[C3PO]
How far advanced down the development line is SP4 now?
[Chris Carter]
SP4 is now a great deal closer to release and will be in its 4th Beta shortly. The amount of fixes and integration of various works has undergone a quantum leap in the last few weeks, due to the formation of a specific data editing group, who have been processing the enormous number of reports generated by the beta testers. We also have our CVS repository in full use, which is enables several key developers to maintain data consistency and a rigorous change control procedure.
We have also had a number of new and past data developers join this group and their contributions have helped enormously. I would also like to mention Gunner , who has taken on the huge task of coordinating the entity testing and has created a superb web site, which allows methodical testing to take place and the results to be fed back to the data group so that they can fix any problems encountered.
The next beta will be evaluated for completeness and it is hoped that we can then produce the first release candidate in a very short timeframe after that. A further amount of "polishing" work will be done and then the public release will occur.
[C3PO]
What improvements over SP3 can we expect to see?
[Chris Carter]
Whilst the functionality of SP4 will be essentially the same as SP3, due to the prohibition on editing and recompiling executables, the data that the executable uses will be greatly enhanced over SP3 in terms of scope and content. The first thing that will be readily apparent to those who develop and play threatres, other than Korea, is the large amount of new entities that have been added in the last year or so. This includes aircraft, ground vehicles, some new buildings, alot of naval vehicles and some new weapons.
A great number of "placeholder" entities in SP3 have been fleshed out with proper data such as loadouts, improved and individualised flight models, making SP4 probably the largest and most complete database of military hardware available in a civilian flight simulation. The amount and detail that has gone into Falcon4 over the years is staggering from all the groups that have been involved in its development.
A large program is also underway to add units (eg squadrons, batallions etc) for the new entities and these should be of interest to the campaign makers for other theatres.
The casual user should notice immediately that many of the SP3 models have undergone an overhaul and that several aircraft have many of their variants implemented. There has been a massive amount of work generated by the modellers and skin artists over the last year and I believe over 1,000 of them have been included in SP4.
The airbase group's work has also been included and although its still work in progress, SP4 users should notice that airbases are richer in object population and some of the new building models will be seen eg. Hardened Air Shelters, new ammo dumps, control towers and perimeter fencing.
Also making its debut is the first iteration of the airbase parking data, which will enable parked and taxiing aircraft. This will have a marked effect in campaigns when combined with the "realistic attrition" setting, as you will often find enemy and friendly aircraft waiting on the ramps either before or after their missions. These aircraft can be destroyed as "targets of opportunity" and will then be deducted from that side's inventory, along with the weapons they were carrying.
The object density has also been given a massive boost for SP4 with towns and cities containing more than a dozen buildings. An explosion in tree population should be more apparent everywhere. This should enhance the environment for those who like to fly low level and aids with distance and altitude perception. To give an idea of the amount of work that has taken place here, the original objectives in 1.08 contained about 5,000 feature entries. The count for SP4 is now over 28,000 and the majority of these have been hand placed by myself and a tireless team of developers who don't seem to know the meaning of the word tedium.
I also did alot of experimentation to add widespread forest cover to SP4, however, the stuttering and framerate hits that these generated were unacceptable even for a high-end PC due to the number of tree objects that have to be tracked in the campaign engine and rendered by the graphics engine. I will keep experimenting but its doubtful we will see large 3d forest cover in the SP3 exe.
There are also weapon enhancements in all arenas, with an overhaul of missile terminal phase characteristics, damage values for various ordnance and the objects they may be attempting to destroy, together with some adjustments in ground unit engagement variables to stop entire batallions firing at single air targets.
There are many other enhancements to other areas such as the ATO, which should be noticeable as soon as you undertake a campaign in SP4.
I could go on further but as you would expect from a year's worth of development the number of changes is huge and although not as apparent as the coding changes completed for SP3, should increase the immersion factor in terms of graphics and the realistic representation of the environment and its inhabitants. I like to think of Falcon4 as an out of focus or blurred alternative universe, which is becoming sharper and more focussed as it is developed.
[C3PO]
Why has it taken so long to bring out SP4?
[Chris Carter]
There have been many reasons for the delay in getting SP4 released and here's a few of them.
Following the release of SP3, which was in essence a largely exe focussed development, the F4UT management and developers became somewhat unfocussed outside of a number of projects. This didn't mean that work stopped altogether but for SP3 the data work undertaken was closely tied to the features being coded and F4UT was a tighter knit organisation because of that single purpose. After SP3 was released the data side essentially had no firm goals or imposed deadlines to work towards, which contibuted to a certain amount of indecision in when to release.
At this time I proposed an 8 week release cycle but due to the amount of work in progress and new entities being created we found it difficult to integrate all of these within that timeframe. Especially as we had no proper change control solution in place, which resulted in all the integration work sitting on one pair of shoulders (more on this later).
The F4UT management (of which I am one) took the view that the Dev group should be closed. This was in hindsight, a mistake but one we took due to the confidential nature of the Viperops group that was being formed at the time. Had the Viperops group achieved even its initial aims, of fixing bugs in SP3 it would have resulted in a far better product for the Falcon4 userbase. Personally I think this was a risk worth taking even though it didn't ultimatley come to fulfilment.
Another factor was that it took us a long time to get a working CVS repository for data changes, that setup work and the accompanying procedures wasn't completed until April. Now that we have a workable CVS repository and change control procedure in place our build meister, Joel Bierling, can run up new beta's very quickly and it also ensures that the data guys have the latest version of files to work from. To illustrate how important running a change control environment of this kind is, I did about 3 months worth of new feature entries in the database only to find out that due to not having the latest version of one of the database files, that work was wasted. Worse than that, I had to do the 3 months worth on the features again to repair the damage (although this time from backups). This kind of problem is now a thing of the past.
During November we also had some reliable information, which I will not elaborate on but which lead us to believe that a release of SP4 at that time would not have been in the interests of the community.
There appears to be a perception on the Falcon Fora that F4UT have essentially been sat on their thumbs for a year but this couldn't be further from the truth and the delay has meant that features, which would not have been available at the beginning of this year are now completed eg. the parking data, which was finished in late February of this year.
There are other reasons for the delay but these are either too political or involve personal events, which I am also not going to elaborate on.
[C3PO]
How much time and effort has gone into SP4?
[Chris Carter]
I dread to think of the number of hours that the various contributors to SP4 and its predecessors have donated. My own contributions haven't been the largest by any stretch of the imagination and everybody who has developed for Falcon4 has differing comitments and therefore differing amounts of time to devote. The point being that, although I have probably been developing for Falcon4 nearly everyday since early 1999, I don't count what I have done for the simulation to be of any more value than somebody who say placed a single terrain texture for the Balkans, there's just more volume.
For an example of the scale of some of the projects involved, I'll refer the reader to the 22,000 additional features placed on Falcon4's objectives (that's town,villages, bridges etc) since SP3. In the early days of that project these had to be placed by hand and you had to think of all the coordinates in reverse. I populated about 100 objectives at that time, with each one taking about 4-8 hours. There was also a team of half a dozen guys doing the same thing and doing a bit of math will show how much that project took in terms of man-hours.
There are also the research hours put in by the guys on the data teams, especially when trying to track down hard to come by information for flight modelling and weapon characteristics. Having spent no small amount of time myself poring through books looking for info on Russian aircraft, (one of my pet areas) I know how soul destroying it is to go through hundreds of pages in say Janes and end up with no more information than you would get buying a plastic model kit.
To quanitify my input into SP4 I would estimate that I have spent well over a thousand hours on Falcon in the last year and I'm into my 4th year of working on it.
[C3PO]
How different will SP4 be over other contemporary Falcon releases?
[Chris Carter]
It will be different in terms of the level of completeness of various components that other releases have contained.
[C3PO]
What's being developed on the airbases side with Falcon?
[Chris Carter]
The airbase enhancements are part of a much larger and more ambitious project to enhance and fulfil the potential for the landscape, that existed in the database when Falcon4 was first released. One thing that's noticeable after flying with the trees enabled and without, is how sterile the original simulation was at ground level. The hardware capabilities at that time were not up to handling the kind of landscape that you see in contempary simulations such as LOMAC, IL2 or Flight Sim 2002. The project myself and others are working on will essentially add altitude and distance perception by adding as many objects to the landscape as the engine and current hardware will allow. We are reaching the limits of what can be achieved for say the features but there maybe work arounds that we can use to continue the progression. Another bonus with a more populous landscape is the sense of speed experienced when flying at low altitude and should suit those guys flying the A10 and other ground attack aircraft.
As an aside one thing that became apparent during this process is the woeful lack of correct scaling used in the terain textures. MPS did a good job at presenting a landscape given the texture resolutions they had to work with but if you look closely and do some calculations you can see for example that individual waves on the sea texture are about 50-60 metres wide - not very pleasant sailing conditions and if they ever became animated would cause a natural disaster on the Falcon shoreline. This isn't the most glaring error and by far the worst is the generic grassland tile used as a background for a whole series of related textures. In those tiles the texture is supposed to represent furrowed grass but those furrows are essentially 40 metres wide (that's a big plough). This results in the distances and altitudes you perceive being vastly compressed, so that hills some 30 odd miles away look more like they are 5 miles distance. Tom Waelti has done a magnificent job of providing correctly scaled terrain textures and although they won't be shipped with SP4 (as they won't work with the hardware set we are currently supporting), they are highly recommended.
The parking points were covered earlier in this interview so I won't bore the reader with another iteration.
The airbase work is still continuing and SP4 will also inlcude an expansion to the numbers and types of airbases, so that more variety is available and realistic layouts created rather than using the generic templates that exist to date. Expect more from this project in the future as SP4 only includes the work "fit for public consumption" at this date.
On a positive note there has been alot of groundwork laid down in SP4 to add much more activity to airbases in terms of what you wil see and how it operates and future airbase group products will realise those goals.
[C3PO]
Where does the Falcon name go from here...where does community modification of patches go from here, in particular regarding the situation with G2 Interactive?
[Chris Carter]
I'm not going to speculate on G2i, its capabilities or its plans. The company has a licence to develop the Falcon product and only time will reveal what results from that licence.
[C3PO]
When do you think exe changes will be permitted, or even if they will be permitted at all?
[Chris Carter]
Whilst G2i hold a development licence and F4UT have no agreement with them as a sub-contractor, or with Atari then no further exe developments will take place.
Anybody doing executable edits not only puts themselves at risk, financially and legally, but also risks Atari becoming more interested in the product than even the data editors would like. It seems to escape notice that apart from data, which can be readily used without modification for other purposes, all the modifications to date are derived products of the intelectual property holders ie Atari. The F4UT coders determined that the risk to them and to the community at large would be too great to ever release an executable without a written agreement from the IP holders.
[C3PO]
Will there be an SP5? If so, what developments are we likely to see there?
[Chris Carter]
Its hard to say what will happen after SP4 at this stage. I personally have a number of long term data projects, which I will continue to work on with the friends who share those goals. However, what will become of F4UT is unknown as to be honest I think:
a. The FreeFalcon release and subsequent disclosure of Viperops has turned many off from working for free for what is an unnappreciative audience (at least that's the perception).
b. The closer we get to a resolution to the G2i issue the less likely it is that developers are going to be motivated to produce work that may be superceded.
From another perspective, the subject has not yet been discussed in either management or on the Dev fora so there is nothing set in stone. I would say that the experiences of the last few months haven't exactly been motivating.
As an aside I think the community should take a long hard look at itself, as it has not acted admirably in terms of its expectations and what it is willing to contribute towards fulfilling those expectations.
Every publically accessible group I have either run or been involved with over the years seems to have a ratio of about 20 to 1 in terms of hangers-on, looking for their next free Falcon fix vs productive members. This is not so much frustrating from the viewpoint that there are people who's concience will allow them to leech other's work for no input but more from the lost potential to complete large projects in short timescales. A prime example would be the Balkans theatre where there are over 300 members on that group and approximatly 30 people contributed to the terrain tiling effort (not technically difficult but time consuming). If all 300 people had taken a small tiling assignment the hand placement of the 750,000 tiles would have taken a couple of months rather than the year plus that it did eventually take.
I'll blatantly rip off a Kennedy quote and say the community and development of Falcon should be more about "ask not what the Falcon community can do for you - ask what you can do for the Falcon community" and that atmosphere is what is lacking these days.
[C3PO]
How has LOMAC changed things for the military flight simulation community? How does Falcon fit into the genre these days?
[Chris Carter]
I'll be frank here, LOMAC is a very good flight simulation and essentially leverages the last 4 years of hardware advancements to provide probably the most realistic looking combat flight sim environment I have ever seen. It will also be the best simulation, for those aircraft that are enabled for human control, available to the civilian PC pilot. I will be buying more than one copy as soon as its available, as its success can only lead to a greater corporate awareness that there is profitability in the genre.
That said there will be inevitable comparisons with Falcon4, as both products model modern jet combat aircraft and there are obvious differences in terms of feature sets and scope. The virtual modern combat pilot has, with Falcon4, been spoilt for the last 4 years with a product that really shouldn't have ever been commercially viable and with the 3rd party modding efforts, have contnued to realise the potential of that product to levels you have today. The problem for any competitor to the Falcon franchise is that at the moment, you have a lot of people willing to spend thousands of hours of their free time doing what any commercial developer would consider financial suicide. For example it would be unlikely to find a commercial developer willing to pay 6 figure salaries to a dozen guys to go and hunt for the data, with which to improve the missile kinematics to a high level, when it is unlikely the average simmer would notice the difference in the resulting behaviour. Ask yourself how many sims previously modelled the AIM120 as a true fire and forget missile, where no mid course corrections are needed before the missile goes autonomous. Nobody kicked up too much of a fuss with those simulations, as the vast majority had no data with which to dispute the behaviour, the internet has largely been responsible for changing that level of information and hence the expectations of the simmer.
The point to all that is that although LOMAC will be a competitive product and will sell well (at least I hope it does) it will be up to its owners and its community as to whether its longevity is ensured. From my point of view I will still continue to develop for Falcon4 because its core engine has more scope for improvement and enhancement than that which I currently perceive LOMAC delivering, at least in its first iteration.
So LOMAC won't affect the development of Falcon but I expect we may see decisions being made about which product is used more on people's machines, dependant on what the simmer's goals and expectations are in a product.
For the second half of the question I would say that Falcon4 still offers the most immersive, most realistic and for me, the most rewarding modern combat flight simulation experience. The multiplayer aspects, even with the existing problems, still offers a feeling second to none when you land with your LAN squadron after a 2 hour mission over Serbia with a human AWACS controller and a decent set of cockpit mock ups. Myself and my squadron have stepped out of our virtual cockpits somewhat physically but definatley mentally drained after a hard mission where anything can and often does happen, within the fully dynamic battlefield environment modelled in Falcon4.
[C3PO]
Give us a vision of flight simulations in ten years' time.
[Chris Carter]
10 years is a long time in the computing industry and there are some fundemental problems associated with the genre but I'll give some positives first:
The additional processing power of both the CPU and GPU, should allow for an almost cinema quality environment to fly in, but after a certain level of graphical improvement the real enhancements are going to come in terms of what developers will be able to model outside of the aircraft you are flying.
You should be able to say model fully animated ground crew all working away on aircraft. You should be able to experience the full startup procedure including visual inspections etc.
The ground environment should experience a quantum leap in terms of what's available now, with no bubble technology needed and therefore a much higher level of fidelity from one end of a theatre to the other. I would also hope that civilians start to get modelled together with the politcal penalties associated with harming them and limited wars, with rules of engagement rather than the all out struggles we enjoy at the moment.
The processing power should be there to start tuning the AI to simulate everything from equipment failures and responses to morale in combat. The inclusion of AWACS is step in the right direction for "controller" interaction but there's alot more that could be done here as well.
Also the advent of broadband internet is going to introduce massivley multi-player elements to flight simming and real sense of character development for your pilot rather than the sexed up login screen we have at the moment. I also see that other genres of combat games could be integrated into the enviroments.
Falcon has the potential to take advantage of all of these advances but it will be upto the IP owner to realise that potential and act accordingly.
There are some fundemental obstacles to overcome:
Flight Sims are a niche market and becoming more so as the computer gaming industry has become almagamated into the "globalised media" where even a third rate film licence is worth more in terms of revenue to a games company than many flight sim products. Some areas of the computer games market needs to be able to regain some of the "small company" environment that existed in the late 80's and early 90's, unfortunatley there are a lot of hurdles in bringing a product to market and the big players don't exactly take intrusions into what they percieve as their marketplace very kindly, when in reality they could profit greatly from the increase in enthusiasm and innovation new products can bring.
There are alot of small volume hardware companies out there making excellent products to enhance the flight sim experience and there are some fantastic products such as the hemi-spherical projection screens. The problem is that these products are not mass market and probably never will be whilst companies fail to market their products to the right audience or people are not able or unwilling to afford to buy the sort of volumes where the prices start to fall.
[C3PO]
And anything else you might want to add or say!
[Chris Carter]
I expect there are enough sleeping simmers out there already so I had better sign off now;-)
[C3PO]
Thanks for taking the time to do this interview.
