Jump to content

HappyHaddock

Members
  • Content Count

    236
  • Joined

  • Last visited

Everything posted by HappyHaddock

  1. Given the wide variation in possibilities with the clouds I've found with my own work that getting any kind of consistent comparison to test frame rate isn't easy so I'd be curious to know exactly what/how much you've tested against what to arrive at any particular percentage change? Has this been a systematic like for like before/after test of each sky.ini file using missions prepared through the mission editor, or simply comparisons of whatever settings came up next through general game-play as I've found (and mentioned before) that within about +/- 5% is kind of the variation you get with the stock clouds anyway so without strict regulation/definition of a base line control it's very easy to get very different results each time you run such tests.
  2. Cheers but I think we were both aware of their being rain applied in the stock settings, the issue was more about getting the sim to apply those effects without having to accept the overcast clouds based upon flat textures
  3. Like you I've tried various options relating to precipitation but the code doesn't seem to produce rain in game, so I was kinda hoping you might figure that one out as it seems hard coded into the flat 2d cloud templates. However I'd adopted the approach of motoadve of deciding overcast doesn't mean the same as raining so use different approaches for different end results.
  4. Don't be too sure of that when I started tinkering with ROF about 5 years ago I didn't think I'd be getting into things as much as I have, but I'm now trying to be disciplined about finishing what I have started before moving onto learning more programming languages to tackle things I'm adding to my "to-do" list.
  5. @rowdyb00t back in July when you first messaged me asking how I was going about producing my own WIP efforts to overhaul the look of the natural world in this sim, and subsequently when you started this, your own thread about your own efforts I "welcomed you to the slippery slope" of interconnected factors that leads from one thing to another. Since then I've periodically checked back in here to see if you've discovered anything new I could "borrow" for my own work; it's been interesting to watch you follow more or less exactly the same steps I myself was taking a year or more ago, starting with the "low hanging fruit", more recently moving onto creating new cloud maps to replace some of the flat 2D overcast clouds with 3D particle based ones... it makes a huge difference doesn't it? If you continue down the same path as myself I suspect your next steps may be those that offer the biggest potential for improvement but at the same point some of the most demanding work fine tuning and balancing things. Any way I wish you well with your endeavours as I continue with my own. To date the 300-400 files I'm currently working on for the on the clouds comprise less than a third of everything I'm trying to work into my own project, though with close to 5GB of "clutter" spread across various experiments it is taking some whittling down and unifying. Who knows the slow progress I'm making trying turn all this into a suite of balanced mods might mean that by the time it is ready, then in one way or another you'll have covered a good chunk of similar content with your own efforts and we'll end up meeting on some sort of common ground in the middle!
  6. The devs have said repeatedly from the start that we should not expect heavy bombers in IL-2, partly because of the amount of AI coding/processing power to model all the crew/gunners in a single aircraft and partly because nobody wants to see a single heavy bomber flying on its own; they operate in large formations massively increasing both the overall development effort to program such things and the subsequent processing demands to then run all this code. With the devs having announced both a pre and post invasion map of Normandy it seems inconceivable that they won't attempt some sort of portrayal of the D-day landings, yet large warships with all their heavy guns and smaller MG's and AA will surely be at least as AI intensive as heavy bombers, plus to model the D-day landings you will not just need several warships but loads of supply ships and smaller landing craft just to give a hint of how much was going on. All of which could lead to much speculation about how the AI and game engine will cope? Moving from the realms of plausible speculation to hopeless fantasy we can say that WWI was a conflict that cost millions their lives and where thousands died in no-mans land during single battles. Given that in Flying Circus we can fly over the front line without ever witnessing any foot soldiers, and that the Normandy beaches could be viewed as a similar environment in which thousands of soldiers were fighting to claim a long narrow strip of land, dare we speculate that some attempt may be made to better model a ground war? With DCS already having released content for Normandy direct comparison will inevitably be made, where it seems unlikely that the devs would attempt this if they didn't feel confident they could offer something better than their competitors... all of which leads to two rather obvious questions. 1)Given what we know about the ability (inability?) of the current game engine to handle large numbers of different units how might what we have cope with modelling the D-day landings? 2)Given the likely answers to question1 how far might people be willing to speculate in terms of possible improvements that might be made to increase the above, especially given that modern CPUs are going ever higher with available cores and threads? Answers to the above will, of course, be idle speculation but dare we dream?
  7. I don't profess to be a great pilot so was pretty pleased to have brought down a Halb CLII without the rear gunner seemingly landing a single shot on me, but for the few minutes following I had a niggling feeling that I couldn't quite put my finger on that something was, if not "wrong" with my camel, somehow different. Consequently I started looking around for signs of damage or obvious smoking, leaks or vapour trails but all looked good. I eventually identified my "concern" as the fact that although my engine didn't sound "bad" and possibly not even "different", I could persuade myself there was a somehow unfamiliar hint of a whine to it. Engine revs seemed more or less OK though perhaps a whisker below what I'd expect but things were running fine, I tried tweaking the fuel mixture but anything more resulted in spluttering and clouds of black smoke from running too rich so I returned settings to where they had been and carried on. Had it been my car in real life I'd have thought to myself I must get the garage to check this out just to be on the safe side, but forgotten about doing anything about it by the time I'd got home. As it was a further 10minutes or so into my flight the engine on my camel spluttered and died leaving me to glide down to land in a suitable field. That a combat flight simulator should model battle damage from combat and/or stress failures from overworking an engine is not surprising, that it should model progressive failure is not news but still a sign of a quality product, but that it should communicate such things in such a subtle manner as to leave you wondering if a stray bullet did hit your engine, or whether you just over-revved/heated it slightly at some point earlier when engaged dogfighting speaks highly of the real quality of the game engine behind Flying Circus. I've said it before and will say it again, there are many "flight simulators" out there that offer endlessly complex button pushing scenarios with vast amounts of digital data being relayed back to banks of cockpit instrumentation, simulators producing bold and dramatic visual effects for catastrophic failures. However for me the real test of any flight simulator is how well it can convey the "feeling" of flying an aircraft without relying on any of that; something that can be very subjective and reliant upon a myriad of subtle and easily overlooked factors. I don't know how subtle some of the sound effects in FC can be or if it was other factors that psychologically convinced me that my engine sounded different, but what ever it was it had that magic ability to create a sense of immersion without ever giving away which "strings were being pulled behind the scenes" in order to do so.
  8. Given the complexities of the modern world I doubt it would be possible even if you wanted to... To cut a long story short my sister is a high achieving exec. in the world of finance based in the city of London. She had a disagreement with the Inland Revenue over the fact they had significantly UNDER calculated her annual income tax bill; she said she didn't want inconveniencing or harassing for the extra money later on in the year when they realised their mistake so she told them that even if the government/tax office didn't think she owed it, she still wanted to "voluntarily" pay them "extra" tax now just to make her life easier later on... They couldn't cope with that and refused to accept the over-payment because they didn't have a system to cope with it.
  9. Well I've not seen any mention of a proposed price for MS Flight Sim 2020 when it releases, but you can bet your bottom dollar that Microsoft are financially far more secure than the team behind IL-2. Nevertheless I'd be very surprised that even with a vastly bigger customer base for their new Flight sim they will still end up charging a lot more... Given the way so much is going with software I wouldn't be surprised if it isn't possible to "buy" MS flight sim 2020 but that you end up paying some inflated monthly subscription fee to access it, a monthly fee that could easily dwarf anything we pay here if actually worked out on a "per hour" basis of enjoyment on investment. What constitutes affordable or good value will always vary from person to person based upon their own circumstances and interests but as others have said compared to other flight sims out there IL-2 is definitely one of the more affordable current generation of flight sims. (as if that is a huge pool to compare against!)
  10. The old pricing structure goes back through several previous titles, maybe it is simply a case that the devs have finally decided to add on a bit to allow for a few years inflation. Many things cost more than they did five or six years ago. Edit @AndyJWestSeems we both posted the same comment simultaneously
  11. It has always been present to a pretty minor degree, I'm just not sure if it has actually got stronger, or simply looks more pronounced on account of other problems appearing less distracting?
  12. My first thought upon seeing that map was that it covered pretty much the ground covered by the maps in ROF (plus a lot more) ... so here's hoping as back-dating a map has to be much easier than creating it from scratch
  13. well I've no particular interest in WWII and always said that possibly the only thing that might tempt me out of the cockpit of the sopwith camel would be a mossie… Is it a sign of weakness to give in?
  14. I know the devs have talked about tweaking the way distant aircraft are rendered against clouds so I'm wondering how much others have noticed the following after today's update? To me at least it looks as if there is now a more pronounced white halo around the edge of distant aircraft, not a problem or even noticeable in front of white clouds, but during evening/twilight missions when the clouds are in shadow and quite dark it become more obvious, plus when aircraft are sufficiently distant as to drop out of view altogether there still seems to be a clear white glow in their location until they have flown even further away. I know many will welcome any improvements that stop pixelated aircraft flickering in and out of view but are others noticing this?
  15. I may be sounding like a stuck record here as I've said this many times before but the real challenge with changing any one thing within the look of the natural world in this sim is that it can impact on so many others; both from a technical point of view through shared parameters within the coding, but also from a perception point of view as the human eye/brain cannot judge things in isolation but only through how they contrast with others around them. This is why you really need to address changing everything to maintain balance as if you simply change one thing you are affectively pushing the problem somewhere else rather than eliminating it.... In many cases the things that can look like problems are often the parts that are most accurate and life like simply being spoiled by the look of others around them. A lot of addressing issues with the lighting in this sim can be counter intuitive, whereby you can get better results not by lightning the thing that looks too dark. but by darkening the stuff around it if it is everything else that is actually too light.
  16. The more I see of your technical changes to the shaders the more impressed I am by the possibilities and the more I realise I need to get to grips with HLSL coding... I just told myself I had more than enough on my plate with the artistic work I've already taken on trying to re-work the natural world in this sim and that I'd try finish what I started before moving on to the whole new topic of shaders. If I thought it was just going to be a few hours to pick up some basics of coding I'd drop what I was doing in flash to learn enough to bring shader re-writes within the scope of my own current project, but I fear learning HLSL coding from scratch is going to be much more than just a few hours to get to grips with... more than 30 years ago when I was considered to be a bit of an IT geek I possibly would have picked up the basics of what I needed to know quickly, but alas any technical understanding from the 1980's and 90's is now about as obsolete as it is possible to get in computing terms.
  17. @LizLemon is this primarily achieved through exaggerating the noise then applying a bit of smoothing or is it more evidence of you re-writing shaders?
  18. Can I assume you've got around to looking at reworking the cloud maps? I've spent ages trying to develop simple tools and add-ons for my graphics packages to aid this sort of work.
  19. I don't profess to remember much of it, but many years ago one of the many topics I was required to study as part of my first degree at University was the basics of meteorology, so I do understand the points you are making, and whilst nothing to do with actually working in meteorology I did for a while have a job which entailed having to routinely monitor changes in atmospheric humidity. I doubt any simulator will ever "properly" replicate the true behaviour of atmospheric humidity, but there's a big difference in not attempting something and having something that attempts to approximate some of the more prominent affects. So as you say there are plenty of ways of generating a "look" for the purposes of improving graphics, but an entirely different ball game modelling all the associated factors from the point of view of their impact on aerodynamics and flights handling. Il-2 FC feels great in away I've never really got any "feel" from other so called simulators (except ROF) , but if MS flight sim 2020 lives up to its hype then it could be something very special, alternatively it could end up like all the other early versions of MS flight sim; a soulless button pushing exercise good only for learning how to program a computer to control a modern fly-by-wire aircraft for you. I know many folk like the big passenger jets but Ms flight sim 2020 won't be worth considering for me until it has some sort of nervous twitchy and unstable wood and canvas bi-plane whose total number of cockpit instruments can be counted on the fingers of one hand. The behaviour of these primitive aircraft are what really show if a simulator is worth the title "simulator"
  20. As somebody who has spent a considerable amount of time taking apart the clouds both within ROF and IL2 I can say that there seems to be a lot that is different about how the various settings and parameters generating the clouds are regulated, but fundamentally it seems to be a largely similar particle based system at the heart of each. The biggest difference was that in ROF you had to physically model each cloud from a whole series of co-ordinates and diameters for each particle making up a cloud. It was a slow process but meant you could create any shape of cloud you wanted. IL-2 grows it clouds from initial maps of humidity; this has the advantage of being far easier and faster to work with but offers less fine control over the actual shaping; something I have been seeking to redress by developing ever more mods/tweaks for the graphics packages I use to paint the humidity maps from which the clouds in IL-2 are grown. Thereafter most of the "issues" people have with IL-2's clouds are not so much with the modelling of the clouds but with the new shaders used to illuminate and render these, something I don't yet have the skills to look into and re-write, but by and large technical solutions re-writing shaders don't really address the core artistic issues but they should be great at fixing the technical problems many others have gone over in depth many times before.
  21. Dynamic weather systems and cloud formations that change and evolve over time with rain forming where relative humidity exceeds a cloud's capacity to carry the water vapour, only for a cloud to then rain itself back out of existence would be a real bonus, and like many I have been following news of MS Flight sim 2020. I don't profess to fully understand all the technicalities behind the coding, nor am I pretending it would be a simple "fix" to implement. However the clouds in IL-2 are already 3D particle based models "grown" from initially programmed starting conditions for "humidity" whilst the ROF and IL-2 game engines have both been long been capable of moving around and disturbing the shapes of 3D particle clouds based upon wind, prop wash etc. My gut instinct is that it wouldn't require starting entirely from scratch but "simply" (or possibly not so simply?) creating some additional new code to tie together various existing functions already within the game engine so as to keep freshly "re-growing" cloud shapes frame by frame based upon their current humidity level, rather than growing their shape only once upon loading a mission and then treating the cloud shapes as static remaining uninfluenced either by their own rain fall or other external factors. It would certainly increase the processing power required and much would depend upon the extent to which the "humidity" values used to grow the existing clouds are actually a useful measure of humidity in the real world, and how much new coding would be needed to allow these values to fluctuate with external factors. Computers are very good at carrying out lots of numerical calculations and this sim already numerically defines wind speeds at different altitudes, has different values for levels of rain fall etc. etc. so I feel all the values needed for modelling evolving humidity values for growing dynamic clouds are , to some degree, already within this sim. However, If the term "humidity" is simply a name given to an abstract and isolated value used to generate the boundary of a cloud volume then without any influence from other factors it would endlessly "re-grow" exactly the same shape of cloud each time. As everyone knows, computers are mindless machines without free will, they just follow the instructions they are given. With sufficient understanding and the time to give them adequate instructions they can and will do whatever you want. The more time I spend "under the hood" of this sim the more I am impressed by the understanding and complexity that went into its creation... and I'm just an old school artist. I am sure there are others who are far more technically capable than me in terms of both appreciating more of what this sim can do, and also understanding how to extract more of it's potential.
  22. I first came across Truesky when I started modding the clouds and sky for ROF several years ago. At the time I was impressed and I still am, but having spent a lot of time poking around within the IL-2 game engine I 'd say much of what truesky has to offer in terms of end results is already possible with what we have. I'd imagine switching to a new technology would be a major workload and very much doubt the devs would consider it, so it would only be worthwhile if it then became massively easier to maximise the potential, otherwise I'd see it as more beneficial to put the effort into getting more out of what we already have. To be honest I feel that major improvement to the look of IL-2 are still to be had by fine tuning, and most importantly "balancing" the various data and textures fed into the existing system, more than actually needing to change the system that information is processed by. In terms of "volume and height" if modelling a single cloud within IL-2 it is not too difficult to create almost any shape /size you want, the real challenge is the sheer amount of work involved in filling multiple skies with a great variety of different looking clouds to create stunning formations. I've gone through several iterations and variations of mods and tools for my preferred graphics packages trying to find a way to speed up production of better cloud formations. I've definitely made some progress but I'm far from being able to achieve the look I really want across every part of a full suite of mods. There are certain effects that I could achieve much more easily with real paint on paper than I can digitally, such that I am now seriously considering hand painting some of the textures I want and then digitally scanning these to see what impact this has when fed back into IL-2.
  23. I too share that dream; in a somewhat crude and simplistic way I started putting out a few mods to enhance what was possible in ROF but soon realised that you could spend forever and a day tweaking little bits here and there putting out new mods on an almost weekly basis endlessly correcting and changing stuff from your own earlier mods. This time around I tried to be far more methodical about building a lot more understanding before finally starting to piece together something far more ambitious, and most important of all, (to me at least) something balanced - Every change you make to one thing impacts upon so many others such that you can't realistically mod "one thing" to make an improvement to the look of the sky and clouds, you have to mod everything en-masse.
  24. It seems a fair while since I last posted any kind of progress update here so I thought I should mention that now FC is officially released I have begun the slow process of amalgamating all the test data I've been collecting over the last year so as to start selectively compiling it into some sort of useful mod, or suite of mods. As such I thought folk might like to see another WIP screenshot of what is currently occupying me; that is trying to balance a variety of effects so that they work all round as a winter sunrise when viewed from any angle or altitude. With a largely white landscape it is amazing how readily the most subtle of changes to colouring and lighting files is reflected and exaggerated on the ground in a way that is just not problematic with the darker and richer spring/summer/autumn maps. The great thing about this is that when everything works and is in balance it can look great, it's just that some of the bolder lighting effects that look good on the clouds can really look out of place and over done on the ground. Anyway as promised another screenshot. HH
  25. last night I did something I've not done in ages; that is after spending hundreds of hours taking IL-2 apart, I actually loaded up a camel in the quick mission generator and just spent half an hour dog-fighting above the clouds. A camel with a full tank of fuel at about 9000 feet isn't the same beast it is with a near empty gas tank at tree top level, and when bounced by a D VIIF I had quite challenge; the new physiology effects limited some of the more extreme moves that used to be possible in the camel and the new AI was actually using its energy advantage against me and wasn't going to be drawn down to the deck in a tight twisting turn fight. The result of this was that after about ten minutes of intense "inaction" in which a lot happened but neither of us got into a position to fire a single shot I knew I couldn't make run for it and had to do something unexpected to break the stale-mate. Consequently I pulled an insanely tight stall turn momentarily losing sight of my opponent. The next instance something flashed in front of me and I instinctively hit the trigger without knowing what I was shooting at or if I'd hit it. With my head lurching left and right I tried to re-gain sight of my opponent then aggressively threw my own plane sideways as my opponent made some sort of insane kami-kazi attempt to ram me. I knew I had the altitude to recover from my resultant spin but half expected to be showered in bullets as I spiralled earthwards.... As I regained control of my camel I found myself coming up behind the D VIIF who was oddly making no attempt to evade me, making it easy to close right up on his tail... the cockpit was empty and the empty plane was just gliding, there was no sign of damage, no vapour trails, smoke or flames and as I looked up the pilot was just drifting down on his parachute, it had been an empty plane that had nearly collided with me. I don't know what I did to cause the AI pilot of the D VIIF to bail out and if more by luck than skill I had actually hit the target with my brief gun burst, but the whole encounter made for a very different dog-fight. What I will say is that if there are any ROF fans out there hesitating about whether FC is worth the purchase price, then I never had such an intense or unexpected encounter in ROF where the AI is just downright predictable and in many respects incompetent. So whilst there are things about FC that I'd like to be better, there is more than enough of it that is better than ROF to make it worth giving a try! HH
×
×
  • Create New...