- Jul 30, 2017
- 833
- 1,086
That error is strange. It says: "Can't open 03_ap_144_31b.png"
I don't have that error, but when I checked the data and I also don't have said file.
My version is the same as the one I sent you.
- Jun 11, 2019
- 4
- 1
- Apr 29, 2020
- 1,685
- 2,813
False positive. You can look at the files in the .7z without opening it. The .xp3 can even be unpacked. It's text files and a PDF. If you could send the actual Defender log with what it says is in there that would be helpful.
Yeah, for other people to see what I updated in the DMs, Madeline got some new CG. I've been getting updated data.xp3 files, but am still running an old image_0.xp3 and image_1.xp3, which is where they are storing the HCG for this game. The scenario files in data.xp3 were updated to reference the new HCG, but for people running a pirated copy and depending on my patches, those HCG don't exist and give that error. I'll have to figure out a way to audit those kind of stealth changes for future patches, but that's quite doable.
So do you plan update cg files? or edit data so the new cg don't appear in game?
Also could you update translation for these lines in future path? I know they are telling something like higher levels better items. Still it would be nice to translate it.
- Apr 29, 2020
- 1,685
- 2,813
Yes, I will fix it properly in 0.8.
And huh. I wonder how that snuck through. I will figure it out.
- Aug/18/2021Correction of typographical errorsBug fixes
- ・旧無印データ利用時に旧無印削除・登録解除した場合の不具合修正
・追加パッチ結合状態追加データ利用時の不具合を修正
-Fixed a bug when using additional patch binding status additional data
- Jul 30, 2017
- 833
- 1,086
Can anyone explain how inheritance stats work in this game? I read reference sheet but still i don't get it.
For test i used Valentina HP:*, Atk: *, Def: **, Mgc: * my target was goblin pld
Gen: 25, CMD:57, HP:111, Atk:111, Def:122, Mgc:62
Gen: 26, CMD:62, HP:111, Atk:103, Def:114, Mgc:62
Gen: 27, CMD:70, HP:106, Atk:95, Def:124, Mgc:61
Gen: 28, CMD:71, HP:113, Atk:104, Def:110, Mgc:63
Gen 24 have been buffed by food and have different seedbed so i don't count him. I didn't use food for gen 25-28. So why my goblins become weaker with next gen not stronger? CMD is growing sure but the goblin stats are becoming weaker.
For test i used Valentina HP:*, Atk: *, Def: **, Mgc: * my target was goblin pld
Gen: 25, CMD:57, HP:111, Atk:111, Def:122, Mgc:62
Gen: 26, CMD:62, HP:111, Atk:103, Def:114, Mgc:62
Gen: 27, CMD:70, HP:106, Atk:95, Def:124, Mgc:61
Gen: 28, CMD:71, HP:113, Atk:104, Def:110, Mgc:63
Gen 24 have been buffed by food and have different seedbed so i don't count him. I didn't use food for gen 25-28. So why my goblins become weaker with next gen not stronger? CMD is growing sure but the goblin stats are becoming weaker.
- Jul 30, 2017
- 833
- 1,086
All stats for goblins born in the same brood are randomized (unlike the first game, which only CMD is randomized) The difference is hard to see in normal goblins, but if you have a brood where there is 2 mutant of the same types, it is more obvious. For example, 2 champions born in the same brood can have their ATK varied wildly, up to ~20 ATK difference.
Rest assured that the average stats are definitely going up.
- Apr 29, 2020
- 1,685
- 2,813
CMD growing was just a fluke. There is a set minimum and maximum for each goblin type, you just got lucky.
Not sure where the variability is in the inheritance. Tried to find it but couldn't.
- Jul 30, 2017
- 833
- 1,086
Ok i get it so even if we got lower stat we in longer run become stronger. And food give us more bonus to stats because goblin will inherited part of it.
I did 2 runs in normal mode (i didn't import my save from GB1 i didn't know how to do it then and the were around 50 gen so no big lost) and i make good goblin's i think
So i try 1 star level. I clear tower in speed run. I capture all the girls without any problem even Amelia, Serena and Eleanor fall easily to my group. And then i hit wall. It's called Florencia. She hit me with her army like a train. And mop all my goblins without any problem. The difference with her and other girls is like heaven and earth. No idea why creators make her that much op compared to others. I change pld with another who have plague of rust (3x def for all) but i didn't see any difference in reducing damage. No idea if that skill even works properly. So the only idea to defeat her is probably higher gen and tower. Or anyone have better idea?. She just to strong (or rather her army) to this level a little broken in my opinion.
I did 2 runs in normal mode (i didn't import my save from GB1 i didn't know how to do it then and the were around 50 gen so no big lost) and i make good goblin's i think
So i try 1 star level. I clear tower in speed run. I capture all the girls without any problem even Amelia, Serena and Eleanor fall easily to my group. And then i hit wall. It's called Florencia. She hit me with her army like a train. And mop all my goblins without any problem. The difference with her and other girls is like heaven and earth. No idea why creators make her that much op compared to others. I change pld with another who have plague of rust (3x def for all) but i didn't see any difference in reducing damage. No idea if that skill even works properly. So the only idea to defeat her is probably higher gen and tower. Or anyone have better idea?. She just to strong (or rather her army) to this level a little broken in my opinion.
- Apr 29, 2020
- 1,685
- 2,813
Okay, found the code and broke it down to explain how the variability works.
Here is the raw code:
In English:
Code Block 1: Just sets it as the base inherited stats from the goblin parent.
Code Block 2: Adds the seedbed inheritance +/-1 from their stated value. This number is passed on to the next generation after this one. The actual stats are calculated next.
Code Block 3: Sets heredity[PRM_AT]. It starts at 0, so this is just assigning the variable.
Code Block 4: Sets maxParentHeredity, which starts really boosting your goblin's stats after 10 generations, and really nerfs them before that because of what happens next.
Code Block 5: The most recent parent gets an additional boost, multiplying their power by maxParentHeredity (max 10), again +/- 1 of their power.
Code Block 6: The final step. The base value for attack is multiplied by the heredity[PRM_AT] and some other factors.
So for example, a basic goblin has an attack of 4. We'll assume his father has +20 Atk heredity so far after 10 generations, and is being bred with Valentina who has one star in Atk.
So a goblin is conceived with between 7 and 10 ATK. Those numbers get much more varied if a Champion is being conceived, who has a base attack stat of 32:
Then any additional stats gained from food, as well as from Tower clears or Ritual clears is added on top of that. This variability can get quite high, especially if you are using stronger seedbeds. It is also not linear, you don't just breed a +1 Atk seedbed and get goblins with +1 ATK, unlike in Goblin Burrow.
One thing I find concerning is that you could in theory get a -1 by breeding seedbeds with a 0 star in a stat, since there is nothing in the intrandom function that prevents it from returning negative values, unless Kirikiri's basic engine treats ints as positive values only. Seems shitty that you could accidentally make your goblins worse with some stellarly bad luck. Might mod that, tbh.
EDIT: Math is hard
Here is the raw code:
Code:
goblinData.atHeredity = f.game.goblin[parentGoblinNo].atHeredity;
...
goblinData.atHeredity += intrandom(tf.charaData[parentHeroinCid].heredity[PRM_AT] - 1, tf.charaData[parentHeroinCid].heredity[PRM_AT] + 1);
...
heredity[PRM_AT] += goblinData.atHeredity;
...
var maxParentHeredity = goblinData.generation - 1;
if(maxParentHeredity > 10){
maxParentHeredity = 10;
...
heredity[PRM_AT] += intrandom(tf.charaData[parentHeroinCid].heredity[PRM_AT] - 1, tf.charaData[parentHeroinCid].heredity[PRM_AT] + 1) * maxParentHeredity;
...
goblinData.at = tf.unitData[goblinData.unitID].at * (1 + heredity[PRM_AT] * 0.1 * 2 / 6);Code Block 1: Just sets it as the base inherited stats from the goblin parent.
Code Block 2: Adds the seedbed inheritance +/-1 from their stated value. This number is passed on to the next generation after this one. The actual stats are calculated next.
Code Block 3: Sets heredity[PRM_AT]. It starts at 0, so this is just assigning the variable.
Code Block 4: Sets maxParentHeredity, which starts really boosting your goblin's stats after 10 generations, and really nerfs them before that because of what happens next.
Code Block 5: The most recent parent gets an additional boost, multiplying their power by maxParentHeredity (max 10), again +/- 1 of their power.
Code Block 6: The final step. The base value for attack is multiplied by the heredity[PRM_AT] and some other factors.
So for example, a basic goblin has an attack of 4. We'll assume his father has +20 Atk heredity so far after 10 generations, and is being bred with Valentina who has one star in Atk.
Code:
goblinData.atHeredity = 20
goblinData.atHeredity = 20 + intrandom(1 - 1, 1 + 1) = (20, 21, 22)
heredity[PRM_AT] = 0 + (20, 21, 22)
heredity[PRM_AT] = (20 : 22) + intrandom(1 - 1, 1 + 1) * 10 = (20, 21, 22) + (0, 10, 20) = (20, 21, 22, 30, 31, 32, 40, 41, 42)
goblinData.at = 4 * (1 + (20 : 42) * 0.1 * 2 / 6) = 4 * (1 + (0.67 : 1.4) = 4 * (1.67 : 2.4) = 6.67, 6.80, 6.92, 8, 8.13, 8.27, 9.33, 9.57, 9.60
Code:
goblinData.at = 32 * (1 + (20 : 42) * 0.1 * 2 / 6) = 32 * (1 + (0.67 : 1.4)) = 32 * (1.67 : 2.4) = 53.33, 54.40, 55.36, 64.00, 65.07, 66.13, 74.67, 75.73, 76.8One thing I find concerning is that you could in theory get a -1 by breeding seedbeds with a 0 star in a stat, since there is nothing in the intrandom function that prevents it from returning negative values, unless Kirikiri's basic engine treats ints as positive values only. Seems shitty that you could accidentally make your goblins worse with some stellarly bad luck. Might mod that, tbh.
EDIT: Math is hard
Last edited:
- Apr 29, 2020
- 1,685
- 2,813
TRANSLATION UPDATE
Also, as a fun fact, I made a script to check for missing HCG that I can run on future updates. Yerum has some HCG planned for a future update I think. There are text cues for more images, but they are commented out.
You must be registered to see the links
is now liveYou don't have permission to view the spoiler content.
Log in or register now.
Also, as a fun fact, I made a script to check for missing HCG that I can run on future updates. Yerum has some HCG planned for a future update I think. There are text cues for more images, but they are commented out.
You must be registered to see the links
1It was virus now it is okay
You must be registered to see the links
virusMy be I must wait something
What operational system do you use and antivirus?
My Windows 10 x64 Professional 21H1 OS build 19043.1165
Windows Feature Experience Pack 120.2212.3530.0
antivirus Microsoft defender
If
You must be registered to see the links
work correctly now, whats a problem with
You must be registered to see the links
.
Last edited:
- Jul 30, 2017
- 833
- 1,086