SoW:BoB 64-bit/Multicore optimised?

[dduff442]

Hi All,

Will SoW:BoB take advantage of 64-bit, quad-core architecture, hyperthreading etc?

The PC gaming industry has gotten seriously out of whack. $800 twin-gpu setups are common, but no more than a tiny handful of games use more than 2 CPU cores, more than 2GB of ram or are multi-threaded. A great opportunity for deep AI and immersive, interactive and open worlds is being totally ignored by the industry. This is all the stranger as RAM and CPUs are relatively cheap compared with high-end graphics cards.

I'm hoping SoW:BoB will make use of the PC features that will be mainstream in the coming years. FSX is the only game that does at the moment (seriously, this is a fact). SoW:BoB is ideal for hyper-threading and the 16GB of RAM the Home Premium editions of Win 7/Vista allow would permit environments of incredible detail to be portrayed, along with intensive layered AI processing etc.

There's a lot of expectation out there for BoB (http://ve3d.ign.com/articles/features/52083/Voodoo-Extreme-2009-Readers-Choice-Awards/p14/c4/), and I'm no different. I held off buying a new PC until it became imminent. (I'm very reassured at the busy look of your offices BTW!) A bit of info about the processing capabilities of BoB at this point might whet people's appetites even more....

Regards,
dduff442
(have played online as Kestrel666)

[Tree_UK]

I think it may be some time yet before we know what system spec will be required to run the sim in all its glory, to date we haven't seen anything that would tax any modern system.

[flyingbullseye]

to date we haven't seen anything that would tax any modern system.

Crysis will.

Flyingbullseye

[Qpassa]

Also FS:X with a lot of extras like : airports like Barajas(Madrid),Heathrow(London) etc

[airmalik]

Crysis will.

Flyingbullseye

I'm pretty sure he meant that he hasn't seen anything from SoW updates that will challenge a modern system. I agree.

[flyingbullseye]

Ahh, got ya. In that case then I agree with Tree.

Flyingbullseye

[Flyby]

I hope Sow_BoB does support 64bit multicore processors. I don't think we've seen anything yet from Oleg that will tax such a system, but perhaps that should worry us a bit? Dynamic clouds and weather. Water, complex AI, plus improvements to FMs and DMs FMBs (if they are in there) may tax a multi-core system. I think it's most of the behind the scenes stuff that will put a hurt on even a modern system. I recall that IL2 taxed my old 2.8P4 and 6800 Ultra in lots of ways. Couldn't have too much flak, avoid flying over big cities, not too many paratroops in the air at once, or watch the slide show. I know. Mere speculation until O.M speaks. Fair enough.
Flyby out

[Flanker35M]

S!

Falcon 4.0 used multicore to an extent and being quite old game ;) So let's hope SoW will use more cores to even out the workload, could finally put my i7 to work!

[Flyby]

It's a sim I'm glad I never sold. I have the much-wanted binder version, and even went out and bought F4AF (though I never loaded it up). I guess I'm a bit nostalgic. :)
Anyway, I hope Oleg is into the multi-core thing. Wouldn't it be crazy if SoW came out without multi-core support? Imagine trying to figure out an affinity trick for it ala DCS-Black Shark! Wait! Isn't that blaspheming? :D
Flyby out

[yarbles]

My only knowledge of apps designed to work in multi-core/cpu environments is stuff like Oracle in Solaris/Red Hat on Enterprise workstations. In order for those apps to be optimized to take full advantage of the 32 cpu workstations w/ 256 GB of RAM - an army of coders were employed. Oracle's business depends on optimized I/O. I think our expectations for BoB are way out of wack. Folks want crazy AI, DoD quality FMs, etc.. it won't happen. There just isn't enough man power in Oleg's lab to code the stuff to fully take advantage of the modern day computing power. I would imagine will get something similar to what we've already seen but w/ better eye candy. We may see BoB have a better handle on how to deal with more planes in the air and better terrian but other than that under a very restrictive budget, I wouldn't expect to see things much more advanced that what's already been done.

[Flyby]

My only knowledge of apps designed to work in multi-core/cpu environments is stuff like Oracle in Solaris/Red Hat on Enterprise workstations. In order for those apps to be optimized to take full advantage of the 32 cpu workstations w/ 256 GB of RAM - an army of coders were employed. Oracle's business depends on optimized I/O. I think our expectations for BoB are way out of wack. Folks want crazy AI, DoD quality FMs, etc.. it won't happen. There just isn't enough man power in Oleg's lab to code the stuff to fully take advantage of the modern day computing power. I would imagine will get something similar to what we've already seen but w/ better eye candy. We may see BoB have a better handle on how to deal with more planes in the air and better terrian but other than that under a very restrictive budget, I wouldn't expect to see things much more advanced that what's already been done.
Wow! Wouldn't that be something if SoW came out behind the power curve instead of ahead of it? You may be on to something, yarbles. Then again, people spent billion$ trying to come up with a gaming system that would run IL2 smoothly with all options turned on (only to be hindered by single-core processors). So it's possible that a lot of items in IL2 that caused stutters, and single-digit fps might actually perform quite well with a dual or a quad. Still, I'm filled with an exquisite dread that when SoW comes out even the most powerful gaming systems out there won't be able to run it full tilt. In LOMAC the clouds could be an fps-killer, not to mention the water (but that was a coding thing too, iirc). I wonder how much of an FPS killer flying over London will be? Wouldn't it be ironic if, in BoB, people avoided missions over London because the city, and the flak brought a modern gaming rig to it's knees? Will Oleg allow that to happen to an area so crucial to the game- play immersion and the title? I think Oleg will have so much going on in SoW that a dual-core will be the minimum requirement. Remember those screen shots of ground objects (that can be effected by damage)? Also, even with a small team, SoW_BoB has been quite a while in the making. So it's possible that there won't be a system that can run it at full tilt. Funny to think about, eh?
Flyby out

[mazex]

Hi All,

Will SoW:BoB take advantage of 64-bit, quad-core architecture, hyperthreading etc?

The PC gaming industry has gotten seriously out of whack. $800 twin-gpu setups are common, but no more than a tiny handful of games use more than 2 CPU cores, more than 2GB of ram or are multi-threaded. A great opportunity for deep AI and immersive, interactive and open worlds is being totally ignored by the industry. This is all the stranger as RAM and CPUs are relatively cheap compared with high-end graphics cards.

I'm hoping SoW:BoB will make use of the PC features that will be mainstream in the coming years. FSX is the only game that does at the moment (seriously, this is a fact). SoW:BoB is ideal for hyper-threading and the 16GB of RAM the Home Premium editions of Win 7/Vista allow would permit environments of incredible detail to be portrayed, along with intensive layered AI processing etc.

There's a lot of expectation out there for BoB (http://ve3d.ign.com/articles/features/52083/Voodoo-Extreme-2009-Readers-Choice-Awards/p14/c4/), and I'm no different. I held off buying a new PC until it became imminent. (I'm very reassured at the busy look of your offices BTW!) A bit of info about the processing capabilities of BoB at this point might whet people's appetites even more....

Regards,
dduff442
(have played online as Kestrel666)

Well, you describe the "strange" scenario well above - how come that so few games use multiple cores effectively but rely on beefy GPU:s? I guess you are not a programmer? It's really hard to use many threads effectively in an application that uses a main render loop that runs the screen updates at very high steady speed (many threads doing the render is even worse - though in some cases possible). When you are going to synchronize/join data for that render thread with a number of separate other threads (AI, network, loading stuff etc), you will have to keep you tounge in the right mouth :) The threads running on different cores really live their own life (and should do so) so the syncronization has to be effective to not get threads waiting for other threads etc (producing stutters and lag). Getting a bunch of threads to dance together while the main render thread runs at a steady 60 loops/sec is like a tango with four wifes - without annoying any one of them ;)

EDIT: It's naturally a grail to be able to be able to write the optimal game engine using all cores available in a system... Even writing a game that runs two threads well can break programmers and companies backs - F4/Microprose anyone?

[Tree_UK]

Getting a bunch of threads to dance together while the main render thread runs at a steady 60 loops/sec is like a tango with four wifes - without annoying any one of them ;)

LOL, got to be quote of the year, nice one Mazex S! :grin::grin:

[JG27CaptStubing]

S!

Falcon 4.0 used multicore to an extent and being quite old game ;) So let's hope SoW will use more cores to even out the workload, could finally put my i7 to work!

There was quite a bit of genius behind the development of Falcon 4. While not perfect by any stretch it was a multithreaded application back in the day of 1998. It basically gave the game a 40 -60% improvement of FPS when over the FLOT. The only time this really came into play was in the campaign mode. It ran the real time 2D world as a separate thread. It also helped out when dealing with the Disaggregation in the 3Dworld. Amazing game even to this day in its current form it has an even better improvement working with multiple cores.

FSX is a bit of a misnomer in terms of utilizing Multicore. The team kind of cheesed it a bit because all it does is deal with the preloading of the game when coming into other areas. Sure it helps out but nothing like leveraging other cores to handle things like AI, and Flight Modeling. To my knowledge there won’t be a real time campaign element to it. But like someone mentioned the sky has opened up in terms of processing power. Lots of memory now and CPU/GPUs to leverage.

[dduff442]

Well, you describe the "strange" scenario well above - how come that so few games use multiple cores effectively but rely on beefy GPU:s? I guess you are not a programmer? It's really hard to use many threads effectively in an application that uses a main render loop that runs the screen updates at very high steady speed (many threads doing the render is even worse - though in some cases possible). When you are going to synchronize/join data for that render thread with a number of separate other threads (AI, network, loading stuff etc), you will have to keep you tounge in the right mouth :) The threads running on different cores really live their own life (and should do so) so the syncronization has to be effective to not get threads waiting for other threads etc (producing stutters and lag). Getting a bunch of threads to dance together while the main render thread runs at a steady 60 loops/sec is like a tango with four wifes - without annoying any one of them ;)

This is all just nonsense. The reasons games don't employ computing power effectively are:

A) The development cycle of nearly all PC games is tied to that of console games. Consoles focus almost totally on graphical bells and whistles to the exclusion of all else.

B) Game devs don't profit (or rather feel they will) from a focus on long-term, evolving codebases. Graphics engines etc are stable technology. Individual game engines on the other hand are throwaway crap for the most part, with few going through more than 2 or three iterations. If Blah I is a classic, Blah II will be derivative, and Blah III is sure to add loads of half-assed junk options as the devs know the wheezing wreck of a codebase is already beyond rescue. Most codebases are dead already if the original design team is disolved. This hasn't been the 1C:Maddox way in the past and I hope it doesn't go down this route now.

C) The development cycle of most games has little to do with engineering. It nearly always starts with the concept art, followed by rounds of meetings by the suits as the company considers distribution, market segment issues, the company's portfolio of other titles etc. The result is never something anyone might love: it's something random disinterested businessmen think is okay, the lowest common denominator creatively speaking.

D) The actual coders are only involved at a low level and only get their say after (C). 'Game engines', where they exist at all, cover graphics and (at best) some scattered elements of the modeled environment. The wheel is reinvented for each new release as far as gameplay is concerned. Why do you think gameplay in FPS games has barely moved forward since Thief in 1998?

Multi threading on multiple cores needn't even be that efficient on all processors with modern PCs to still make a big difference to gameplay. Not all processing need be done on a frame-by-frame basis. For example, top-level AI heuristics could provide direction to lower-level manoeuvre AI which could in turn guide AI behaviour on each frame. Synchronisation of the first two strands isn't critical and memory bandwith issues etc. would not be significant.

Developer commitment to high-quality engineering and a sustainable codebase can be very profitable as the Oracle example cited elsewhere proves. The reason it features so little in the gaming industry is that it no more guarantees profitable games than do good lighting and camerawork in the movies, so it takes a back seat to the often stupid ideas of the business 'creatives'. This attitude is short-sighted, however.

A stable, tightly-knit dev team committed to incrementally improving a sustainable product could blow away the competition in many areas of gaming. Flight sims are evidence of this. Image processing and compression tech are other good examples. Look at the humble jpeg, basically the technology that made the internet possible. JPEG (and MPEG, MP3, MP4 and other derivatives) will be 18 this year. That sort of thing is real technology, not the throwaway stuff pumped out by the games industry.

Regards,
dduff

[mazex]

This is all just nonsense. The reasons games don't employ computing power effectively are:

A) The development cycle of nearly all PC games is tied to that of console games. Consoles focus almost totally on graphical bells and whistles to the exclusion of all else.

B) Game devs don't profit (or rather feel they will) from a focus on long-term, evolving codebases. Graphics engines etc are stable technology. Individual game engines on the other hand are throwaway crap for the most part, with few going through more than 2 or three iterations. If Blah I is a classic, Blah II will be derivative, and Blah III is sure to add loads of half-assed junk options as the devs know the wheezing wreck of a codebase is already beyond rescue. Most codebases are dead already if the original design team is disolved. This hasn't been the 1C:Maddox way in the past and I hope it doesn't go down this route now.

C) The development cycle of most games has little to do with engineering. It nearly always starts with the concept art, followed by rounds of meetings by the suits as the company considers distribution, market segment issues, the company's portfolio of other titles etc. The result is never something anyone might love: it's something random disinterested businessmen think is okay, the lowest common denominator creatively speaking.

D) The actual coders are only involved at a low level and only get their say after (C). 'Game engines', where they exist at all, cover graphics and (at best) some scattered elements of the modeled environment. The wheel is reinvented for each new release as far as gameplay is concerned. Why do you think gameplay in FPS games has barely moved forward since Thief in 1998?

Multi threading on multiple cores needn't even be that efficient on all processors with modern PCs to still make a big difference to gameplay. Not all processing need be done on a frame-by-frame basis. For example, top-level AI heuristics could provide direction to lower-level manoeuvre AI which could in turn guide AI behaviour on each frame. Synchronisation of the first two strands isn't critical and memory bandwith issues etc. would not be significant.

Developer commitment to high-quality engineering and a sustainable codebase can be very profitable as the Oracle example cited elsewhere proves. The reason it features so little in the gaming industry is that it no more guarantees profitable games than do good lighting and camerawork in the movies, so it takes a back seat to the often stupid ideas of the business 'creatives'. This attitude is short-sighted, however.

A stable, tightly-knit dev team committed to incrementally improving a sustainable product could blow away the competition in many areas of gaming. Flight sims are evidence of this. Image processing and compression tech are other good examples. Look at the humble jpeg, basically the technology that made the internet possible. JPEG (and MPEG, MP3, MP4 and other derivatives) will be 18 this year. That sort of thing is real technology, not the throwaway stuff pumped out by the games industry.

Regards,
dduff

Well, calling other peoples opinions nonsense is just troll behavior and I won't swallow that bait...

[dduff442]

Well I'm sorry I upset you but your remark about guessing I'm not a programmer got me going. I think enough reasons other than technical grounds were given for the near total lack of either memory- or processor-intensive games was given in any rate. Other fields in computing use these features routinely.

[Feathered_IV]

SoW better have 250+ aircraft in the air at any given time.
If it pretends to be anything more than the piddling skirmishes of Il-2, it had better take advantage of 64bit and multi core systems.

[dduff442]

SoW better have 250+ aircraft in the air at any given time.
If it pretends to be anything more than the piddling skirmishes of Il-2, it had better take advantage of 64bit and multi core systems.

I'd love to see this happen. That would truly elevate SoW to a level way above the competition. With that many a/c involved, squadron leaders would need to make important decisions in the air which would be novel.

[Blackdog_kt]

I've had ridiculous amounts of aircraft present at the same time in one mission back when i was still flying European Air War. That game is older than IL2 and it could do 255 aircraft on screen at once without problems. Of course, it wasn't as detailed as IL2.
However, despite the fact that graphics, FM and DM were inferior, it was superior in the amount of random events during missions (returning from an escort mission you could stop time compression/autopilot and fly manually and you'd find convoys to strafe), the way you commanded the AI (and they actually listened and executed) and the amount of aircraft in the air.

I'll be very excited if we can have something similar with SoW quality FM/FM and graphics, because it adds tons of immersion. In that mission, i had 12 FW190s under my command and going up against 36 heavy bombers and 12-24 escorts. Instead of engaging i sat back and radioed for reinforcements twice and lucky me, i got them both times (this didn't happen all the time, sometimes you didn't even get one flight). So, we were now 3 flights of 12 on the LW side (36 planes), against 36 bombers and 2 flights of 12 escorts (36 heavies+24 escorts=60 total USAAF). This was a grand total of 96 aircraft. The thing is, there was a second group of 36 bombers a couple of miles behind the first one.

FM/DM and graphics aside, just seeing 132 aircraft going at it in a small part of the sky, following realistic tactics (ie only the inexperienced enemy AI escorts gave chase to the deck, the rest would just try to spoil your attack runs and stuck with covering the bombers like they should, and you could see top, close and lead cover for each bomber group) and watching flak puffing all around kind of hammers the point home in a very effective manner.

[julian265]

So true dduff. That sort of engineering management is certainly commonplace, in much more than just the gaming industry.

[Les]

...If Blah I is a classic, Blah II will be derivative, and Blah III is sure to add loads of half-assed junk options as the devs know the wheezing wreck of a codebase is already beyond rescue...This hasn't been the 1C:Maddox way in the past...

:lol: Sorry, not having a go at you, but I find that funny, as I think 'IL-2 Sturmovik', 'IL-2 Sturmovik: Forgotten Battles', 'Pacific Fighters' and 'IL-2: 1946' fit your description quite well. We're just lucky there was more to it all than that.

About the 64-bit/multicore thing. After watching and waiting for SOW this long, I'd be disappointed if it's performance isn't more responsive to hardware upgrades than the IL-2 series. In fact I just can't imagine how it couldn't be.

The IL-2 series fell into the same unavoidable trap that the Microsoft Flight Sim series fell into (and is still in with FSX). And that is, when the sims were first built the only foreseeable (bankable) development path for CPU's was in terms of sheer single-core speed increases, and the software had to reflect that. But that path came to a shuddering, overheating, physically limited halt and we all went off down the multi-thread/multi-core path instead, leaving the single-core oriented software behind. Only, we didn't leave it behind, did we? It's still here. For the last few YEARS we've had the hardware and practically no software to make use of it. A whole generation of CPU's (dual core) has been superceded (by quad-cores), and now there are six-core chips about to come out and practically no game or sim has been coded to take full advantage of all those extra cores.

Now, I can't say whether it's even possible, practically speaking, for SOW to make use of those cores or not, as I don't know enough about it. But I'm much more sure about the fact that if there are still scenes in 1946 that a Core i7 920 @ 3.8 GHz can't handle, then SOW is going to be dead in the water judging by what I've seen and heard about it so far.

I honestly don't know how the thing's going to run, let alone expand, if it doesn't follow the hardware. Surely it would be madness to code for the past and not the future.

The fact that they've switched over to DirectX gives me hope that they're willing to make major changes to the way they do things to ensure the future viability of the series. And I'm left thinking, if they can (if anyone can) make it a truly multi-core sim, they will.

Again though, if it turns out to be code-limited like the IL-2 series is, which is to say, if parts of it remain unplayable even six or seven years after its release, I don't think I'll be the only one feeling a bit ripped off. Adding more (and faster) CPU cores and more powerful video-cards, (and therefore more RAM too I guess) must have a beneficial on-screen effect, for the duration of the whole SOW series (if extra features are added along the way, otherwise there must be a reachable point where the sim can be 'maxed out' while still getting a good frame-rate).

Sorry for the rant/over-reaction, but I just realized there is a possibility SOW might not be the combat flight sim I've been waiting for afterall, if it's neutered from the start and left with nowhere to go in terms of being able to take advantage of current and future hardware.

But who knows, maybe none of this multi-core/64-bit stuff is necessary anyway, it just seems illogical though that more in this case wouldn't be better...

Les.

[mazex]

Well I'm sorry I upset you but your remark about guessing I'm not a programmer got me going. I think enough reasons other than technical grounds were given for the near total lack of either memory- or processor-intensive games was given in any rate. Other fields in computing use these features routinely.

Fair enough, being insinuated as a non-programmer is serious stuff ;)

Still - after spending 15+ years of writing multi threaded c++ code, and at the same time writing some games I sometimes get tired when people that are not programmers "buy" the Intel/MS propaganda that you must have 64-bit OS with at least 4 cores for your entertainment PC - and then demand that the game developers must start using their hardware. We all know that Intel would love to increase the CPU speed instead of doing a more complex chip with many cores - but they have reached a technology barrier with the current production process - so they went for the multi core strategy instead.

I first suspected you might be a guy that had ditched his old overclocked E6700 to a new i7 and had realized it was not faster in games - but obviously you do work with software development in some way? You arguments are interesting and I agree with most of them...

In my opinion, the problem is that normal games /that does not have that many AI objects) manage to end up being GPU limited instead of CPU limited, even on the Core 2 family of CPU:s... Is that because the AI is to simplified then? Maybe, but fact is that as very few games are CPU limited - why do the extra work of trying to do an efficient multi threaded engine?

Sure, as the market is now going to a 4+ cores per socket in every new computer, and the raw processing power of each core is not increasing that much - the multi threaded approach is the way to go for the future. But if your engine is not CPU limited today - why do the extra work if you have a tight budget (like most non blizzard projects)? Sure, some obvious candidates like threads for strategic AI and preloading textures to memory etc are candidates today to reduce "stuttering" in the game - but the main render loop is still responsible for a very large portion of the CPU cycles used...

What is your proposal for the multi threaded strategy for games?

Regards /Mazex

[dduff442]

This might be an appropriate moment for a slight tactical withdrawal on my part. While I have extensive experience in databases, my applications experience is fairly limited. I do understand that whereas it's possible to produce a *proven* design using the old-fashioned functional approach (restricted to a single thread/core), it is not possible to do this with multi-threaded applications. Nonetheless, a commitment to quality engineering can produce tremendous benefits.

There are certainly plenty of calculation problems in flight sims. One I've given a fair degree of thought to is perception and a detailed model would probably break the field of vision down to at least three cones. Relevant factors include angle-off from center of vision, angular velocity of target vs field, surface area of target (a complex calculation that might need simplification), contrast vs background, line of sight etc. Some of these factors apply to sensor calculations also.

This kind of calculation is amenable to parallel processing and isn't always time critical. Of course it should be possible to spread the calculation load of a/c physics over several threads as well.

As it happens, I did by an i7 system recently though I did understand at the time that there weren't really any games to stretch even dual-core machines. After 7 years without buying a new PC, I found the job pretty frustrating. There's not much point on spending €850 on a good gaming PC that will choke on applications in 2 years and has limited upgrade potential. On the other hand, there's really no return for gamers in more expensive systems at the moment.

My new PC cost €1800, excluding the screen; I got i7/920, 12GB RAM, HD5870 plus a couple of extras like a TV tuner and sound card. This machine won't get out of 2nd gear for years on any standard game. On the other hand, the €400 graphics card will probably be struggling already in 2 years time. According to NVIDIA fancy PCs are a waste, but they have it backwards. Fancy graphics cards are a waste when lashings of cheap RAM and CPU power that could vastly enrich the gaming experience are available but are ignored by developers.

It looks like PC gaming is on the way out except for niches like flight sims. This is a pity because PC games allow much more interesting games than the clunky game-controllers permit.

Regards,
dduff

[dduff442]

:lol: Sorry, not having a go at you, but I find that funny, as I think 'IL-2 Sturmovik', 'IL-2 Sturmovik: Forgotten Battles', 'Pacific Fighters' and 'IL-2: 1946' fit your description quite well. We're just lucky there was more to it all than that.

About the 64-bit/multicore thing. After watching and waiting for SOW this long, I'd be disappointed if it's performance isn't more responsive to hardware upgrades than the IL-2 series. In fact I just can't imagine how it couldn't be.

The IL-2 series fell into the same unavoidable trap that the Microsoft Flight Sim series fell into (and is still in with FSX). And that is, when the sims were first built the only foreseeable (bankable) development path for CPU's was in terms of sheer single-core speed increases, and the software had to reflect that. But that path came to a shuddering, overheating, physically limited halt and we all went off down the multi-thread/multi-core path instead, leaving the single-core oriented software behind. Only, we didn't leave it behind, did we? It's still here. For the last few YEARS we've had the hardware and practically no software to make use of it. A whole generation of CPU's (dual core) has been superceded (by quad-cores), and now there are six-core chips about to come out and practically no game or sim has been coded to take full advantage of all those extra cores.

Now, I can't say whether it's even possible, practically speaking, for SOW to make use of those cores or not, as I don't know enough about it. But I'm much more sure about the fact that if there are still scenes in 1946 that a Core i7 920 @ 3.8 GHz can't handle, then SOW is going to be dead in the water judging by what I've seen and heard about it so far.

I honestly don't know how the thing's going to run, let alone expand, if it doesn't follow the hardware. Surely it would be madness to code for the past and not the future.

The fact that they've switched over to DirectX gives me hope that they're willing to make major changes to the way they do things to ensure the future viability of the series. And I'm left thinking, if they can (if anyone can) make it a truly multi-core sim, they will.

Again though, if it turns out to be code-limited like the IL-2 series is, which is to say, if parts of it remain unplayable even six or seven years after its release, I don't think I'll be the only one feeling a bit ripped off. Adding more (and faster) CPU cores and more powerful video-cards, (and therefore more RAM too I guess) must have a beneficial on-screen effect, for the duration of the whole SOW series (if extra features are added along the way, otherwise there must be a reachable point where the sim can be 'maxed out' while still getting a good frame-rate).

Sorry for the rant/over-reaction, but I just realized there is a possibility SOW might not be the combat flight sim I've been waiting for afterall, if it's neutered from the start and left with nowhere to go in terms of being able to take advantage of current and future hardware.

But who knows, maybe none of this multi-core/64-bit stuff is necessary anyway, it just seems illogical though that more in this case wouldn't be better...

Les.

While I agree with most of what you say, IMO there are encouraging and discouraging signs from the recent vids and screenshots.

If I had to guess, I'd say what we're seeing is elements of il-2 and SoW spliced together for testing purposes. If that guess is right, it probably means SoW is further off than it might seem from vids of newly modeled a/c in flight. OTOH, it would show that the il2 codebase is still workable whereas most game developers would have had to start from scratch due to poor quality control.

Regards,
dduff

[Igo kyu]

It looks like PC gaming is on the way out except for niches like flight sims. This is a pity because PC games allow much more interesting games than the clunky game-controllers permit.
I think some on the hardware side wanted this, but it isn't working out that way.

The bog-standard Laptop with the integrated graphics is dying, the netbook that replaced it is getting minimally game-capable graphics (nVidia Ion), and cheaper gaming capable laptops seem to be appearing.

All in all, there was a decline in PC gaming for a while, due in part to the competitive graphics on the new consoles. As the consoles have aged and PC graphics have overtaken them, PC gaming is coming back, not strongly yet, but I suspect that before the next generation of consoles arrive it will have recovered more substantially.

Then there will probably be another temporary decline, and if the new consoles come with upgradeable graphics, or PCs don't any more that might not be so temporary, but that's apparently two or three years off yet.

[Flyby]

"It is better to be thought a fool than to open one's mouth and remove all doubt." But here goes. Compared to most of you, I'm a babe in rough waters when it comes to any thing beyond just using a PC. I've been without a gaming PC for some time now (as well as without my beloved IL2). I understand, from what you've discussed, the limitations the coding has had on prior PC games, especially our flight sims. But hasn't it been already stated that SoW_BoB will support multi-core processors? I also wonder if the new generation of GPUs will be able to use their computing powers in SoW as well? Is that a coding issue too? Seems to me we are talking about how far ahead of the hardware curve Oleg's team is or is not. Meanwhile I'd be disappointed if my i7-920 processor is overkill when SoW is released (upon a trembling, unsuspecting world). Better for me to believe that in four years, when I'm more able to afford one, a six-core socket 1366 processor will finally tame SoW_BoB!!! :D Oh, let's not forget the video cards too (Crossfire, SLi, or Hydra configs may get some SoW love too).
Flyby out
PS I hope PC gaming is not in decline. Look at all the cash spent by AMD and Nvidia on product development. Look at all the hardware tech sites that still test CPUs and GPUs not only by running apps, but also PC games like Crysis to test the mettle of those components. Look at the companies that sell gaming PCs. Perhaps they know something we don't? ;)

[Igo kyu]

I've been without a gaming PC for some time now (as well as without my beloved IL2).
IL*2 should run on pretty much anything PC. I noted elsewhere that FB ran on an Athlon (32 bit) 1GHz from 2000ad, and on a Geforce 2 GTS with 64M of ram on it. That was a top end GF2 at the time, but it's nothing now, a PCIe card to beat it would cost less than a month's internet connection.

I also wonder if the new generation of GPUs will be able to use their computing powers in SoW as well?
I wouldn't want that. If the GPU is doing calculations, or physics, it's not doing graphics, and I like graphics, it's what I buy GPUs for. If people are prepared to fund my graphics habit by buying GPUs to make supercomputers from them, that's fine, but it's not what I want from GPUs.

[Flyby]

IL*2 should run on pretty much anything PC. I noted elsewhere that FB ran on an Athlon (32 bit) 1GHz from 2000ad, and on a Geforce 2 GTS with 64M of ram on it. That was a top end GF2 at the time, but it's nothing now, a PCIe card to beat it would cost less than a month's internet connection.


I wouldn't want that. If the GPU is doing calculations, or physics, it's not doing graphics, and I like graphics, it's what I buy GPUs for. If people are prepared to fund my graphics habit by buying GPUs to make supercomputers from them, that's fine, but it's not what I want from GPUs.
My last system ran IL2 pretty well (p4-2.8, 1.5gig of HyperX DDR 3500, and Nvidia 6800 Ultra), but too much flak from ships, for example really slowed it up big time, and the Kamikaze trk hit a low of 7fps. I recall a COOP mission I made where the Allied objective was to shoot down 8 Ju-52 before they could reach their drop point for their paratroops. Well, what a slide show that was when those guys all jumped! LOL!! All this at 12x9 rez.
Good old AGP! But I agree with you about leaving the GPU to do graphics.
Flyby out

[Les]

... We may see BoB have a better handle on how to deal with more planes in the air and better terrian but other than that under a very restrictive budget, I wouldn't expect to see things much more advanced that what's already been done.

At the end of the day, I think this is a realistic expectation.

Oleg has stated we probably won't be able to switch views from plane to plane like we can in IL-2, so as to avoid having to render all planes in such detail, and therefore have more of them in the sky at the same time. I don't think a measure like that would be made if it were possible to keep that feature in by using some multi-threading technique. Or maybe that's more of a current graphics-power issue/limitation.

Whatever the case, in relation to this coding for multi-core thing, it's interesting for me personally to realize I probably have been sucked in by the Intel marketing/hype/spin. I actually believed them when they said games could be coded to make use of multi-cores. When in reality, that's probably just not practical for the most part, if possible at all.

It's interesting to note too that all the new SOW features we've seen on recent updates can be had as a result of the increases in graphics power and CPU core power (not the number of cores) that have taken place since the IL-2 days. And the old SOW features we saw and heard about ages ago, all the meaty nuts and bolts stuff, was never quaranteed to make it to the final version anyway. It was all, we can do this, we can do that, now we've just got to go find a way to put it all together and still have it run at more than 10fps. That's why no promises are made. It's all got to be made to fit, and no single feature can be allowed to drag the whole thing down.

Not saying SOW is just going to be IL-2 all cleaned up graphically, with just enough new CPU-intensive features to make even the most powerful release-date system crawl, but...it's probably just going to be IL-2 all cleaned up, with just enough new CPU-intensive features to make even the most powerful release-date system crawl. With less planes. :grin:

And if it is, well, okay, it's just a game, or a sim, or a way to learn about history and what some people have experienced for real in the past. So, it's no big deal.






But if they can get that multi-core thing working...! :-P

(That would make SOW as innovative as the original IL-2 was, and that, all these years later, would be amazing.)

Les.

[mazex]

Whatever the case, in relation to this coding for multi-core thing, it's interesting for me personally to realize I probably have been sucked in by the Intel marketing/hype/spin. I actually believed them when they said games could be coded to make use of multi-cores. When in reality, that's probably just not practical for the most part, if possible at all.


Well. It's not all spin ;) But it's a hell of a lot harder than making an Oracle database use all the cores/sockets for all those parallel queries. Falcon 4 is a good (and old) example where the multithreaded code works rather well with the ground war running on a separate thread. FSX also implemented a usable multithreaded code in SP1 where the texture loader was placed on a separate thread so that the loading of ground textures did not pause (stutter) the main render loop (much more fluid gameplay after that). Most modern block buster FPS games are multithreaded to some extent, but using it with mixed results... The bottom line is that it's not the salvation that Intel tries to make it (at least for gaming). Games using 4 cores in a way that really improves performance significantly does not exist as far as I know...

So sure, I guess with a lot of work the overall strategic AI in SoW could be placed on a separate thread handling the stuff outside of "the bubble" around the player, and stuff like object loading and textures. Maybe that won't be that important in a game that does not use unique satellite textures for the ground. I guess most of it will be loaded at mission load (especially if it uses the extra memory that the process may use on a 64-bit system ;)).