Sunday, 28 June 2015

Almost 20 years of Longbow

After another project hiatus I feel brave enough to come back and tackle some of the problems that defeated me early this year. I'm posting back on this blog as I've been unable to re-gain my editor access at since last Holiday season.

Combat-Helo relies on a flight model that performs well across all modes of flight. Somehow I'd introduced an error into the vertical stabiliser forces, I know this is to do with my conversion and/or how our forward vector points down the minus z axis (as a result a lot of code relies on this arrangement). The FM doesn't fly well with tail forces enabled. It proved near impossible to debug it adequately and plans to release an early beta were thwarted.  After waiting for these problems to fix themselves (they didn't oddly) I do what I usually do when and went off and did more interesting things and hone my professional testing credentials on things I can fix. Inspiration would surely strike? (It didn't)

My plan B (other than abandoning the entire thing and re-building it in another engine that is developer friendly) is to write a testing framework around the flight model and a UI to step through it while in-flight. And unit test the crap out of it. There's the other glaring problem that the 3D engine is no longer supported, no source is available to fix problems in the physics integration. A very difficult situation.

I have tinkered with migrating to other engines but it all comes down to the same blocking issue. The 3D assets, the kicker is I don't have any of them in the original MAX format. It's a matter of taking the Leadwerks GMF versions, converting to OBJ (which looses a lot of information such as groups, smoothing) then massaging the model in a third party editor to export as an FBX.

It should be possible to automate that process using the Leadwerks GMFSDK and an FBX SDK, a dedicated tool to export them. That's the sensible thing to do.  However I'm still driven to complete the existing code-base and get a demo working well enough to sit on it's own. Vastly reduced in scope but it's time to move on to something better. We started in 2009 (yes check those early blog entries) and next year is the 20th anniversary of Longbow 2. It would be nice to mark it somehow.

Development resumed on the old Leadwerks code in April, all I've got to show for it is a build with a broken flight model and Hellfires that miss most of the time unless you find the sweet spot. Not good enough.

Source code for the project now resides in a secure remote GIT repository. I've been going through the source cleaning as I go, this makes collaboration easier. The caveat with GIT is the problem of large media files. There are set size limits. Trialling the LFS (large file storage) service which seemed fine until it tried to deal with a 384MB raw height-map that wan't used by the project but left from a previous terrain conversion. It caused the whole thing to choke beyond repair and I had to clean out my local repository and start again.

Wednesday, 29 October 2014

Ye Olde Leadewerks blog

Long time since I posted an update on my personal blog. And ironically it's about YET another blog, but this is devoted to old school Leadwerks which is still fun and extensible. I can't promise regular updates as I'm usually busy elsewhere. But there will be something useful for Leadwerks 2.x users in every post.

If you like Leadwerks 2 and all things Blitz and C++ then it'll be worth following. There will be posts featuring engine mods and shaders. Stuff we used in Combat-Helo and distilled knowledge from the community.

Tuesday, 4 February 2014

Outlining and other 3D UI tropes

The all new official web site for Combat-Helo (are we still going with the hyphen?) will launch real soon (this week). All future blog entries will be posted there, this blog will serve as a personal and more technical insight into development for other projects. Where is this awesome new web site? I'm not telling as it's not live yet (check back Thursday).

In the meantime I've been working on 3D user interface code that will work with Leadwerks 2.51 (depreciated). All the bits needed to enhance the GUI such as outlining of selected entities, see the Hellfires in the free-roaming arming mode shown in screen-shots below. 

Once upon a time creating this kind of effect was easy in OpenGL using stencil buffers, or some flipping normals trick combined with object rendering order. However in Leadwerks, complex scenes require a little extra work; the creation of a new World, new Buffer, a Camera, post-shader. Hmmm, exactly the same way the cockpit is rendered.

In the main draw method, once we've done the bulk of the scene mesh rendering, we render this new outline world as a post process prior to the deferred lighting stage. The output buffer of this outline world is then alpha-blended over the Leadwerks gbuffer, a frag shader processes the output buffer adding the highlight colour.

Once that modification to the main render method was completed, we can outline any world entity by copying it into the outline world. This is what we do during arming or other menu mode. We had to use EnityUserData() on created outline world entities to maintain links to the main world entity (you can't parent between worlds without some kind of book-keeping).

In addition to this we needed 3D UI objects to facilitate area marking for things like triggers and unit formations during editing. So I built a class that generates the geometry to create these animated "force-field" looking objects that form fit to the terrain using a feature of the built-in vegetation shaders.


Creating groups of units and assigning them as primary targets, or special formations can be done by area rather than linking at design time. Simply throw down a bunch of vehicles in proximity and ring-fence them as a unit. The Formation manager will try and do the rest. The game portion is starting to come together, I'm sure we can find ways to improve WYSIWYG mission editing.

You can add a dozen oil storage tanks, wrap them in a target-zone and set the zone as a "primary target". Destruction of a percentage of primary targets are considered for mission evaluation. We also have secondary, tertiary and specials.

Sadly what you don't have is an infinite supply of ammo...unless you enable cheats and type "/rearm" into the console window.

A Call for Arms

We're looking for new models, specifically more radar and SAM units such as SA-10s, Rapier batteries etc. to make deadly spaces (the Hills have Eyes, or so I'm told). I'm still working on the ASE demonstration video, I stopped to add some features to the avionics when recording but I'll pick it up again this weekend once we've launched the new site.

I would love to add more threats to our release, if you've been in touch already then expect an email from me in the next few days. With so much going on I'm behind as usual.

See you at the new site at in a few days and clear skies. You have control.

Sunday, 26 January 2014

Roadmap - 2014

Another down to earth post about the Combat-Helo road-map and release schedule (minus an actual date). We're going to adopt a rolling release schedule, as such we feel it's necessary to tell you what to expect and roughly in what order.

What is a rolling release?

A rolling release is a continuous development model. We release the so called 'early access' version of the software to respond to feedback and fix issues that inevitably crop up on a wide hardware footprint. Also it means you get some hands-on time before we roll out more features.

First, welcome aboard this FlexAir blog post. When we release the first version, this won't be the complete version as we're skipping traditionally important things like; infernal (sic) beta testing, quality control, pre-flight checks, maintenance, radar. All the things you'd expect even from the tightest of budget airlines. So it's important to understand that crashing before you reach your destination is a real possibility. Should cabin suffer a sudden exception, a bug report form will deploy from the overhead compartment. Thank you for flying FlexAir.

I'll add that exceptions in the code are actually pretty rare and when they do happen it's typically some configuration or engine/driver issue. If there's anything that's a game stopper we will move heaven and earth to identify, fix and roll out into another release ASAP.

"Gunnery" Roadmap - 2014

Gunnery is a discrete game. It is considered a single title that is a prequel to future releases under the same IP. Gunnery will be release with a free-flight mode, limited number of pre-set missions and the Gunnery range map. When you buy Combat-Helo during the "Early Release" this is what you can expect.

  • Release 1 (Early Wingnut Release)
    - Gunnery Range Map
    - Free Flight Mode
    - 30mm Cannon Enabled
    - FFAR Rockets
    - Radar Hellfires (laser to come)
    - Static Targets
    - Player stats/score-board
    - Active SAMs
  • Release 2 (The Apologetic Release)
    - Laser Guided Hellfires
    - Autopilot enhancements
    - Bug fixes and Enhancements
    - Mission Files
    - ATC and AI traffic
  • Release 3 (Milestone Release)
    - Multiplayer enabled
    - Training Mode (Instructor Pilot audio and training missions)
    - Bug fixes and Enhancements
    - Additional "NTC" desert map
    - Radio stack, battlefield "modem" and TSD upgrade
  • Release 4 (Now We're Cooking Release)
    - Scramble (Dynamic mission type)
    - Bug fixes and Enhancements
    - Mission builder enhancement
    - New Map (tbd)
  • Release 5 and beyond (Are We There Yet Release)
    - Additional upgrades and new maps

Some of these items may get re-prioritized, items like AI traffic depend on availability of resources for coding and testing.

A new official website and forum will be launched early Feb and will be the defacto stop-off for solid information. This blog will remain for unofficial musings, ramblings. Eventually winding down.

Beyond "Gunnery"?

The engine we started with, Leadwerks 2.x has been discontinued, support is now limited. We don't anticipate creating a major follow up using this engine. We will however release our original North West Afghan map featuring armor in a hypothetical crisis situation under the banner "Combat Helo - Deployment" (or "Crisis" for want of a better title) and we will look at adding the COIN simulation as an update to this.

We have been evaluating new engine technology and found at least two that are both amazing in their own way and both totally suited to the larger theater scale we want to bring to Combat Helo. We'll jump those fences when we reach them.

If you have queries about any of this, post a comment below or send us a message on Facebook. We'll compile a Q&A sheet.

Monday, 20 January 2014

Official world on DCS World

I see more and more requests for us to create the Apache Longbow for DCS World. For sake of clarity we thought we'd put out an official response.

For now, the answer is simply no. We talked about it, thought about it and started to look at what might be involved. In the end we had to draw a line and decide what we wanted to do.

It is not for lack of interest on my part. And I don't want to fully stamp on it as a future possibility (if we're here in 3 years time and someone still hasn't stepped up to build one). If someone with experience will take it on I'm sure it would be a superior study aircraft given the pedigree and changes in the Edge engine. We don't yet have the man-power or experience needed to create a quality aircraft for that series. With the unfortunate delays in production of Combat-Helo, taking on yet another project would be near impossible without skilled individuals with suitable experience. Even with crowd-funding it's hard to justify it as a spin-off project as it doesn't fit in with what we set out to do. Dave and I set out to create the kind of game nobody was making anymore which is the stand-alone combat-sim. And for all my talk, I've discovered that it's not as easy as it sounds.

Scott Elson (former EA and Microprose senior developer) reminded me way back that creating a flight simulation was not simply a matter of making an aircraft and flying it around. There's a lot more work involved, weapons, audio, UI, animation, events, triggers, file systems. All the glue that holds a stand-alone game together which would stretch a five man team, never mind two.

David and I began work on this in 2009. This started out as a full-time project and when the money ran out we had a choice, carry on as best we could working in our free time or abandon it. It was never going to be ideal but we carried on regardless, I had to trim features as time went by to pick up the slack to keep the end in sight. Eventually my output was struggling to keep pace with David's output and problems with our choice of game engine threw a spanner into the works. You don't find some problems in a technology until you really stress it, our issues were not apparent until much later in the project. Hindsight is awesome.

Recently, volunteers have allowed us to push past some of the older problems we had with Leadwerks 2.x. which has been fantastic (again props to Macklebee, Klepto and Shadmar who contributed). Technology has moved on and our ambition while humbled, stubbornly remains. It was clear we couldn't deliver the game we set out to make, the dynamic Afghan campaign will for now remain incomplete on our hard-drives, but from a future development point of view we now have more options available to us. The game development landscape has changed substantially thanks to crowd-funding. This is something we're currently preparing for. But more on that later.

We're still committed to a long road-map for Combat-Helo starting with the release of "Gunnery" in the near future (we're working really hard to release this quarter). Our focus is still creating an echo of the combat simulation scene that existed on the PC during the mid-90s. Building on foundations we've laid down. While frameworks like DCS World are an ideal platform for pure simulation, it's still a very different kind of experience which is well catered for.

Combat-Helo will rely on dynamically generated missions intermixed with narrative and maybe some comment. It has never just been about the aircraft. We are creating foundations for a worthy successor to legendary games like Janes Longbow. It may take a few iterations to perfect and we mail fail but it won't be for lack of trying. To carry us on beyond "Gunnery" we'll be looking to you, the community to help us take it further. What direction that takes depends on what happens in the next few months. Whatever happens, 2014 will be an exciting year for us and I hope you'll keep the faith and join us for the ride.

- Signed Richard Hawley, Director Tricubic Studios Ltd.

That done....incoming fire...deploying countermeasures!

Saturday, 28 December 2013

Happy Holidays

I somehow managed to hit "publish" instead of save earlier. I apologize for the curiously abrupt email if you are among those that elect to receive updates in this fashion.

Let's try again.

Yesterday I was in my living room having family time, up on our wall sized projector screen I was browsing YouTube for a number of classic helicopter games. These included DI's Apache Longbow, Gunship 2000 and Jane's Longbow. Since it's the end of the year and it's the popular thing to do a little retrospective, here's some Longbow themed naval gazing followed by a little project update.

Let the old games begin

Starting with Gunship 2000 from my old chums at Microprose. This was a sequel to Gunship, offering a choice of three helicopters, the Apache, Blackhawk and Cobra.

Gunship 2000 "Do a barrel roll"

I remember having the Amiga version, only experiencing the superior PC version later. Some of you younger folks might be looking at this and thinking it's total crap. But imagine you hadn't seen a 3D patchwork landscape with filled polygons before. Ever. Some simulations had a basic mountain (pyramid) or two but this was starting to look interesting. There's a lot going on, apart from the hideous teeth grinding sound effects of the day. There's a battlefield, mission structure and a menu screen that my step-son said looked like "Theme Hospital" (meaning the clinically white base office).

It's hard to go backwards. You can't go back to these games anymore than you can revert to using a one button joystick. Evolution of game controllers and the complexity of combat simulations go hand in hand. Early versions taught us the language of these games and a generation has grown with them.

We have adopted more complex controllers and games. We demand more from them, expect more from them. At the same time we appear to have lost what was fun about these games. It's not not to see whimsy in the 'action themed' intro for Gunship.

The Nintendo Wii was successful as it reset the clock for gaming. Everyone including grandma could pick up the controller and understand it. You didn't need to be taught how to use a joypad or have any previous gaming experience. Is it possible to reset the clock for combat simulations? Apparently so, we are already seeing this, and they have proven to be very successful examples. World of Tanks and War Thunder are not deep combat simulations but they are teaching a new generation about the language of military simulations. A proportion of this generation will want more complex experiences, or at least different ones (we're so easily bored today).

Moving on....

Post Gunship 2000 in all it's VGA glory we had Digital Intergration's Apache Longbow. This was more like a real helicopter with authentic looking symbology. The format for Apache Longbow and the successor Hind remained mostly unchanged in games that shared their DNA, Enemy Engaged - Apache Havoc and Comanche Hokum.

Apache Longbow contained a training scenario which is similar in concept to Combat-Helo Gunnery. Hand on my heart, I swear this was unintentional. It was Dave's idea, maybe my subconscious remembered this aspect of DI's game. This is a video game approximation of a live fire exercise. A video is presented below. The fidelity of the flight model still stands on it's own. Smooth.

DI Apache Longbow Training Mission - Fort Hood

DI's games were more serious, throwing out the jingoism and had more in common with current dry simulations. Less game, more pain. Today, most of the hi-fidelity simulations come from eastern Europe and out of Russia. And it has to be noted, the best middle-ware too. Something to do with the higher education system, I just had a Russian friend stop by for two days (he's currently attending a game degree course at Abertay university and something of a math prodigy). His description of how they teach mathematics in Siberia made the hairs stand up on the back of my neck. It sounded more like "know everything" or you're out, only the mentally fittest survived. He then proceeded to diagram an algorithm for personal happiness which I totally failed to comprehend. Must be a Russian thing. Just give me some cheese and a glass of Port (I'm easily pleased).

Jane's Longbow 2

Time to review some Longbow 2 footage. A great game in the day (1996/7) and one I continue to reference as the ideal balance between difficulty and replayability. Sure it's lacking ground detail. Everything we would love to fit into Combat-Helo is here. This is our aim, a stand-alone simulation game. While we'll have some basic live fire exercises for the firing range, I want to squeeze in a "point-to-point" scenario, a random placement of air defense units and a primary / secondary target. There's a nice area of hills on the gunnery range ideal for this.

Here's a trip down memory lane, the "instant action" mode in Longbow 2.

It still contains a good balance of difficulty and replay-ability. The multiplayer (when it worked) added a huge thrill. Dual-seat co-operative play, multi-ship flights and different mission roles. Plus you were never very far from mission objectives. Campaigns could be played cooperatively spread over days or weeks.

The avionics of the Longbow were significantly reduced in complexity, even allowing missions to be mostly flown on autopilot. The rich pool of radio transmissions that played in the background from both ground forces and AI flights made every game a rich "combat" experience. Even if you just stat idle at the FARP you were reminded that there was a real battle going on out there. And you mattered. You could go and find those guys sending out those maydays and help them out. Or not. It was up to you. So much depth in such a "small" package, not many combat simulations today have been able to come close to this. Some games script it, but it's scripted and it shows.

Hardware Of The Year - The Year of the Oculus?

The most talked about hardware of the year has to be the Oculus Rift. Despite it not yet available as a commodity item everyone was talking about it. Being a victim of VR technology in the 90s I'm still on the fence. Convenience still trumps quality in all sectors and the reasons why VR didn't explode 20 years ago have not changed that much. I suspect like Kinect, it will sell like hot cakes but not get much use in the home.

I keep getting emails about Oculus support in Combat-Helo. Short answer, no. That's mostly down to Leadwerks Engine and not having a dev kit. I don't see a compelling reason to get one for flight simulations, when I play a sim I'm often looking at keyboard and switches. I own a MIG welder and welding goggles for really serious PC repair (and bodywork), it's a pain having to flip them goggles up and down all the time. Guess I need them goggles with an automatic LCD plate. I've seen a simple WWII sim working really well with these so I guess it's down to what you can comfortably learn to do on your HOTAS.

We've been looking at Unigine recently with a couple of other engines for a post Combat-Helo Gunnery project which do support Oculus. If it's easy to do we'll do it. Simple as that.

CastAR from Technical Illusions is a slightly different story. Often overlooked as it's not as sexy as VR, despite being fully capable of VR. With head tracking built into a simple pair of glasses and two micro-projectors your natural vision is not impaired, you can see all your keyboard and joystick controllers. Making your visible screen bigger is as simple as adding more retro-reflective material around your cockpit space, painting MFDs onto a physical cockpit layout. I'm just spit-balling here as I've not seen how good the native resolution is going to be. MFDs will need 512x512 resolution minimum and has to be readable from 2 to 3 feet away.

NaturalPoint should take a tip from these new Kickstarter projects for their TrackIR consumer division. My LED Antlers have been laying withered and broken on my desk for 2 years. Hear that NaturalPoint? How about adopting some new technology that's compatible with your DLL? You can buy these things called "accelerometers" in chip form that cost next to nothing. You can ditch the whole IR baggage and call it TrackER.

In terms of hardware released this year the LeapPad was both exciting and disappointing. LeapPad is a small mysterious black rectangle that sits on your desk with nothing but a micro-usb cable running to your PC. It uses three IR LEDs and two angled cameras to read gestures in it's field of view. The smart stuff is all done in software so it's a fairly inexpensive add-on. My experiments for integrating it into a simulation cockpit were not great. Limited by it's field of view and only from below, basically pointing a finger at an MFD was all I managed before giving up in frustration. As a virtual pilot I want to hold my hand in a vertical position to perform hand gestures, rather than a horizontal position for playing air piano. To do that I'd have to mount the device on a nearby wall or stand.

Saitek made us raise a collective (no pun intended) eyebrow with the announcement of a new HOTAS. The X55 pictured below.

I don't have the inside scoop on this, it was a real (but nice) surprise. Since I LOVED utterly the metal build quality of the X65F but hated the force sensing (for helicopters they don't work). So I've been using my tried and tested X52 just because of it's movement. This X55 looks like it's the best of both worlds, the same flexible all-round no cross-centering spring movement with the metal build. It comes with extra gizmos that are perfect for manual start-ups, adjusting cockpit lighting, symbology adjustment.

Work update

Keep running into silly things that slow you down. Recently the tail wheel assembly of the Apache. We got the springy suspension working but the model hierarchy for tail wheel articulation isn't correct. The crew limbs couldn't be animated as we need them in a T-pose to apply the bones. Also I think the crew scale is slightly wrong. Check this pic out....

Crew scale against cockpit
Winter is coming. One of the lasting impressions I have of my early combat gaming comes from the original Microprose M1 Tank Platoon, the cold winter landscapes juxtaposed with flying HEAT rounds and burning wrecks. It provided a welcome change of scenery. A few weeks ago I started applying a snow shader that uses the up normals on a model to apply a snowy texture. It's a lot of work to go through every model material definition to update them. The results of this shader applied to the M1 tank is shown below.

Once applied you can set the snow coverage in real-time from 0% to 100% by setting a shader uniform. This almost never made it as a feature.

It presented a small problem for vegetation models since Leadwerks creates billboards which are generated on a first use basis. Once you apply a snow texture to a model you need to update the billboard. Fortunately you can get access to Leadwerks vegetation billboards like this (where the index 0 is the vegetation layer 0-15).

TTexture tex = leadwerks.engine.TVegetationLayer.billboardtexture[0]

With a little shader magic this allows real-time manipulation. You can see a glimpse of these experiments in an unlisted "Thank you" movie I sent to the team on Christmas day (I'll link that at the bottom of this blog). Oh to have an engine that does all this stuff for you.

Right now we're wrapping up additional game effects such as flares. Mack came up with a pretty tight working concept after watching some of the impressive air show displays where Apaches equipped with flare dispensers go 4th of July crazy. From these videos he got a feel for how they behave in flight, coming up with a working prototype. Just like our 30mm shell ejection we're using physical bodies for flares so they bounce off objects roll around. Pretty neat. The Leadwerks particle effects won't win any awards.

So while I'm supposedly finishing off the ASE gear, I've also been tinkering with the landing gear dynamics that Mack fixed up. It's easy to get distracted when nobody is looking over your shoulder.

This devs test map - random screen-shot
Next on the list is adding the AI to get AIRDEF units to search and destroy. Mack is back on figuring out how to tackle the damage modelling. He's got some ideas on swapping out child objects in the vehicle hierarchy. It's the sort of fudge you get used to with Leadwerks.

By way of a thank you and Merry Christmas to the team I sent them this short music video containing just some cockpit g effect footage and tree billboard tests.

Happy holidays from "Mossie" (or Mossy if you're from the wetlands). You'll have come across this chap on our Facebook page.

And happy holidays from me and the rest of the team at Combat-Helo. Here's to finally shipping in 2014.

Sunday, 15 December 2013

ASE, Master Warning and Master Caution

Sales brochures are always a good place to start when you want some screenshots of avionics (that and Google images).

ASE or Aircraft Survivability Equipment - is the name given to a suite of devices that interact, they act as the ears of the aircraft. Our combat game mechanic relies on paranoia, avoid being painted by SAM radar, pinpoint enemy launch systems before they spot you, and eliminate them using your stand-off missile capability.

Listening for enemy radars and presenting this information to the crew is the job of the ASE page. This has inputs for controlling active decoy systems such as the IR Jammer, flare package and radar jamming equipment.

ASE page from sales brocure
My version is looking similar, I can't quite fit everything on as I made the decision a while ago to increase the symbol size by 20% to make MPDs readable in the upright position for typical screen resolutions. The Autopage logic is going to be fun but perhaps not as much fun as the flare dispenser programming. The logic for flares should allow for salvos, interval between salvos and flares per salvo. This should look awesome with the Leadwerks engine lighting. I'm guessing flare programming ends up on the ASE UTIL page in concordance with other page types.

My take on the ASE Page

ASE alongside AIR SURV mode, no rings

I can't tell what the range circles are in scale. Will figure that one out shortly but I'm going to go out on a limb and say 2km per ring....or it's arranged by type with the biggest threat on the inner rings. Easy enough to try both. After trying the ASE with range circles alongside the Air Radar mode it looks confusing. For now I'll leave the rings off the ASE page.

To test the radar detection I'll have to script some radar entities, we don't have any model assets for radar units other than the SA-9. They may end up as deadly green cubes for now. The symbology appears to be the same as the TSD, easy enough.

This just leaves what to do with the symbols for different states. Time to crank out Longbow 2 to get a refresher. I'm looking for:

  • Radar non active
  • Active Radar (uh oh)
  • Tracking Radar (about to have a bad day)
  • Launch Tracking (bad day)
  • Laser warning (LWR)
  • Small Arms Acoustic signature

I realise the final one is a bit of stretch but I'm interested in how well this works after reading about UK Apaches fielding acoustic sensors that not only picked out small arms fire but determine the direction. If it's like equipment tested elsewhere it should be able to determine what weapon is being fired. Useful in a support role. And perhaps too much like voodoo.


More warning tones have been added to the cockpit which are part of the WAC warning and caution system. These audio warnings go off at a regular 5 to 10 second interval until you acknowledge them by pressing the appropriate warning lamp. The Z and X keys have been mapped to the warning and caution ack switches accordingly. If it does it's job it'll have you swearing at it every time it goes off, something the wife can join in with too (and I speak from experience).

Very happy with the team who are finding time to squeeze in what they can especially over the holiday period. Sérgio has been working hard on a new web site and forums for Combat-Helo. No eta on when it goes live. When it launches this blog will be migrating to the new site. To date, my personal dev blog has had nearly half-a-million views, golly.

I'll be sad to see this go but the new site will have everything under one roof. News, updates and forums. Not just for Gunnery but everything we have planned beyond.

Alpha testing will start soon. That's quite a big thing for me. It represents letting go of something I've been really protective about (paranoid even?) for years. More than a few nights crying on sofas with the stress of juggling workloads and expenses. Looking forward to getting a chunk of outstanding work items done over the holidays. If we can cross off a few more big ticket items early next year we can then focus on content.

And now for something slightly different.

Unigine Helicopter Demo

I cant post a blog update this week without mentioning the other amazing thing I saw. Everything I ever wanted in an engine. This week Unigine released video footage from their CIGI compatible Rescue Helicopter demonstrator at I/ITSEC 2003 Orlando Fl. The level of detail is pretty amazing. A physical cockpit with instrument integration, multi screens (client server based) and a totally immersive simulator. Just watch it. And if you've watched it already, watch it again. Kudos to the team that put this together.

The Washington state dataset is about 60GB. Yikes. Not that you really need this level of detail everywhere. I've had a look through it when playing around to take some screen-shots and you can tell there's a lot of data that's been imported from geophysical databases. I look forward to seeing a GROME plugin to export terrain tiles for this engine in the near future. It looks like it should be pretty compatible in terms of workflow and how the tiles are exported. I'm guessing World Machine was used (it was used in the creation of The Valley demo and benchmark).

Letters are powerful. For example, the letter "H" can attract helicopters.*
* Joke stolen from Milton Jones