So. I've been mainly screwing up my craft tests this evening. Way to go me.
The tests that I did get right gave scary similar results. You'll see in a bit. First my beaaauuuutiful failures:
- Craft 2000 Unity guns (We all need some unity, don't we?)
So.
Don't do that. Bad idea. My inventory wasn't happy with those 2000 guns at all. I could've simply thought about that. But. Yeah. I figured disassembling was going to take waaaaaaaay too long. (took between 5 - 10 seconds do disassemble 1 item)
Starting: 6229
Without disassembly: 7429
A bit of a useless stat, though nice to realise that items actually take up some fair space.
- Craft 50 psycho flexiweave. I feel like a psycho right now. Don't judge me. Ran out of mats half way through what is going to be my new crafting/disassembling format:
Craft 50 items, disassemble 50 items. Brilliant.
5 runs:
- Starting: 6470
- Items crafted and disassembled: 250 (duh)
- Ending: 6482
That's not too bad. Is it? only 12. (yeah it's still bad, but wait for the other results)
5 runs:
- Starting: 6470
- Crafted: 121
- Ending: 6514
Yeah that's right. I screwed up, okay?! I forgot about resources and I didn't have enough.
No matter though! I did get a little bit of data out of this. Because the bloating of the file by 12KB is nice to note. The Psycho clothing is clothing without any random modifiers. Now look what happens when you craft an item that does have random modifiers:
Crafting some more Unity guns. Same format 50/50 times 5
5 runs:
- Starting: 6229
- Crafted: 250
- Ending: 6290
5 runs:
- Starting: 6290
- Crafted: 250
- Ending: 6351
5 runs:
- Starting: 6351
- Crafted: 250
- Ending: 6411
5 runs:
- Starting: 6411
- Crafted: 250
- Ending: 6470
So. That's scary right!? Every 250 items with a single random modifier increases the file size by 60KB (sometimes off by 1, but honestly? Even if you continuously save the game the saves differ by 1KB. So I'm going to ignore that.)
I am not happy with the amount of crafting I did on wearable items without modifiers and I actually want to see if items with more modifiers change the results. We know modifiers matter, but we don't know if the amount of modifiers matter.
Now down to the question that
@pagefault404 stated: What about the fricking unique identifiers.
Well. I think... If we take the unique identifiers into account and we say that they are especially heavy on items with random modifiers, then I think my Unity test might actually be usefull.
Let's see.
(65.000/250) * 60 = 15.600
If I got this right, which I definitely may not have: if I divide 65000(total) by 250 (used in my test) we get 260. So we can fit 260 of my tests into the save file to get close the highest possible unique identifier. If I get 60KB times that by 260 we get 15.600 and divide that by 1024 just to be cool. We all want to divide by 1024. 15MB. I don't think it's something we need to worry about just yet.
Unless I am getting something wrong here. Do let me know.