Let the rumours fly, everyone and his grandson will know someone who knows someone who is related to someone who once met someone who knows someone who might have once worked for someone who knows someone who did some contracting for a supplier of a supplier of Microsft at some point in the past and is therefore an inside source...It was said by such people (of course without stating the source which was their own imagination) that FS2004 would have a completely new engine and that no existing addons would work.The same was said of FS2002.No doubt the same was claimed for FS2000 as well and all the way back to FS2 if people would have had the capability to run the rumour mill the internet is.No doubt there will be changes, old code replaced with new.But I severely doubt they're going to completely rewrite everything from scratch.Not only is the risk too high, the cost would be astronomical as well.Longhorn is the first major rewrite of Microsoft operating system code in a decade and is already 2 years over planning and I don't care to calculate how many millions of dollars.There they have a perfectly valid reason, removing all the crud of a decade of updating and putting in new features, consolidating all that while removing dead code and closing down holes and potential holes.To do the same for an entertainment title with an installed base that's only a fraction of a percentage point of the installed base of Windows would drive up the unit cost to astronomical levels which would make the project economically unsustainable.Why do you think 90% of all first person shooters out there use one of a handfull of commercially available engines and don't use their own?It's simply because the cost of creating even such a relatively simple engine is prohibitive in today's game market. To create an entirely new engine for FS would cost several times as much and have a far more limited commercial viability (essentially there's only 2 products that could use it) and installed base.You'd have to pay hundreds of dollars for your FS2006 (or more likely FS2010 because of the delays), which would drive the product out of the market and end the enterprise.

no, it's because customers complain and send death threats to developers when the dihedral of the wing is perceived to be a tenth of a degree wrong, or when a static vane is a millimeter out of place.Users complain that the FDE is incorrect when an aircraft handles differently from what the THINK it should.People were up in arms against the first POSKY 744, they claimed it wasn't slow and sluggish enough. After all, such a big aircraft has to be slow... They were of course completely wrong.The same is repeated over and over again to the point where smart developers just ignore all complaints because the percentage of valid ones isn't worth the effort of wading through the reams of whining by self-proclaimed experts who know nothing yet whose claims you'd have to verify to your own cost in time, resources and money.So forget about FDEs, 99.9% of users wouldn't know the difference anyway.And for those that do know the difference, they also know the limitations of simulating a real aircraft and all its complexity within the limited resources of a $1000 computer when even a million dollar full motion sim can't do it perfectly.

The problem with the CFS3 engine is that it

Hi, "The problem with the CFS3 engine is that it

It's just like the "default" Caravan for FS2002. Users complained that it didn't seem sluggish enough in roll. Then we find that the real Caravan has spoilers to aid in roll which made it feel more like a 182. And that fact was verified by a writeup in Flying Magazine.And I won't even mention all the simmers that don't believe an airplane is flying as fast as it should.......... due to altitude, indicated airspeed,etc.L.Adamson

>If the code was optimised, the resolution updated and a new>engine created, I for one would be prepared to endure the>learning curve providing MS were upfront about continuing>development WITH the community.>I don't know about that ........ personallyFor instance, X-Plane litterly has new updates & many changes monthly. But then scenery, graphics, and flight modeling seem to become incompatible overnight. The only way to keep up is by making the program your full time hobby & a lot of reading in various forums. I read a lot, but don't care to keep up & make numerous changes to make the sim work. For some, its great, & takes place of boredom I suppose...........but not me!L.Adamson

No need talk to ms will take your word on it ;)Well I think a new graphics engine would allow you to make better use of the video cards we will use in two years time. Perhaps optimize for the X800:s and 6800 as lowest recommended spec or something while still having great scalability.I can

Jeroen for once I have to give it to you. Your post here has made the most since I've heard all day....

Nice comments here in this thread,And just a few to add...MS have continued to trim old code, there is a ton of old code that was cut just from FS2k2 to FS2k4 alone. Much code cutting was to happen going into FS2k2, but as developers produced new products late in the game due to lack of proper SDK info and having to come up with "back door" solutions thankfully (in some cases we at least had a year to enjoy some of our add-ons before having to re-develop) MS delayed some of the changes that we were "warned" would occur (just read the FS2k SDK docs) but to keep things moving forward much old code was cut from FS2k4.Can FS be improved? sure, many good comments in here that no doubt get echoed in the FS developer team conference room, how far will they go?Well, I sure can't say I have a clue, but would say they have probably "timed" things very well So that it will good all around.There are probably a lot of great things in store, and threads like this are what will help move things along, but as far as a whole new engine? Not sure that is a good idea...Look at it this way, If right now we got an all up to date FS graphics engine that supported the latest Shader3.0 and DX9c code, how many people would have a machine to run it, so as to buy it, to keep FS going? OK drop the specs back a bit, or go forward in time enough to where that is not such a huge issue...How long before developers can get there hands on the tools that can be used to produce add-ons? And then how long before they are made and get produced do you think? 12-18 months or more based on the past record.Buy that time, the same thing could be said as is now "New game engine for FS11?" yep, it would all be old again and we would still be waiting for our add-ons, and all of you who say no gain without pain would probably get impatient and drop FS like a hot potato, perhaps not but I think it goes without say that there certainly would be a huge drop in interest.Anyhow enough of my rambling, Speaking of the latest, has anyone seen this? >http://www.nvidia.com/object/IO_14172.htmlStart saving yer pennies!

Aside from the technical dos & don'ts of new code for the venerable FS franchise, the price of evolution is a painful thought, but that's the price to be paid everytime this happens (which isn't many, except the FS2K2 debacle that you point out).Imagine where we would be if everyone was content to max out the FS95 code? Imagine FS9 on an OpenGL platform? Imagine this & imagine that.....I agree the biggest roadblock to development are 3rd party developers....Not because of the 3rd party developers (thank God for them) but because Microsoft have come to rely on them, to lean on them as it were. This does & will effect the marketplace as you describe, on the turn of any new code design from Microsoft.The "simulation experience" isn't really complete without 3rd party developers & surely isn't complete with what Microsoft puts on the shelf alone. The franchise has really turned into an unspoken co-operative of sorts. All the more reason that developers should be extremely inclusive & cohesive with alpha & beta development of the simulation itself. This would damper the "time to market" factor of a new release & would ultimately put more money in the pockets of Microsoft. Think of how many people would actually purchase FS9, WITHOUT the ability to add anything to it at all ! Hopefully not too many.....NVIDIA: SLI is long overdue; I just hope Intel & Samsung etc... can scale accordingly....All we need now is Intel to return the IDE channel they stole from their i9XX chipsets & a triple peer PCI bus (including 64 bit) for scsi.Cheers.

Paul makes valid points but remember, MS has ALWAYS optimimsed their code, their `game` or sim, call it what you will for tomorrows technology, not todays. Through their crystal ball, MS must make a decision on the next development based on the emergent technologies. The mostr significant of which is 64-bit computing, which like 32-bit computing before it will explode on the market after a slow, stuttering start as it requires architecture (hardware) change. BUT IT WILL HAPPEN. The latest shaders and DX allow features not previously available and that will NEVER be available by tweaking the existing code. So why not change now? If not now, then it will have to be for the generation after that, and as we know it takes time and effort to get the aftermarket up to speedm and for the bugs to get ironed out, it actually makes complete sense to do it now, while we are at the cusp of new technology streams that will impact our computing experience for years to come.And look more closely at that SLI solution - it requires a lot more than just money, it needs a dual processor, workstation board, twin PCI-Express sockets - and heaven only knows how many power sockets!Very expensive, and given the pace of development, that `90%` gain in performance will be achieved by standard desktop hardware within 18 months. Pixar and Evans & Sutherland use only, I suspect!Allcott

"And look more closely at that SLI solution - it requires a lot more than just money, it needs a dual processor, workstation board, twin PCI-Express sockets - and heaven only knows how many power sockets!Very expensive, and given the pace of development, that `90%` gain in performance will be achieved by standard desktop hardware within 18 months. Pixar and Evans & Sutherland use only, I suspect!"<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>Let's face it, if many had their way on this forum, that might be the only way we'll ever see 50FPS by 2008. But look on the brightside, atleast we'll be able to fly in space, have 10 metre mesh around the globe & every other feature conceivable (albeit with a $500 per month power bill) a 2 ton computer & a second mortgage.....Cheers.

Well, I don't think there's anything wrong with that. Tomorrows technology is pricey today. But then I just upgraded with the package I first priced up 4 months ago, when it was THREE times the price I actually paid! So the hardware upgrade is one of the least painful, as MS can be expected to target hardware that will be middle-market by the time FS2006 has matured in the marketplace.It seems the biggest problem is the simmer letting go of something they've become familiar with. Roll out the new, sooner, and we get familiar with it quicker! That and MS having to cater, as you quite rightly say, to the increasingly bizarre fringe requirements of the diehard simmer, one definition of which seems to be their studied ignorance of the needs and requirements of others. Makes you wonder whether they could make MS FS 2006 as a collection of sub-packages you could buy and bolt-together to make the sim of choice, to spread the financial load over time? Taking the SLI idea a stage further - what about two similar installs of FS working in conjunction, one with the atmospherics, one with the scenery, SLI'd? The `old` computer could then be used to take some of the load off the new, giving it a networked lease of extra life. Or adaptive multiplayer, where the communal server could offer temporary online upgrades for the duration of that sim session - One could create different `Communities` with enhancements for each - low level VFR scenery and grassstrips for bush flyers, more detailed airports and advanced ATC and IFR procedures for jet bus drivers, highly realistic flight models for the 1%ers, that kind of thing. The simmer could then fly in a many different environments as they wanted, with dedicated features for each but without the cost in terms of hard drive space or system overheads carrying features that are not to be used for that session. I would love FS to have a better menu system before it boots that allows one to select which features one wants for that session, so as to be able to push resources around in the best way possible.All of this and much more could be achieved if they started from scratch. Sticking with the old, we won't even get HT and full DX9 compatibility next time out.Allcott

Yep, Yep & Yep....It's quite a shame that one could truly overcome a lot of the shortfalls of the current engine with SLI in place (if even for big bucks); but I'm gonna bet my bullocks to a barn dance that Nvidia will use nothing short of DX9 for their "SLI-enabling" software "miniport" drivers.This would actually ADD to the argument FOR a 21st century core logic evolution of the current simulator core. For not only could MS enable DX9, but they could also factor in the Nvidia SLI optimisations that are purported to be in the works at this time. Throw in HT (at a bare minimum) & possibly for the first time in history you would have a trifecta for an MS development team.....At least with respect to successfully playing catchup with EVERY other developer in the marketplace !!!They might even think about replacing the clumsy (9 year old) menu interface that continues to pollute on the whole experience.But that's all a pipe dream; it's interesting that right now as Microsoft runs surveys (fictitious or not) so many "inside the box" thinkers will suggest things like better scenery & better this & improved that.....Count on Microsoft doing as little as possible to apease the masses in this regard (ensuring Intels roadmap will coincide with the release). This way, Microsoft can be sure there will be enough hardware in the marketplace to PUSH the FS9.X engine to keep up with the extra features they have thrown in......I don't mean to sound so harsh here (to anyone reading this); but it's really hard to be optimistic when history is my guide....I really hope I'm wrong & some of those pipe dreams I mentioned come to fruition. But I believe there remains little argument as to whether FS needs to be "rethunk" all over again.cheers.

>Yep, Yep & Yep....>>It's quite a shame that one could truly overcome a lot of the>shortfalls of the current engine with SLI in place (if even>for big bucks); but I'm gonna bet my bullocks to a barn dance>that Nvidia will use nothing short of DX9 for their>"SLI-enabling" software "miniport" drivers.SLI doesn't care what version of DX you use.>This would actually ADD to the argument FOR a 21st century>core logic evolution of the current simulator core. For not>only could MS enable DX9, but they could also factor in the>Nvidia SLI optimisations that are purported to be in the works>at this time. Throw in HT (at a bare minimum) & possibly for>the first time in history you would have a trifecta for an MS>development team.....At least with respect to successfully>playing catchup with EVERY other developer in the marketplaceSoftware doe not have to be SLI enabled nor be "optimized", SLI has already been programmed into NV45 and to use it you will need a dual PCIx16 slot MB (not dual CPU) a good Powersupply and the Drivers take care of the rest. :-)Titles like FS won't see as much gain as others as they rely more heavely on CPU and memory subsystem but it will certainly do a great deal.>They might even think about replacing the clumsy (9 year old)>menu interface that continues to pollute on the whole>experience.>>But that's all a pipe dream; it's interesting that right now>as Microsoft runs surveys (fictitious or not) so many "inside>the box" thinkers will suggest things like better scenery &>better this & improved that.....>>Count on Microsoft doing as little as possible to apease the>masses in this regard (ensuring Intels roadmap will coincide>with the release). This way, Microsoft can be sure there will>be enough hardware in the marketplace to PUSH the FS9.X engine>to keep up with the extra features they have thrown in......>>I don't mean to sound so harsh here (to anyone reading this);>but it's really hard to be optimistic when history is my>guide....Well perhaps a few versions ago, but certainly not the last two.>I really hope I'm wrong & some of those pipe dreams I>mentioned come to fruition. But I believe there remains little>argument as to whether FS needs to be "rethunk" all over>again.FS has always evolved, and not to take sides here but try to come up with a scheme that would make any better sense that what MS FS team did and I dare anyone to find a better way to have done a better plan business balanced to the desires and whims of everyone in the hobby. I believe MSFS team did an outstanding job, if it hadn't of sold as well as it did we wouldn't even be talking about what the next version should be and you can thank the "old" code that allows it to be played on DX7/8 cards for that. :-)I guess I am just saying I think you aught to tone it down a bit.