Does anyone think PC simulators will ever look this good ?

[kgwanchos]

Sorry if you have seen them before.... anyone know what/who made them and what for.........

http://www.youtube.com/watch?v=nqulS54IURI&feature=related

http://www.youtube.com/watch?v=2FQ5UK1XPmo&feature=related

[GF_Mastiff]

Looks Like stuff from Battle field 360 engine.

It's Japanese so you would need to find some one from Japan to elaborate on.

[Feuerfalke]

That's pure CGI. There are a lot of movies from the same source, often using the same models.

Yes, games will look like this someday. When I started flying simulations, my screen looked like this:
http://www.thelegacy.de/pics/screen/f/Fighter_Pilot.gif

This is what it can look like, now (and this is only the old engine...)
http://www.digitalcombatsimulator.com/img/products/large/e69b4e1b32acd08aa590536f3e74538d.jpg

[Urufu_Shinjiro]

Yes, one day.

[proton45]

This guys stuff comes up on the forums every so often, and he is genius...

Here is his home page...he even has some tutorials on modeling a KI-61 in "Light Wave"

http://www.k4.dion.ne.jp/~suppon/

His web page has some links to more of his work if you are curious.

p.s. Oh, by the way (sorry) his name is Masaru Tochibayashi.

[|ZUTI|]

Hm, not related to original question, but that's Sakai in those movies, isn't he? I remember the second one from his book, not sure about first one, but the first one is too "geenral".

Very nice CGI.

[Tata]

Amazing! Very good made videos.

And to ansver to first question, I hope yes, one time they (sims) will look like these videos, and maybe better, just question when ;)

[mondo]

Check out Ace combat 6 for some seriously impressive graphics. Yes, its an arcade game and doesn't contain some of the complex calculations a sim contains but visually, its stunning.

[Jughead]

Oleg says that the SoW engine will look "cinematic". So SoW ought to look as good as these vids. Unless ofcourse Oleg is exagerating. But just by looking at some of the graphics in the new games coming out, Oleg could be damn close with his statements.

[virre89]

Graphics will definantly looks like this someday, but this is as said b4 pure CGI.
Already graphics are somewere close to this if simulators were to use all the latest graphical technolegy etc im sure they'd not look far from this :P, anyway BOB will still be a Huge leap from the IL2 engine and look just awsome in comparance.

Btw gr8 cgi movies just awsome and love the sounds.

[Tbag]

From a technical point of view probably in 10 to 15 years. Nothing more than a guess of course. But who is going to model a whole flightsim to that degree of detail? Not sure if that would pay of. Hope Oleg stays in buisiness!

[IceFire]

We're not that far off really...what you see there isn't really stunningly impressive. IL-2 almost does all of that already. There are some effects that I hope to see in Storm of War which will take us closer or past that with maybe some small exceptions.

[BadAim]

Aggreed. BOB, I think will look damn close to that from the start, assuming of course a state of the art machine. I really doubt capability has anything to do with it, realtime computing power is the biggest obstacle, next is manpower (read money).

So really it's about power and money. Can't get away from it can we?

[Golden_Eagle_FM]

I am ready to bet that in ten years from know real time interactive raytrayced animation of the quality of what we have on those short films will be available at less than 1000 US$ (today value) for general purpose desktop machines.

This is if CPU design evolution goes as planned that is the massive multicore CPU's of the Larabelle type that Intel has already demonstrated with a raytraced version of a very popular FPS game. We are talking of CPU's with 80 cores and plus.
We are now at the stage were we have quadcore CPU's with a max frequency around 3GHz available on the market and in general use. I do not believe the frequency will increase much, but a breaktghrough in technology may allow some major steps. Shrinking the size of CPU components in combination of high voltages for high frequencies, and high heat dissipation problems is already starting to hit some fundamental physical barriers that generate many unwanted effects as for ex. electron migration. We see that it is a few years know that with INTEL and AMD have stalled between 2.5 and 3.2 GHz. Maybe we get to 4GHz but it is not really of much use. On the other hand the fundamental limits to core multiplication in the CPU are much further away.

If we apply Moore's law in conservative manner that is a doubling every 24 months then in 2009 we have 8 cores in 2011 16 cores, in 2013 32 cores, in 2015 64 cores and in 2017 128 cores. The advantage of this approach is that you can lower the frequency, and run say at 1.5 GHZ max and you can get full use of the shrinkage of components as you do not have any more the heat issue. So all the real estate can go to low heat dissipating and simpler architectured cores and so also with even more shrinking without major problems (but for sure there is also an atomic phisical limit here).
The final result is promising. Better 80 cores at 1 GHz then 4 at 4 Ghz. Roughly speaking and to have it simple you get 80 Ghz of power instead of 16.

So it is better to go the lower frequency+low heat dissipation+simpler architecture and higher multiple-core path then the other way round.

If these multi-multicore CPU's do get on the market then there will be enough processing power to get all the software rendering algorithms executed in realtime (60FPS).

The problem is that the whole polygon based culture and tools in game design will have to be readapted to optimally generate raytraced images.
But also the whole software generation community will have to be geared to use massively parallel hardware. This last point may be the most difficult issue.

So to finish I would say that we will have in ten years the hardware capability on the table for sure. I will not bet even a dime that we will be able to use that capability. That will be the major roadblock.

Gold

[Tbag]

Agree 100% Eagle! Well, let's say I agree as far as I can follow :D

@IceFire: Can you please send me a copy of the IL2 you play? My copy can't keep up with these vids!

[virre89]

Think he means in realism.
Anyway it won't look like this counting cinematic wise i hope everyone understands the amazing details going into charachter looking over his shoulder etc just won't happen really, that's part of CGI(MOVIES).

Even back when games like Medal of Honor Allied Assault and BF 1942 was released they had CGI intros which look beyond todays graphics more or less so, still remember this is actually a type of MOVIE.

[kgwanchos]

Yes accept what you are saying Virre ,that is a movie we are seeing totally scripted etc etc. I guess a few of the things which might apply to BOB would center around a more immersive environment. For eg the sky and clouds, terain and water are very impressive and thrown in the mix really start to add to the the photorealistic effect. The character animations to that level I agree are really not nessesary (i think they let the whole thing down in fact) but little touches can make all the difference... remember how cool it was when you first realised that pilots in IL2 where tracking their opponents with their head movements (still miss the rain drop rivulets on the canopy in the first demo anyone ??)
For my money i still think the IL2 series looks pretty amazing at times especially when you consider how long its been around. I think BOB will be a stunner and well worth the extra overtime to pay for a super computer (Im waiting for the game before upgrading).....

[virre89]

Yes accept what you are saying Virre ,that is a movie we are seeing totally scripted etc etc. I guess a few of the things which might apply to BOB would center around a more immersive environment. For eg the sky and clouds, terain and water are very impressive and thrown in the mix really start to add to the the photorealistic effect. The character animations to that level I agree are really not nessesary (i think they let the whole thing down in fact) but little touches can make all the difference... remember how cool it was when you first realised that pilots in IL2 where tracking their opponents with their head movements (still miss the rain drop rivulets on the canopy in the first demo anyone ??)
For my money i still think the IL2 series looks pretty amazing at times especially when you consider how long its been around. I think BOB will be a stunner and well worth the extra overtime to pay for a super computer (Im waiting for the game before upgrading).....

Hehe :),
As for computer i'd not wanna hesitate or anything but by upgrading to say Intel Core Duo or Quad and one of the 8800GTS you'd be close to safe for the release of BOB i dare to almost promise. But indeed i mean no reason to upgrade yet since the game ain't around unless you need it :), altho i bought my new comp 3 months ago but yet again i also play a`lot of Cod4, Crysis etc.

Looking forward to BOB quite abit, it will be worth the wait for sure.

[mondo]

Even back when games like Medal of Honor Allied Assault and BF 1942 was released they had CGI intros which look beyond todays graphics more or less so, still remember this is actually a type of MOVIE.

Its funny to go back to play games from the last 15 years that have used CGI intro's, like BF1942 and thinking at the time that the graphics could never be replicated in real time. BF1942 is only 6 years old as well which shows just how quickly yesterdays CGI quality graphics appear as todays real time graphics.

[Bearcat]

Touchy is the guy with the 1942 thing.. he does some great stuff...

I am ready to bet that in ten years from know real time interactive raytrayced animation of the quality of what we have on those short films will be available at less than 1000 US$ (today value) for general purpose desktop machines.

This is if CPU design evolution goes as planned that is the massive multicore CPU's of the Larabelle type that Intel has already demonstrated with a raytraced version of a very popular FPS game. We are talking of CPU's with 80 cores and plus.
We are now at the stage were we have quadcore CPU's with a max frequency around 3GHz available on the market and in general use. I do not believe the frequency will increase much, but a breaktghrough in technology may allow some major steps. Shrinking the size of CPU components in combination of high voltages for high frequencies, and high heat dissipation problems is already starting to hit some fundamental physical barriers that generate many unwanted effects as for ex. electron migration. We see that it is a few years know that with INTEL and AMD have stalled between 2.5 and 3.2 GHz. Maybe we get to 4GHz but it is not really of much use. On the other hand the fundamental limits to core multiplication in the CPU are much further away.

If we apply Moore's law in conservative manner that is a doubling every 24 months then in 2009 we have 8 cores in 2011 16 cores, in 2013 32 cores, in 2015 64 cores and in 2017 128 cores. The advantage of this approach is that you can lower the frequency, and run say at 1.5 GHZ max and you can get full use of the shrinkage of components as you do not have any more the heat issue. So all the real estate can go to low heat dissipating and simpler architectured cores and so also with even more shrinking without major problems (but for sure there is also an atomic phisical limit here).
The final result is promising. Better 80 cores at 1 GHz then 4 at 4 Ghz. Roughly speaking and to have it simple you get 80 Ghz of power instead of 16.

So it is better to go the lower frequency+low heat dissipation+simpler architecture and higher multiple-core path then the other way round.

If these multi-multicore CPU's do get on the market then there will be enough processing power to get all the software rendering algorithms executed in realtime (60FPS).

The problem is that the whole polygon based culture and tools in game design will have to be readapted to optimally generate raytraced images.
But also the whole software generation community will have to be geared to use massively parallel hardware. This last point may be the most difficult issue.

So to finish I would say that we will have in ten years the hardware capability on the table for sure. I will not bet even a dime that we will be able to use that capability. That will be the major roadblock.

Gold

Yes well just consider where sims were in 1998 CFS1 (http://www.youtube.com/watch?v=F2qvc2wdjEM&feature=related)... compare that with 46. Back then Pcs hadnt crossed the 233MHZ mark yet....


Its funny to go back to play games from the last 15 years that have used CGI intro's, like BF1942 and thinking at the time that the graphics could never be replicated in real time. BF1942 is only 6 years old as well which shows just how quickly yesterdays CGI quality graphics appear as todays real time graphics.



Remember that video for CFS3 (http://www.youtube.com/watch?v=OKSK9AvLUK4)? That was what 2002.. LMAO... I was so souped up .. I thought the sim would look like that... Boy was I disapointed... LOL.

[Golden_Eagle_FM]

If I understand you right Bearcat I see two points in your reply:

1) More computing power does not necessarily mean better quality (visually) games.

2) We are far from having used all the existing hardware power available today.

I totally agree. One path which will also bring gains and which I have not mentioned is to invest into better usage of existing hardware. But this is a human/financial factor.

I think there are two issues with that:

a) The graphic hardware chain from CPU/GPU up to and included the display as well as the sound system does allow us to have a home cinema wide-screen type environment to play with our simulators. But for such an experience we need more and more details.

On the "primitive" systems of 15 years ago the LOD had to be crude for multiple reasons like CPU/GPU performance, low resolution etc. etc. to get to an acceptable framerate.
My personal experience is that framerate, sound and right control are THE three things for the immersiveness and the success of this very dynamic and interactive type of games like flight simulation. Maybe the fact of being an active pilot makes me feel like this. From the hardware side we have all these three things available today to some extent.

Allow me a small diversion here:
(I allways put the visual quality to a level were I get enough FPS, but I agree then that visually the result can be poor. So I had allways to buy the most powerful system available to improve display quality. In fact I fully change my system every year and I allways have the most powerful GPU available so I may switch graphic board every six months or so. A pretty expensive hobby. My systems are entirely assembled by myself which allows me to pick the best components.)

But a high LOD to be visually interesting means also to "really" put more details in the scenery, the aircraft models etc. etc. This is hours and hours of work because more and more REAL details have to be put in. More CPU power means more refined physical models, flight models and so on. But again all this has to be painstakingly "hand"crafted in. In the past when the polygon budget was so tight, crude and approximative models were okay. Today a Nvidia GX2 board can generate > 1 billion textured pixels in real time (at 60 fps). This translates in at leat 5 million polygons (transformed, textured, shaded etc.) in real time. All this is fine but to now you must model the locking pin on the wheel axle (i am not kidding) , the screws which are along the canopy frame, the fluid tube that runs along the landing gear and so on. The texture has to be also with high LOD with all the small scratches etc. etc. When I see the LOD for the aircraft that are being modelled for SOW I start to wonder if this game will ever come out. Oleg is practically doing the work that an aircraft manufacturer may do to build a real plane with all the inner details. We must add also that this work is not enough, the same LOD has to be put in the scenery, ground detail, ships, transport, buildings, trees, the sound effects, lighting, meteoroligical events, wind, clouds etc. Nothing is worse than beautifully modelled airplanes moving around in a very crude environment and interacting with crude targets like ships, artillery etc..

This brings me to the second comment:

b) The economical financial factors. Software and coding throughput does not, very unfortunately evolve at the same speed as the hardware does. The evolution in software is more along the lines of generation changes of software engineers. Tools have improved and automating some coding tasks too, but still there is a huge gap between software and hardware.
I am afraid that the financial economics of human code production may simply limit the LOD (visual, physical etc.) that can be put in a game that is sold for max 40 US$ (in US) and around 80 $ (in Europe). Either you can have a budget like hollywood has for CGI movies or somhere you are stuck.
On the other hand flight simulators (I do not speak for Microsoft Flight Simulator which is a special case) like IL2 do adress a low volume niche market.

So one conclusion:

First we must recognize that Oleg made miracles with the IL2 saga, and I really wonder what his business model is.

Second we should be ready to accept to spend not 30 but 300 US$ for our passion/hobby and maybe then there is a chance we get the best possible game making the most use of the available hardware. 300 is not so much after what we really spend on hardware in general. Very strangely people are ready to spend much more on hardware but start to cry and shout when they must pay 10 more bucks on a software that is extremely complex and difficult to produce and that allow them to play for 6 years with free support. Even the hardware does not last such long time.

Software has not the visibility of hardware (No nice PC case with led lights flashing and blinking, coolers spinning, display, keyboard etc. etc.). Software feels immaterial and as such is being considered of little value (Worth just the CD or DVD) and should be granted. It is a completely wrong perception. This perception and total lack of respect for software is in my opinion the base why it is so common to pirate, steal, copy software.

The truth value lies much more in software then in hardware. We should all start thinking about it if we want to tip the balance of economics in our favor.

Gold

[mondo]

Remember that video for CFS3 (http://www.youtube.com/watch?v=OKSK9AvLUK4)? That was what 2002.. LMAO... I was so souped up .. I thought the sim would look like that... Boy was I disapointed... LOL.

You and me both and probably a whole load of other sim fans. I watch that video in awe and then played CSF3...

[Luffe]

Remember that video for CFS3 (http://www.youtube.com/watch?v=OKSK9AvLUK4)?

Wow. That Intro is awful. So many wrong details, I don't even know where to start lol.

And, yes, I bought into the hype as well...

[mondo]

Looking back at it now, the video is crap ain't it.

[Bearcat]

Eagle what I meant was that between the sim technology and the hardware that level of graphical qulaity may not be too far off.

[Avala]

I think that we will have something like that with SOW BoB, just without human figures. Must I say "remember my words?" :)

[Xiola]

I am ready to bet that in ten years from know real time interactive raytrayced animation of the quality of what we have on those short films will be available at less than 1000 US$ (today value) for general purpose desktop machines.

This is if CPU design evolution goes as planned that is the massive multicore CPU's of the Larabelle type that Intel has already demonstrated with a raytraced version of a very popular FPS game. We are talking of CPU's with 80 cores and plus.
We are now at the stage were we have quadcore CPU's with a max frequency around 3GHz available on the market and in general use. I do not believe the frequency will increase much, but a breaktghrough in technology may allow some major steps. Shrinking the size of CPU components in combination of high voltages for high frequencies, and high heat dissipation problems is already starting to hit some fundamental physical barriers that generate many unwanted effects as for ex. electron migration. We see that it is a few years know that with INTEL and AMD have stalled between 2.5 and 3.2 GHz. Maybe we get to 4GHz but it is not really of much use. On the other hand the fundamental limits to core multiplication in the CPU are much further away.

If we apply Moore's law in conservative manner that is a doubling every 24 months then in 2009 we have 8 cores in 2011 16 cores, in 2013 32 cores, in 2015 64 cores and in 2017 128 cores. The advantage of this approach is that you can lower the frequency, and run say at 1.5 GHZ max and you can get full use of the shrinkage of components as you do not have any more the heat issue. So all the real estate can go to low heat dissipating and simpler architectured cores and so also with even more shrinking without major problems (but for sure there is also an atomic phisical limit here).
The final result is promising. Better 80 cores at 1 GHz then 4 at 4 Ghz. Roughly speaking and to have it simple you get 80 Ghz of power instead of 16.

So it is better to go the lower frequency+low heat dissipation+simpler architecture and higher multiple-core path then the other way round.

If these multi-multicore CPU's do get on the market then there will be enough processing power to get all the software rendering algorithms executed in realtime (60FPS).

The problem is that the whole polygon based culture and tools in game design will have to be readapted to optimally generate raytraced images.
But also the whole software generation community will have to be geared to use massively parallel hardware. This last point may be the most difficult issue.

So to finish I would say that we will have in ten years the hardware capability on the table for sure. I will not bet even a dime that we will be able to use that capability. That will be the major roadblock.

Gold


Surely there is a size limit here, 4 cores is already close to filling a normal size CPU, so 10 cores will mean that the size of the CPU as a whloe will need to grow.

As we have already hit limits with electron migration, then the size of the core caanot go much smaller than it already is with the limits as we know them. The paths are already so small as to have individual electrons escaping and causing errors.

Wouldnt a 80 core CPU be massive? (with the limits as we now know them?)

[mondo]

I think before we get so many cores on a single chip we'll go back to what was taking off yet disappeared 5 or 6 years ago; motherboards for enthusiast PC's with support for 2 CPU's. SLI made a come back after years being undeveloped so motherboards with multiple CPU support might also make a comeback for the home PC.

Don't count out GPU's either. Looking at the specs of AMD's new ATI card, its not hard to imagine video cards with 2 GPU's containing multiple cores in systems with multiple CPU's each with multipe cores, or the amalgamation of GPU and CPU like IBM's Cell.

[BigC208]

What i find an interesting developement is Nvidia acquisition of Ageia's physics technology. That will find it's way into mainstream video cards and should help with more realistic ingame physics without dragging the cpu/gpu down. As far as realistic aircraft modeling goes, just look at some of the aftermarket products for FS2004/X. Real Air simulations Spitfire package and Shockwave's WW fighters look and fly realisticly. If Oleg releases anything looking and flying close to that I'll be a happy simmer for a long time to come.

[Golden_Eagle_FM]

Eagle what I meant was that between the sim technology and the hardware that level of graphical qulaity may not be too far off.

Cross fingers Bearcat. Be the gods with you and make us have this quality shortly.

Gold

[Golden_Eagle_FM]

Surely there is a size limit here, 4 cores is already close to filling a normal size CPU, so 10 cores will mean that the size of the CPU as a whloe will need to grow.

As we have already hit limits with electron migration, then the size of the core caanot go much smaller than it already is with the limits as we know them. The paths are already so small as to have individual electrons escaping and causing errors.

Wouldnt a 80 core CPU be massive? (with the limits as we now know them?)


Xiola here are some links that could be of interest on the topic:

http://www.news.com/Intel-shows-off-80-core-processor/2100-1006_3-6158181.html

http://www.pcworld.com/article/id,128924-c,intel/article.html

http://news.zdnet.com/2100-9584_22-6158181.html

http://softwarecommunity.intel.com/articles/eng/3724.htm

Gold

[nearmiss]

When I consider the limitations of the current IL2 series one thing comes to mind as the principal issue... visuals.

IMO, a reasonably priced virtual reality type hood would provide the type of visuals we need for a combat flight sim.

[fireflyerz]

In answer to your question... remember Pac man... and look where we are now.:cool:

[mazex]

Yes, games will look like this someday. When I started flying simulations, my screen looked like this:
http://www.thelegacy.de/pics/screen/f/Fighter_Pilot.gif


Ahh, Fighter Pilot - don't know how many hours I've spent with that old game in the mid-80:ies. Those where the days when we dreamed of upgrading our Spectrums from 16k to 48k instead of graphics cards with more memory than the server park of NASA in those days... I remember thinking that the graphics where great ;) The most impressive thing is that the trusty Spectrum only managed to pull something like 4 fps while running those simplistic graphics (if I dare call those vectors that) - and no one complained! I recently fired up my old speccy (my father found it on the attic) and booted Fighter Pilot. Had to use a cable from the sound card of my PC with a downloaded "recording" as could not find a working MC player (have the orginal tape so no accusing of pirating please!). After "flying" that game with the mushy rubber keys of the Spectrum for a few minutes you get some perspective on how good our current games really are ;)

/Mazex

[Moritz]

The Japanese video is the story of Saburo Sakai. Recognize the markings and the great rendition of his face. Love to see the whole video!:grin:

[4H_V-man]

Mazex hit the nail on the head. Compared to what was available to the general public just a few years ago the products we have available to us now are amazing! And then you go to a forum and over half the posts are people %issing and moaning about how much they want this or how much they deserve that. Were I in Oleg's position I'd be tempted to say 'To hell with it' and walk away.

[Xiola]

Xiola here are some links that could be of interest on the topic:

http://www.news.com/Intel-shows-off-80-core-processor/2100-1006_3-6158181.html

http://www.pcworld.com/article/id,128924-c,intel/article.html

http://news.zdnet.com/2100-9584_22-6158181.html

http://softwarecommunity.intel.com/articles/eng/3724.htm

Gold

Really interesting reading, thanks mate!

[lbuchele]

First of all: Are you folks sure that SOW will looks even close to this videos?
Second:I´m praying for that not be true,because may damage marriage and social life...
Third:Agree with Golden Eagle, some point, in our natural search for professional level simulations, will be necessary to pay more for this sims (look for the time to do new sims and size of the developing teams comparing to ten years before,easy to understand)

[Codex]

Xiola here are some links that could be of interest on the topic:

http://www.news.com/Intel-shows-off-80-core-processor/2100-1006_3-6158181.html

http://www.pcworld.com/article/id,128924-c,intel/article.html

http://news.zdnet.com/2100-9584_22-6158181.html

http://softwarecommunity.intel.com/articles/eng/3724.htm

Gold

Hehehe, I'll see your 80 core chip and raise you to 512 cores.

http://www.engadget.com/2006/11/06/japanese-researchers-build-512-core-math-coprocessor/

[usr]

Does anyone think PC simulators will ever look this good ?

The rendering quality is only a few years from now, if technological development keeps it's current pace (which is not to be taken for granted, just think about how much aviation technology improved from 1900 to 1950 and how little it changed in the next 50 years, in contrast to the first five decades).

But (someone said it before, but i'll repeat in support): even with the best rendering quality you will never see the dramatic camera-work. Maybe in arcade games, but never in a simulation. The current IL-2 engine used with dramatic camera would look much more impressive than the rendering quality of these clips applied to strict first person gameplay graphics.

Another, totally different aspect: Youtube compression tends to blur CGI videos (including computer game videos) in a way that does a very "good" job at making it more difficult to distinguish from real life video footage. In the game, you always notice the jagged polygon edges, sharp shadows and total of focal blur. On youtube, all these problems are replaced by the general youtube fuzzyness.