Jump to content

Do you guys think the game engine needs an upgrade?


Recommended Posts

Question in the title. What's your opinion guys? From what I understand BoS/BoM use the same engine as RoF without any major changes. Do you think an upgrade is necessary? I personally think better multi-core optimization and maybe some modern tech like Dx 11/12 (12 optional. i don't think it's super important) and maybe some eye candy like PBR (which makes everything look better, even War Thunder planes look kinda nice with it  :biggrin: ). What do you guys think? I think it's an important topic, maybe even more than the endless discussions about "what planes should we get next"

I'm curious to hear your opinions

Edited by I./JG3_Asgar
Link to post
Share on other sites

Well first of all there are actually a bunch of improvements in BoS compared to RoF. In the very long term I'm sure such an upgrade can be considered but probably not soon. Honestly for a DX9 game BoS looks and runs fantastic. I know they have mentioned ways to optimize the load on the CPU which seems to be the chief performance limitation. Obviously there are many benefits to going with something like DX12 but I'm guessing the games will need to sell a lot in order to justify the time and cost of creating a new engine. Digital Nature from what I've heard was created as an entirely new engine back when flight sims seemed big business. The genre would have to see a really decent resurgence to fund making a new one.

Link to post
Share on other sites

We need to concentrate on improving what we have at the moment .

MP is still the key to keeping this game alive .

I could make a long list of what i would like changing . But that isn't down to me .

Link to post
Share on other sites

I think an upgrade would be justified. But it would come with a price. Either a long delay between releases or a substantial price increase to defray cost of development. People an create polls or post opinions all they want. Money to fund it is all that matters.

Link to post
Share on other sites

A new engine is always nice, as long as it is thought out and planned appropriately. It's not going to happen over night, so your new Christmas technology won't be taken advantage of in a long time. Engines take years to build. Look at how long DCS is taking.

 

The old Il-2 series went a decade, and still going, with the same engine. This one still has plenty of capabilities and life in it.

Link to post
Share on other sites

Honestly I think this game looks better than War Thunder

The aircraft themselves, absolutely. And the terrain too.

It's a real credit to the graphics in this game that you're basically looking at white painted planes flying over white snow and you can see them. There's a nice range of color and shaders etc. In some other games those planes would just be invisible, white against white.

  • Upvote 2
Link to post
Share on other sites

One thing that looks really really bad in BOS are the smoke effects from damaged airplanes. Smoke from cities and burning vehicles are good however the transition from 3d smoke to 2d smoke for damaged airplanes(ca 50m behind the aircraft) look REALLY outdated and could be massively improved! For the rest I think BOS looks really good!

  • Upvote 2
Link to post
Share on other sites

Do you guys already have an idea, who is willing to invest ?  It would need a few millions, I guess. Better invested would be some ideas, how to get at least ten times as much players into MP. How many would come due to better eyecandy,

allthough it is already the best looking WW2 sim ?

  • Upvote 1
Link to post
Share on other sites

Do you guys already have an idea, who is willing to invest ? It would need a few millions, I guess. Better invested would be some ideas, how to get at least ten times as much players into MP. How many would come due to better eyecandy,

allthough it is already the best looking WW2 sim ?

I think there is no need for more eye candy, just performance upgrade to take care of multithreading and access more memory.

Link to post
Share on other sites

I'm pretty sure the devs have said that the move to 64 bits is in the works for '16. As far as multi core utilization; I checked using afterburner just yesterday and all 4 cores were being extensively used. Each was consitently in the 75% to 95% usage range. That's pretty darn good.

 

Memory usage was pretty low through out though. It is my understanding that this is due to the 32 bit system. Hopefully the move to 64 bit will help.

Edited by SYN_Mike77
Link to post
Share on other sites

Seems to show good usage of four cores to me?

 

what do people mean by better multithreading? better use of six and eight core processors? not enough market share... better use of hyperthreading? majority of people have i5 which does not use it

 

or are people talking solely about Dserver?

 

Cheers Dakpilot

  • Upvote 1
Link to post
Share on other sites

Graphics are a matter of customer opinion, performence is not. While certainly not the prettiest compared to other modern games and somwhat strange usage of lightning / overdone fog for maps it's still looking ok. Heck even RoF looks still good to me (with SweetFX) and sometimes suprises me with nicely lookign scenes. Also aircraft cockpits look way more realistic in RoF than in BoS due to better reflection shaders (my opinion).

 

What is not opinion based is performence though. It's been an issue for more than a year now that every second patch performence is hit for a certain group of players (most obvious in 1.104). Tahn it takes a month to correct it and the circle repeats. With the big goals both the devs and the comunity have in mind with this series the engine definetly should be optimised / revised to make room for the bright future everybody wishes.

 

Once that happens new graphic technologies may be implemented as time and money allows.

I'm pretty sure the devs have said that the move to 64 bits is in the works for '16. As far as multi core utilization; I checked using afterburner just yesterday and all 4 cores were being extensively used. Each was consitently in the 75% to 95% usage range. That's pretty darn good.

Memory usage was pretty low through out though. It is my understanding that this is due to the 32 bit system. Hopefully the move to 64 bit will help.

It squeezes slightly more usage out of 2 of my 4 cores than the others. Also CPU usage is very inconsitent, in hangar it can go up to 92-96 (!) while flying in midair (3000m) it's around 75%.

 

I'm using an i5 4570 3,2 Ghz (capped).

Edited by Stab/JG26_5tuka
Link to post
Share on other sites

Seems to show good usage of four cores to me?

 

what do people mean by better multithreading? better use of six and eight core processors? not enough market share... better use of hyperthreading? majority of people have i5 which does not use it

 

or are people talking solely about Dserver?

 

Cheers Dakpilot

A multi-threaded solution would be able to use many cores. Imagine we have hundreds of objects that have their state evolving over time. Some of them use sophisticated physical model. Mainly the aircraft we fly and also AI pilots fly. Each of them may evolve its state in parallel. Even a single-core processor is able to run many threads. It's the task of the task manager to assign logical threads to physical cores or the task of the thread pool to solve the logic. There is no need to predict number of physical cores on each software solution layer. Also there is no reason to limit 8 or even 16 core PCs because someone else can't afford so powerful PC. Many of us do not want to end up with canonical "640kB should be enough for everyone" today adjusted for "4 cores ... for everyone".

Link to post
Share on other sites

here is where Han said they are moving to 64 bits. It's on page 39 of the "questions to devs thread"

 

 

 

NN_RugbyGoth, on 27 Nov 2015 - 09:02, said:

 

When will the soft be built in 64 bits? Is it an objective?

 

Any plan to use an earlier version of DirectX?

 

 

 

We plan to use x64

 

We don't plan to use DirectX 8

Link to post
Share on other sites
MP is still the key to keeping this game alive .

 

There are around 14000 people with "founder" status and most likely a few thousand more people who bought the game but don't have founder status. That against maybe 400 regularly active MP players (even that number is probably an exaggeration).  MP is not what's keeping this sim alive.

 

 

I think moving to a new engine is the most important thing right now and i'm pretty sure that's one of the big tasks for the devs this year. 

Edited by Matt
  • Upvote 3
Link to post
Share on other sites

What will keep the game alive is thousands upon thousands of normal casual players who might play for a few hours, check out the graphics, make some stuff go boom and then move on to another game where they check out the graphics and make some stuff go boom.

That funding can support those of us who will play this game for hundreds of hours. Maybe some of those casual players will stick around too.

Pretty much that's what keeps every game alive

:-D

Edited by SharpeXB
  • Upvote 3
Link to post
Share on other sites

It feels like there is a lot of untapped computing resources that the engine could use:

  • 64-bit addressing allows to use more than 4GB. Nowadays, having 8GB or 16GB isn't uncommon. That would allow for more textures to fit in memory at the same time, which would allow for e.g. nicer damage rendering, more fine-grained cockpit textures and skins. It also reduces the need to swap textures in and out of memory from the disk. Reduces stuttering, flashing plane skins, blurry landscape textures. It also simplifies code, which in turns means fewer bugs and crashes.
  • multi-core allows to simulate more things: more AI tanks, more/better AAA
  • 32-bit is becoming obsolete, which means that 3rd-party libs that the game uses are old versions, with bugs fixed in more recent versions not being available to the devs. On the other hand, it's sometimes precisely because these libs have been slow to make the move to 64-bit that games are forced to stay in 32-bit.
  • The game once worked pretty well with Occulus Rift, I believe. Not any more, because OR dropped support for dx9.

In addition to that, some glitches show the game code is probably not very clean and could do with some cleaning-up: Triple-screen setups show a number of inconsistencies in GUI layout and formatting: The player list is only centered in standard screen ratios. In triple-screen setups, it's so far off-center is not even visible. The strength of the blinding effect by the sun depends on your screen ratio, which is weird. Play in triple-screen setups, and the screen is 3 times more blinding. The bombing UI is completely unusable in triple-screen setups, three times too high. Same problem for the in-game map. I think in 1.104 some planes had an ESC menu that showed "end fight", others had "end flight". Looks like there might some unnecessary code or data duplication going on.

 

Although the game does a good job at rendering a pretty environment, some details show weaknesses in the rendering engine:

  • Grass is rendered in a circle around the player. That's a pretty rough and insufficient optimization: it should be rendered on the portion of terrain that player looks at. In other words, the optimization shouldn't just use the X and Y coordinates, it should also use the elevation, the orientation of the player and the zoom factor. The math is more complicated, but it's worth the gains. Forests and ground objects currently use the full optimization (you can check that by zooming: the tree and house "front" goes forward as you zoom in). I'm surprised foliage doesn't use that technique.
  • Transparency of tree textures does not use proper blending: The edges are rough, as if transparency is all-or-nothing. But typically, the alpha channel has 256 values available. That should allow for much softer edges.
  • Transition between far and near tree textures uses dithering. I'm not even convinced you need two sets of different textures, but assuming you do, there's got to be a more discrete way of doing it.
  • The tree density, while realistic, does not mix well with the billboarding technique used: overlapping trees cause popping of textures. Spread out the trees, or brake them up in smaller chunks. On that front, I think War Thunder does things better. I know people like to diss that game, but on the landscape rendering techniques I think it's got a definite advantage: better water, better foliage, better trees. IL-2 makes up for it by pretty good artistic techniques (palette, light effects).

At the moment, it does not feel like the dev team has a rendering specialist full-time on the payroll, which sounds unreasonable for a 3d game that uses an in-house engine. I can only hope they have one or more, silently busy on implementing the next generation of the engine.

 

Regarding the costs: I know the game is not cheap, but think of how much money people are investing in upgrading hardware to keep the game running at a decent-ish 45 FPS flying over a pretty still landscape with a handful of planes in the sky. It's excellent business for intel and nvidia, but 1CGS is missing out on the money flow. People throw over $2000 at hardware, how much do the devs get? In the end, an optimized engine should pay for itself by lowering the barrier to entry (for those students, jobless, low-wage recipients, husbands with watchful wives... who can't afford throwing thousands of dollars at hardware makers).

  • Upvote 14
Link to post
Share on other sites

It feels like there is a lot of untapped computing resources that the engine could use:

  • 64-bit addressing allows to use more than 4GB. Nowadays, having 8GB or 16GB isn't uncommon. That would allow for more textures to fit in memory at the same time, which would allow for e.g. nicer damage rendering, more fine-grained cockpit textures and skins. It also reduces the need to swap textures in and out of memory from the disk. Reduces stuttering, flashing plane skins, blurry landscape textures. It also simplifies code, which in turns means fewer bugs and crashes.
  • multi-core allows to simulate more things: more AI tanks, more/better AAA
  • 32-bit is becoming obsolete, which means that 3rd-party libs that the game uses are old versions, with bugs fixed in more recent versions not being available to the devs. On the other hand, it's sometimes precisely because these libs have been slow to make the move to 64-bit that games are forced to stay in 32-bit.
  • The game once worked pretty well with Occulus Rift, I believe. Not any more, because OR dropped support for dx9.

In addition to that, some glitches show the game code is probably not very clean and could do with some cleaning-up: Triple-screen setups show a number of inconsistencies in GUI layout and formatting: The player list is only centered in standard screen ratios. In triple-screen setups, it's so far off-center is not even visible. The strength of the blinding effect by the sun depends on your screen ratio, which is weird. Play in triple-screen setups, and the screen is 3 times more blinding. The bombing UI is completely unusable in triple-screen setups, three times too high. Same problem for the in-game map. I think in 1.104 some planes had an ESC menu that showed "end fight", others had "end flight". Looks like there might some unnecessary code or data duplication going on.

 

Although the game does a good job at rendering a pretty environment, some details show weaknesses in the rendering engine:

  • Grass is rendered in a circle around the player. That's a pretty rough and insufficient optimization: it should be rendered on the portion of terrain that player looks at. In other words, the optimization shouldn't just use the X and Y coordinates, it should also use the elevation, the orientation of the player and the zoom factor. The math is more complicated, but it's worth the gains. Forests and ground objects currently use the full optimization (you can check that by zooming: the tree and house "front" goes forward as you zoom in). I'm surprised foliage doesn't use that technique.
  • Transparency of tree textures does not use proper blending: The edges are rough, as if transparency is all-or-nothing. But typically, the alpha channel has 256 values available. That should allow for much softer edges.
  • Transition between far and near tree textures uses dithering. I'm not even convinced you need two sets of different textures, but assuming you do, there's got to be a more discrete way of doing it.
  • The tree density, while realistic, does not mix well with the billboarding technique used: overlapping trees cause popping of textures. Spread out the trees, or brake them up in smaller chunks. On that front, I think War Thunder does things better. I know people like to diss that game, but on the landscape rendering techniques I think it's got a definite advantage: better water, better foliage, better trees. IL-2 makes up for it by pretty good artistic techniques (palette, light effects).

At the moment, it does not feel like the dev team has a rendering specialist full-time on the payroll, which sounds unreasonable for a 3d game that uses an in-house engine. I can only hope they have one or more, silently busy on implementing the next generation of the engine.

 

Regarding the costs: I know the game is not cheap, but think of how much money people are investing in upgrading hardware to keep the game running at a decent-ish 45 FPS flying over a pretty still landscape with a handful of planes in the sky. It's excellent business for intel and nvidia, but 1CGS is missing out on the money flow. People throw over $2000 at hardware, how much do the devs get? In the end, an optimized engine should pay for itself by lowering the barrier to entry (for those students, jobless, low-wage recipients, husbands with watchful wives... who can't afford throwing thousands of dollars at hardware makers).

Well thought and well stated.

Link to post
Share on other sites

Very true. I just bought 16GB of PC2400 to replace my 8GB of PC1600. At $65, it isn't exactly a crippling financial burden. In any case, BoS/BoM should have been 64bit and DX11 from day one. I understand why that move wasn't made, but 32bit and DX9 are horribly obsolete and incapable of doing what needs to be done. Every time I load up BoS/BoM, I cringe a little at the low quality textures and abysmal visibility ring around the aircraft once I get up off the deck. But those things can't change without 64bit support and the performance won't improve without DX11/12. For comparison, I get about 90 fps average flying over the desert in DCS 2.0 alpha on high settings at 1080p. On similar settings in BoS/BoM, I get about 45 average. And the graphics quality in DCS2 is far superior to what BoS/BoM offers. Incidentally, DCS2 eats up 10-11GB of memory for me, hence that RAM upgrade. And those FPS numbers are with DCS constantly using a page file on my SSD due to a lack of RAM. That RAM is still in the mail. :lol:

 

I sincerely hope that this series moves onto more modern tech. It has great flight modeling and absolutely destroys DCS in terms of damage modeling, AI and focus (for WW2), but the tech running it just isn't up to the challenge.

Edited by King_Hrothgar
Link to post
Share on other sites

Thinking big picture... The Digital Nature engine that they built is a pretty solid platform from what I can tell. Its been able to support a very sophisticated set of aircraft from WWI and WWII eras and we even have things like tank warfare working at pretty high fidelity as a side project. That's pretty impressive to me although maybe its the thing to expect these days in complex simulation engines.

 

Any change to the core engine technology would probably be something that 1CGS would be spreading out amongst all of their products as well as fit into whatever their future long term plans are. Rise of Flight and ILYA Muromets as well as IL-2: Battle of Stalingrad/Battle of Moscow. It'd be a strategic error not to be far thinking but it would also be a tactical mistake if the tech update significantly impacts their current products and revenues.

 

I hope they go all out - DX12, multi-threaded, support for VR (for those who want it), and maybe along the way they can build a more intuitive mission system or at least put the hooks in place for one.

Edited by ShamrockOneFive
Link to post
Share on other sites

Seems to show good usage of four cores to me?

 

what do people mean by better multithreading? better use of six and eight core processors? not enough market share... better use of hyperthreading? majority of people have i5 which does not use it

 

or are people talking solely about Dserver?

 

Cheers Dakpilot

In my experience, no one really knows what exactly they're talking about. A lot of people type buzz words like "dx12", "multithreading", and "optimization" but really don't take into account what they're asking.
  • Upvote 2
Link to post
Share on other sites

It feels like there is a lot of untapped computing resources that the engine could use:

  • 64-bit addressing allows to use more than 4GB. Nowadays, having 8GB or 16GB isn't uncommon. That would allow for more textures to fit in memory at the same time, which would allow for e.g. nicer damage rendering, more fine-grained cockpit textures and skins. It also reduces the need to swap textures in and out of memory from the disk. Reduces stuttering, flashing plane skins, blurry landscape textures. It also simplifies code, which in turns means fewer bugs and crashes.
  • multi-core allows to simulate more things: more AI tanks, more/better AAA
  • 32-bit is becoming obsolete, which means that 3rd-party libs that the game uses are old versions, with bugs fixed in more recent versions not being available to the devs. On the other hand, it's sometimes precisely because these libs have been slow to make the move to 64-bit that games are forced to stay in 32-bit.
  • The game once worked pretty well with Occulus Rift, I believe. Not any more, because OR dropped support for dx9.

In addition to that, some glitches show the game code is probably not very clean and could do with some cleaning-up: Triple-screen setups show a number of inconsistencies in GUI layout and formatting: The player list is only centered in standard screen ratios. In triple-screen setups, it's so far off-center is not even visible. The strength of the blinding effect by the sun depends on your screen ratio, which is weird. Play in triple-screen setups, and the screen is 3 times more blinding. The bombing UI is completely unusable in triple-screen setups, three times too high. Same problem for the in-game map. I think in 1.104 some planes had an ESC menu that showed "end fight", others had "end flight". Looks like there might some unnecessary code or data duplication going on.

 

Although the game does a good job at rendering a pretty environment, some details show weaknesses in the rendering engine:

  • Grass is rendered in a circle around the player. That's a pretty rough and insufficient optimization: it should be rendered on the portion of terrain that player looks at. In other words, the optimization shouldn't just use the X and Y coordinates, it should also use the elevation, the orientation of the player and the zoom factor. The math is more complicated, but it's worth the gains. Forests and ground objects currently use the full optimization (you can check that by zooming: the tree and house "front" goes forward as you zoom in). I'm surprised foliage doesn't use that technique.
  • Transparency of tree textures does not use proper blending: The edges are rough, as if transparency is all-or-nothing. But typically, the alpha channel has 256 values available. That should allow for much softer edges.
  • Transition between far and near tree textures uses dithering. I'm not even convinced you need two sets of different textures, but assuming you do, there's got to be a more discrete way of doing it.
  • The tree density, while realistic, does not mix well with the billboarding technique used: overlapping trees cause popping of textures. Spread out the trees, or brake them up in smaller chunks. On that front, I think War Thunder does things better. I know people like to diss that game, but on the landscape rendering techniques I think it's got a definite advantage: better water, better foliage, better trees. IL-2 makes up for it by pretty good artistic techniques (palette, light effects).

At the moment, it does not feel like the dev team has a rendering specialist full-time on the payroll, which sounds unreasonable for a 3d game that uses an in-house engine. I can only hope they have one or more, silently busy on implementing the next generation of the engine.

 

Regarding the costs: I know the game is not cheap, but think of how much money people are investing in upgrading hardware to keep the game running at a decent-ish 45 FPS flying over a pretty still landscape with a handful of planes in the sky. It's excellent business for intel and nvidia, but 1CGS is missing out on the money flow. People throw over $2000 at hardware, how much do the devs get? In the end, an optimized engine should pay for itself by lowering the barrier to entry (for those students, jobless, low-wage recipients, husbands with watchful wives... who can't afford throwing thousands of dollars at hardware makers).

 

 

Coco,

    I agree with everything you say here, but personally I am very happy with the quality of the graphics.  All the improved processing, multi-threading, Multi Core processing need to be focused toward improving the rendering of objects and improving the Net code.  What I see is some really great potential not being realized in terms of the number of ground units and AAA that should be on the map.  Right now a unit of 20 or so linked objects will crush a mid-range system, and deliver significant frame rate reductions.  Too much AAA to protect your Over-Vulched field?  Sorry...too much AAA causes the same problem.  

 

     Also since this is a very international community working to improve latency so the client/server connection isn't necessarily the bottleneck would go long way to improving what you are seeing from the 109 about to gun you.  Timing is everything, and right now you have to maneuver too early to avoid guns if you hope to be on time based on latency...

 

    I think it is a always a fine balance between stunning graphics and playability (ie. smooth FPS).  I think we have amazing graphics now, we don't need to necessarily improve them.  But hey if DX11 fixes the issues I layed about here, then I am for it.

Link to post
Share on other sites
Multi-threaded optimization. Newer DX versions. x86-to-x64 upgrade.

This sounds good. :salute:

 

 

A new engine is always nice, as long as it is thought out and planned appropriately. It's not going to happen over night, so your new Christmas technology won't be taken advantage of in a long time. Engines take years to build. Look at how long DCS is taking. The old Il-2 series went a decade, and still going, with the same engine. This one still has plenty of capabilities and life in it.

Thank you very much, you have explained to us, that such a thing takes a long time.

 

And i would pay for a upgrade to dx12.

Edited by L3Pl4K
Link to post
Share on other sites

It feels like there is a lot of untapped computing resources that the engine could use:

  • 64-bit addressing allows to use more than 4GB. Nowadays, having 8GB or 16GB isn't uncommon. That would allow for more textures to fit in memory at the same time, which would allow for e.g. nicer damage rendering, more fine-grained cockpit textures and skins. It also reduces the need to swap textures in and out of memory from the disk. Reduces stuttering, flashing plane skins, blurry landscape textures. It also simplifies code, which in turns means fewer bugs and crashes.
  • multi-core allows to simulate more things: more AI tanks, more/better AAA
  • 32-bit is becoming obsolete, which means that 3rd-party libs that the game uses are old versions, with bugs fixed in more recent versions not being available to the devs. On the other hand, it's sometimes precisely because these libs have been slow to make the move to 64-bit that games are forced to stay in 32-bit.
  • The game once worked pretty well with Occulus Rift, I believe. Not any more, because OR dropped support for dx9.

In addition to that, some glitches show the game code is probably not very clean and could do with some cleaning-up: Triple-screen setups show a number of inconsistencies in GUI layout and formatting: The player list is only centered in standard screen ratios. In triple-screen setups, it's so far off-center is not even visible. The strength of the blinding effect by the sun depends on your screen ratio, which is weird. Play in triple-screen setups, and the screen is 3 times more blinding. The bombing UI is completely unusable in triple-screen setups, three times too high. Same problem for the in-game map. I think in 1.104 some planes had an ESC menu that showed "end fight", others had "end flight". Looks like there might some unnecessary code or data duplication going on.

 

Although the game does a good job at rendering a pretty environment, some details show weaknesses in the rendering engine:

  • Grass is rendered in a circle around the player. That's a pretty rough and insufficient optimization: it should be rendered on the portion of terrain that player looks at. In other words, the optimization shouldn't just use the X and Y coordinates, it should also use the elevation, the orientation of the player and the zoom factor. The math is more complicated, but it's worth the gains. Forests and ground objects currently use the full optimization (you can check that by zooming: the tree and house "front" goes forward as you zoom in). I'm surprised foliage doesn't use that technique.
  • Transparency of tree textures does not use proper blending: The edges are rough, as if transparency is all-or-nothing. But typically, the alpha channel has 256 values available. That should allow for much softer edges.
  • Transition between far and near tree textures uses dithering. I'm not even convinced you need two sets of different textures, but assuming you do, there's got to be a more discrete way of doing it.
  • The tree density, while realistic, does not mix well with the billboarding technique used: overlapping trees cause popping of textures. Spread out the trees, or brake them up in smaller chunks. On that front, I think War Thunder does things better. I know people like to diss that game, but on the landscape rendering techniques I think it's got a definite advantage: better water, better foliage, better trees. IL-2 makes up for it by pretty good artistic techniques (palette, light effects).

At the moment, it does not feel like the dev team has a rendering specialist full-time on the payroll, which sounds unreasonable for a 3d game that uses an in-house engine. I can only hope they have one or more, silently busy on implementing the next generation of the engine.

 

Regarding the costs: I know the game is not cheap, but think of how much money people are investing in upgrading hardware to keep the game running at a decent-ish 45 FPS flying over a pretty still landscape with a handful of planes in the sky. It's excellent business for intel and nvidia, but 1CGS is missing out on the money flow. People throw over $2000 at hardware, how much do the devs get? In the end, an optimized engine should pay for itself by lowering the barrier to entry (for those students, jobless, low-wage recipients, husbands with watchful wives... who can't afford throwing thousands of dollars at hardware makers).

+1  :cool:

We are awaiting it.

Edited by LAL_Wolf
Link to post
Share on other sites

What will keep the game alive is thousands upon thousands of normal casual players who might play for a few hours, check out the graphics, make some stuff go boom and then move on to another game where they check out the graphics and make some stuff go boom.

That funding can support those of us who will play this game for hundreds of hours. Maybe some of those casual players will stick around too.

Pretty much that's what keeps every game alive

:-D

 

 

I understand, (somehow I think the same, sadly, when I see so many people here asking for what's next) but I'm not sure this reasoning is valid... right now, in the context of flight simulation games, where do we have examples of a flexible accessibility that doesn't mean near-fatal compromises in its design?

 

To stay on topic, I think CPU performance optimisation is more than essential. I really hope to hear news on what has already been mentioned, using simplified FM/physics for AI and/or assigning physics to its own thread.

Then I can hope to see decent and complex cockpit shadowing... (yes, I've got a thing for shadows and right now they're almost embarassing, along with certain textures :ph34r: ).

Edited by Picchio
Link to post
Share on other sites
Guest
This topic is now closed to further replies.
×
×
  • Create New...