Heroes of Might and Magic Community
visiting hero! Register | Today's Posts | Games | Search! | FAQ/Rules | AvatarList | MemberList | Profile


Age of Heroes Headlines:  
5 Oct 2016: Heroes VII development comes to an end.. - read more
6 Aug 2016: Troubled Heroes VII Expansion Release - read more
26 Apr 2016: Heroes VII XPack - Trial by Fire - Coming out in June! - read more
17 Apr 2016: Global Alternative Creatures MOD for H7 after 1.8 Patch! - read more
7 Mar 2016: Romero launches a Piano Sonata Album Kickstarter! - read more
19 Feb 2016: Heroes 5.5 RC6, Heroes VII patch 1.7 are out! - read more
13 Jan 2016: Horn of the Abyss 1.4 Available for Download! - read more
17 Dec 2015: Heroes 5.5 update, 1.6 out for H7 - read more
23 Nov 2015: H7 1.4 & 1.5 patches Released - read more
31 Oct 2015: First H7 patches are out, End of DoC development - read more
5 Oct 2016: Heroes VII development comes to an end.. - read more
[X] Remove Ads
LOGIN:     Username:     Password:         [ Register ]
HOMM1: info forum | HOMM2: info forum | HOMM3: info mods forum | HOMM4: info CTG forum | HOMM5: info mods forum | MMH6: wiki forum | MMH7: wiki forum
Heroes Community > Heroes 7 - Falcon's Last Flight > Thread: An Empirical Study of MMH7 Memory Consumption and Time Complexity
Thread: An Empirical Study of MMH7 Memory Consumption and Time Complexity This thread is 6 pages long: 1 2 3 4 5 6 · NEXT»
frostymuaddib
frostymuaddib


Promising
Supreme Hero
育碧是白痴
posted August 23, 2016 09:58 PM
Edited by frostymuaddib at 22:17, 23 Aug 2016.

An Empirical Study of MMH7 Memory Consumption and Time Complexity



'ello 'eroes

As many 'eroes attacked me for not playing Heroes 7 I decided to borrow uplay account of my friend that actually bought Heroes 7 in order to do some testing. So, let us start with my pc specs:

i7 processor,
16 gb RAM
2 gb video RAM

DISCLAIMER Before I get attacked I just want to say that these results are on my pc. Some of the forum members had different experience!

Now, the first thing I did is installing Herpes 7 on my poor PC, and It takes 16,7 gb of my hdd. Now, in order to do real test, I had to measure some basic things. So here is what I got on my pc:
-average game startup times: 2 mins 13 secs (started the game 10 times, I ignored the first startup as it did some additional installing)
-generating random map (128x128, no underground, mixed terrain, normal richness and normal monsters, 4 players) 32 seconds average (10 times generating)
-loading auto save: average 23 secs (10 loads)
-exiting game to main menu: 10 secs average (10 exits)

Now, the next measurings were made during the gameplay:
-game memory: initially 2.6 gb, went up to 4gb during the gameplay
-battle loading screens (calculated for 36 combat starts, stopped after) -average is 6 seconds
-ai turns: in the beginning it was 2 secs, after 34 turns went over 10 secs (all 3 ai players alive), Rough average is 6 secs


Now the test:

I generated a map (as specified above) and started the game. During the hour long gameplay (34 turns) I had four crashes That means that I had to load the game 4 times.

So, I had to start game once in the beginning, plus the four times after the crushes, total of five, and avg of game starting is 133:
5*133=665 seconds wasted on starting the game
Now, I had to load the game 4 times, with average of 23 secs loading
4*23=92 seconds wasted on loading
As I said, I calculated average for battle loading screens with only 36 combats (stopped afterwards). So, I will simplify the calculation and say that I had only 36 combats (they were against neutrals):
36*6=216 seconds wasted on waiting for combat starting
Now for AI turns - this is a very rough estimation
I had 36 turns, that means I waited for AI 36 times. average wait was 6 (very rough!),36*6=216 for ai turns.

So, total seconds wasted (including ai time) is 1189. That is 19 minutes 49 seconds wasted during the one hour gameplay.

I also tinkered a little with editor, but not as much as with the game.
-average startup (10 runs) is 1 munte 45 seconds
-memory usage: started with around 7 gb, but stabilized around 6.5 gb
-saving map: for some strange reason, every time I try to same the map editor crushes after some time, so this can be ignored
-mode switching - first time crushed after two minutes, subseqent ones are all aroud 1 minute 30 secs

I know, I know, you will throw spiders at me sayin' that I'm no 'eroe as I haven't played Heroes 7 enough and that I do not have rights to say anything.
That is why I called for backup! My pal LizardWarrior did some testing as well! And as you all know he did H7 modding (poor guy...) and spent a lot of time with this ... game. So here are his findings:

LizardWarrior said:
Greetings, noble nwcists and filfthy ashanites, I have seen that some of you question the memory consumption of the gruesome spider-beast known as Herpes 7, spawn of the spider messiah, Erwin de Brettonia and Professor Arthropologist Xavier Penin. So I have decided to use my noobish knowledge of programming, computer science and game engines to prove once and for all that you should never ever let your RAM in close proximity of might and magic heroes 7 or either it's map editor. Aside as it is very prone to crushes, it will utterly destroy your PC, infect it with spiders and have it smolder thick smoke, so any volunteers?

PROJ said:

I can honestly say if you think homm vii is hard on your computer you must not do much modern gaming.



Oh, hello there, ashanite, prepare your mind to be blow, as I show you how modern programs intensive programs, as graphic software, game engines and 2016 games won't have the audacity to obscenly devour memory and blatantly rape mistreat your CPU. So below you will see, as I fired up a 3ds max, a 3d modelling software, newest version of photoshop, the Unreal Development Kit, or UE3, the engine which h7 has been made, even UE4 which rocks on full quality (EPIC) graphics at more than 60 FPS on my decent PC, and even Dark Souls 3, a game from spring 2016, whose graphics I set up to max, only consumes about 1.6 GB out of the full 8 GB RAM I got.



But now you'll claim "H7's better!" and scream it in unison. But let me tell you there's an error, I would not dare to open both game and editor, because my PC would fry up, just look at them, over 3 GB each one has, both editor and game. Let's sum them up, shall we? Dark Souls 3:1.6 GB + UE4: 0.7 GB + UDK: 0.4 GB + 3ds max: 0.5 GB + PS: 0.3 GB = 3.5 GB, while only h7 has more than them, 3.6 GB. The editor is more of a mess, when udk only has 0.4 GB consumption, how do you dare defend this?


PROJ said:

Especially with UE4 games starting to come out, if you don't have a pretty beast machine good luck even starting some of the games coming out in the next couple of years, let alone getting a playable framerate.


Oh, I see you are talking about UE4, an engine I happen to have used, so using my noobish programming skills, I have decided to put make a test and see how far can I push the memory consumption, so in this case, the worse the programmer, the better! So, I have decided to do something foolish, something so bad that a programmer would be insane to introduce into a product made for the market or even for free distribution. I will make a spawn volume and set it to generate fairly high-poly meshes for a game on an infinite loop, every once at random interval. So I programmed the spawn volume, please excuse if I made noobish mistakes:


ASpawnVolume::ASpawnVolume():
MinDelay(0.1f), MaxDelay(0.2f)
{
// Set this actor to call Tick() every frame.  You can turn this off to improve performance if you don't need it.
PrimaryActorTick.bCanEverTick = false;

//create box component to represent spawning location
SpawnLocation = CreateDefaultSubobject<UBoxComponent>(TEXT("Spawning Location"));
RootComponent = SpawnLocation;

}

// Called when the game starts or when spawned
void ASpawnVolume::BeginPlay()
{
   Super::BeginPlay();
   SpawnDelay = FMath::FRandRange(MinDelay, MaxDelay);
   GetWorldTimerManager().SetTimer(SpawnTimer, this, &ASpawnVolume::SpawnObject, SpawnDelay);

}

void ASpawnVolume::Tick( float DeltaTime )
{
   Super::Tick( DeltaTime );

}

FVector ASpawnVolume::GetRandomPointInVolume()
{
   FVector SpawnOrigin = SpawnLocation->Bounds.Origin;
   FVector SpawnExtent = SpawnLocation->Bounds.BoxExtent;

   const FVector BoxMin = SpawnOrigin - SpawnExtent;
   const FVector BoxMax = SpawnOrigin + SpawnExtent;
   return FMath::RandPointInBox(FBox(BoxMin, BoxMax));
}

void ASpawnVolume::SpawnObject()
{
   //check if we got something to spawn
   if (ObjectToSpawn)
   {
       //check if there's a valid world
       UWorld * const World = GetWorld();
       if (World)
       {
           //set spawn parameters
           FActorSpawnParameters SpawnParams;
           SpawnParams.Owner = this;
           SpawnParams.Instigator = Instigator;

           //get a random spawn location
           FVector SpawnLocation = GetRandomPointInVolume();

           //get a random rotation
           FRotator SpawnRotation;
           SpawnRotation.Yaw   = FMath::FRand() * 360.0f;
           SpawnRotation.Pitch = FMath::FRand() * 360.0f;
           SpawnRotation.Roll  = FMath::FRand() * 360.0f;

           //spawn the object
           APickUp * const Object= World->SpawnActor<APickUp>(ObjectToSpawn, SpawnLocation, SpawnRotation);

           SpawnDelay = FMath::FRandRange(MinDelay, MaxDelay);
           GetWorldTimerManager().SetTimer(SpawnTimer, this, &ASpawnVolume::SpawnObject, SpawnDelay);
       }
   }
}



Then I selected a smoothed Erwin head which has 25k triangles as my 3d mesh to spawn. But what if this isn't enogh? Well, it should be pretty heavy on memory, rendering and spawing a 25k triangle object every once at 0.1 - 0.2 seconds, especially since I didn't set any LOD for it and I will just use my camera at point blank of the spawing point. Let's use a crazy idea, not even the h7 team can screw it that much, let's make an array of arrays that pushes 1k long double values, worth 64 bits each for every erwin head every 0.1 second, so a total 100k values every second per erwin head. So I wrote this function and override the tick to trigger once at 0.1 seconds in blueprints, ignoring the good and mandatory practice of freeing the memory, because I want to make a memory leak.


void APickUp::Tick( float DeltaTime )
{
   Super::Tick( DeltaTime );
   TArray<long double*> temp;

   for (auto i = 0; i != 0x3D0+0x22; ++i) //I really wanted to make the 3do joke
       temp.Push(new long double(i));
   some_array.Push(temp);

}


But now you may say, I'm insane! Did your pc blew up? Let me tell you friends, that's not even close, but first let me tell you the facts! I let it run to spawn around 1000 erwin heads, so 25k triangles * 1k = 25 million tris or 12 million polygons, plus that I let his push up values, I think I hit some billions, as if I let it run just 10 seconds it would be over a million... per erwin head! so let's say I could catch a trillion. But damn, ue4 lags, but it's a beast, this engine, the lions roars, it may lag, but my framerate grows to 60 as soon as I turn the camera in another direction. But let's see how much memory would this consume, the task manager will tell... oh my, oh my, only 0.8 GB!? Heroes 7 would just kill it.


PROJ said:

I can't take these criticisms seriously, especially since many of the people here clearly understand nothing about computers.


Then I must say, I can't take you seriously neither, as I have proved with this emprical evidence and your arguments aren't backed up.



PS: Heroes 3 with HD mod uses 0.06gb of memory. H7 uses 3.6gb, so you can run 60 Heroes 3 processes with HD mod instead of H7
____________
"Occam's shuriken: when the answer is elusive, never rule out ninjas." -- Dr. Gordon Freeman (Freeman's Mind)
"lol" -- VERRIKER VON ERWINSSEN

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
LizardWarrior
LizardWarrior


Honorable
Legendary Hero
the reckoning is at hand
posted August 23, 2016 10:00 PM

I am Lizard and I approve this message
____________

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
Neraus
Neraus


Promising
Legendary Hero
Pain relief cream seller
posted August 23, 2016 10:19 PM

Those heads... I can't help but feel responsible for such suffering and agony...
____________
Noli offendere Patriam Agathae quia ultrix iniuriarum est.

ANTUDO

 Send Instant Message | Send E-Mail | View Profile | PP | Quote Reply | Link
LizardWarrior
LizardWarrior


Honorable
Legendary Hero
the reckoning is at hand
posted August 23, 2016 10:21 PM

Don't be ashamed, you deserve our eternal spidery praise for your tremendous work
____________

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
cleglaw
cleglaw


Famous Hero
posted August 23, 2016 10:25 PM

guys, can you please drop the "in your face!!" attitude? i really wanna read, but i also dont.

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
PROJ
PROJ


Known Hero
posted August 23, 2016 10:26 PM
Edited by PROJ at 22:27, 23 Aug 2016.

while this is cute, 8 gb ram has been shown to be rough on a UE4 game like lawbreakers, which also dips to under 100 fps on my i7-4790k and gtx 970 machine frequently.  I'm also an internal playtester for midair (another UE4 game) that dips framerates.  Hell, homm vii runs better on my old i3-2100 and hd radeon 6950 pc than overwatch does.  That's only a couple of games from several I've tested (including UT4 and savage resurrection).  

Lets use actual UE4 games instead of strawmanning it, shall we?

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
LizardWarrior
LizardWarrior


Honorable
Legendary Hero
the reckoning is at hand
posted August 23, 2016 10:52 PM
Edited by LizardWarrior at 11:26, 24 Aug 2016.

PROJ said:
while this is cute, 8 gb ram has been shown to be rough on a UE4 game like lawbreakers, which also dips to under 100 fps on my i7-4790k and gtx 970 machine frequently.  I'm also an internal playtester for midair (another UE4 game) that dips framerates.  Hell, homm vii runs better on my old i3-2100 and hd radeon 6950 pc than overwatch does.  That's only a couple of games from several I've tested (including UT4 and savage resurrection).  

Lets use actual UE4 games instead of strawmanning it, shall we?


First heroes VII uses udk, not UE4.

Second, heroes VII doesn't use udk's full rendering capacity.

Third, heroes VII shouldn't crunch 10 trillion numbers at once, not even saying pushing them up on a dynamic array, which is kinda costly since you have to allocate the whole array every time you add values.

Fourth, you don't even know what strawmanning means. It is an extreme test case, it isn't something you would dare to put in a program, spawning continuously high-poly objects and creating a high memory leak on propose. And dark souls III which is a game newer and far more graphic intense runs smoothly using only 1.5 GB of RAM, so your claim that I'm building a strawman doesn't quite stand.

Fifth, compare h7 with a game like ark survival, or lawbreakers as you say. Look at the render quality, reflections, shaders, then compare them with the washed out h7. Then lawbreakers has a minimum requirement of 4 GB and since it's in a testing phase, it is more normally to have memory leaks, unlike h7.

Sixth, you are using an appeal to authority fallacy, don't you think I played games made in UE4? If I would do the same, I'd say that my opinion is more important because I both played games in UE4 and made games in it. As well as I have spent more time modifying h7, even trying to hex edit and bind dlls to it. Well, I played Ark Survival, which is made in UE4, it indeed uses 4 GB of RAM, and sometimes has a leak and jumps to 6, but just stay a minute and compare them. Unlike Heroes VII, Ark Survival has some top-notch killer graphics and amazing gameplay.
____________

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
Stevie
Stevie


Responsible
Undefeatable Hero
posted August 23, 2016 10:53 PM
Edited by Stevie at 22:56, 23 Aug 2016.

PROJ said:
while this is cute, 8 gb ram has been shown to be rough on a UE4 game like lawbreakers, which also dips to under 100 fps on my i7-4790k and gtx 970 machine frequently.  I'm also an internal playtester for midair (another UE4 game) that dips framerates.  Hell, homm vii runs better on my old i3-2100 and hd radeon 6950 pc than overwatch does.  That's only a couple of games from several I've tested (including UT4 and savage resurrection).  

Lets use actual UE4 games instead of strawmanning it, shall we?


Dude, when you have a machine with these specs:

CPU: Intel Core i7-5930K @ 4.0ghz.
RAM: Corsair Vengeance LPX 16GB DDR4-2800
BOARD: ASUS Rampage V Extreme
COOLER: Corsair H110i
GPU: 2x Nvidia Geforce GTX 1080
SSDs: 3x Sandisk Extreme Pro 480gb, 1x Samsung 950 PCI-E SSD.
OS: Windows 7 x64 Ultimate.
DISPLAY: Asus ROG SWIFT PG278Q 1440p 144hz G-sync monitor.

and run a game with these specs:

MINIMUM:
OS: Windows 7 SP1 or Windows 8/8.1 (64bit only)
Processor: Intel Core i5 660 3.3 GHz or AMD Phenom II X4 955 @ 3.2 GHz
Memory: 4 GB RAM
Graphics: nVidia GeForce GTX460 or AMD Radeon HD5850 (1024 MB VRAM)

RECOMMENDED:
OS: Windows 7 SP1 or Windows 8/8.1 (64bit only)
Processor: Intel Core i5 2400 @ 3.1 GHz or AMD FX-6100 @ 3.3 GHz
Memory: 6 GB RAM
Graphics: nVidia GeForce GTX670 or AMD Radeon HD7870 (2048 MB VRAM)

it is entirely reasonable to expect a good performance. If what you get instead is frequent frames drop and crashing then the game is poorly optimized / coded, and no amount of balderdash can change that fact. End of story.
____________
Guide to a Great Heroes Game
The Young Traveler

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
PROJ
PROJ


Known Hero
posted August 23, 2016 11:07 PM

I don't have those issues on a weaker machine.  Are you going to provide data instead of anecdotes?

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
PROJ
PROJ


Known Hero
posted August 23, 2016 11:12 PM - penalty applied by Maurice on 24 Aug 2016.
Edited by Maurice at 12:12, 24 Aug 2016.

LizardWarrior said:
PROJ said:
while this is cute, 8 gb ram has been shown to be rough on a UE4 game like lawbreakers, which also dips to under 100 fps on my i7-4790k and gtx 970 machine frequently.  I'm also an internal playtester for midair (another UE4 game) that dips framerates.  Hell, homm vii runs better on my old i3-2100 and hd radeon 6950 pc than overwatch does.  That's only a couple of games from several I've tested (including UT4 and savage resurrection).  

Lets use actual UE4 games instead of strawmanning it, shall we?


First heroes VII uses udk, not UE4.

Second, heroes VII doesn't use udk's full rendering capacity.

Third, heroes VII shouldn't crunch 10 trillion numbers at once, not even saying pushing them up on a dynamic array, which is kinda costly since you have to allocate the whole array every time you add values.

Fourth, you don't even know what strawmanning means. It is an extreme test case, it isn't something you would dare to put in a program, spawning continuously high-poly objects and creating a high memory leak on propose. And dark souls III which is a game newer and far more graphic intense runs smoothly using only 1.5 GB of RAM, so your claim that I'm strawmaning it's entirely unbased.

Fifth, compare h7 with a game like ark survival, or lawbreakers as you say. Look at the render quality, reflections, shaders, then compare them with the washed out h7. Then lawbreakers has a minimum requirement of 4 GB and since it's in a testing phase, it is more normally to have memory leaks, unlike h7.

Sixth, you are using an appeal to authority fallacy, don't you think I played games made in UE4? If I would do the same, I'd say that my opinion is more important because I both played games in UE4 and made games in it. As well as I have spent more time modifying h7, even trying to hex edit and bind dlls to it. Well, I played Ark Survival, which is made in UE4, it indeed uses 4 GB of RAM, and sometimes has a leak and jumps to 6, but just stay a minute and compare them. Unlike Heroes VII, Ark Survival has some top-notch killer graphics and amazing gameplay.



yawn

you're going to have to be a lot more intelligent than your posts over the last several months have indicated if you actually want to stand a chance in this one

What games have you made in UE4?  What games have you tested?  

You do realize slightly long loading times and memory usage being high when memory is relatively cheap commodity has pretty minor gameplay implications compared to low framerates in an fps, right?  This isn't even really a technical issue as much as it is you try to comically and desperately latch on to any criticisms you can despite them being in the scheme of things, relatively minor.  If you want to deal with major problems, talk about their lockstep networking bugs or various other logic bugs.  This is just a desperate strawman.  

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
blob2
blob2


Undefeatable Hero
Blob-Ohmos the Second
posted August 23, 2016 11:18 PM

Instead of writing this walls of text, let's just summerize it in one simple sentence:

For a game this ugly the game runs like s*it.

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
cleglaw
cleglaw


Famous Hero
posted August 23, 2016 11:23 PM

game is not well optimized. its not a secret.

im really disturbed by these aggressive words in a heroes forum.

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
Elvin
Elvin


Admirable
Omnipresent Hero
Endless Revival
posted August 23, 2016 11:35 PM

Nothing we didn't know but numbers are always appreciated.

Though this is more of a 'come at me bro' thread
____________
H5 is still alive and kicking, join us in the Duel Map discord server!
Map also hosted on Moddb

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
The_Polyglot
The_Polyglot


Promising
Supreme Hero
Nuttier than squirrel poo
posted August 23, 2016 11:40 PM

Even accounting for the blatant provocation, this is pretty hostile. Please do tone it down guys.
____________
Sanity through drugs. Order yours today!

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
GenyaArikado
GenyaArikado


Bad-mannered
Supreme Hero
posted August 23, 2016 11:42 PM

I didnt need a thread to know H7 loads slowly and its poorly optimized, thanks for the numbers i guess, I couldnt be arsed to read since you couldnt help the tacky jokes lol.

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
Stevie
Stevie


Responsible
Undefeatable Hero
posted August 23, 2016 11:44 PM

PROJ said:
I don't have those issues on a weaker machine.  Are you going to provide data instead of anecdotes?


Data has already been provided in this very thread. Are you going to provide a refutation or rhetoric?
____________
Guide to a Great Heroes Game
The Young Traveler

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
olorin
olorin


Adventuring Hero
tophatchild.blogspot.com
posted August 23, 2016 11:45 PM



cleglaw said:
game is not well optimized. its not a secret.

im really disturbed by these aggressive words in a heroes forum.


No agression whatsoever. People need to know before investing their bucks in flawed products.

In light of this BIG issue, the future of H7 is clearly compromised.

As always, I end up playing Heroes 3, Heroes 2, Master of Magic, AoW Shadow Magic... and more recently Heroes V 5.5. You don't need a high-end machine for playing those: only imagination and some wits
____________
Top Hat Child dixit: http://tophatchild.blogspot.com.es/

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
LizardWarrior
LizardWarrior


Honorable
Legendary Hero
the reckoning is at hand
posted August 24, 2016 12:00 AM
Edited by LizardWarrior at 11:40, 24 Aug 2016.



yawn

you're going to have to be a lot more intelligent than your posts over the last several months have indicated if you actually want to stand a chance in this one

What games have you made in UE4?  What games have you tested?  

You do realize slightly long loading times and memory usage being high when memory is relatively cheap commodity has pretty minor gameplay implications compared to low framerates in an fps, right?  This isn't even really a technical issue as much as it is you try to comically and desperately latch on to any criticisms you can despite them being in the scheme of things, relatively minor.  If you want to deal with major problems, talk about their lockstep networking bugs or various other logic bugs.  This is just a desperate strawman.  


I love it when they ran out of arguments and start spitting up insults

Please, back-up your arguments and not start insulting, I lost a day I could go play something good instead of testing this stuff. First address the arguments I gave in the main post. Please don't ignore everything I say and just call it a strawman, I know it takes some effort to build up a competent argument, but in this kind of conversation it is needed.

This being said, the following is gonna be my final well-thought post for you, as you clearly seem pretty uninterested in the points I have to make and just dismiss them entirely without counter-arguments. If you refuse hold up a civilized discussion, then there's no point in continuing it.

First, let me go with you into some programming terms, as you seem to lack knowledge in those. Heroes 7 has a problem called a memory leak, memory leak are bug, they aren't improving anything, should I tell you the definition of a bug now too ? I will just go straight to memory leaks. Memory for variables can be allocated in two ways, either dynamically on a heap, which can be freed for later use, or stack which is allocated for the whole life of the program. A memory leak is allocating basically using heap memory then forgetting to free it or simply the variable becomes inaccessible and it stays on the heap for the rest of the program's lifetime. As fellow HC members may confirm, the game starts using more and more RAM as they play, this means it has a memory leak, as memory builds up on the heap and isn't freed. This is because of bad programming, not because it helps improving the framerate. And of course the framerate will suffer because of the increasing memory usage.

You claim that those are "pretty minor", that's just untrue, crashing a game while just normally playing and not doing something remotely considered as unexpected behavior. Crashes may have multiple causes, either they run out of memory and that's a direct cause of memory leaks or you have an unhandled exception and the code does some unexpected behavior. Handling exception is a concept in programming, where a portion of code is run through a protected block and if something that shouldn't happen happens (called an exception), then another block of code is called (called a catch block) to deal with the exception. For example dividing by 0 will give you an unexpected behavior and you must handle it, but that's just a simple exception, in a game there are thousands of variables used.

Now definitely, games like Dark Souls 3, GTA V and Witcher 3, are much more graphic intense, just look at them, from higher poly models, to better shaders and lighting. While h7 is washed down and has relatively poor rendering settings. But look at AoW 3, a game which is still pretty new and even more complex than h7. But why is does it perform better than h7 then? Even the graphics look better and it still uses less resources. In programming this is called spaghetti code, more code doesn't always means better and most of the times it will be slower and use more memory. This doesn't mean that h7 has a better framerate, it means it's slower because it's badly programmed.

Now using those kind of attacks, like what games have I made? What game have I tested? First, I said I have worked with ue4 and haven't published games, because unlike other I'm not gonna put a game on steam unless it's something that can compete, but I have made AI with A* pathfinding, an advanced expanded class for damage, more noise systems, an abstract character class for RPG type games, including backpack and stat systems and weight limit. Now this point is entirely irrelevant and more on the side of ad hominem. Now I could ask you too, but I guess the answer is 0, playing a beta, testing a game, isn't remotely related to actual programming, every one can do that, for example I got a friend who enters in all kind of beta tests, but he actually doesn't know programming. As for testing, I have tested and played Proto Star and Ark Survival, for udk I have played and tinkered with Painkiller Hell and Damnation and Rise of Argonauts, and well, even modded h7 and actually using the regular udk. Believe me or not, I got more experience with h7 than you do, not actually playing, but more on the technical side. I didn't ignore your arguments because of this, but since you seem to use this point, even though it's a logical fallacy, it looks like I got a better ground to stand on.

As I said, I won't continue this conversation with you, as it won't come to fruition, unless you actually bring up actual arguments that are worth discussing. Now I know you will answer with a one-liner or accusing me of building a straw-man again. As clearly, we are talking about a memory leak, not a memory trade-off for better performance.


GenyaArikado said:
I didnt need a thread to know H7 loads slowly and its poorly optimized, thanks for the numbers i guess, I couldnt be arsed to read since you couldnt help the tacky jokes lol.

The_Polyglot said:
Even accounting for the blatant provocation, this is pretty hostile. Please do tone it down guys.

Elvin said:
Nothing we didn't know but numbers are always appreciated.

Though this is more of a 'come at me bro' thread

cleglaw said:
game is not well optimized. its not a secret.

im really disturbed by these aggressive words in a heroes forum.


It's called a retaliation, I thought this was a Heroes forum . That's why I chose only to respond to Proj's post, sadly he couldn't actually provide counter-arguments or strike a friendly conversation . Maybe it was a bit harsh, I wanted to make my part as objective as possible at first, but well, this sparks a lot more interest. Well, there are some people that even if defend h7, they are pretty chill most of the time, like Cleglaw and Antalyan. If you like h7, then all the power to you, don't let memory leaks to hinder you, but don't start outright lying about how well optimized is h7 and how all other games are just like it.

____________

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
Salamandre
Salamandre


Admirable
Omnipresent Hero
Wog refugee
posted August 24, 2016 04:57 AM

Starting game and reloading save should be instant. For starting, if game is really worth I could force myself to wait ~5 seconds, but for reload, especially if tbs Heroes-like game, where battles sometimes require dozens of reloads, anything beyond 1 second delay in out of my games list. So still playing Master of Orion 2, Imperialism 2, AoW 2, Civilization 3, Heroes 2-3 and others from that time, until developers are back to reason and common sense. And having fun too.
____________
Era II mods and utilities

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
lotihoti
lotihoti


Famous Hero
posted August 24, 2016 08:46 AM
Edited by lotihoti at 08:47, 24 Aug 2016.

Well i have to be lucky - i didnt had a single crash till now (around 3 days gameplay time in total) (campaign and single player - didnt test multiplayer).

Maybe the game runs better on amd machines
3 secs to go into a battle here

Or maybe im just lucky - we will never know

 Send Instant Message | Send E-Mail | View Profile | Quote Reply | Link
Jump To: « Prev Thread . . . Next Thread » This thread is 6 pages long: 1 2 3 4 5 6 · NEXT»
Post New Poll    Post New Topic    Post New Reply

Page compiled in 0.1064 seconds