XWA vs XWAUCP OPT Differences

Here we can have in-depth discussions about which missions need to be adjusted because they are broken either by original design or by XWA models.

XWA vs XWAUCP OPT Differences

Luke_Skywalker
Cadet 1st Class
Posts: 160
Joined: Fri Mar 11, 2005 12:01 am

Post by Luke_Skywalker » Sun Sep 06, 2020 6:55 pm

While futzing around to find out why the TIE X2 is so seemingly overpowered, I found some differences. So far I've found that the Otana has eight turret hardpoints to match the modeled barrels, so that threw mission balance completely out the window. (I've since fixed this and firing behavior seems to match vanilla now. I'll post it if anyone wants me to) I've also found that weapon setups sometimes don't match vanilla. ([EDIT] - X2s kill me just as fast as they do in Vanilla. I don't remember this. I used to be good at this game, I guess.)

For example:
https://i.imgur.com/H2Pq67y.png

Both new and old OPTs for TIE X1 have SuperEmpireLaser as a mesh hardpoint, yet they don't have matching weapon entries.

Furthermore, what would happen if I tried changing the scale values of the Millennium Falcon for a slightly less broken/janky Death Star run? Currently you kind of have to jam the cockpit up against a far wall between pipes, effectively threading the needle, or suffer a horrible fate at the hands of collision damage. The rather imprecise nature of the game refusing to let you be at 0 on any axis (or be perfectly centered on a model in general, it seems) makes this a bit of a game of luck, rather than skill.
When 900 years old you reach, look as good, you will not. Hmm?" -Yoda

User avatar
Ace Antilles
Admiral (Moderator)
Posts: 7829
Joined: Sat Jan 22, 2000 12:01 am
Contact:

Post by Ace Antilles » Wed Sep 09, 2020 10:41 am

There has been talk in the past of scaling the Lsndo Falcon down for Endor use.
If you want to have a go at that please do. Will be interesting to see how it comes out
Chief XWAU Team annoying nitpicker.
Ace Antilles - The X-Wing Outpost
Image

Exiled
Recruit
Posts: 8
Joined: Thu Jul 09, 2020 6:33 pm

Post by Exiled » Wed Sep 09, 2020 12:41 pm

Instead of downscaling the Falcons, why not upscale the Death Star Tunnels?

I read something a while back in another thread Darksaber suggested upscaling the the tunnel parts
I've done this, first I compared the original Corts to the New Corts, and found the New Cort's are around 40% bigger than the original, so I scaled all the DS Tunnel parts making them 40% bigger and tested in game, to my surprise it worked the tunnel parts still fit together, my only problem is that I have no clue where I'm going and can't seem to test it to completion I just keep going around in circles.

Here are the upscaled DS Tunnels parts, just copy and replace the original Death Star opt parts and don't forget to back up the original opts :)
Last edited by Exiled on Thu Sep 10, 2020 4:20 am, edited 1 time in total.

User avatar
Will T
Galactic Empire
Posts: 1371
Joined: Thu Aug 19, 2004 11:01 pm

Post by Will T » Wed Sep 09, 2020 1:43 pm

So this mission is a weird one. It's basically hard coded.

The DS tunnel parts aren't placed in the mission. I don't know how they're set up to connect, but it might well be that it's just a list of opts to use in a certain order, and coded so that each subsequent part connects edge to edge. If so, that might explain how scaling them up still allows them to connect.

There are two problems, though. 1 is the placement of the mission objects like the reactor core, DS lens etc, E-WEB turret etc. They ARE set to specific locations in the mission file, and so might be out of place if the DS corridor has shifted and extended in length.

The other is that if the tunnel parts are scaled up, the run itself is going to require flying a longer distance and take longer. This could be a problem during the escape, where the mission is timed. Take too long and the reactor explosion catches you and kills you. If the run is going to take longer, it may become impossible to go fast enough to escape.

One other thing to consider is that the scaled up DS parts will now be bigger relative to the other ships in that mission. It shouldn't affect the AI, but I wouldn't say for sure that it won't.


I might take a look at those edited DS parts at some point, but I suspect the issue might be that certain mission objects are now in the wrong place.


Either way, I think it's far, far easier to just scale the Falcon down for that mission, if that can be done.
Formerly known as The 95 Headhunter

Exiled
Recruit
Posts: 8
Joined: Thu Jul 09, 2020 6:33 pm

Post by Exiled » Wed Sep 09, 2020 2:46 pm

Your being quiet negative about this and condemning the idea before even testing them, have you tested them?

Yes I know that they are not placed in the mission, they are not hard coded either, yes there must be something the places them close together, I'm not sure what it is, but they actually connect with the use of hardpoints, each section has a number of AccStart and AccEnd hardpoints on each opt similar to the connection rod which forms a train of containers.
There are two problems, though. 1 is the placement of the mission objects like the reactor core, DS lens etc, E-WEB turret etc. They ARE set to specific locations in the mission file, and so might be out of place if the DS corridor has shifted and extended in length.
Have you tested them?
The other is that if the tunnel parts are scaled up, the run itself is going to require flying a longer distance and take longer. This could be a problem during the escape, where the mission is timed. Take too long and the reactor explosion catches you and kills you. If the run is going to take longer, it may become impossible to go fast enough to escape.
Yes that maybe true, the distance may have changed, and yes the time it takes maybe longer, this is why it needs testing! Have you tested this yet?

So to play that one mission, before you do, your still going to have to install a downscaled Falcon, before you start, then reinstall the correctly scaled falcon once finish the death star Run, seem a little daft to me.

My only question is, have you tested the upscaled tunnel opts yet?

I think it might be a valid option, but you personally seem to have made up you mind already.

If tested and found to be no good then I'll apologise then, until then, please test!

It maybe an idea if as many people test it as possible, instead of only one persons opinion of the testing.

User avatar
Will T
Galactic Empire
Posts: 1371
Joined: Thu Aug 19, 2004 11:01 pm

Post by Will T » Wed Sep 09, 2020 5:18 pm

Right. Okay, I think you may have taken what I was saying the wrong way there. Sorry if you felt I was being negative.

I was literally just theorising. All I wanted to do was think aloud a bit to try and pre-empt any potential issues.

I don't have time to test it just yet, but I do intend to make the time.
Exiled wrote:
Wed Sep 09, 2020 2:46 pm
Yes I know that they are not placed in the mission, they are not hard coded either, yes there must be something the places them close together, I'm not sure what it is
The 'something' that places them together is the hardcoding ;)

All I mean is that it's something that's clearly handled by the exe, and not determined by an editable file that we have access to.
Exiled wrote:
Wed Sep 09, 2020 2:46 pm
but they actually connect with the use of hardpoints, each section has a number of AccStart and AccEnd hardpoints on each opt similar to the connection rod which forms a train of containers.
Yep, and that's what I was assuming must be the case. But like I said, that doesn't apply to the mission objects, which use regular opts that don't have those hard points. This includes things like the Nav Buoy that's used as a proximity trigger for certain messages, or the Gun Platform that's hidden underneath the E-WEB opt to allow it to fire (the E-WEB doesn't have any weapon hardpoints and isn't counted as a craft in the game). They are placed in specific locations on the mission map, and if the structure you can actually fly through has changed, they're going to be in different places relative to the tunnel.

That's apparent just looking at the mission file.

All I'm doing here is speculating why you might have found yourself flying round in circles. It's possible that expanding the tunnel opts has caused some mission triggers to not work.
Exiled wrote:
Wed Sep 09, 2020 2:46 pm
Yes that maybe true, the distance may have changed, and yes the time it takes maybe longer, this is why it needs testing! Have you tested this yet?
No, I'm just stating ahead of time something that will need to be deliberately paid attention to in testing.

That's how good testing works. You identify potential problem areas that might need focus and iteration to fully test out.
Exiled wrote:
Wed Sep 09, 2020 2:46 pm
So to play that one mission, before you do, your still going to have to install a downscaled Falcon, before you start, then reinstall the correctly scaled falcon once finish the death star Run, seem a little daft to me.
No that's not what I'm suggesting at all.

Obviously that would be a lot of work on the part of the user. The suggestion would be to scale down the Falcon only for that mission, and to adapt the files needed to make it seamless.

An easy and currently doable approach is to create a new Falcon opt that is identical in stats, just downsized, and switch the mission to use the new opt. The only problem is that this takes up a valuable spare craft slot. That's the only reason this hasn't already been done.

The preferable approach would be to do it with a mission hook that swaps the opts being used. That's not currently possible, I believe, but it might be with a bit of work.
Exiled wrote:
Wed Sep 09, 2020 2:46 pm
My only question is, have you tested the upscaled tunnel opts yet?
Look, if we're being pointed about this then my only question would be why haven't you tested it yet?

You said you ended up going in circles, but that doesn't really make sense to me. The tunnel run is basically a linear path. There's a branch at one point, but one branch goes to a dead end you're supposed to do something and turn round in. And that branch is completely optional. Assuming you don't roll your craft from the starting position, I'm pretty sure if you pick the right fork you should eventually come to the reactor room.

My point is that if you're struggling to complete the mission then maybe there's a reason for that.
Exiled wrote:
Wed Sep 09, 2020 2:46 pm
I think it might be a valid option, but you personally seem to have made up you mind already.
I really, really haven't. I hope it works! That would solve a lot of problems.

I'm just pessimistic because I've tried picking this mission apart before and it's weird. There are all sorts of things that could potentially be warped just by increasing the size of the opts. The timing of the blocks through the accelerator, for example.
Exiled wrote:
Wed Sep 09, 2020 2:46 pm
If tested and found to be no good then I'll apologise then, until then, please test![.quote]

If it doesn't work there'd be no reason to apologise. It's cool that someone's actually taken the time to give us testable opts for this. I couldn't ever be bothered to sift through the pile of DS opts.

I just think it's good practice to figure out the potential breaking points before diving in.
Formerly known as The 95 Headhunter

Exiled
Recruit
Posts: 8
Joined: Thu Jul 09, 2020 6:33 pm

Post by Exiled » Wed Sep 09, 2020 6:47 pm

You do have some valid points :)

to answer your question, I've tried playing the mission, but like I said I just get lost, it's been years since I played the mission and I no longer have a clue with it and just get lost, I even tried playing the mission with original DS Tunnel opts and Falcon opt, just to see if there was a difference but I just haven't a clue where I'm going anymore. lol :D
Obviously that would be a lot of work on the part of the user. The suggestion would be to scale down the Falcon only for that mission, and to adapt the files needed to make it seamless.

An easy and currently doable approach is to create a new Falcon opt that is identical in stats, just downsized, and switch the mission to use the new opt. The only problem is that this takes up a valuable spare craft slot. That's the only reason this hasn't already been done.

The preferable approach would be to do it with a mission hook that swaps the opts being used. That's not currently possible, I believe, but it might be with a bit of work.
Yes I've looked into this also, scaling down the Falcon so it's the same size as the original Falcon but there is a problem

The Game uses the smaller Falcon (MiniFalcon) in the mission fine, the problem is the POV of the pilot, it's still using the same POV as the larger Falcon so your POV is outside the Falcons Cockpit (See Image below)

I tried adding a Cockpit Hardpoint, which usually overides the Cockpit POV co-ordinates, but it didn't work

I adding these lines to 1b7m4w.ini

[Objects]
FlightModels\MillenniumFalcon2.opt = FlightModels\MiniFalcon.opt
FlightModels\MillenniumFalcon2Cockpit.opt = FlightModels\MiniFalconCockpit.opt
FlightModels\MillenniumFalcon2Exterior.opt = FlightModels\MiniFalconExterior.opt
.
flightscreen5.jpg
.
.

I might have a solution though but it would need Jeremy to implement it
To solve this problem, I think it would need an addition to one of the Hooks (perhaps Hangar Hook) which you could add the "*.ini "files, which overrides the POV coded into the game for each pilot POV

At present the Falcon POV is X=454, Y=352, Z=62

But if Jeremy was able, he could add something similar like the Hangar Map settings to the Objects / HangarObject section of the ini file, something like

model index, pov X, pov Y, pov Z

model index = 59, pov X =334, pov Y =220, pov Z =49

so it would look like

[Objects]
FlightModels\MillenniumFalcon2.opt = FlightModels\MiniFalcon.opt
FlightModels\MillenniumFalcon2Cockpit.opt = FlightModels\MiniFalconCockpit.opt
FlightModels\MillenniumFalcon2Exterior.opt = FlightModels\MiniFalconExterior.opt
59, 334, 220, 49

This way the idea of scaling down a version of the Falcon, would make scaling of the DS Tunnel opts unnecessary, and you would be able to play with the normal size falcon in any other mission

How does that idea sound? :D
You do not have the required permissions to view the files attached to this post.

User avatar
Will T
Galactic Empire
Posts: 1371
Joined: Thu Aug 19, 2004 11:01 pm

Post by Will T » Wed Sep 09, 2020 8:25 pm

Exiled wrote:
Wed Sep 09, 2020 6:47 pm

to answer your question, I've tried playing the mission, but like I said I just get lost, it's been years since I played the mission and I no longer have a clue with it and just get lost, I even tried playing the mission with original DS Tunnel opts and Falcon opt, just to see if there was a difference but I just haven't a clue where I'm going anymore. lol :D
Okay, I'll give you the very brief guide for how to get through.

Hitting the O key at any time brings up the reactor core. You can sometimes use that as a very brief guide.

The first part of the run should be straightforward, you take the left branch when you hear "great, the primary route is blocked" - it's not hard to miss because the way forward is indeed completely blocked. When you come to the container accelerator rings, you just go right through them.

Shortly after you hear "We're losing too many ships", you'll come up to a fork in the tunnel. Flying dead ahead will crash you into a wall, but you should see it coming in plenty of time. Start cycling through your targets when you hear the dialogue, and you should come across the laser node object. Whichever way your radar is showing you it is, take the opposite fork. I think I was wrong earlier, and if you haven't rolled at all it should be the left fork you take. Targeting the node will show you for sure though.

Follow the tunnel until you come to a big, open room. You'll hear dialogue about "repair crews, they must be getting desperate". At this point, you should be able to target the E-WEB. I forget exactly what it's called, but you'll recognise it. Fly towards it. Try and maintain your orientation (don't roll too much). You should find it's on a little platform. At the back of that platform is the rest of the tunnel.

Once you're at that point, you shouldn't have any more choice about where to go, it's just following the tunnel.
Exiled wrote:
Wed Sep 09, 2020 6:47 pm
Yes I've looked into this also, scaling down the Falcon so it's the same size as the original Falcon but there is a problem

The Game uses the smaller Falcon (MiniFalcon) in the mission fine, the problem is the POV of the pilot, it's still using the same POV as the larger Falcon so your POV is outside the Falcons Cockpit (See Image below)

I tried adding a Cockpit Hardpoint, which usually overides the Cockpit POV co-ordinates, but it didn't work

I adding these lines to 1b7m4w.ini

[Objects]
FlightModels\MillenniumFalcon2.opt = FlightModels\MiniFalcon.opt
FlightModels\MillenniumFalcon2Cockpit.opt = FlightModels\MiniFalconCockpit.opt
FlightModels\MillenniumFalcon2Exterior.opt = FlightModels\MiniFalconExterior.opt
Yeah, this is what I mean about it not currently being possible.

The object replacement hook isn't really meant to be used on the player's ship. Cockpit POV is written into the exe, not the opt.
Exiled wrote:
Wed Sep 09, 2020 6:47 pm
I might have a solution though but it would need Jeremy to implement it
To solve this problem, I think it would need an addition to one of the Hooks (perhaps Hangar Hook) which you could add the "*.ini "files, which overrides the POV coded into the game for each pilot POV

...

How does that idea sound? :D
Yep, this is probably exactly what would be needed to make it happen. Not sure if thats doable.

I think Ace has mentioned he has an idea that might sort it. I guess just watch this space.


In the meantime, I had half an hour to play around with the scaled up DS parts

Perhaps predictably, it was both better than I expected and still a weird mess.

For one thing, I had a pretty bad crash on one run. Nice glitchy green screen, some messed up audio and an immediate reboot. This might be completely unrelated - my PC is very new and I built it, so there could be some hidden problems there. I've been doing a lot of XWA testing this week and this is the only crash though, so.....

Other observations:

The friendly AI ships with you die really quickly. I'll need to remember to record a run for the film room so I can observe them. I'm not sure if they're crashing into obstacles or being shot down, but something kills all of them and brings invincible Wedge down to 1% hull before I even make the acclerator room, which I'm sure doesn't happen in vanilla.

Something got really screwy with my joystick. Maybe that's a problem unique to me, but it works fine on other missions. It felt like the control fighting you get if you point at one of the 'vertexes' of the map (usually the very top or bottom). Where you just can't fly straight at them. Some parts were better than others, but some places were really bad. Way worse than I've experience on any other mission. I died several times going through the accelerator ring (which shouldn't happen - the ring locks you in place then fires you straight forward), I kept veering into the wall of the tunnel or crashing into the stop ring itself. I'll need someone else to verify that. Like I said, it could be something to do with my joystick setup making it worse - the z-fighting in normal missions seems a bit worse on this install than I remember it being.

The random obstacles are too big. This is the sort of problem I saw coming, it's one of those classic problems of messing with scale. The random pipes and tubes that get in your way and can be destroyed probably did need to be scaled up, otherwise they'd have got too easy to dodge. But being scaled up in all dimensions means now a lot of them clip into each other. It's also possibly the reason the ally fighters die so much quicker. Without being able to change the code that controls the spacing between the obstacles, there's nothing that can be done to accommodated the bigger size.

I might be crazy, or I might have just inverted the ship at one point but I swear all the connections have been flipped. Everything past the first point that should be to the right is now to the left and vice versa. I'll need to confirm that one. No idea how it could have happened either.

In some ways, the up scale makes the run easier than vanilla. Because the sections are now longer, but the Falcon moves through them at the same speed, you're actually moving through them more slowly. It loses some of the edge of your seat feel that you get in vanilla.

I feel like maybe the scaling factor is actually too much? What was 40% based off?

It's really only the diameter of the tunnel parts compared with the Falcon being wider that's the problem. How much wider is the XWAU Falcon? Is that where the 40% value came from?
Formerly known as The 95 Headhunter

User avatar
JeremyaFr
XWAU Member
Posts: 3921
Joined: Mon Jan 18, 2010 5:52 pm
Contact:

Post by JeremyaFr » Wed Sep 09, 2020 9:21 pm

WIP

Hello,
Here is a WIP version of the mission objects hook.

You can now replace the cockpit pov for a craft via the craft's ini file.

Suppose that the new craft is "FlightModels\[Model].opt".
To define the cockpit POV, set CockpitPovX, CockpitPovY and CockpitPovZ in a file named "FlightModels\[Model]Object.txt" or create a section named "[Object]" in "FlightModels\[Model].ini".

Code: Select all

;CorellianTransport2Object.txt or [Object] in CorellianTransport2.ini
CockpitPovX = 174
CockpitPovY = -121
CockpitPovZ = 33

EDIT: link removed
Last edited by JeremyaFr on Thu Sep 10, 2020 3:56 pm, edited 2 times in total.

User avatar
Will T
Galactic Empire
Posts: 1371
Joined: Thu Aug 19, 2004 11:01 pm

Post by Will T » Wed Sep 09, 2020 9:29 pm

Jeremy, you are an absolute machine!

This is awesome, thanks for adding this.

Just so I'm clear, how would you use this on a per mission basis?

You'd add the object hook to the mission ini to swap the normal Falcon for the mini Falcon, then the mini Falcon would have its own ini file where the cockpit POV is set and the mission will read the POV from the target swapped craft's ini?

Exiled wrote:
Wed Sep 09, 2020 6:47 pm

Yes I've looked into this also, scaling down the Falcon so it's the same size as the original Falcon but there is a problem

The Game uses the smaller Falcon (MiniFalcon) in the mission fine, the problem is the POV of the pilot, it's still using the same POV as the larger Falcon so your POV is outside the Falcons Cockpit (See Image below)

I tried adding a Cockpit Hardpoint, which usually overides the Cockpit POV co-ordinates, but it didn't work

I adding these lines to 1b7m4w.ini

[Objects]
FlightModels\MillenniumFalcon2.opt = FlightModels\MiniFalcon.opt
FlightModels\MillenniumFalcon2Cockpit.opt = FlightModels\MiniFalconCockpit.opt
FlightModels\MillenniumFalcon2Exterior.opt = FlightModels\MiniFalconExterior.opt
Exiled, could you attach your modified mini Falcon opts?

I want to give this a go, and if you've already shrunk the Falcon down I may as well use that rather than making one myself.
Formerly known as The 95 Headhunter

User avatar
JeremyaFr
XWAU Member
Posts: 3921
Joined: Mon Jan 18, 2010 5:52 pm
Contact:

Post by JeremyaFr » Wed Sep 09, 2020 9:44 pm

In "1b7m4w.ini":

Code: Select all

[Objects]
FlightModels\MillenniumFalcon2.opt = FlightModels\MiniFalcon.opt
FlightModels\MillenniumFalcon2Cockpit.opt = FlightModels\MiniFalconCockpit.opt
FlightModels\MillenniumFalcon2Exterior.opt = FlightModels\MiniFalconExterior.opt
and in "MiniFalcon.ini":

Code: Select all

[Object]
CockpitPovX = 334
CockpitPovY = 220
CockpitPovZ = 49

User avatar
Will T
Galactic Empire
Posts: 1371
Joined: Thu Aug 19, 2004 11:01 pm

Post by Will T » Wed Sep 09, 2020 9:49 pm

Perfect. That's what I was assuming.

Thanks for clearing that up.
Formerly known as The 95 Headhunter

User avatar
Ace Antilles
Admiral (Moderator)
Posts: 7829
Joined: Sat Jan 22, 2000 12:01 am
Contact:

Post by Ace Antilles » Wed Sep 09, 2020 10:23 pm

Will T wrote:
Wed Sep 09, 2020 9:49 pm
Perfect. That's what I was assuming. Thanks for clearing that up.
Ok so I did a little testing and it seems to work ok.

I took and X-wing and copied the files and renamed them.
Ace.ini
Ace.opt
AceCockpit.opt
AceExterior.opt

Then I put this code in the Ace.ini

Code: Select all

[Object]
CockpitPovX = 0
CockpitPovY = -31
CockpitPovZ = 36
Then I made 1b7m4w.ini and put in this code.

Code: Select all

;1b7m4w.ini

[Resdata]
Resdata\Planet2.dat

[Objects]
FlightModels\MilleniumFalcon2.opt = FlightModels\Ace.opt
FlightModels\MilleniumFalcon2Cockpit.opt = FlightModels\AceCockpit.opt
FlightModels\MilleniumFalcon2Exterior.opt = FlightModels\AceExterior.opt
Viola. I was flying an X-wing in the tubes.
Now that's not perfect obviously as it was still behaving like a Falcon and not good for hangars.
But a mini Falcon on the last mission only should behave the right way and perform right.

The only thing is to be careful as Mxted lists it as X Z Y so easy mistakes can be made when typing codes.

Another option well executed Jeremy. :) Even if it probably only has one use.
Chief XWAU Team annoying nitpicker.
Ace Antilles - The X-Wing Outpost
Image

Exiled
Recruit
Posts: 8
Joined: Thu Jul 09, 2020 6:33 pm

Post by Exiled » Thu Sep 10, 2020 8:36 am

First your welcome for the suggestion and idea lol :)

@Ace, Just wondering why you have the lines

[Resdata]
Resdata\Planet2.dat

in your 1b7m4w.ini? The mission doesn't have the backdrops included so these lines are redundant in the ini file

@WillT thanks for the heads up :)
The object replacement hook isn't really meant to be used on the player's ship. Cockpit POV is written into the exe, not the opt
Isn't it? I thought that was the whole point of having a [HangarObject] section, [Objects] and now an [Object] section which probably would be better if it was called it [ObjectPOV] or simply [POV] to stop any confusion between "Objects" and "Object"

[POV]
CockpitPovX = 333
CockpitPovY = 219
CockpitPovZ = 45

True the POV is written to the exe, but as I explained a hardpoint called "Cockpit" can be added to the Opt (not recommended though), basically you place the Hardpoint where you think the pilots head might be, this hardpoint overrides the POV written to the exe and becomes the POV in game, but this didn't work for the minifalcon as your only exchanging the opt with the hook, so the POV for the larger Falcon takes president, therefore my request to Jeremy for the POV Hook addition.

Also just so we're on the right track, on and off I've been editing the Xwing Games for over 20 years now, I appreciate the quick game guide, (put the crappy game play down to age) and thanks I don't need everything explaining :D

@Jeremy Thanks for the POV addition it does work well, thank you for implementing it

@WillT Here is the Mini Falcon, please let me know when you have downloaded it, then I'll delete the files
.
.
EDIT: File Deleted
Last edited by Exiled on Fri Sep 11, 2020 5:42 am, edited 1 time in total.

User avatar
JeremyaFr
XWAU Member
Posts: 3921
Joined: Mon Jan 18, 2010 5:52 pm
Contact:

Post by JeremyaFr » Thu Sep 10, 2020 3:57 pm

WIP

Hello,
I've renamed [Object] to [CockpitPov].

EDIT: link removed
Last edited by JeremyaFr on Fri Sep 11, 2020 11:39 am, edited 1 time in total.

Exiled
Recruit
Posts: 8
Joined: Thu Jul 09, 2020 6:33 pm

Post by Exiled » Thu Sep 10, 2020 6:10 pm

Cheers Jeremy, that's much better :)

User avatar
JeremyaFr
XWAU Member
Posts: 3921
Joined: Mon Jan 18, 2010 5:52 pm
Contact:

Post by JeremyaFr » Fri Sep 11, 2020 11:41 am

Hello,
I've merged the WIP version of the mission objects hook into the stable version.

Luke_Skywalker
Cadet 1st Class
Posts: 160
Joined: Fri Mar 11, 2005 12:01 am

Post by Luke_Skywalker » Fri Sep 18, 2020 8:58 am

Ha, I come back to this after forgetting about it and find out progress was made without me. :D
When 900 years old you reach, look as good, you will not. Hmm?" -Yoda

Luke_Skywalker
Cadet 1st Class
Posts: 160
Joined: Fri Mar 11, 2005 12:01 am

Post by Luke_Skywalker » Fri Sep 18, 2020 12:53 pm

Will T wrote:
Wed Sep 09, 2020 8:25 pm
Something got really screwy with my joystick. Maybe that's a problem unique to me, but it works fine on other missions. It felt like the control fighting you get if you point at one of the 'vertexes' of the map (usually the very top or bottom). Where you just can't fly straight at them. Some parts were better than others, but some places were really bad. Way worse than I've experience on any other mission. I died several times going through the accelerator ring (which shouldn't happen - the ring locks you in place then fires you straight forward), I kept veering into the wall of the tunnel or crashing into the stop ring itself. I'll need someone else to verify that. Like I said, it could be something to do with my joystick setup making it worse - the z-fighting in normal missions seems a bit worse on this install than I remember it being.
Something I've noticed after playing XWA, both vanilla and upgraded: the more ships in the mission, the screwier the controls. I don't know if it's a joystick value thing that XWA doesn't know what to do with (Joystick Gremlin and a Virpil WarBRD input with a greater-than-8-bit range, for instance) or if it's an actual engine thing. The deadzone/gimbal lock at the 0,0 mark is much more pronounced when compared to vanilla, and that was something I never, ever saw way back in the days of Win98. It's only recently that I've noticed, but that could be something with me just noticing more issues and hacks when it comes to older engines. The first time I experienced it I was taken by complete surprise, and as it turns out it is a massive pain in the ass because, naturally, the AI is not at all bothered by going straight up, forcing you to disengage and open yourself up to the fighter you were chasing. But more than that, I've noticed that the more ships or polygons are in the mission, the more jittery the controls get. There are times where the perfect yaw of transports isn't so perfect, as I'll see spiking in the Y axis and the ship will slightly pitch up. I'll also have a devil of a time trying to center my sights on an enemy fighter at times, as it feels like the engine does not want me to be pointed directly at that TIE fighter's cockpit, and in some instances I'll be fighting with the controls trying to stick to a direction because the nose of the ship is floating around what I want to fly towards. It's very odd, and the small joystick movement hook helps a little bit, but it's still a problem.

I've noticed the issue with getting yeeted into the wall when going through the cargo acceleration area, as well. You have to nail it just right or else you're toast. There is definitely something off with velocities and trajectories in the upgrades, but I honestly have no idea where to even start with looking into or rectifying this.
When 900 years old you reach, look as good, you will not. Hmm?" -Yoda

User avatar
Will T
Galactic Empire
Posts: 1371
Joined: Thu Aug 19, 2004 11:01 pm

Post by Will T » Fri Sep 18, 2020 1:35 pm

Luke_Skywalker wrote:
Fri Sep 18, 2020 12:53 pm

Something I've noticed after playing XWA, both vanilla and upgraded: the more ships in the mission, the screwier the controls. I don't know if it's a joystick value thing that XWA doesn't know what to do with (Joystick Gremlin and a Virpil WarBRD input with a greater-than-8-bit range, for instance) or if it's an actual engine thing. The deadzone/gimbal lock at the 0,0 mark is much more pronounced when compared to vanilla, and that was something I never, ever saw way back in the days of Win98. It's only recently that I've noticed, but that could be something with me just noticing more issues and hacks when it comes to older engines. The first time I experienced it I was taken by complete surprise, and as it turns out it is a massive pain in the ass because, naturally, the AI is not at all bothered by going straight up, forcing you to disengage and open yourself up to the fighter you were chasing. But more than that, I've noticed that the more ships or polygons are in the mission, the more jittery the controls get. There are times where the perfect yaw of transports isn't so perfect, as I'll see spiking in the Y axis and the ship will slightly pitch up. I'll also have a devil of a time trying to center my sights on an enemy fighter at times, as it feels like the engine does not want me to be pointed directly at that TIE fighter's cockpit, and in some instances I'll be fighting with the controls trying to stick to a direction because the nose of the ship is floating around what I want to fly towards. It's very odd, and the small joystick movement hook helps a little bit, but it's still a problem.

I've noticed the issue with getting yeeted into the wall when going through the cargo acceleration area, as well. You have to nail it just right or else you're toast. There is definitely something off with velocities and trajectories in the upgrades, but I honestly have no idea where to even start with looking into or rectifying this.
This is both interesting and a bit worrying because I've noticed every single one of those issues too. The worst one is definitely when your reticle bounces around a certain target. Really frustrating. And yeah, the top and bottom z-fighting is definitely more pronounced with XWAU than in vanilla.

I hadn't really twigged that it was worse when there are more ships on screen, but that does make a lot of sense.

I found I noticed it more on certain missions, and doing some testing on B4M5 and M6 yesterday, it was really problematic. Those, of course, are both missions featuring the SSD and a ton of fighters.



Agreed, this is something I also have absolutely no idea how to even test and reproduce consistently, let alone diagnose or fix.

One thing I've noticed is that I've never had to calibrate my T1600m. But back in the day of XWA, joystick calibration was basically mandatory. Do you suppose there's any chance it's a calibration issue? Does XWA need to read some calibration data from somewhere?


Another thing I've just remembered off the back of your comment about velocities and trajectories - I've noticed some really weird behaviour at very low speeds. Dropping down to <10% throttle to manually fly into a hangar, for example. I've found that if I'm a bit too low at those speeds, pitching up doesn't get me into the hangar. I don't fly straight in the direction I'm pointing and so move diagonally up. Instead I just sort of drift along forwards relative to the plane of the mission with my nose pointed up and I still crash into the bottom of the hangar opening. The same sort of thing can happen when yawing is well.
JeremyaFr wrote:
Any thoughts on this at all? Is there anything in the exe that might cause a link between control handling and ships / polycount on screen?

Or as Luke suggested is there any chance this is a hardware issue caused by XWA not being able to get full input from certain joysticks?
Formerly known as The 95 Headhunter

User avatar
JeremyaFr
XWAU Member
Posts: 3921
Joined: Mon Jan 18, 2010 5:52 pm
Contact:

Post by JeremyaFr » Fri Sep 18, 2020 2:05 pm

Hello,
The game engine computes the distance as follow:

Code: Select all

int deltaX = ...;
int deltaY = ...;
int deltaZ = ...;

deltaX = abs( deltaX );
deltaY = abs( deltaY );
deltaZ = abs( deltaZ );

int distance;

if( deltaX > deltaY && deltaX > deltaZ )
{
    distance = deltaX + deltaY / 4 + deltaZ / 4;
}
else if( deltaY > deltaX && deltaY > deltaZ )
{
    distance = deltaX / 4 + deltaY + deltaZ / 4;
}
else
{
    distance = deltaX / 4 + deltaY / 4 + deltaZ;
}
Maybe this can cause an issue.
Last edited by JeremyaFr on Fri Sep 18, 2020 4:45 pm, edited 1 time in total.

Luke_Skywalker
Cadet 1st Class
Posts: 160
Joined: Fri Mar 11, 2005 12:01 am

Post by Luke_Skywalker » Fri Sep 18, 2020 2:52 pm

"delatX"

is that a typo by you or the coders? everything else has delta spelled correctly.
When 900 years old you reach, look as good, you will not. Hmm?" -Yoda

User avatar
Will T
Galactic Empire
Posts: 1371
Joined: Thu Aug 19, 2004 11:01 pm

Post by Will T » Fri Sep 18, 2020 3:31 pm

Luke_Skywalker wrote:
Fri Sep 18, 2020 2:52 pm
"delatX"

is that a typo by you or the coders? everything else has delta spelled correctly.
It's weird, but it shouldn't matter.

The variable is declared as DelatX, and referenced consistently as such. It would only be an issue if the typo was on the variable declaration, but the usage in the code was spelled correctly.

If that is a TG mistake, it's a funny little insight into their coding practices though :D

JeremyaFr wrote:
Fri Sep 18, 2020 2:05 pm
Hello,
The game engine computes the distance as follow:

Code: Select all

int delatX = abs( delatX );
int deltaY = abs( deltaY );
int deltaZ = abs( deltaZ );

int distance;

if( delatX > deltaY && delatX > deltaZ )
{
    distance = delatX + deltaY / 4 + deltaZ / 4;
}
else if( deltaY > delatX && deltaY > deltaZ )
{
    distance = delatX / 4 + deltaY + deltaZ / 4;
}
else
{
    distance = delatX / 4 + deltaY / 4 + deltaZ;
}
Maybe this can cause an issue.

What does the distance referred to here mean Jeremy? Is that render distance?

If so then yeah, maybe there's something weird going on here where a strong motion in a particular axis causes feedback?
Formerly known as The 95 Headhunter

User avatar
JeremyaFr
XWAU Member
Posts: 3921
Joined: Mon Jan 18, 2010 5:52 pm
Contact:

Post by JeremyaFr » Fri Sep 18, 2020 4:46 pm

Luke_Skywalker wrote:
Fri Sep 18, 2020 2:52 pm
"delatX"

is that a typo by you or the coders? everything else has delta spelled correctly.
It is a typo.

User avatar
JeremyaFr
XWAU Member
Posts: 3921
Joined: Mon Jan 18, 2010 5:52 pm
Contact:

Post by JeremyaFr » Fri Sep 18, 2020 4:53 pm

Will T wrote:
Fri Sep 18, 2020 3:31 pm
If that is a TG mistake, it's a funny little insight into their coding practices though :D
It is not a typo in my code or in TG code. It's only a typo in the above post.
Will T wrote:
Fri Sep 18, 2020 3:31 pm
What does the distance referred to here mean Jeremy? Is that render distance?

If so then yeah, maybe there's something weird going on here where a strong motion in a particular axis causes feedback?
It is the distance used for proximity checks. For example, when a craft is near a hangar.

Post Reply