Jump to content
Sign in to follow this  
tavichh

Does engine 4.0 improve framerates at all / Why does Combat Mission Lag?

Recommended Posts

Hello there!

So I own SF, BN, and BS and I have the same problem on all of them I can't get a decent framerate to which I feel as ignorance to release games like combat mission in 2010's and have them not run on modern computers. Basically I see people on YouTube with pure **** running as systems and combat mission runs with 60fps with maxed out settings and yet I can't obtain thirty. I've tried everything; Installed it on SSD, overclocked CPU and GPU and absolutely nothing does anything to my frames. So my question is as it says in the title: will upgrading to engine 4.0 increase my frames at all? Also, from a geek's perspective, I can't put my finger on what is causing lag on Combat Mission? All I can bring it down to is poor optimization. Games like Command run perfectly on my system and that game is doing much more behind the scenes than Combat Mission.

If anyone wants to help troubleshoot my frame issue; specs are below:

GPU: R9 290 CPU: Intel i5 4690k Ram: Kingston 8gb 2133mhz SSD: Kingston HyperX 120gb

Share this post


Link to post
Share on other sites

You didn't state the processor speed. I'd be surprised if it were 3.0 Ghz which would explain your problem with slow fps.

 

Edit: I checked the 4690k and see that meets that speed. It's rather puzzling in that case. Unless it's the motherboard.

 

Edited by The Steppenwulf

Share this post


Link to post
Share on other sites

I don't believe 4.0 includes any additional optimization, but I could be wrong. I personally have not noticed any change, but then I haven't had any particular issues running CM 

Share this post


Link to post
Share on other sites
18 minutes ago, The Steppenwulf said:

What size monitor are you running your CM games on and what is the resolution? 

Both monitors are 1920x1080 at 23" so playing on a different monitor doesn't have much any different impact

29 minutes ago, sburke said:

I don't believe 4.0 includes any additional optimization, but I could be wrong. I personally have not noticed any change, but then I haven't had any particular issues running CM 

So the 4.0 features on the front page are the only additions to CM?

34 minutes ago, The Steppenwulf said:

You didn't state the processor speed. I'd be surprised if it were 3.0 Ghz which would explain your problem with slow fps.

 

Edit: I checked the 4690k and see that meets that speed. It's rather puzzling in that case. Unless it's the motherboard.

 

Motherboard was switching from AMD AM3+ Asus Sabertooth 990fx to LGA 1150 PC Mate

Reason for the switch was alot of games don't run well on AMD as they have poor multi-core support so I went with Intel. I thought CM would be one of those games that would be fixed during my switch of motherboards. It wasn't.

Share this post


Link to post
Share on other sites
2 hours ago, thryckz said:

Both monitors are 1920x1080 at 23" so playing on a different monitor doesn't have much any different impact

 

That's the resolution I'm playing on with the best of the R7 card range - same size monitor. 

 

2 hours ago, thryckz said:

Motherboard was switching from AMD AM3+ Asus Sabertooth 990fx to LGA 1150 PC Mate

Reason for the switch was alot of games don't run well on AMD as they have poor multi-core support so I went with Intel. I thought CM would be one of those games that would be fixed during my switch of motherboards. It wasn't.

Your motherboard could still be bottle-necking with the bus speed. No good having a turbo 512-bit card if your motherboard is restricted to 128-bit for example. I'm not saying it is this, but it's worth investigation. That said you'd likely see the issue in other games too..

I'd take out any texture mods you have installed. I do have lots of mods installed but some of the BN mods for example are HD textures and quite demanding on my last system so I'm appreciative of the potential demand some mods have.

You are quite right about the poor optimisation though for CM. It's well known that the engine wasn't coded for multi core handling. On the other hand it isn't the only game out there that suffers from this fundamental disadvantage in data processing 

Share this post


Link to post
Share on other sites
On 12/27/2016 at 4:03 PM, The Steppenwulf said:

That's the resolution I'm playing on with the best of the R7 card range - same size monitor. 

 

Your motherboard could still be bottle-necking with the bus speed. No good having a turbo 512-bit card if your motherboard is restricted to 128-bit for example. I'm not saying it is this, but it's worth investigation. That said you'd likely see the issue in other games too..

I'd take out any texture mods you have installed. I do have lots of mods installed but some of the BN mods for example are HD textures and quite demanding on my last system so I'm appreciative of the potential demand some mods have.

You are quite right about the poor optimisation though for CM. It's well known that the engine wasn't coded for multi core handling. On the other hand it isn't the only game out there that suffers from this fundamental disadvantage in data processing 

Fresh install of windows/cm.

Also I don't think its the bus speed since 128bit era games run perfect but its not comparable 100% imo to cm due to cm being much more demanding on calculations than other games so I don't believe that it would be the mobo but yeah thats a bottleneck.

Also is there a reason behind the development process of cm? I feel like they are bleeding money with expansions and standalone when they should make a game that will run on computers made within the past five years. I don't get them at all.

Share this post


Link to post
Share on other sites

AMD's OpenGL drivers are not very optimized and CM games probably tend to run a little slower on them compared to equivalent hardware and drivers from Nvidia. However if you're expecting 60fps with CM, then I'll tell you that your expectations are TOO HIGH. The size of your battle, the size of the map, what is on the map (buildings, forests, significant/numerous elevation changes, units, etc.), the display resolution and the viewpoint of the camera can make significant differences to the framerate. Admittedly CM isn't as graphically optimized as a number of other games, but there are also other details about how those games render and limit what is on the screen.

When you mention 'Command' as a game to compare to are you speaking of MatrixGames 'Command Live' series or something else ?

The biggest graphical performance optimizations for CMx2 came with the 2.0 Upgrade, which made some significant changes in the models and some of the graphic routines (still utilizing OpenGL). There have been changes here and there since, but nothing matches the amount of change that the 2.0 Upgrade instituted.

OpenGL performance trails DirectX because the video card driver developers concentrate on DirectX bugs and performance due to its popularity (a vast majority of games on the PC utilize this API). OpenGL was chosen by Battlefront because it is used on both Windows and MacOS platforms; DirectX isn't. Battlefront has only one programmer currently (and this has been so for the majority of its existence), so it is an unrealistic expectation to develop the game with more than one API. Game engines take a long time to develop, so it is much easier to create content and slowly update the game engine than it is to re-write the game engine. The last significant update to the game engine was done back in 2003-2007 with CMSF being the initial game release. During this period no new games (for Battlefront) were released (though other games were distributed by them). On top of this CMSF had a bit of a painful birthing process to becoming a decent game, which may have taken about 2 more years of additional development time with patches fixing various issues (and modules being released). CMBN's release showed off an enhanced game engine that became the standard for the rest of the CMx2 series (though CMSF and CM Afghanistan are part of the CMx2 family, there have been some important improvements since their release).

CM's engine isn't as advanced as some other games in the marketplace. This is definitely one of the problems of having only one programmer to develop everything. For awhile there were two programmers, but the second one moved on to the more lucrative fields of AI finance programming (from where he came from). When you have a team (or several teams) of programmers you can spread the load out and have individuals or teams concentrate development time on a variety of optimizing features or other feature sets of the game engine. This can also lead to quicker development of a game (one or two years, depending on where you start). A lot of game companies will also license a 'game engine' to base their game design on. These licensed game engines are then customized for their game, which allows them to avoid a lot of heavy ground work that can delay development of a game by years (even with several programmers or teams) if a game engine were designed from scratch. Licensing a game engine really wasn't a financial possibility for Battlefront and CM and there may have been no desirable engine to develop with on the market at the time. The CM series is a very impressive development considering that only one programmer came up with a vast majority of the code (there are others who do the art work/models and data).

Developing a game engine takes a significant amount of time and simultaneously reduces the number of new products to sell and make a living off of. So creating anything significantly new is treacherous ground for a game developer. If a new game engine is too ambitious it could take too long to develop, causing financial difficulties for the developer and possibly force them to shutdown if there isn't a decent revenue stream to live off of. So in the case of Battlefront any new game engine may have to be developed in conjunction with new releases of existing product. This lengthens the development time of the new engine, but increases the possibility that a game may eventually result from it. 

Share this post


Link to post
Share on other sites
12 hours ago, Schrullenhaft said:

AMD's OpenGL drivers are not very optimized and CM games probably tend to run a little slower on them compared to equivalent hardware and drivers from Nvidia. However if you're expecting 60fps with CM, then I'll tell you that your expectations are TOO HIGH. The size of your battle, the size of the map, what is on the map (buildings, forests, significant/numerous elevation changes, units, etc.), the display resolution and the viewpoint of the camera can make significant differences to the framerate. Admittedly CM isn't as graphically optimized as a number of other games, but there are also other details about how those games render and limit what is on the screen.

When you mention 'Command' as a game to compare to are you speaking of MatrixGames 'Command Live' series or something else ?

The biggest graphical performance optimizations for CMx2 came with the 2.0 Upgrade, which made some significant changes in the models and some of the graphic routines (still utilizing OpenGL). There have been changes here and there since, but nothing matches the amount of change that the 2.0 Upgrade instituted.

OpenGL performance trails DirectX because the video card driver developers concentrate on DirectX bugs and performance due to its popularity (a vast majority of games on the PC utilize this API). OpenGL was chosen by Battlefront because it is used on both Windows and MacOS platforms; DirectX isn't. Battlefront has only one programmer currently (and this has been so for the majority of its existence), so it is an unrealistic expectation to develop the game with more than one API. Game engines take a long time to develop, so it is much easier to create content and slowly update the game engine than it is to re-write the game engine. The last significant update to the game engine was done back in 2003-2007 with CMSF being the initial game release. During this period no new games (for Battlefront) were released (though other games were distributed by them). On top of this CMSF had a bit of a painful birthing process to becoming a decent game, which may have taken about 2 more years of additional development time with patches fixing various issues (and modules being released). CMBN's release showed off an enhanced game engine that became the standard for the rest of the CMx2 series (though CMSF and CM Afghanistan are part of the CMx2 family, there have been some important improvements since their release).

CM's engine isn't as advanced as some other games in the marketplace. This is definitely one of the problems of having only one programmer to develop everything. For awhile there were two programmers, but the second one moved on to the more lucrative fields of AI finance programming (from where he came from). When you have a team (or several teams) of programmers you can spread the load out and have individuals or teams concentrate development time on a variety of optimizing features or other feature sets of the game engine. This can also lead to quicker development of a game (one or two years, depending on where you start). A lot of game companies will also license a 'game engine' to base their game design on. These licensed game engines are then customized for their game, which allows them to avoid a lot of heavy ground work that can delay development of a game by years (even with several programmers or teams) if a game engine were designed from scratch. Licensing a game engine really wasn't a financial possibility for Battlefront and CM and there may have been no desirable engine to develop with on the market at the time. The CM series is a very impressive development considering that only one programmer came up with a vast majority of the code (there are others who do the art work/models and data).

Developing a game engine takes a significant amount of time and simultaneously reduces the number of new products to sell and make a living off of. So creating anything significantly new is treacherous ground for a game developer. If a new game engine is too ambitious it could take too long to develop, causing financial difficulties for the developer and possibly force them to shutdown if there isn't a decent revenue stream to live off of. So in the case of Battlefront any new game engine may have to be developed in conjunction with new releases of existing product. This lengthens the development time of the new engine, but increases the possibility that a game may eventually result from it. 

1) Command from MatrixGames is the one I'm speaking of. IMO its on par with CM in terms of the calculations it's doing.

2) I didn't know it was just one developer. Do you know why isn't he just reaching out for other developers for hire? CM could benefit so much from faster development.

3) Why isn't CM on steam? I have a developer/publisher account on steam and I can tell you the takeaway amount from steam that you hear on the internet is 100% bull****. However, I can't say the exact amount due to NDA but if Combat Mission were to be on steam and at least 100 people buy it; It would be flourishing.

4) I noticed it was OpenGl when I tried recording a video with it then later saw mac versions available on the store. I guess that/you answered that question.

5) I know AMD is **** with OpenGL (bought this card last summer because there was a crazy 50% deal going on as I went from Nvidia to AMD) but I feel as if there is at least some optimization to be had on Battlefront's part. Better garbage collection would be nice regarding CM. Also, due to the popup of Linux and Mac games on steam AMD is just going to release better OpenGL drivers which in turn will nullify the reasoning of any error on AMD's part. But in the mean time; Better optimization would be great from CM where and when it can happen.

I know there's probably a thousand posts about this but I don't want to go into 2017 knowing I still cannot run Combat Mission.

Share this post


Link to post
Share on other sites

I don't know what YouTube videos you're referring to but no one gets consistent 60 fps in CM. 20-30 in large scenarios is typical on even the highest end machines. I play on a laptop I bought in 2009 and I don't much care about FPS but to each their own.

Share this post


Link to post
Share on other sites

tavich -

1) The Command series appears to be quite a bit simpler in its graphics, though it is probably doing quite a bit of calculations 'under the hood' with spotting routines. The CMx2 series performs A LOT of spotting checks, though there is a limit on how frequent those are since the game is NOT multi-threaded. However there can be A LOT of units needing LOS/LOF checks in the game.

2) Being a game developer of non-FPS wargames is a limited market (despite what some customers may think). I don't know the financial flows of Battlefront, but I would assume that they may not be as big as some people are estimating. To have a larger staff you have to make certain predictions on how good your future sales are of the products you intend to develop. If you're off significantly, then you may have to shut down or curtail operations. Also hiring a significant number of staff may require some sort of funding in place before you develop the game. This is something most small developers can't come up with (if you want full time staff). If you do, then you're dealing with investors who often want a large return on their investment (in a certain time frame to boot). This can cause significant stress on time tables and content development choices ('make an FPS game or an MMORPG, etc. since that will generate more revenue...'). These are the kind of choices and pressures that Battlefront wanted to stay away from. The second programmer was someone who came on when financial flows appeared to be good and he had a certain track record and qualifications that Battlefront was already familiar with (and he was able to accommodate the financial limitations of Battlefront's business model).

3) Steam has been a point of contention between Battlefront and some of its customers who are fans of the Steam model. In the past Battlefront looked at the Steam offerings at the time, but decided that they could make more money doing it themselves (more income per copy sold). This wasn't just some sort of decision based on greed, but the idea that they didn't need to sell more copies to 'break even' compared with non-Steam sales (which was working for them already, but did have limitations). While Steam may get you 'more eyes' on the game there is also the issue with how sales are performed on the platform and the expectation from MANY customers (especially ones that may not be familiar with the product) that prices will ALWAYS drop at some point. With the expected number of sales being limited in Steve's eyes, the sales model didn't make sense since they would have to sell far more copies to, again, 'break even' with non-Steam sales. There are several threads on this subject and you may not agree with some of the impressions or decisions of Steve on this matter, but I don't think that they are considering a Steam presence in the future.

4) It took awhile before a Mac version came out, but that was the primary reason the OpenGL API was used. The CMx1 series was programmed in DirectX (and RAVE for the Mac, now since deprecated with OS X).

5) While this specific Knowledgebase article mentions the Radeon R9 290x it might be applicable to your situation, though I don't know how much, if any, improvement may result from these recommendations. I don't know the details of CM's memory management. It is a 32-bit game, so something is happening in that regard or the game would probably quickly error out, especially with the larger battles. One other difference that larger game developers may have with their games is the number of LODs (Level of Detail) models and textures that they have. CM definitely has some (or the performance would be even more 'unoptimized'), but other developers may have quite a few more levels (since they have the staff to create them). DirectX 10 and above may offer some more optimized graphical calls that can't quite be duplicated in OpenGL, but I don't know the performance advantage such calls may have (since some may involve routines that CM is not trying to utilize). A lot of AMD's recent efforts have been aimed more at DirectX 12 than any other API (though they did have a significant hand in the development of Vulkan with their own Mantle API). CM's open environment (in terms of the camera views) does necessitate some compromises in performance optimization, especially with some shading routines (which would result in very low performance).

AMD, along with Nvidia, slowly improves their drivers. However they often 'break' things in trying to optimize the performance of their drivers. This can result in some drivers being 'broken' for quite awhile before the issue gets addressed. Again DirectX gets the lion's share of development time and optimization. Bugs that the larger developers come across get addressed much quicker (though it can still be a number of months or longer, even for AAA titles). So things do improve, but usually not by very large margins. Usually a performance improvement involves a little image quality tradeoff, the problem sometimes is that the occasional tradeoff results in a bug for some applications.

Edited by Schrullenhaft

Share this post


Link to post
Share on other sites
Sign in to follow this  

×
×
  • Create New...