecting old threads.... (you didn't, you just saw it now that it is live again)

My point was simply that effectiveness didn't increase in a linear progression. A ship that does 3 times as much damage may in fact be more than 3 times as effective for the reasons you mention...or it might not due to shield mitigation of the opponent hurting it more than a ship that only does 1x damage. As for survivability, more HP does not mean more mitigation per se, but it offers more opportunity to mitigate more damage overall before dying....which I'm not sure if it makes much of a difference since mitigation affects all ships across the board, unless of course the survivability of the ship includes higher mitigation levels. But I agree, when you increase power and survivability of a ship, it becomes much more effective

A great example of this is the Stargate Races mod....some of the Ancient ships are juggernauts...and when they reach level 10, they can fight entire fleets by themselves...which is kinda fun.

Successfully updated karma reason!

The time to get these upgrades is when you're sitting at an upkeep threshold. An upgrade enables you to improve your present fleet without exceeding present upkeep levels. If you have unused command slots, filling them is definitely a priority.

Successfully updated karma reason!

The time to get these upgrades is when you're sitting at an upkeep threshold

That's my feeling towards upgrades in one sentance.

Successfully updated karma reason!

Quoting Raging Amish,

reply 14

The time to get these upgrades is when you're sitting at an upkeep threshold
That's my feeling towards upgrades in one sentance.

And also: upgrades cost more than the resources listed; they also cost time. In the same way that you should never have a massive quantity of resources sitting unspent, I think you should try not to have a panel in the tech screen sitting with nothing researching on it.

If you sit and wait for your fleet to be as big as these calculations imply before upgrading, while your opponent has been upgrading consistently as they play, then you're going to be pretty far behind technologically. Important since the last upgrade in a chain tends to be a doozy. An interseting comparison would be to take two fleets, seperated in size by 1 fleet cap upgrade, and work out the cost in upgrades needed to make the small one comparable to the large, and then compare that to the resources generated over time by the disparity in fleet tax.

Successfully updated karma reason!

80 Illums is 480 fleet (or command) points. That's 3 1/2 fleet upgrades! (Tier 3, Normal fleet settings).

I brought this old post back up because: its about the only good quantification of upgrade costs that I've seen, and we were discussing worthwhile techs elsewhere. (Here is a link to another similar discussion).

Quoting Howdidudothat,

reply 8

I would have to say that, in general, you should always tech the first fleet supply before upgrading your ships. However...

I'd generally have to agree with How (& Sith in reply #1), to wait. I stand by my observation that most players err by buying the "measly" 5% upgrades too soon. Remember 5% is only 1/20... so at a minimum it would take 20 ships to offset the opportunity cost. Another factor to consider: the upgrade costs get substantially more expensive as you move up the tech tree. According to my observations, a few 'pros', like JJ, rarely buy the upgrades - until later.

Successfully updated karma reason!

Some thoughts:

First: Increasing your maximum health (or whatever) without increasing your regeneration rate only means it will take longer to recover a given loss. The amount of loss is a function of your opponent's capabilities. If you can survive a battle because you have more health, and then have time to regenerate at the slower rate, then increased regeneration has no meaningful effect.

Second: It is not only the choice to add additional ships to an existing fleet or upgrade the fleet but also it affects the number of replacements that will be required to keep the current fleet levels. This goes in hand with upgrading instead of adding additional fleet capacity.

Third: Time needs to be considered in this form of analysis - how long to complete all the research compared to just building some more fleet. Researching a command level and building more ships (possibly with mulitple factories) is likely to be much quicker than doing all the research.

If we wish to expand this analysis we should focus strictly at command thresholds and techology levels (i.e., all level 1, 2, 3, etc... upgrades). Since there are more command levels then upgrade levels the goal would be to determine which command/technology levels should be paired together for the fastest and least expensive upgrade path. Can also add in capital ship crews into the analysis to help eat some of the fleet capacity. The goal wouldn't be so much to find an optimal path but instead should help point out where trade-offs are being made (say between short-term fleet size increase and a longer time having the increased maintenance and decreased stats from delayed research).

I may play with some of this as time permits but figured I'd throw it out there for general consideration and feedback.

Successfully updated karma reason!

Welcome Guest! Please take the time to register with us.

There are many great features available to you once you register, including:

Richer content, access to many features that are disabled for guests like commenting and posting on the forums.
Access to a great community, with a massive database of many, many areas of interest.
Access to contests & subscription offers like exclusive emails.
It's simple, and FREE!

View all recent posts

Mark all posts as read Delete cookies created by the forum Return to Top

Stardock Forums v Server Load Time: Page Render Time: