Actual Damage Spreadsheet

[Ladadoos]

16 minutes ago, SilenceKiller99 said:

Nice! Will you also change the description on AllCards?

I could, but the thing is, I am not 100% sure what the speed is. As you just saw by Stormsinger, the speed can also have decimal numbers, meaning that the speed could be also be 1.8 or 1.9. Not only that, but the description in AllBFCards wouldn't be the same as in the game, which might confuse people... I guess I could change it but I'm not sure, what do you think ?

 

@Chimaka( ͡° ͜ʖ ͡°) You're indeed correct, the description is wrong. It shoots every 3 seconds and knocks medium units as well (according to ingame description).

Edited February 21, 2016 by Ladadoos

[SilenceKiller99]

6 minutes ago, Chimaka( ͡° ͜ʖ ͡°) said:

Isn't the skycatcher's attack rate too high? 1 second atk rate isn't true, just watching the video and seing 2 sec(maybe 2,5 sec) between attacks

It definitely doesn't look like 1 second yeah xD If someone gives me the correct (tested) rate of fire, I will recalculate and change the value.

5 minutes ago, Ladadoos said:

I could, but the thing is, I am not 100% sure what the speed is. As you just saw by Stormsinger, the speed can also have decimal numbers, meaning that the speed could be also be 1.8 or 1.9. Not only that, but the description in AllBFCards wouldn't be the same as in the game, which might confuse people... I guess I could change it but I'm not sure, what do you think ?

You would have to search the game code to find the exact attack speed of every unit. Maybe someone is able to do this.

For know I think you should leave the description how it is because we don't know the exact value, but you could add a warning at the description about this.

[Ladadoos]

7 minutes ago, SilenceKiller99 said:

It definitely doesn't look like 1 second yeah xD If someone gives me the correct (tested) rate of fire, I will recalculate and change the value.

You would have to search the game code to find the exact attack speed of every unit. Maybe someone is able to do this.

For know I think you should leave the description how it is because we don't know the exact value, but you could add a warning at the description about this.

I will have a .txt where I keep track these kind of changes (how fast units actually attack) I make so that if anything I can easily remove/change it. I've made some slight adjustments to Commando's description. That alright ?

[Ultrakool]

25 minutes ago, Chimaka( ͡° ͜ʖ ͡°) said:

Isn't the skycatcher's attack rate too high? 1 second atk rate isn't true, just watching the video and seing 2 sec(maybe 2,5 sec) between attacks, and also he knockbacks M units, Proof ->

https://youtu.be/1vwkpwmUshU?t=11m41s

Look right to anim of Skycatcher who fires into the BanditStalker. There is 1 sec for the first atk, but between atks there are 2 or 3 sec.

*Have Just counted right now. His atk is 4333, if we use correct attack time - 3 sec - like the Giant Wyrm's. So this number is near as on the card 4333 ~~~ 4660

Also LSS too legit...

indeed.

 

[SilenceKiller99]

Do all the ingame descriptions are 100% correct? Or would it be worth to test some of these?

[Chimaka( ͡° ͜ʖ ͡°)]

Also to clear up the amount of LSS, I've watched a vid. The LSS fires 2(!only) rockets with 2(?) sec period. This is hard to count, but let's try - em... so we know that LSS fires only 2 rockets on each side of ship. There 4 pairs of them. In 20 sec he fires 10 rockets. Summing their dmg we have 10*((115+140)/2)=1275. I can't say that this is right math, but try someone too.

[ImaginaryNumb3r]

1 hour ago, Eirias said:

@Mental Omega @SilenceKiller99

Might I suggest a U3 spreadsheet in addition? U3 is generally more important than U0.

My spreadshot has both, U0 and U3

And @SilenceKiller99 the attack value of the card increases too. You can see that by promo cards which have a different attack value than their normal counterparts (promo cards are U3 to begin with).
 

[Ultrakool]

2 minutes ago, SilenceKiller99 said:

Do all the ingame descriptions are 100% correct? Or would it be worth to test some of these?

I'm confident they are 100% correct. I've tested quite alot and havent found any anomalies so far.

[Ladadoos]

1 minute ago, SilenceKiller99 said:

Do all the ingame descriptions are 100% correct? Or would it be worth to test some of these?

Well, it's pretty clear that not all descriptions are correct. I asked Kelvin and there might be a way to view the code which has the attack speeds, so I wouldn't test anything right now (not that you can, but talking to other Alpha members). 

 

1 minute ago, Ultrakool said:

I'm confident they are 100% correct. I've tested quite alot and havent found any anomalies so far.

How would you explain the 1.8-2.0 sec attack rate of commando's ?

[Chimaka( ͡° ͜ʖ ͡°)]

1 minute ago, Ultrakool said:

I'm confident they are 100% correct. I've tested quite alot and havent found any anomalies so far.

Test out LSS please, the 11k can't be, like seriously.

[ImaginaryNumb3r]

1 minute ago, Ultrakool said:

I'm confident they are 100% correct. I've tested quite alot and havent found any anomalies so far.

Maybe not 100% but 99%. Just look at Commandos, as Ladadoos told us they actually shoot about every 2 seconds (the ingame description suggest every second).

Ladadoos also did me a favour some time ago and tested some other cards (like Gunner) and from that it really looks like you can trust that information in general.
 

[Eirias]

9 minutes ago, Mental Omega said:

My spreadshot has both, U0 and U3

And @SilenceKiller99 the attack value of the card increases too. You can see that by promo cards which have a different attack value than their normal counterparts (promo cards are U3 to begin with).
 

Oh, I see now. The numbers on the right. Sorry.

What are the 3 colored columns?

[SilenceKiller99]

7 minutes ago, Eirias said:

What are the 3 colored columns?

@Mental Omega

I was wondering that too. I think they are ratio's, by looking at the formula that is used. 

[ImaginaryNumb3r]

10 minutes ago, Eirias said:

Oh, I see now. The numbers on the right. Sorry.

What are the 3 colored columns?

Awww, I knew I forgot something.

The red one basically is the attack value per power, while the green one is HP per power. The blue one is general combat effectiveness per power (or how much bang you get for your buck).

I found this extremely helpful to find out which units are better in a stat by stat comparison. For example Bandit Sorceress has 4.56, while DA have 6.91. That makes it easy to see how difficult it is for her to effectively counter DA. But this also allows you to compare stats of units which have cost different amounts of power.

 

Edited February 21, 2016 by Mental Omega

[Eirias]

16 minutes ago, Mental Omega said:

Awww, I knew I forgot something.

The red one basically is the attack value per power, while the green one is HP per power. The blue one is general combat effectiveness per power (or how much bang you get for your buck).

I found this extremely helpful to find out which units are better in a stat by stat comparison. For example Bandit Sorceress has 4.56, while DA have 6.91. That makes it easy to see how difficult it is for her to effectively counter DA. But this also allows you to compare stats of units which have cost different amounts of power.

 

Oh, that's nice. I think more labels might clear up the confusion (btw, you know you can fix the top row to stay there, right?)

[ImaginaryNumb3r]

9 minutes ago, Eirias said:

Oh, that's nice. I think more labels might clear up the confusion (btw, you know you can fix the top row to stay there, right?)

I was pretty sure I made the labels but apparently I forgot :P. I guess I've worked with it so many times that I became blind to the details ^^

It was a rushed release where my main intent was just to provide the raw stats so SilenceKiller (and everybody else) has a reference sheet and doesnt need to make everything on their own.

Regarding fixing that top row I know that you can do that but I forgot the way you do that... could you tell me how?

 

Edited February 21, 2016 by Mental Omega

[Eirias]

5 minutes ago, Mental Omega said:

I was pretty sure I made the labels. I've worked with it so many times that I became blind to the details ^^

It was a rushed release where my main intent was just to provide the raw stats so SilenceKiller (and everybody else) has a reference sheet and doesnt need to make everything on their own.

Regarding fixing that top row I know that you can do that but I forgot the way you do that... could you tell me how?
 

Select a the row underneath the ones you want to fix. Go to view-->freeze panes-->freeze panes.

This will freeze all the rows above the cell you selected. For instance, if you select the Shadow cell, you will freeze the first 2 rows.

For your general effectiveness, did you do (damage*health)/(power cost*1000)? I'm not sure "general effectiveness" (even just in terms of stats) is that simple...

[ImaginaryNumb3r]

19 minutes ago, Eirias said:

Select a the row underneath the ones you want to fix. Go to view-->freeze panes-->freeze panes.

This will freeze all the rows above the cell you selected. For instance, if you select the Shadow cell, you will freeze the first 2 rows.

For your general effectiveness, did you do (damage*health)/(power cost*1000)? I'm not sure "general effectiveness" (even just in terms of stats) is that simple...

The formula for power efficency is HP * ATK / power and this is a univeral formula which works for basically all games. You need to think about it this way, a unit will deal damage as long as its alive. HP basically equats to time, as a unit with more health will generally stay alive for a longer time.

In the excel sheet efficency is just calculated ATK ratio * HP ratio, but that is justified since the ratios are: ATK / root (power). When multiplying the ratios you square the root, subsequently restoring the original power cost. As for efficency I also divide this by a constant factor so I get a lower number that is easier to compare.

For example if a unit with 200 ATK and 800 HP fights a unit with 800 HP and 200 ATK, both will (or should) die. In that sense ATK is interchangeable with HP. However, the more health a unit has the longer a fight will last and the more macro oriented a fight becomes.

Of course that does not work with units that makes heavy use of passive or active abilities. Ashbone Pyro, Shadow Mage, Nox Trooper etc. are hardly comparable. But this is no complete science and you simply don't need to have exact values. It also doesnt matter if you are a few percent off since in reality there are many other factors which pay a huge rule (like 50% counter damage, CC, abilities, knockback, spells etc.).

This is just supposed be a simple unit by unit comparison. For example, if you would want to compare Silverwind Lances (80e )with Bandit Lancers (60e) . The problem here is that they have require different amounts of power, so you can't compare them directly.

So, if 3 Silverwind Lances (240e) would attack 4 Bandit Lancers (240)e the Silverwinds should actually win (for now we disregard the fact that Bandit Lancers have life-steal). I can say that simply because Silverwinds have a much higher efficiency value with 29,4 whereas Bandit Lancers only have 21,23. As I said, other factors also come into play but from a mere mathematical standpoint without context this is as far as it gets and it works.

As I've said many times already, I am modding a different RTS game and I have an excel sheet with all my units which helps me balancing the game.

This is perfect to sort out whether a unit is (too) good or bad. Simply by the efficency we can conclude that units like Bandit Sorceress or Lost Priest are abysmal in combat.

Yes, I also did divide by 1000 but this is a simple constant which is not related to the formula per se and done for every unit. If you really do calculate ATK * HP / power you get quite a big number.
On the last sheet you can just replace the factors by 1 and get the original results if you wish.
 

[Eirias]

12 minutes ago, Mental Omega said:

The formula for power efficency is HP * ATK / power and this is a univeral formula which works for basically all games. You need to think about it this way, a unit will deal damage as long as its alive. HP basically equats to time, as a unit with more health will generally stay alive for a longer time.

 

I understand this. I just mean that there are very few BF units for which this calculation is useful. Unit-on-unit fighting is rare outside of t1, because both parties usually have different objectives. Usually one player wants to spawn units that kill other units (and counters become very important here) while the other player is trying to do something else, like gain ground presence or drop a well. I  can't think of a single scenario where 3 silverwind lancers and 4 bandit lancers would be fighting each other. There are also a ton of abilities that matter much more than the real stats, like the firedancer range and siege.

 

Also, for many units, attack or defense is more important. For instance, skyfire drake has a very low efficiency because it has high attack but low health. For skyfire drakes, especially since many things can't hit them, attack is probably 2 or 3 times more important than health.

 

And even in the most useful scenario: for instance, comparing burrowers to vileblood to firestalker. Sure, you can (mostly) put firestalker out of contention because of its low efficiency, but vileblood and burrowers aren't easy to compare, and they are both used for the same purpose. Burrowers are better because they have swift and--costing less power-they can hit multiple areas at once. That said, vileblood has advantages in things like 2v2 because buffs work better on him.

 

I'm not saying efficiency is meaningless, but there are so many factors that are much more important. I just worry that some inexperienced player might see these efficiency values and make deckbuilding decisions from them, which would be a very bad idea.

[ImaginaryNumb3r]

Of course there are more important factors, especially for a game like Battleforge that is mainly about microing unit, spells and abilities effectively. I was very well aware that the scenario with the Lancers fighting each other was a forced example. I wanted to point out the fact that, despite equal resources being invested one unit can come out more beneficially than another.

Also, as for comparing effiency in a fight that does not really make sense to begin with, as the player which makes better use of counters and spells normally gets an edge to begin with. So I grant you that.
However, imagine that you are a shadow splash and just want a beefy XL with good stats you could choose between Lost Grigori and Soulhunter. Both have their unique abilities but let's assume you just want to play pve and want a unit with good stats.

In this example you can't just compare them stats by stats because Soulhunter is 30% more expensive. To get a meaningful comparison you need to compare them on a per power basis and this is where the efficency comes into play. Soulhunter: 69.23, Lost Grigori: 52,50. Done.

This is not so much a problem for pure element players who normally only have one unit that fulfills a role to begin with, but the more choices you have the more interesting possibilities you could try to make work. Sure, often some units will have different key abilities like swift and you can't directly compare Vileblood and Burrower.

However, imagine Vileblood and Burrower did have similar efficency. It would show more clearly than anything else that Vileblood just is no valid choice in pvp and you won't even need to play to say that for sure. It's the ame story for many T4 units which are simply inferior to the most popular choices. There is a reason why nobody sees Void Maw...

Actually, the only reason I did this spreadsheet was to have an overview of the efficency in general. It helps a lot when you see the game from a conceptual approach and is a big help for balancing. I like participating in the daily card dicsussion and having a quick reference sheet helps a lot. It provides a quick overview why certain key units are seen as very strong while others are too so weak they are never used. Sure, some units are known to be bad (did I mention Bandit Sorceress?) but knowing it and having evidence that a unit has terrible stats is a difference.

But in the end, I guess it is just a personal matter of taste. I like knowing the guts of a game and doing number crunching.
 

[Ultrakool]

On ‎2‎/‎21‎/‎2016 at 10:25 PM, Chimaka( ͡° ͜ʖ ͡°) said:

Test out LSS please, the 11k can't be, like seriously.

By my calculations, it indeed isn't. The way I did the testing was. I would start the timer(20s) at the time LSS fired its first shot, and counted the number of turrets which would shoot out at a certain enemy(Only 1 Enemy). I repeated this 3 times, and it seems like when the timer hits zero, the average number of turrets shot out would be 20.

1 turret= 80- 100 damage, so we take the value of 1 turret= 90. The card description now becomes irrelevant since we timed 20s, however, it has to be wrong, and should read "every 2 seconds, 2 turrets shoot out ......." instead of "every 2 seconds, 8 turrets shoot out ...."

On to damage calculation: 20 x 90= 1800 damage per 20s, actually much less than the 2960 displayed on the card. It is a real mystery how they came to that number, but apparently LSS isn't as OP as the description makes it out to be. 

[Ladadoos]

4 minutes ago, Ultrakool said:

1 turret= 80- 100 damage, so we take the value of 1 turret= 90. The card description now becomes irrelevant since we timed 20s, however, it has to be wrong, and should read "every 2 seconds, 2 turrets shoot out ......." instead of "every 2 seconds, 8 turrets shoot out ...."

Did you keep in mind that the ship shoots a lot faster when there are more enemies, basically doing a lot more damage if there are more enemies to be shot at ?

[Ultrakool]

Just now, Ladadoos said:

Did you keep in mind that the ship shoots a lot faster when there are more enemies, basically doing a lot more damage if there are more enemies to be shot at ?

Really? Haven't really tested that yet. But if so, that would be another thing that would belong in the "hidden info database" from @MrXLink

[Ladadoos]

Just now, Ultrakool said:

Really? Haven't really tested that yet. But if so, that would be another thing that would belong in the "hidden info database" from @MrXLink

If there are more enemies the ship uses more of it's turrets to shoot. You can test it yourself.

[Ultrakool]

8 minutes ago, Ladadoos said:

If there are more enemies the ship uses more of it's turrets to shoot. You can test it yourself.

That's true, I counted ~30 by 20s, the interesting thing about LSS is that the closer it gets to a swarm of enemies, the more turrets fire, but at the same time, it gets really hard to count exactly how many are dispensed per 20s.