[request] sacrificial knife

[Stan`]

1 hour ago, MrLux said:

Okay, I'll rethink the silhouettes.

How do I instantiate a specific variant in the atlas ?

Unfortunately not yet. The easiest way is to disable the others or to spam the actor viewer by going back and forth on two objects

[vladislavbelov]

20 minutes ago, Nescio said:

For comparison, the falcata.xml uses 5 variant meshes, gladus.xml 6, xiphos.xml 5, and kush_nubian_mace.xml 15, and I daresay their shapes are more similar than the 5 knives in @MrLux's screenshot. Besides, those other weapons are used by only a few actors, whereas the knife is used by the slaughter attack animation of all infantry and female citizens.

I'd prefer to have low number of knifes for all civs

21 minutes ago, Nescio said:

I don't know how much it affects performance, though I wonder whether it's significant in this case.

The problem is that you can't notice the difference by eye on modern hardware like 1050/1660+, but we have a lot of meshes each of them increases number of GL state changes which decreases performance (so in total they affect performance significantly). Also more different meshes makes instancing less applicable.

[wowgetoffyourcellphone]

2 hours ago, Nescio said:

For comparison, the falcata.xml uses 5 variant meshes, gladus.xml 6, xiphos.xml 5, and kush_nubian_mace.xml 15, and I daresay their shapes are more similar than the 5 knives in @MrLux's screenshot. Besides, those other weapons are used by only a few actors, whereas the knife is used by the slaughter attack animation of all infantry and female citizens.

I don't know how much it affects performance, though I wonder whether it's significant in this case.

Honestly, at most each of those weapons should only need about 3 variants for proper visual variation. 

[Nescio]

14 hours ago, vladislavbelov said:

The problem is that you can't notice the difference by eye on modern hardware like 1050/1660+, but we have a lot of meshes each of them increases number of GL state changes which decreases performance (so in total they affect performance significantly). Also more different meshes makes instancing less applicable.

Do meshes that are not used affect perfomance? E.g. are kush assets loaded into a mace vs pers match? Also, surely it matters how often something is drawn? E.g. swordsmen have swords most of the time, whereas knives appear only when gathering meat?

Anyway, more important than the number of variants is that the fact the dagger is replaced by a knife, and even that's only a minor detail visible when zooming in and paying attention (yet still an improvement, hence this thread).

[vladislavbelov]

On 27/09/2020 at 11:32, Nescio said:

Do meshes that are not used affect perfomance?

If they're not used during all game - yes, they shouldn't affect performance. But if they were placed (or even drawn) then they might. Because you need to load all needed resources (cost memory). You need to draw at least one frame then you need to make some state changes that also affects performance of the frame.

[Nescio]

On 11/10/2020 at 16:08, vladislavbelov said:

If they're not used during all game - yes, they shouldn't affect performance. But if they were placed (or even drawn) then they might. Because you need to load all needed resources (cost memory). You need to draw at least one frame then you need to make some state changes that also affects performance of the frame.

Gathering meat is an activity far fewer units are typically involved in than farming or wood cutting. If a player assigns only a single unit to slaughter animals, then only one variant has to be drawn, right?

Also, as you pointed out, the prop is rather small, surely that affects performance too?

[vladislavbelov]

2 hours ago, Nescio said:

Gathering meat is an activity far fewer units are typically involved in than farming or wood cutting. If a player assigns only a single unit to slaughter animals, then only one variant has to be drawn, right?

Yes, but people use not the only one unit. There might be 2 or more.

2 hours ago, Nescio said:

Also, as you pointed out, the prop is rather small, surely that affects performance too?

Unfortunately (especially on modern HW) in our game just a number of drawcalls matters (doesn't matter how big a drawcall is, "big" in reasonable bounds), the number of drawcalls and related state changes is the bottleneck now.

[Nescio]

2 hours ago, vladislavbelov said:

Yes, but people use not the only one unit. There might be 2 or more.

Unfortunately (especially on modern HW) in our game just a number of drawcalls matters (doesn't matter how big a drawcall is, "big" in reasonable bounds), the number of drawcalls and related state changes is the bottleneck now.

Thanks for the clarification, it's good to know what the bottleneck is. (What is a “drawcall”? Drawing a single object?) How much worse is having e.g. five instead of three variants of a knife for the game as a whole? And would deleting variants from props and actors speed up the game significantly?

On 27/09/2020 at 10:32, Nescio said:

Anyway, more important than the number of variants is that the fact the dagger is replaced by a knife, and even that's only a minor detail visible when zooming in and paying attention (yet still an improvement, hence this thread).

 

[vladislavbelov]

On 13/10/2020 at 11:32 PM, Nescio said:

Thanks for the clarification, it's good to know what the bottleneck is. (What is a “drawcall”? Drawing a single object?) How much worse is having e.g. five instead of three variants of a knife for the game as a whole? And would deleting variants from props and actors speed up the game significantly?

Draw call is a single command to draw a mesh (usually) - a part of some object.

If it's not rendered then it costs only memory. But if you see all these knife types, then it might have an impact. Yes, 3 or 5 knifes won't make the performance noticeably worse. But when you have a whole map of different variations it might become noticeable.

How it speedups you might check with bb (entityvariants) and wraitii (instancing) patches.

[Nescio]

11 hours ago, vladislavbelov said:

If it's not rendered then it costs only memory. But if you see all these knife types, then it might have an impact. Yes, 3 or 5 knifes won't make the performance noticeably worse. But when you have a whole map of different variations it might become noticeable.

Out of curiosity, is having say three knives used by all civs worse than each civ one knife only they use?

[vladislavbelov]

5 minutes ago, Nescio said:

Out of curiosity, is having say three knives used by all civs worse than each civ one knife only they use?

It doesn't matter how many cives you haves, the number of different meshes (their variants) per frame matters.