Compileable version of Alien Breed 3D II - The Killing Grounds

[abu_the_monkey]

Originally Posted by Karlos

I just added an experimental crosshair and have discovered something I suspected previously: the weapon aim is significantly exaggerated relative to the view. That is to say, as you look up, your shots go further and further above the crosshair, and as you look down, they go further and further below it.

I can't decide if this is a bug or not. It could be intentional in order to make the most use of vertical aim when you only have a limited amount of look up and down space.

is that in both small and full screen?

maybe it relates to the max/min look values I was looking into if it is ok in small but off in full screen?

[Karlos]

Actually... Maybe it is a bug after all. In small screen, shots aim towards the centre unless distracted by an enemy target.

[Karlos]

It's even weirder than you can imagine. Firing a rocket in fullscreen with the overshot vertical aim, corrects itself if you switch to 2/3 screen while it's in flight. Equally a correctly aimed rocket in 2/3 will suddenly be on a different trajectory when you switch to fullscreen.

That's nuts! Will raise an issue for that.

[abu_the_monkey]

it sounds like I will have to revisit the max/min look then

is there a version available with the experimental crosshair?

[Karlos]

Originally Posted by abu_the_monkey

it sounds like I will have to revisit the max/min look then

is there a version available with the experimental crosshair?

I'll push it in a few. Don't get excited, it's just a single dot. And you can't turn it off lol.

[abu_the_monkey]

that's cool

[vroom6sri]

Originally Posted by abu_the_monkey

in The Zone!

Fang Q!

So the results from the original are woeful at best. Replicating the same spinning around tests I saw these results in V1.1 of the WHDload build in full screen:

INSIDE FIRST ROOM (Level A)
3-4 FPS (5 in 2/3 mode)

IN FRONT OF THE BIG RED DOOR
3 FPS (6-7 in 2/3 mode)

Incidentally using the reduced quality renderer make no difference to the FPS.

[Karlos]

Pushed to misc/experiments branch

[Karlos]

Originally Posted by vroom6sri

Fang Q!

So the results from the original are woeful at best. Replicating the same spinning around tests I saw these results in V1.1 of the WHDload build in full screen:

INSIDE FIRST ROOM (Level A)
3-4 FPS (5 in 2/3 mode)

IN FRONT OF THE BIG RED DOOR
3 FPS (6-7 in 2/3 mode)

Incidentally using the reduced quality renderer make no difference to the FPS.

I think the whdload version may be faster than the team17 version too.

[Karlos]

@abu

If there is any chance you could retest two player with the refactoring branch that would be great. It's a pretty sizeable chunk of work I'd like to get back in. For whatever reason I just can't get the two instances to connect locally, they just don't talk.

[abu_the_monkey]

Originally Posted by Karlos

@abu

If there is any chance you could retest two player with the refactoring branch that would be great. It's a pretty sizeable chunk of work I'd like to get back in. For whatever reason I just can't get the two instances to connect locally, they just don't talk.

I have tested the one from 5 hours ago and all seems ok.

It was only a quick test so I will do a more in-depth test after I have eaten

[AlphaAmiga]

Currently standing just pass the end corridor of the first level facing the big red door, not moving.

TF1260 @ 50Mhz

Full screen - 9.4 FPS
2/3 Screen - 18.5 FPS

TF1260 @ 84Mhz

Full Screen - 13.5 FPS
2/3 Screen - 27.0 FPS

This is with normal walls (not simplified). Turning walls to simplified appears to increase the registered FPS by about 1 in all situations.

Question though, should I be seeing a halving of the FPS when going from a 66% to a 100% screen size?

Is it also possible to have a screen size between 100% and 66% - e.g. 75%?

Either way, the game is now definitely noticeably smoother/more responsive/playable with the 060 clocked at 50Mhz, and is very nice when I clock to about 80Mhz.

Great job everyone involved!!

[Karlos]

Originally Posted by AlphaAmiga

Currently standing just pass the end corridor of the first level facing the big red door, not moving.

TF1260 @ 50Mhz

Full screen - 9.4 FPS
2/3 Screen - 18.5 FPS

TF1260 @ 84Mhz

Full Screen - 13.5 FPS
2/3 Screen - 27.0 FPS

This is with normal walls (not simplified). Turning walls to simplified appears to increase the registered FPS by about 1 in all situations.

Question though, should I be seeing a halving of the FPS when going from a 66% to a 100% screen size?

Is it also possible to have a screen size between 100% and 66% - e.g. 75%?

Either way, the game is now definitely noticeably smoother/more responsive/playable with the 060 clocked at 50Mhz, and is very nice when I clock to about 80Mhz.

Great job everyone involved!!

Other sizes are a bit problematic. The 2/3 size seems to be "native" and to go fullscreen involved a lot of kludgy scaling by fixed fractions.

[abu_the_monkey]

@Karlos

[ Show youtube player ]

level A 2 player in WinUAE (sorry for the poor quality)

[Karlos]

That seems to work then! I realised the player shoot code is also duplicated but it's probably better to get this merged and tackle that as a second PR. This one at least seems to prove that the refactoring is the right way to go.

[Karlos]

Player code refactoring merged into main.

[abu_the_monkey]

top stuff

I will upload a short clip of your experimental branch with the max look mod shortly for you to give your opinion on.

[Karlos]

Originally Posted by abu_the_monkey

top stuff

I will upload a short clip of your experimental branch with the max look mod shortly for you to give your opinion on.

I'll be headed AFK for a few hours in a moment.

[abu_the_monkey]

no worries, it will be waiting for you on your return

it's nothing special just wanted to know if it shows any improvement on what you were seeing and is worth pursuing further

[abu_the_monkey]

Max look mod short clip

[ Show youtube player ]