Sorry your browser is not supported!

You are using an outdated browser that does not support modern web technologies, in order to use this site please update to a new browser.

Browsers supported include Chrome, FireFox, Safari, Opera, Internet Explorer 10+ or Microsoft Edge.

FPSC Classic Product Chat / Maybe this will help the lag.

Author
Message
Klick
18
Years of Service
User Offline
Joined: 24th Jan 2006
Location: ANGEL COMMANDing...
Posted: 13th Mar 2006 02:50
I`ve got an idea. Since FPSC can`t support huge worlds, then, can we resize everything to miniature size? So, like a metre long field will become a kilometre to the player and so on. But the problem will be how to resize the weapons when player is carrying them. Any suggestions to takle the problem?

Oh and also, a hill would become a mountain, so the player can roam around more freely. I got the idea from playing one of the counterstrike maps, something about the teams roeming around miniature sized a kitchen.

All suggestions and pointers will be deeply appreciated,
A.C.


[A]NJL`s [N]on-[G]oody-goody [E]xtermination [L]eague [C]arrying [O]ut [M]ass [M]urdering of [A]NJL`s [N]efarious [D]evotees -Essence of ANGEL COMMAND
Bloodeath 6 6 6
19
Years of Service
User Offline
Joined: 5th Nov 2005
Location: Sierra vista in indonesia
Posted: 13th Mar 2006 02:59
not a bad idea but how would we resize all the weapons and media and everything withoutit taking huge amounts of time?

Nunticaelitusphobia---im scuurrred of the internet
uman
Retired Moderator
20
Years of Service
User Offline
Joined: 22nd Oct 2004
Location: UK
Posted: 13th Mar 2006 04:58 Edited at: 13th Mar 2006 05:09
Its been suggested.

Can be done easily at the click of a button in some engines.

I dont believe its currently practical in FPSC anyway as you have no way of scaling the world relative to the camera (player) which if thats the case the camera (player) would be the only giant as it cant be scaled or repositioned at least not easily as far as I know. A correct camera view at least without resort to the source code may be not possible? Perhaps I am wrong about that and will stand corrected.

Effectively everything else could be rescaled with enough effort in theory but then again you must remeber that FPSC only works the way it does as it works to a fixed tile based system with fixed tile and level height and so on - everything works and is based on that principle from the ground up.

If someone wants to give it a go it would be a lot of work and the results unknown and possible unpredicatable. Knowing FPSC I would not give it much chance of success but who knows until someone does it.

In those engines that I have known to allow world rescale command a 50% scale certainly speeds up everything - though thereafter again the results can mean instability.

I personally doubt anyone at the user end can do anything with any certainty about the lagg issue as its I believe inherantly linked to lets say issues inside the engine being an erractic behavioural trait liable to crop up in wide ranging circumstances and it moves around with a life of its own. If you dont know what causes it then you cant ascertain what you can do to aleviate it. You can ask TGC what causes it if you like - dont know if they will resppond or even if they know exactly what the reason is and if so what can be done to ovecome it. I believe its linked to AI thinking but may be incorrect.

Re-sacle of the world may not help with lagg but it would obviously allow a larger world if you have the enegry to set it all up - its a massive task unless you can click and play it all to scale.

Better you or TGC take it on than me.

I wish you luck with that.




=ChrisB=
19
Years of Service
User Offline
Joined: 23rd Jun 2005
Location: starring into a viewfinder
Posted: 13th Mar 2006 07:45
Two smilies??? OMG, uman are you sick?
LoL.

This i an excelant idea, but yes resizing would be a bit trickey, especialy cause the player doesnt have a hight setting. Stupid hard code

God damnit kenny!
Klick
18
Years of Service
User Offline
Joined: 24th Jan 2006
Location: ANGEL COMMANDing...
Posted: 13th Mar 2006 13:23
Wasn`t the FPSC Source Code posted around here to help users to get around the
Quote: "Stupid hard code"
thing?

,
A.C.

[A]NJL`s [N]on-[G]oody-goody [E]xtermination [L]eague [C]arrying [O]ut [M]ass [M]urdering of [A]NJL`s [N]efarious [D]evotees -Essence of ANGEL COMMAND
uman
Retired Moderator
20
Years of Service
User Offline
Joined: 22nd Oct 2004
Location: UK
Posted: 13th Mar 2006 19:26 Edited at: 13th Mar 2006 19:43
Ay,

But getting involved with the source is going to be a nightmare. Changing or overiding any of the default settings in code as relative to the whole concept of how FPSC works is going to have major impact on everything else.

Believe me and I have not even seen the source - starting messing around in there is a recipe for a programmers life and not gamemaking one. It for those that love messing around with sorce codes.

Believe it or not - despite any external thinking and issues or bugs in the source - overall FPSC works the way it does good or bad for real reasons.

Any serious change to the way FPSC works in source and you might as well re-write the whole thing as it impacts on everything across the board including Collision and Physics - a massive task.

CellBlock deserves our admiration. Perhaps he or another could advise as I am no expert and just comment from a general viewpoint on how I see the the FPSC engine as an outsider.

It may be easy to achieve if theres any such thing associated with FPSC - but I very much doubt it.

I would think its easier to fix the lagg and culling and any other of the issues which cause slowdowns and fps drops which nothing else like rescaling the world will correct or the problems may well be made worse. Until the speed issues are fixed then anything else is a non starter for 10. And those issues may never be fixed. Its seems unlikely that TGC are prepared to tackle these issues and have made indications that they wont at least in anything like a near timescale if ever but improvements to the DBpro situatuion may help that thinking - I just doubt it.

The best and only real solution is to fix the speed and culling bugs and issues or overcome them by whatever method in source adhereing to the way in which FPSC is designed to work to free up fps and raise the overall max fps limit. i.e. improve the performance by any given factor achievable - then increaee the default world size.

Like I said you would be almost forced to re-write at least much of the whole - perhaps? Maybe I'm completely incorrect - Ask someone else if anyone knows eaxctly.

Chris - only one smiley coming

Les Horribres
19
Years of Service
User Offline
Joined: 20th Nov 2005
Location: My Name is... Merry
Posted: 14th Mar 2006 05:20
Uman... get to the point.

I think this is kinda wrong. Scaling something smaller does not mean it won't render. And THAT is the problem. Easier just to create crappy low poly versions of objects.

Mosillivo: Fires Rage, Earth Rumble, Evil Reigns, Cities Tumble
Join the NJL: The War Has Begun, Which Side Are You On?
Nunticaelitusphobic (Scared of Internet)
Van B
Moderator
22
Years of Service
User Offline
Joined: 8th Oct 2002
Location: Sunnyvale
Posted: 14th Mar 2006 08:09
I'm stumped as to how scaling the level would improve performance - I mean, if it acts as a polygon reducer then I'd understand that, but I don't see how scaling a FPSC level would do much except destroy it.


Van-B

Put away, those fiery biscuits!
uman
Retired Moderator
20
Years of Service
User Offline
Joined: 22nd Oct 2004
Location: UK
Posted: 14th Mar 2006 11:04
Mosillivo,

I dont understand your post I did not mention anything about rendering.

Van B,

Quote: "I'm stumped as to how scaling the level would improve performance."


(I did have an engine which could do that within the parameters I mentioned above. No idea how - something to do with BSP worlds - complex and finicky things which I know absolutely nothing about).

Me neither, I agree, its a non starter for various reasons and best left as a suggestion which is what I will do now and leave it to the experts.

transient
19
Years of Service
User Offline
Joined: 22nd Apr 2005
Location: Australia Zoo
Posted: 14th Mar 2006 12:25
Maybe this isn't a great performance-enhancing idea, but I like the prospect of playing Godzilla.

I remember a game called Giants which had some cool building-smashing gameplay.

instinct is more valuable than intelligence.....
Reality Forgotten
FPSC Reloaded TGC Backer
18
Years of Service
User Offline
Joined: 28th Dec 2005
Location: Wichita Falls TX
Posted: 14th Mar 2006 12:46 Edited at: 14th Mar 2006 12:47
I think there are a few things that worked to kill lag for me and I have a few threads on them and Uman has helped me kill lag quite a bit (there is a thread for that as well) It seems that there are a lot of possibilities causing lag. I say tinker with how you put things in the level, and watch what you put in the level.

"I am ready to meet my Maker. Whether my Maker is prepared for the great ordeal of meeting me is another matter."
-- Winston Churchill
SpyDaniel
18
Years of Service
User Offline
Joined: 4th Feb 2006
Location: United Kingdom
Posted: 14th Mar 2006 14:22
Scaling the level wont reduce the polys, it just makes them smaller. So its not reducing them at all, you would still have the same lag issues, and probably a really ugly game with bad texturing.
uman
Retired Moderator
20
Years of Service
User Offline
Joined: 22nd Oct 2004
Location: UK
Posted: 14th Mar 2006 16:12 Edited at: 14th Mar 2006 16:13
With reagrd to Lagg and not Rescaling.

We can do a lot even as FPSC stands to increase fps or maintain it at an acceptable level throughout at around say 30 fps constantly no matter how complex a level - but its not available by default - the users unfortunately has to do a lot of additional work from what they get out of the box - but it seems it can be done.

There are threads about this and personally I am still extending and trying to find additional optimisation methods to ensure this.

The load and unload entities method I am still evaluating - it seems it works well and I have increases of around 8 fps in some areas despite taking advantage of this method and adding additional quanities of dynamic entities. This is now even more important as it seems that dynamic entities do not shut themselves down as they should by default when the player is far away from them but they remain Always Active.

Even in outdoor levels which I am shortly to work more intensively on this can work effectively and I can maintain constantly acceptable fps. More work testing needs doing in outdoor levels.

The (serious) lagg issue however is a different matter. Thus far I am unable to find an exact single cause for this and there may be multiple influences. TGC (as they do) have not really acknowledged this as an issue, nor made any helpful comment as to what the cause is in any helpful way or detail. They simply rarely comment.

Unfortunately there is no way to remove the serious lagg issue from a level so despite all you do in optimisation this problem which can seriously impact on level design and gameplay is one you will have to live with. Work as hard and as cleverly as you like you wont kill it. Only TGC can do that.

While this issue exists it will remain the only serious issue affecting end gameplay which can make the difference between a professional quality level of gameplay and a poor one. The rest by and large is down to the end user - You can do what you can to limit its impact by limiting or changing your game design and it can have a serious influence on that - but you cant fix it.

Given that the serious lagg issue was fixed or could be overcome and the end user was to call upon all the optimisation methods available and yet to be discovered I am of the opinion that fps could be maintained throughout a game at high enough fps to meet with modern acceptable standards of 30fps (excepting normal momentary fluctuations) - even in fairly open outdoor type levels if the design is well thought out.

I have 30fps overall in outdoor levels that are quite open and complex - though lagg spots are impossible to remove so have to be worked around - affecting the level design unduely.

More testing and optimisation needs to be done in all levels - possibly a fix or workaround by isolation of the exact or multiple causes could be found to the lagg issue but its seemingly inherently an inbuilt engine inefficiency issue so a real end user fix seems to be impossible.

If I find an anwser I will post deatils at the forum. Dont expect any solution any time soon though.

Just a note on optimisation - it seems to me that FPSC does indeed take into account any info/data instructions from any source that is attached to any entity or object. So for insatnce dont presume that it does not take account of uneeded attached/referenced script files or settings in editor or anywhere else : i.e. simply switching off Physics = No is not enough FPSC still seems to take account of all the other settings below that and has to check everything against that entity constantly. Thats just one example so the morale is if its not needed - remove its reference to the object and help save vital fps.

Reality Forgotten
FPSC Reloaded TGC Backer
18
Years of Service
User Offline
Joined: 28th Dec 2005
Location: Wichita Falls TX
Posted: 14th Mar 2006 16:22 Edited at: 14th Mar 2006 16:24
You could remove the main and destroy .fpi's from all static entities since they are not needed. They are just simply there as level filler. So for instance, in my latest level I have 589 static entities and I deleted both main and destroy .fpi's thus removing 1178 scripts that are not needed and freeing up the engine to do other things. and yes the level works perfectly fine.

I am now trying the process of elimination method and removing as much info on an entity as possible before the engine rejects it. I will post more when I find more out.

"I am ready to meet my Maker. Whether my Maker is prepared for the great ordeal of meeting me is another matter."
-- Winston Churchill
Van B
Moderator
22
Years of Service
User Offline
Joined: 8th Oct 2002
Location: Sunnyvale
Posted: 14th Mar 2006 20:00
Sounds like some clever tweaks Reality, be sure to post your findings .


Van-B

Put away, those fiery biscuits!
Reality Forgotten
FPSC Reloaded TGC Backer
18
Years of Service
User Offline
Joined: 28th Dec 2005
Location: Wichita Falls TX
Posted: 14th Mar 2006 20:35
I will

"I am ready to meet my Maker. Whether my Maker is prepared for the great ordeal of meeting me is another matter."
-- Winston Churchill
Les Horribres
19
Years of Service
User Offline
Joined: 20th Nov 2005
Location: My Name is... Merry
Posted: 14th Mar 2006 21:25
Except for one small fact, FPSC only uses the textures off of the original model data. Other then that, static entitys are just apart of the universe.dbo. Any script attached won't run, and since you need to be dynamic to load the script... It doesn't really make sense.

As for the amount of entitys, you can have hundreds of entitys dynamic, static, the works. And as I have said many times before, it all depends on how your map is made.


Uman, I ment stop typing essays. I prize myself on reading those essays... but now I would rather you cut it short.

Mosillivo: Fires Rage, Earth Rumble, Evil Reigns, Cities Tumble
Join the NJL: The War Has Begun, Which Side Are You On?
Nunticaelitusphobic (Scared of Internet)
Reality Forgotten
FPSC Reloaded TGC Backer
18
Years of Service
User Offline
Joined: 28th Dec 2005
Location: Wichita Falls TX
Posted: 14th Mar 2006 21:52
ok but does it hurt to remove them? not really. you know as well as I do that things that are supposed to happen with the media and scripts for this engine are supposed to react a certain and some of them don't. with that said who really knows what the engine is doing with the supposedly dead scripts for static entities. the fact of the matter is they are still there and can be and probably are read by the engine.

"I am ready to meet my Maker. Whether my Maker is prepared for the great ordeal of meeting me is another matter."
-- Winston Churchill
Les Horribres
19
Years of Service
User Offline
Joined: 20th Nov 2005
Location: My Name is... Merry
Posted: 14th Mar 2006 22:11
I am farly certain that that data is ignored. If you watch the game build, it will say making entitys with a big number. And then applying scripts, with a smaller number... or that is what occurs over here.

Mosillivo: Fires Rage, Earth Rumble, Evil Reigns, Cities Tumble
Join the NJL: The War Has Begun, Which Side Are You On?
Nunticaelitusphobic (Scared of Internet)
Reality Forgotten
FPSC Reloaded TGC Backer
18
Years of Service
User Offline
Joined: 28th Dec 2005
Location: Wichita Falls TX
Posted: 14th Mar 2006 23:19
well the engine is supposed to do alot of things.

"I am ready to meet my Maker. Whether my Maker is prepared for the great ordeal of meeting me is another matter."
-- Winston Churchill
Les Horribres
19
Years of Service
User Offline
Joined: 20th Nov 2005
Location: My Name is... Merry
Posted: 15th Mar 2006 01:46 Edited at: 15th Mar 2006 01:46
That is true... it is suppose to make my uber game for me! (Noobism 101)

Mosillivo: Fires Rage, Earth Rumble, Evil Reigns, Cities Tumble
Join the NJL: The War Has Begun, Which Side Are You On?
Nunticaelitusphobic (Scared of Internet)
Reality Forgotten
FPSC Reloaded TGC Backer
18
Years of Service
User Offline
Joined: 28th Dec 2005
Location: Wichita Falls TX
Posted: 15th Mar 2006 02:04
lol

"I am ready to meet my Maker. Whether my Maker is prepared for the great ordeal of meeting me is another matter."
-- Winston Churchill
uman
Retired Moderator
20
Years of Service
User Offline
Joined: 22nd Oct 2004
Location: UK
Posted: 15th Mar 2006 05:36 Edited at: 15th Mar 2006 06:13
Mosillivo,

The point is as far as optimisation of levels are concerned we can all understand that static entities should not call any data.

That removing references to unecessary scripts on dynamic entities should help with lightening the load the engine has to cope with goes without saying.

There are quite a number of things known and possibly many other things that FPSC does that it should not by design or otherwise do and account for at run time.

If no one tries to eliminate all methods of removing all possible unecessary sources of calls to data the engine might make at run time then no one will ever know what data processes might be eliminated. Not even TGC know or understand everything thats going on far from it. Many of the discovered issues would not exist if they did and had eliminated them.

Again we all know we can have many hundreds of world objects, static and dynamic entities because we have them - just as many or more than you have - with optimisation the aim is to have more of them - thats not an issue at all - what is an issue is that all of those things have a bearing on fps and the more of everything you have the greater the burden on the engine and the lower the fps.

You dont have to do only what you wish to do - others are seeking to find methods of maximising optimisation by methods known or yet to be discovered that you may not wish to embrace.

Only by doing will results be achieved - for the individual no harm can come from experimentation but a loss of ones own labour but that can lead to a benefit for all.

It is from my insistance of removing all unecessary possibility of data reading by FPSC that I guess scripts attached to static entities are in question perhaps when its uneeded - though it cant do any harm.

Personally I would accept that its very probably unecessary - but if time is available removing the refs is worthwhile as an additional measure in an effort to aid optimisation. Static entities rendered as world objects really should not be seen through by the engine either but its almost certain that at least in some instances FPSC sees through them, around them or just ignores them completely as if they dont exist so dont talk about what static entities or indeed any other entities are supposed to do - its what FPSC actually does that matters and who knows excatly what that is realtive to everything in the world at compile or run time.

As a further possible means of optimisation - there are yet further things I can think of that may be possibly done relating to static entities and optimisation which I have not yet spoken of here if they become necessary to be called upon.

Who knows until these things are tried.

One things certain current methods being used particularly the loading and unloading of entities on the fly and removal of unecesary dynamic entities attached scripts as well as minimising and streamlining the actual data content within the scripts themselves is in my case at least having the desired effect by returning an increase of an average of 7fps in low spots pushing them back to the 30fps mark where previously they were minus that 7fps. Additionally optimisation has allowed this to be done whilst adding in a single level a large number of additional dynamic entities.

The proof of the pudding is in the eating.

Its extremely difficult to be certain of exactly what benefits each method of optimisation individually might achieve and it does not matter - the whole combined effort and end result is what counts.

And I dont write posts long or short for your benefit or otherwise - I write what I need to explain fully to those who wish to learn something from what I do and undersatnd to be the case. If you already know the stuff or dont want to read them then skip the posts.

On a final note you will be pleased to hear that I wont be writing so much for a while giving you all well deserved break - I really must get on with the level building.

Hayer
19
Years of Service
User Offline
Joined: 4th Nov 2005
Location: Norway
Posted: 15th Mar 2006 15:32
btw the Counter-Strike map is named Rat(and some more..) xD

Visit our site : www.freewebs.com/tpfpsc
Reality Forgotten
FPSC Reloaded TGC Backer
18
Years of Service
User Offline
Joined: 28th Dec 2005
Location: Wichita Falls TX
Posted: 16th Mar 2006 00:08
what are you talking about?

"I am ready to meet my Maker. Whether my Maker is prepared for the great ordeal of meeting me is another matter."
-- Winston Churchill
Klick
18
Years of Service
User Offline
Joined: 24th Jan 2006
Location: ANGEL COMMANDing...
Posted: 17th Mar 2006 14:38
*BUMP*

I mean just resizing the entities, blah blah, before the level is even created, would that REALLY demege the - oh yeah it will.
Bleh, wasted post.


[A]NJL`s [N]on-[G]oody-goody [E]xtermination [L]eague [C]arrying [O]ut [M]ass [M]urdering of [A]NJL`s [N]efarious [D]evotees -Essence of ANGEL COMMAND
Disturbing 13
19
Years of Service
User Offline
Joined: 12th Apr 2005
Location: Murder Capital of the World
Posted: 17th Mar 2006 21:38
@ Uman- I do know why this worked in that "other engine" you used. It worked because it accually did reduced pollys. A close look at one of the walls in the other engine revealed that the bigger the wall the more faces were accually added to the wall and vice versa. Thats because it used easily resizeable brushes, not models like FPSC.

The segments retain their pollycount reguardless of what size you make them. The only way I see this resizeing haveing any function is by building levels in a modeler at half the size so reduceing 4 FPSC squares into 1 and placeing the model as an entity . The only drawbacks with this is you would have to hand line up al your rooms. The only thing you would gain from this is more workspace and not a framerate boost, if anything you may get a framrate decrease for haveing all those pollys crunched into one area.

JEEZ!You people just STFU! You waste more space complaining about people wasting space than the people your bashing! Man, I thought I had no life.
Klick
18
Years of Service
User Offline
Joined: 24th Jan 2006
Location: ANGEL COMMANDing...
Posted: 21st Mar 2006 14:33
@Disturbing 13

Hole-In-One! That`s what I was talking about. Although it seemed that this thread already died the moment it started

[A]NJL`s [N]on-[G]oody-goody [E]xtermination [L]eague [C]arrying [O]ut [M]ass [M]urdering of [A]NJL`s [N]efarious [D]evotees -Essence of ANGEL COMMAND
uman
Retired Moderator
20
Years of Service
User Offline
Joined: 22nd Oct 2004
Location: UK
Posted: 21st Mar 2006 15:54 Edited at: 21st Mar 2006 15:59
Disturbing 13,

Quite.

The poly counts in FPSC and how they are handled by the engine is a separate issue and not so both one and at the same time. As it cannot ever be discounted.

As culling and poly issues in FPSC allow no leeway at all - being always at the forefront of speed issues due to the engine inefficiencies - a user always has too many polys from the very start for FPSC to handle efficiently. Thats polys in game not that should be in view and rendered - polys from whatever source of object type.

Engine inefficiency issues are so overiding that they instantly become a factor in whatever anyone does or might suggest doing.

From the moment a user starts building a level they have to consider every inclusion and how it may affect fps. Or ignore it of course until it becomes a serious problem.

Unfortuneately it seems this situation is unlikely to change - so users wishing to see more as in larger levels or more complex levels will have no option but to either limit their level complexities accepting the engines limited capability or alternatively be prepared to spend a great deal of time and I mean a great deal of time - finding ways of achieving their objectives via workarounds or serious amounts of frustrating labour time in optimising every aspect and inclusion to their levels - without resorting to reducing quality unduely. Not an easy task but necessary.

FPSC is quite slow at doing everything and constant efforts optimising and testing (and we all know how long testing and compile of complex levels perhaps thousands of times per level can take) will in such cases certainly add a massive and significant amount of development time to anyones efforts to make a large complex game. In such games the engine is not of much help - you have to help yourself.

Thus again back to the original topic increasing the world size one way or another is a really great idea in principle as we all know how small the levels are - but they are already too big for FPSC to handle efficiently in its current state. How then would one expect FPSC to handle a much larger level size full of content partuicularly in respect to outddor levels where one might expect to see wide open spaces - a major reason for requiring larger levels.

The option of larger levels would be nice though I doubt we could get away with it. More than likely we will never know as no one will ever get to test such levels out as I doubt TGC will ever introduce a larger level size to FPSC.

Nice to dream and ask for such but again TGC are aware of both level size issues and the speed issues that make giving them to users a non stater of an idea.

As with most things - its down to the FPSC product and TGC I'm afraid to integrate improvements to the engine - the end users ability to influence "effectively" many of the engine issues is either very limited or non existant. We all know the source code (half the product) is openly available and I wish anyone wanting to try their hand with that good luck as they need it.

A larger world size is asked for and desirable - TGC are aware of it - you would have it now or TGC will give it to you if it was possible in the future. Just dont ever expect to see it.

We dont really need to tell TGC that larger levels would be a valuable addition they know that - and they wont pay any attention anyway unless it was possible - which its not (well maybe 100 x 100 so we can have some terrain outside to the horizon beyond the 40 x 40).

Red Ocktober
20
Years of Service
User Offline
Joined: 6th Dec 2003
Location:
Posted: 21st Mar 2006 16:37
can i ask a few questions, i don't have a license for fpsc, so i can't test it out to find out what i'm about to ask...

1- can you hide level geometry and disable the script and collision for it...

2- is it possible to loa multiple level gemoetry in a single level or 'mission'...

thx

--Mike
uman
Retired Moderator
20
Years of Service
User Offline
Joined: 22nd Oct 2004
Location: UK
Posted: 21st Mar 2006 18:30
Red Ocktober,

Dont get me wrong - FPSC can handle a lot of stuff of all kinds whether it be world objects or static and dynamic entities as well as much other media. i.e. large texture counts, audio quanities and so on.

Thats not in question. The point being that it requires as with any engine a great deal of planning and optimmisation and some times much uneeded changes to level design to get the best from it - and additionally that is not helped or aided by default by the engine which suffers from the well known and documented issues of problems with culling, AI think time drain and the Dreaded serious lagg issue.

If you can work around those issues with a lot of perhaps from the user point of view unecessary effort - then FPSC can handle a lot of complex content.

As to your specific points :

You cannot hide as far as I am aware - world objects - i.e. the basic walls and prefabs of construction - which is what FPSC should do sensibly during the BSP compile and thus throughout the gameplay in what it sees in camera view. - That it does not do - correctly or at least efficiently in varying - apparently erratic circumstrances where it makes its calculations of what should and should not be seen - specific to each and every users individual level designs. Thus its difficult for anyone to appreciate the apparently random variations in fps and serious lagg that another user - may encounter in both widly varying - simple and complex level design scenarios.

You can certainly load and unload on the fly world objects of any kind as dynamic entities - so you can use such as part of actual building construction rather than interactive game elements if you like. I do with all large groups of dynamic entities, enemies - even loading and unloading game audio and elements of that nature. Some like audio entities have no collision - they are invisible and play no part other than carry the relative script commands. Of course all dynamic entities loaded in such way have some kind of basic commands which require a simple script - though not so simple when you consider their combined drain on the engine. Remeber though its a localised effect in some ways - rather than the default global that FPSC has. Dynamic entities of the load unload variety are destroyed and respawned when necessary and as far as I can tell have no important influence on the gamplay whilst their existance is None i.e. between being destroyed any further calls to respawn.

Its now also apparent that if the situation calls for it (look for excessive distant polys in wireframe} and its done carefully, one can gain fps by re-turning static entities into dynamic ones and using the load and unload entities on the fly method of optimisation. This can in given situations remove large numbers of non world object static entity polygons from FPSC view and release much needed fps.

You could of course build much of your world outside of FPSC and place as dynamic entities and load and unload though I would not reccommend going quite to that extreme of trying to make whole levels that way. In fact their is a cut of point somewhere where a balance has to be kept. Something like a law of diminishing returns - eventually theres a limit to how much optimisation can give you back.

I cant really say a great deal regarding default - world objects used for the building blocks - as they are part of the BSP compile and supposedly all solid begnign objects in game so I cant see how one could have much influence on them?

I doubt you could load or unload any as they dont have any AI as such.

Sure is complicated. If only someone know exactly what FPSC was doing much of its time?

All of the things a user can do to speed up gameplay are if they want to push FPSC to its limits a matter of much trial and error and sometimes unecessary backtracking as no one knows what those limits are in any given situation until someone breaks them. Sometimes the effort just results in failure. If you dont try you will never know.

Again the point being that FPSC is lacking in what one would normally expect to see in a modern engine in some areas such as culling polys and positively a major hindrance in the case of the serious lagg issue.

Users can have a lot of skill and put a great deal of effort and labour into a project doing their part to maximise gameplay quality including issues of speed and fps during gameplay - it would be nice if FPSC did a better job on its part than it does in support of those efforts. As it looks like that may not be the case for the product, then we have to find solutions and master them so we work somewhat smarter not harder.

Login to post a reply

Server time is: 2024-11-24 13:39:14
Your offset time is: 2024-11-24 13:39:14