da12thMonkey

Nah, it has happened to several of our team members too. Game seems to randomly use the lowest LOD for the Mk.11 or Mk.11 EC sometimes (when flipping through multiple weapons in the Arsenal before arriving at the Mk.11s usually, but seems to only effect one model or the other rather than both). But other times the game will load them all fine. Has been mentioned that it seemed to start when BIS began implementing the LOD Blending tech, but we're not certain of the cause since it seems to occur so randomly.

We have changelogs: AFRF USAF GREF But a kind of summary of new and important stuff from memory:

To my knowledge the face lighting/shadow face properties are now defunct and performed by renderFlags and texture formatting, so I don't think that's the problem. The BIS Sample house doesn't have anything special on the windows in Face Properties anyway. I did just notice that your glass .rvmat has in Stage 4: texture = "#(argb,8,8,3)color(0,0,0,1,AS)";Which means your procedural ambient shadow on the glass is black Where the BIS one is: texture = "#(argb,8,8,3)color(1,1,1,1,AS)";So that'd be all white.

Shouldn't really matter where in the header the renderFlags[] array goes. This is glass.rvmat from structures_f_data\Windows, and yours is virtually the same ambient[] = {1.0, 1.0, 1.0, 1}; diffuse[] = {1.0, 1.0, 1.0, 1}; forcedDiffuse[] = {0, 0, 0, 0}; emmisive[] = {0, 0, 0, 1}; specular[] = {0.066667, 0.062745, 0.058824, 1}; specularPower = 2000; renderFlags[] = {"NoZWrite"}; PixelShaderID = "Super"; VertexShaderID = "Super"; class Stage1 { texture = "#(argb,8,8,3)color(0.5,0.5,1,1,NOHQ)"; uvSource = "tex"; class uvTransform { aside[] = {1, 0, 0}; up[] = {0, 1, 0}; dir[] = {0, 0, 0}; pos[] = {0, 0, 0}; }; }; class Stage2 { texture = "#(argb,8,8,3)color(0.5,0.5,0.5,1,DT)"; uvSource = "tex"; class uvTransform { aside[] = {1, 0, 0}; up[] = {0, 1, 0}; dir[] = {0, 0, 0}; pos[] = {0, 0, 0}; }; }; class Stage3 { texture = "#(argb,8,8,3)color(0,0,0,0,MC)"; uvSource = "tex"; class uvTransform { aside[] = {1, 0, 0}; up[] = {0, 1, 0}; dir[] = {0, 0, 0}; pos[] = {0, 0, 0}; }; }; class Stage4 { texture = "#(argb,8,8,3)color(1,1,1,1,AS)"; uvSource = "tex"; class uvTransform { aside[] = {1, 0, 0}; up[] = {0, 1, 0}; dir[] = {0, 0, 0}; pos[] = {0, 0, 0}; }; }; class Stage5 { texture = "a3\structures_f\data\windows\glass_smdi.paa"; uvSource = "tex"; class uvTransform { aside[] = {1, 0, 0}; up[] = {0, 1, 0}; dir[] = {0, 0, 0}; pos[] = {0, 0, 0}; }; }; class Stage6 { texture = "#(ai,64,64,1)fresnel(9.6,0.9)"; uvSource = "tex"; class uvTransform { aside[] = {1, 0, 0}; up[] = {0, 1, 0}; dir[] = {0, 0, 0}; pos[] = {0, 0, 0}; }; }; class Stage7 { useWorldEnvMap = "true"; texture = "a3\data_f\env_land_co.paa"; uvSource = "tex"; class uvTransform { aside[] = {1, 0, 0}; up[] = {0, 1, 0}; dir[] = {0, 0, 0}; pos[] = {0, 0, 0}; }; }; I assume the glass alpha texture you are using is properly formatted - saved as 32-bit and named filename_ca.tga before converting to filename_ca.paa? Make sure you remove P:\ from your .rvmat paths btw - ingame it'll look for the files on your P:\ drive rather than the ones in the .pbo (so wont work for people who download your mod)

Glass generally has RenderFlags[] = {"NoZwrite"};in the .rvmat to stop shadows casting on it.

VME has a visible laser like this, so maybe it's that.

Yes, you can use 2:1. They're not uncommon in the game files. I guess it's just that textures files in the game tend to be 1:1 that people might think they have to be 1:1 I've used 2:1 .paa textures on 3D models and as the splash screen for the mods/expansions menu (in fact the biki says 2048*1024 is the standard for these https://community.bistudio.com/wiki/Mod_Presentation but I used 1024*512 for mine) and all inventory images for weapons in the game are 2:1. Never had a problem with 2:1 so long as the dimensions are based on 2^n pixel values

Texture files can be 1:1, 2:1, 1:2 even 4:1 IIRC But AFAIK the only permissible image sizes for any .paa in 2:1 aspect ratio are e.g. 128*256, 256*512, 512*1024, 1024*2048 etc. Indeed, if you use the newer official imageToPAA conversion tool instead of the older texView tool (texView really is a legacy tool aimed more at just viewing .paa files now, rather than being the primary conversion tool for Arma 3), it will actually refuse to convert any image that uses unacceptable dimensions. And shows error messages in the GUI console when it encounters an image that is not right for the game.

What size are the images? They have to have pixel dimension that are order 2^n - so 64, 128, 256, 512, 1024, 2048, 4096

The way I'm beginning to interpret this from the wording used is that BIS do generate optimised shadowVolume LODs (i.e. they're still kind of needed for the sake of performance rather than using complex res LOD geometry to generate the ingame shadow) but because they are using shadowbuffer all the time on terrain objects and not stencil shadows, they merely don't need to carefully (manually) generate meshes that are closed, sharp, triangulated and closely follow the geometry of the visual LODs. Automated optimisation tools in 3D software are often rather bad at generating clean, closed topology since they merge verts. But IIRC shadowbuffer has never really shown up problems like this that you get on stencil shadows when a mesh isn't properly closed and triangulated. So it's probably easier to optimise a mesh for shadowbuffer than it is for stencil shadows. But still could do with a proper explanation in terms from the art team or something.

He's commenting more on the fact that 99% of the time, people buying models don't bother to optimise them for Arma in any way, because they don't have proper understanding of creating 3D assets for games and just assume that all models are created equally. Models available from 3D vendors are by and large not game-ready since they are normally targeted at other forms of media (selling 3D assets to TV production companies, print media, marketing companies etc. is far more lucrative than selling to video game hobbyists), but most buyers are not aware of that. The end results being shittily made addons, and a lack of interest in learning how to do things properly (people just want a model ingame to consider it "job done"), is what is bad about it. There are people around that use models from Turbosquid but make significant effort to make them work for Arma - Firewill's planes and NightIntruder's CH-46 are examples of that. They don't get heat for it, because they approach the task with prior understanding of how a game-ready model should be, choose the projects wisely (not just get whatever model looks coolest in the store), and actually put significant work in to convert the commercial model to something that fits Arma's requirements. Improving upon the original model in significant ways with newly built geometry.

Those are the proxies for various things such as the gun muzzle flash, crew positions, flags. They act as reference positions for loading other .p3d objects on to the vehicle model in game. they should all have selection names like proxy:Path\To\Some\Other\Model.00N You can remove them while you're editing the model, but you'll want to copy at least some of them back in from the original .p3d once you're done editing the vehicle. Otherwise some of the vehicle's features wont work.

No, like I said it was the .455 Auto ammunition from the Webley Scott self-loading pistol rather than the .455 Webley revolver https://en.wikipedia.org/wiki/.455_Webley#.455_Webley_Auto

Was in WW1, so it was actually before the RAF was formed (Royal Flying Corps at the time). They were designed to use the same ammo as the Royal Navy's Webley & Scott .455 IIRC the RAF did order more M1911s in .455 after 1918 as well though.

I don't think anyone has worked out the exact hard limit yet, but I've seen a few people say they were able to import models with 100,000 faces But the practical poly limits for models in the engine haven't changed. The fact that fewer Tanoa buildings have interiors in order to improve performance, and the fact that LOD transitions appear at closer distances now is testament to the fact that the game will run like shit if you have lots of high poly models and no optimisation of the meshes. The Apex weapon models are heavily optimised in their first resolution LOD compared to their view pilot LOD, and seem less detailed than previous Marksmen DLC weapons. So don't be too excited by it. You should probably still aim to build things that are comparable to existing Arma assets in terms of detail and generally sit under the old vertex limit in their base configuration. The extension of the poly limit seems to be more about removing the need for proxying a few extra ancillary items and weapon turrets on vehicles (especially the viewGunner LODs), rather than for adding shit loads of detail to the overall mesh.

Think you need the step and xStep values to be bigger than zero https://forums.bistudio.com/topic/165004-walk-animation-dont-move-the-player/

Doesn't mean they're not using optimized shadowVolume LODs meshes instead of the full resolution LOD. It just means they're using shadow buffer for all terrain objects now for the sake that the new lower resolution shadow buffer performs better than CPU shadows. This was mentioned in a SITREP, and they explained why weapons, vehicles etc. still use GPU stencil shadows. Shadowbuffer can be generated from LODs specified with the sbSource named property https://community.bistudio.com/wiki/Oxygen_2_-_Manual#Properties(scroll down a bit) Trees etc. used hybrid shadows to generate shadowbuffer for the foliage alphas even if they have a shadow LOD but the hybrid property generally isn't used on objects unless they have a specific need for it. Old example of using hybrid: https://forums.bistudio.com/topic/104643-camo-net-shadow-lod-problems/

That has been fixed internally on the HMMWVs. There was an issue with autoCenter in the models

Yeah, zeroing2 is obsolete. Use zeroing.1 to animate stuff for the second muzzle now. https://community.bistudio.com/wiki/Model_Config#Animation_sources

Yes, the Specter 2D/3D variants are handled by the preference menu now. Rather than having separate 2D and 3D entries for each optic in the Arsenal like we did before. Switching the optics type preference between 2D and 3D (and PiP in the case of the ACOG and M145) on the fly during play works, as demonstrated in the video.

Euroscepticism has been pretty strong for well over a decade so I don't think it's possible to simply say it's a protest against established politics. Subsequent governments have failed to tackle immigration though, in spite of many promises to bring numbers down that end up rising every year. So I guess in a way, nobody has seen the two main parties (Labour and Conservative) as being truly interested in dealing with an issue that has concerned a lot of people (rightly or wrongly), for many years. However, people don't vote for third parties because things are too partisan between the main two. Nobody wants to split the vote and concede victory to the party they definitely don't want to win. So party policies don't have to run extremely close with what the electorate would say they want. As Prof said, the Tories ended up calling for a referendum because they didn't want to lose Conservative voters to UKIP the same way as Labour ended up losing voters in Scotland to the SNP, and risking a loss or another hung parliament where the SNP would side with Labour. The fact that the leading ministers for both parties supported Remain coupled with the results, may also be seen to be affirmative of how London-centric a great many people in England perceive politics here to be (and other parts of the UK though they'd determine it as England-centric in as a whole, not just London-centric specifically). I'd say though, that rather then them having any actual belief that the Leave vote is a swipe at established politics; some Labour MP would suggest that changes at the top of the parties are needed because within that party there is a split over the leadership of Jeremy Corbin and people are maneuvering to remove him. I don't know which MP you're referring to though, so I wouldn't know where they stand on Corbin's leadership.

I see this being trotted out a lot at the moment, but the claim that seems to be going alongside it: that people voted to leave without knowing what the EU was. Well, that seems counter-intuitive to me. It's an acknowledged fact that the older populace was the driving force behind the Leave vote. But the older populace is never the computer-literate crowd, searching Google at 1am. Seems far more likely to me that the "What happens if we leave the EU" search was driven by younger voters who, as the results were coming in and being broadcast in the middle of the night; were encountering an outcome from the referendum that they did not expect. Voter turnout in key areas that were expected to see large votes for Remain (such as Newcastle very early on) failed to materialize in many cases. And to me that smacks of many young people being assured in themselves that "Leave" stood little chance, so had no contemplation that "BrExit" might be a very real thing today. i.e. they've been the ignorant ones. Anyway. I voted Remain; but for economic reasons. Not because of principle confidence in the EU's ability to run Europe. I sit here looking at the "Leave" result and the celebrations, regrets and lamentations among those who voted, and can only think: - If there is a majority from those that sincerely wanted to leave, when combined with those who are concerned about the EU enough that that they conflictedly decided (includings those who have expressed regret now) that maybe life outside the EU was worth a gamble. - Then there are people like me who only really voted to remain because the economic risk was so high. And that the most vocal lament from Remain voters today reflects that so many voted purely on the basis that they believe it is racist/xenophobic to hold anti-EU beliefs (because of the "Leave" focus on immigration). Then, how many people are actually left among the "Remain" vote who truly believed in the institutions of the EU and had confidence in its current form? The percentage who are actually fundamentally happy with the EU must be very small. I think lack of transparent understanding and subsequent enthusiasm for the long-term goals of the EU is commonplace across Europe. Yet so far there is nobody looking seriously at reforms that might bring greater parity with what the people want, rather than what Brussels and the economists want from Europe's future.

You have to declare it as being compatible with weapons. Either you have to manually add it to the compatibleitems[]= array for every weapon you want to add it to - which is a pain in the arse and means it wont be compatible with most custom community weapons. Or you can make it compatible with Joint Rails which is much easier, but adds a CBA dependency to your addon. The latter uses the improved class compatibleItems system rather than the old array method that BIS use. Example of manually adding compatibility for a custom optic, in this case for BIS' 4Five pistol: class CfgPatches { class RKSL_PMII_4Five { units[] = {}; weapons[] = {"RKSL_optic_PMII_312_sunshade"}; requiredVersion = 0.1; requiredAddons[] = {"A3_Weapons_F_Pistols_Pistol_heavy_01"}; }; }; class CfgWeapons { class Pistol; // External class reference class Pistol_Base_F : Pistol { class WeaponSlotsInfo; // External class reference }; class hgun_Pistol_heavy_01_F : Pistol_Base_F { class WeaponSlotsInfo : WeaponSlotsInfo { class CowsSlot { compatibleItems[] += {"RKSL_optic_PMII_312_sunshade"}; }; }; }; }; Example making the scope compatible with all weapons with a Joint Rails dependency: class CfgPatches { class RKSL_PMII { units[] = {}; weapons[] = {"RKSL_optic_PMII_312_sunshade"}; requiredVersion = 1.0; requiredAddons[] = {"A3_Weapons_F","A3_Weapons_F_Acc","cba_jr"}; }; }; class asdg_OpticRail; // External class reference class asdg_OpticRail1913: asdg_OpticRail { class compatibleItems { RKSL_optic_PMII_312_sunshade = 1; }; };

Can I ask how the goggles know whether to have the raised or lowered position now? Is it for specific "NVG" and "TI" entries in the array, or just any index > 0 in the array? I did some experimenting before the Apex content came out and concluded that I'd be super happy if BIS could add a way to have "Normal" vision mode that allows goggles lowered. Would this be possible with a visionMode[] = {"Normal","Normal"}; entry in the config now, or would it require an extra vision mode option to be programmed?

I think this one is fixed on the current dev Branch actually. I noticed it yesterday on the Preview branch, haven't seen it when I switched back to dev today after the weapons and vehicles were included.