Welcome

Please register for Total War Access to use the forums. If you're an existing user, your forum details will be merged with Total War Access if you register with the same email or username. For more information please read our FAQ’s here.

Categories

Total war warhammer 3 engine.

Deep_echo_soundDeep_echo_sound Registered Users Posts: 601
Hello.
Sorry of English.
"Total war warhammer" games currently are in desperate need of 2 things. From the technical point of view.
Heavy multi threading and appropriate API.
By the time game 3 is out, which is 2020 or 2021, 8 cores and 16 threads will be like a minimum for gaming machines. We have 12 and 16 cores today. In 2 years this will only escalate. And game 3, with combined map will have long life.
Right now game is very dependent on speed of 1 core. More cores helps at least at 4 to 6 or 8, but 1 core is always 100% loaded.
This is the answer for users who wonder why their new processor and ultra powerful graphics card still give then 10-20 fps in 40 vs 40 massive battles on ultra unit scale. It is hard CPU limit. 1 core limit.
Game engine needs total rework for very good multithreading. For like any amount of cores (32 is really easy to buy now, half-pro platforms.)
Second thing, may be even more important in some cases, - DX12 or Vulcan exclusive game. No more "windows 7" and DX11 in 2020-2021 gigantic strategy game. The most important things about these APIs are very good optimization for draw calls and possibility to handle huge amount of interactive objects on the field with good performance. And no other game needs this that much like a total war titles! And it is very obscurely, why this is not implemented when this limit total war games so much? Unit sizes and actual performance. No more than 2 vs 2 stacks battles in campaign because, again, mainly for performance reasons.
And this lag game, in ultra unit sizes, when graphics are on all extreme setting but still a card loaded for 35% maximum, and you still have 10-14 fps for 80 units fighting (at zoom in).
This is top technical importance. It is not about balance or atmosphere where you can have "different feelings". This is about tech side of a product of the company. And it must have good quality.
Owning a game with all DLCs is not that cheap now, and customers expect great tech quality. And, at least future game 3, with a new or really modernized engine with much-needed new APIs exclusivity.
These things alone will totally cut the dreaded "end turn times" too. And personally I am totally fine with them, but many other players do not. And that opens a possibility for real all world map fog game 3 "mortal empires". With all continents in full scale. Not shrink and cut for better turn times.
It is easy - tech limitations limit game. And this limit the players impressions and emotions from these impressions.
And from the game perspective, easy and achievable short campaign victory, realistically by turn 50, will make impressions and gameplay not overwhelming, by the size of the map, for players. (Take 2 to 4 key provinces, may be far and strong, like Altdorf for orcs, but not 10. And destroy 2-3 factions. Period.) But this is all second. Engine is a tech base of the game. Good, new API engine - great possibilities!

Comments

  • ThirdeyeThirdeye Registered Users Posts: 251
    What do you think a windows 10, 8 cores 16 threads machine with DX12 or Vulcan and new API engine will cost?
    How soon do you think we will see such a machine?
    How long do you think it would take CA to rework the game engine and all current games for “very good multithreading”?
    How much would such a game cost?
    Would we have to buy the same game again?
  • Steph_F_DavidSteph_F_David Junior Member Registered Users Posts: 1,730
    Beside, if CA wants to keep some compatibilty with what they already did for game 1 and 2n they may not be able to do this jump.
  • mightygloinmightygloin Karaz-a-KarakRegistered Users Posts: 4,015
    Do games of today really use more than 4 cores? Let alone 6 or 8 cores? Does anyone have a sure answer for this?

    But one thing i really observed is WH2 doesn't really take advantage of better CPU when it comes ME turn times.
  • Deep_echo_soundDeep_echo_sound Registered Users Posts: 601
    For the visual understanding. To know what it is all about, and what a resulting performance (for CPU) can be with these things implemented.

    https://youtube.com/watch?v=uCY3vff2cvs

    https://youtube.com/watch?v=lIB7bLUhZ1A

    https://youtube.com/watch?v=P_I8an8jXuM

    Videos are not about legacy API's (like "mantle" when "Vulcan" and DX12 are better) or games, or engines. It is about performance optimization.
    Total war games are made for such improvements.
    The last one is especially painful to watch, because it reminds you so much of the current state of the great game, which is hugely limited by the engine. Let's fix it in game 3!
  • Deep_echo_soundDeep_echo_sound Registered Users Posts: 601
    Thirdeye said:

    What do you think a windows 10, 8 cores 16 threads machine with DX12 or Vulcan and new API engine will cost?
    How soon do you think we will see such a machine?
    How long do you think it would take CA to rework the game engine and all current games for “very good multithreading”?
    How much would such a game cost?
    Would we have to buy the same game again?

    Cost of windows 10, 8 cores 16 threads machine with DX12 or Vulcan will be far less than 1k USD in 2020 and after.
    We see such machines from the early 2017 for mass market. DX12 cards in all price segments are here for like 5 years at least?

    It is company's work to constantly improve. Millions of sales - new or heavily reworked engines at least once in 5-6 years. Not an engine without major improvements for a decade and more.

    Cost enough to make profit.

    Why? We have game 3 already? To buy it "again"? 1 time buy and have final game 3 with combined map and all DLCs from previous games, like "mortal empires" in game 2 now.

    All can be done if there is a will behind it. And not just a desire to make a profit with minimum effort and ancient tech.
    "They will eat anyway" - is a disgusting practice for a company. Hope CA will not do this. Some companies get hate for a reason. And I do not even once mentioned denuvo... Because new or heavily reworked engine will overcome penalty caused by this parasitic software too.
  • GoatforceGoatforce Registered Users Posts: 5,845
    I am a little sceptical that 12 and 16 core CPUs will be the norm in 2-3 years (which will cover a reasonable amount of game 3's main lifecycle). I have an i7-8700K and it runs like a charm, and I haven't even overclocked the CPU yet (which I could easily get around 700-800MHz extra out of if I wanted), I really don't think that extra power would get a huge amount more performance from the WH game - after all this is an interlocking trilogy, they are not going to radically overhaul the engine as that will throw a wrench in the works transferring stuff from game 1 and 2.

    However, in answer to the whole multi-threading thing, well, it has already been done. Have you seen the performance gain using the "laboratory" mode? As I recall that is using an experimental system for multithreading, or something to that effect. I am really hoping that they integrate that software into game 3. Even if they don't scale things up the improvement to gameplay will be more than worth it - and it may well massively improve the experience of ME.2.
  • psychoakpsychoak Registered Users Posts: 3,337
    Lab mode? Please. I get down to 20 FPS running that Skaven benchmark with a 1080 Ti at 1440p. It's atrocious, doesn't even manage to maintain half load on the GPU. It sounds to me like you just don't mind dipping down to 20-30fps in big battles.

    Warhammer II already uses 18 cores. I have an i9 9980XE, it's quite nicely spread out across the cores for the most part.

    This isn't actually about using 12 cores though, it's about efficiently using 4 or more.

    When you're running any DX11 video game, with sufficient draw calls to make you CPU bound, that's only two cores.

    This can be accomplished by having high frame rates in even relatively simple game engines. My i9 9980XE, a two thousand dollar processor, is massively insufficient for the current capabilities of video cards. 144hz monitors? Useless. High object count games? Useless. High object count games on 144hz monitors? Cue the depraved giggles of a mad man.

    If you play at 1080p on ultra, you can max out with a 1080, you can even max out with a 1070. A 2080 Ti? You have to be playing at 4k to avoid being CPU bound in just a run of the mill battle. OC to 5Ghz or so, and you can probably actually use a 144hz monitor to good effect. In small battles. 40vs40 with lots of high count infantry units? Nope, still need two or three Ghz more.

    FPS is crippled by single thread limitations. Plain and simple. The only way you get around this, without upgrading to DX12 or Vulcan, is by optimizing around using the CPU. You just can't do that for lots of separate units that have to be independently rendered and animated.
  • jamreal18jamreal18 Senior Member Registered Users Posts: 10,676
    How about next major game after 3k?
  • UgandaJimUgandaJim Registered Users Posts: 316
    Thirdeye said:

    What do you think a windows 10, 8 cores 16 threads machine with DX12 or Vulcan and new API engine will cost?
    How soon do you think we will see such a machine?
    How long do you think it would take CA to rework the game engine and all current games for “very good multithreading”?
    How much would such a game cost?
    Would we have to buy the same game again?

    Thats called "investing" and thats what every company needs to do if they wanna stay competitive.
  • UgandaJimUgandaJim Registered Users Posts: 316
    psychoak said:

    Lab mode? Please. I get down to 20 FPS running that Skaven benchmark with a 1080 Ti at 1440p. It's atrocious, doesn't even manage to maintain half load on the GPU. It sounds to me like you just don't mind dipping down to 20-30fps in big battles.

    Warhammer II already uses 18 cores. I have an i9 9980XE, it's quite nicely spread out across the cores for the most part.

    This isn't actually about using 12 cores though, it's about efficiently using 4 or more.

    When you're running any DX11 video game, with sufficient draw calls to make you CPU bound, that's only two cores.

    This can be accomplished by having high frame rates in even relatively simple game engines. My i9 9980XE, a two thousand dollar processor, is massively insufficient for the current capabilities of video cards. 144hz monitors? Useless. High object count games? Useless. High object count games on 144hz monitors? Cue the depraved giggles of a mad man.

    If you play at 1080p on ultra, you can max out with a 1080, you can even max out with a 1070. A 2080 Ti? You have to be playing at 4k to avoid being CPU bound in just a run of the mill battle. OC to 5Ghz or so, and you can probably actually use a 144hz monitor to good effect. In small battles. 40vs40 with lots of high count infantry units? Nope, still need two or three Ghz more.

    FPS is crippled by single thread limitations. Plain and simple. The only way you get around this, without upgrading to DX12 or Vulcan, is by optimizing around using the CPU. You just can't do that for lots of separate units that have to be independently rendered and animated.

    Yeah but DX12 is poorly implemented in Game2. I have big performance losses when i activate it ...
  • Deep_echo_soundDeep_echo_sound Registered Users Posts: 601
    edited August 2019
    "Warhammer 3" is the next major game after "3 Kingdoms".
    And to "cost of the game". Base of units, stats, spells, effects in data tables can be ported to updated engine like a charm. Engine code must be updated. Not all graphics, all game data, etc.
    And this highly parallel engine will serve for all future titles as well. Major games sells in 2-3 millions copies each, at least. And continue to sell and sell over years. DX12 and Vulcan are the future. Heavy multithreading is the future.

    And, yes, today's games uses more than 4 cores. You can see how "battlefield 1" (same may 2016 title, just like "total war warhammer") uses 8 cores effectively. And in that "battlefield" game they do not need it that much. But in "warhammer" they do.
    For any amount of cores. 1, 2, 4, 6 etc., they all must be utilized by close to 100% for every core. This change in engine will bring far better performance for everyone. It is not about pure core count.
    It is all about great scalability.

    And 12 cores, 24 thread CPUs for 250 USD in 2021 - absolutely. Zero skepticism here. We had 8 cores, 16 thread CPUs in the early 2017 for 500 USD at release, and for 180-200 USD now (used and almost new and easy to buy). And a new 8-16 CPUs, close to 200-250 USD, on discounts, from time to time, in major online shops for a good time already. Problems?
    Post edited by Deep_echo_sound on
  • Deep_echo_soundDeep_echo_sound Registered Users Posts: 601
    edited August 2019
    UgandaJim said:



    Yeah but DX12 is poorly implemented in Game2. I have big performance losses when i activate it ...

    They "emulate" DX12 in game 2. Game is DX11 in core. Emulation is very costly for performance.
    Pure right implementation of DX12 or Vulcan (latter the better for porting for other OS) will bring a huge jump in CPU performance and efficiency.
  • darkgaia01darkgaia01 Registered Users Posts: 357
    a very good example is wargaming when they upgraded their engine for world of tanks a couple of years ago as it allowed them to do so much stuff while keeping the gameplay pretty much the same and even improved in alot of areas.

    the major problem i have with CA is they don't communicate with the tech improvements they do for the game or when they do it just some text on forum which pretty much most of the playerbase would never seen. what they need to do is whe na major game comes out do a video showing what tech improvements they done not just show videos on units or gameplay all the time.
  • GoatforceGoatforce Registered Users Posts: 5,845
    psychoak said:

    Lab mode? Please. I get down to 20 FPS running that Skaven benchmark with a 1080 Ti at 1440p. It's atrocious, doesn't even manage to maintain half load on the GPU. It sounds to me like you just don't mind dipping down to 20-30fps in big battles.

    Warhammer II already uses 18 cores. I have an i9 9980XE, it's quite nicely spread out across the cores for the most part.

    This isn't actually about using 12 cores though, it's about efficiently using 4 or more.

    When you're running any DX11 video game, with sufficient draw calls to make you CPU bound, that's only two cores.

    This can be accomplished by having high frame rates in even relatively simple game engines. My i9 9980XE, a two thousand dollar processor, is massively insufficient for the current capabilities of video cards. 144hz monitors? Useless. High object count games? Useless. High object count games on 144hz monitors? Cue the depraved giggles of a mad man.

    If you play at 1080p on ultra, you can max out with a 1080, you can even max out with a 1070. A 2080 Ti? You have to be playing at 4k to avoid being CPU bound in just a run of the mill battle. OC to 5Ghz or so, and you can probably actually use a 144hz monitor to good effect. In small battles. 40vs40 with lots of high count infantry units? Nope, still need two or three Ghz more.

    FPS is crippled by single thread limitations. Plain and simple. The only way you get around this, without upgrading to DX12 or Vulcan, is by optimizing around using the CPU. You just can't do that for lots of separate units that have to be independently rendered and animated.

    I haven't tested the lab on my new computer, but on my old one I had the opposite experience, my frame rate was far higher in standard 20 vs 20 (but with the entity count increased) and was workable much in larger battles (it was a laptop, albiet a good one, so suffered from its CPU not being incredibly fast). It seemed to me the performance gains were significant. I doubt, again considering the fact this is an interlocking trilogy, that one should expect more significant overhauls between game 2 and 3 than this, as it would likely be a nightmare to transfer assets between titles - we already had Norsca, this would probably be far worse.
  • ThirdeyeThirdeye Registered Users Posts: 251
    edited August 2019
    UgandaJim said:

    Thirdeye said:

    What do you think a windows 10, 8 cores 16 threads machine with DX12 or Vulcan and new API engine will cost?
    How soon do you think we will see such a machine?
    How long do you think it would take CA to rework the game engine and all current games for “very good multithreading”?
    How much would such a game cost?
    Would we have to buy the same game again?

    Thats called "investing" and thats what every company needs to do if they wanna stay competitive.


    Its also called “BOAT” (Bust-out-another-thousand) to up-grading my rig again, to get the full game experience. Sorry, not all of us are made of money. There needs to be a balance between game performance and accessibility.
  • ITA_Vae_VictisITA_Vae_Victis Senior Member ItalyRegistered Users Posts: 1,657
    Better multi-core optimization would definitely be welcome, and it's probably not a daunting task on the technical and compatibility side if CA decides to put the time and resources into it.


    On the other hand, OP's do-or-die stance is completely unjustified. 16 threads will NOT be the norm in 2 years, it will be the high-end of the spectrum. Same goes for DX 12 only support, which would leave a lot of people (aka customers, aka sales) behind.

    According to the lastes Steam Survey, about 1/3 of the users still don't have a DX12+Windows 10 setup, and the overwheling majority of people have a CPU with 4 cores or less: https://store.steampowered.com/hwsurvey/Steam-Hardware-Software-Survey-Welcome-to-Steam?platform=pc

    And keep in mind that this is a voluntary survey, which probably means most people who even just take the time to answer it are already from the "top" strata of the gaming market, and therefore these data represent a more "enthusiast" slice of the total gaming population, not the real average Steam user.


    TL;DR Optimization should keep progressing over time, as it has been doing so far, but a sudden and drastic jump in technology would be unwarranted for and probably more damaging than anything else for a good portion of the user base.
  • Deep_echo_soundDeep_echo_sound Registered Users Posts: 601
    edited August 2019
    What? 6c-12t or 8c-16t is the norm for gaming machines today. By gaming I mean gaming. Not a web-browsing low cost.
    Well, from the late 199x gaming CPU of 200 USD or more is the norm. Like i5 2500k or i7 2600k for the 2011. It is not about 6 years old i3 budget. AAA-games are not like a "darkest dungeon" segment. (Which is great, but yeah! Another segment.)
    And gaming P.C.s are not consoles. They are not "400 USD for 7 years".
    Gaming machines updates at least twice in console generation. So every 3-4 years. You can have low, middle or top end, but gaming P.C. is about upgrading. Your specs are about playing the game on low, medium or ultra depending on your budget. That is it.
    Just tell me how many were crying because "Witcher 3" had new engine for a time to give a new good graphics and world?

    Strategies are dying already! The worst genre situation now. From the 199x to early 20xx they were super great. Right now I can not even name just one game, outside of "total war", which has great graphics.
    That's why "age of empires 2" is very popular even today. And improved game will come with what? Great improved graphics and other updates!
    And some other series, global multi-nation game, I will not name it, have degradated to complete trash. From 5 to 6 part not improving in visual is a crime against sophisticated gamers. This tablet casual graphics is just an insult. I hate it so much. I will not even pirate this piece of....
    So, please, do not ask to turn "total war" in that kind of garbage. That will kill non-budget strategy games genre totally.

    I really do not like to write brands and product names, but this is for factual material.
  • GoatforceGoatforce Registered Users Posts: 5,845

    What? 6c-12t or 8c-16t is the norm for gaming machines today. By gaming I mean gaming. Not a web-browsing low cost.
    Well, from the late 199x gaming CPU of 200 USD or more is the norm. Like i5 2500k or i7 2600k for the 2011. It is not about 6 years old i3 budget. AAA-games are not like a "darkest dungeon" segment. (Which is great, but yeah! Another segment.)
    And gaming P.C.s are not consoles. They are not "400 USD for 7 years".
    Gaming machines updates at least twice in console generation. So every 3-4 years. You can have low, middle or top end, but gaming P.C. is about upgrading. Your specs are about playing the game on low, medium or ultra depending on your budget. That is it.
    Just tell me how many were crying because "Witcher 3" had new engine for a time to give a new good graphics and world?

    Strategies are dying already! The worst genre situation now. From the 199x to early 20xx they were super great. Right now I can not even name just one game, outside of "total war", which has great graphics.
    That's why "age of empires 2" is very popular even today. And improved game will come with what? Great improved graphics and other updates!
    And some other series, global multi-nation game, I will not name it, have degradated to complete trash. From 5 to 6 part not improving in visual is a crime against sophisticated gamers. This tablet casual graphics is just an insult. I hate it so much. I will not even pirate this piece of....
    So, please, do not ask to turn "total war" in that kind of garbage. That will kill non-budget strategy games genre totally.

    I really do not like to write brands and product names, but this is for factual material.

    Err, are strategies dying? TW doing well, Paradox games, Civ still going strong, Age of Wonders, on the more tactical side Xcom 2 is great. More niche titles like Dominions are really good too. There are many others I haven't played like Ashes of the Singularity, Homeworld: Deserts of Kharak (and the Homeworld remakes are gorgeous, though they are remakes). The first few I mentioned as I recall did very well, on Steam reaching their Gold ranking on Steam with some of them I recall.

    Also, since when has strategy required high tier graphics? I mean yes good graphics is always nice, but are not nearly the most important thing in strategy - take the aforementioned Dominions, which uses sprites, but is one of the deepest strategies I have played. It is quite amusing to see you call lack of graphics improvements an insult to "sophisticated gamers", since graphics are the most superficial part of a game - hardly sophisticated.

    Strategy has always been fairly niche, however I would say it is in a very good position - probably as close to mainstream as it has ever been. I would say the only issue in it atm with lack of competition in sub-genres of strategy hindering innovation.
  • Deep_echo_soundDeep_echo_sound Registered Users Posts: 601
    edited August 2019
    Yes. They are dying. "Total war" are the last good strategy games.

    Was never interested in Paradox games, sorry. Except for a great "King Arthur: the role-playing wargame" which one is close to "total war warhammer", but older. But that game was developed by "NeocoreGames", and had "Paradox" only as a publisher. Really great, impressive graphics for the time, by the way!!
    Civilization now is just a piece of... Cake. Bad cake. I like "civilization 3" with great graphics and high quality for the time. The last one that was actually good.
    So. Which were the good strategy games?
    "Warcraft" 2 and 3 + expansions, "starcraft" + "broodwar", "age of empires" 1 and 2, "stronghold" + "crusader", "civilization 3", "settlers 3", "heroes of might and magic" 3 and 4, "Rome: total war" and "medieval 2", "disciples 2", "red alert 2", "majesty", "spellforce" 1 and 2 + expansions, "king's bounty: the legend", "king Arthur".
    75% of them came before 2005.
    First "X-com" was great, but that is more of a tactic. Now they are all low graphics, low budget.

    And yes. All of these games had great graphics for a time. If a game had bad graphics, I do not liked it that much. Just like "heroes 1" or "warcraft: orcs & humans" are not in the list. And "age of wonders" are the same. "Pixelated something" compared to "heroes 3" or "civilization 3".
    Every game is about graphics. Not only that, but first you see graphics. Game must be beautiful for a time or I do not care about it. If it is not, this is just forces low-budget impression.
    Just like newly released game by that "Paradox". Ugly 2007-ish graphics. 12 years later, looks like a financial fraud. When we have "resident evil 2" (2019), "hellblade: Senua's sacrifice" and "Detroit: become human". Difference in graphics with "Paradox" games is just unimaginable.


    The only last one series that survives and do not have bad graphics is "total war". That's why I am here. The last strategy games that worth the time. All other games are dead or took this behavior: "we can look ugly because we have strategy".
    Big "no" to this.
  • psychoakpsychoak Registered Users Posts: 3,337

    Better multi-core optimization would definitely be welcome, and it's probably not a daunting task on the technical and compatibility side if CA decides to put the time and resources into it.


    On the other hand, OP's do-or-die stance is completely unjustified. 16 threads will NOT be the norm in 2 years, it will be the high-end of the spectrum. Same goes for DX 12 only support, which would leave a lot of people (aka customers, aka sales) behind.

    According to the lastes Steam Survey, about 1/3 of the users still don't have a DX12+Windows 10 setup, and the overwheling majority of people have a CPU with 4 cores or less: https://store.steampowered.com/hwsurvey/Steam-Hardware-Software-Survey-Welcome-to-Steam?platform=pc

    And keep in mind that this is a voluntary survey, which probably means most people who even just take the time to answer it are already from the "top" strata of the gaming market, and therefore these data represent a more "enthusiast" slice of the total gaming population, not the real average Steam user.


    TL;DR Optimization should keep progressing over time, as it has been doing so far, but a sudden and drastic jump in technology would be unwarranted for and probably more damaging than anything else for a good portion of the user base.

    Switching to a modern API that actually functions with modern processors, isn't even remotely drastic.

    DX12 has been put into Windows 8, and Microsoft even caved and let multiple games run in DX12 mode in Windows 7.
    Windows 7 completely leaves support this year. It's not a current operating system. If you're still on it, upgrade.

    Vulkan? No such limitation to start with. Works on anything newer than XP, including Linux and . If you're still on XP, you're not playing Warhammer to start with.

    Core count? Same deal. It doesn't matter if 50% of Steam's users still have 4 cores. DX12 or Vulkan would only be "bad" for people with 2 cores, which is arguable because DX11 already has high overhead for threading with the inactive state. People who do only have two cores, can't play Warhammer for **** already. They need to upgrade.

    DX12, or Vulkan, is a performance improvement, not some kind of nightmare scenario for the guy with a five year old computer. They're actually the ones that get the most benefit, because without it they can't even feed the newest video cards.
  • GoatforceGoatforce Registered Users Posts: 5,845
    edited August 2019

    Yes. They are dying. "Total war" are the last good strategy games.

    Was never interested in Paradox games, sorry. Except for a great "King Arthur: the role-playing wargame" which one is close to "total war warhammer", but older. But that game was developed by "NeocoreGames", and had "Paradox" only as a publisher. Really great, impressive graphics for the time, by the way!!
    Civilization now is just a piece of... Cake. Bad cake. I like "civilization 3" with great graphics and high quality for the time. The last one that was actually good.
    So. Which were the good strategy games?
    "Warcraft" 2 and 3 + expansions, "starcraft" + "broodwar", "age of empires" 1 and 2, "stronghold" + "crusader", "civilization 3", "settlers 3", "heroes of might and magic" 3 and 4, "Rome: total war" and "medieval 2", "disciples 2", "red alert 2", "majesty", "spellforce" 1 and 2 + expansions, "king's bounty: the legend", "king Arthur".
    75% of them came before 2005.
    First "X-com" was great, but that is more of a tactic. Now they are all low graphics, low budget.

    And yes. All of these games had great graphics for a time. If a game had bad graphics, I do not liked it that much. Just like "heroes 1" or "warcraft: orcs & humans" are not in the list. And "age of wonders" are the same. "Pixelated something" compared to "heroes 3" or "civilization 3".
    Every game is about graphics. Not only that, but first you see graphics. Game must be beautiful for a time or I do not care about it. If it is not, this is just forces low-budget impression.
    Just like newly released game by that "Paradox". Ugly 2007-ish graphics. 12 years later, looks like a financial fraud. When we have "resident evil 2" (2019), "hellblade: Senua's sacrifice" and "Detroit: become human". Difference in graphics with "Paradox" games is just unimaginable.


    The only last one series that survives and do not have bad graphics is "total war". That's why I am here. The last strategy games that worth the time. All other games are dead or took this behavior: "we can look ugly because we have strategy".
    Big "no" to this.
    Every game is about graphics.... well that says a lot. Graphics are the skin, they are superficial. Yes it is good to see a game with a pretty skin, but that is secondary. Graphics also take a lot of time and effort, the game Dominions uses sprites because it is a very small developer and good graphics would mean not being able to provide a deep strategy game, companies have to prioritise what they are going to focus on, and the gameplay should always trump graphics, especially for strategies. CA is big enough to do both, however.

    Paradox has its problems, but graphics isn't one of them. Stellaris is quite pretty for a start. Secondly they are producing focussed grand strategy, which doesn't need visual flair for the most part as the scale is too large for that to be important. The games are about the deep and complex mechanics. If you can't appreciate a strategy for its mechanics, and only care about how they look, then perhaps strategy games aren't for you.
  • ITA_Vae_VictisITA_Vae_Victis Senior Member ItalyRegistered Users Posts: 1,657
    psychoak said:

    Better multi-core optimization would definitely be welcome, and it's probably not a daunting task on the technical and compatibility side if CA decides to put the time and resources into it.


    On the other hand, OP's do-or-die stance is completely unjustified. 16 threads will NOT be the norm in 2 years, it will be the high-end of the spectrum. Same goes for DX 12 only support, which would leave a lot of people (aka customers, aka sales) behind.

    According to the lastes Steam Survey, about 1/3 of the users still don't have a DX12+Windows 10 setup, and the overwheling majority of people have a CPU with 4 cores or less: https://store.steampowered.com/hwsurvey/Steam-Hardware-Software-Survey-Welcome-to-Steam?platform=pc

    And keep in mind that this is a voluntary survey, which probably means most people who even just take the time to answer it are already from the "top" strata of the gaming market, and therefore these data represent a more "enthusiast" slice of the total gaming population, not the real average Steam user.


    TL;DR Optimization should keep progressing over time, as it has been doing so far, but a sudden and drastic jump in technology would be unwarranted for and probably more damaging than anything else for a good portion of the user base.

    Switching to a modern API that actually functions with modern processors, isn't even remotely drastic.

    DX12 has been put into Windows 8, and Microsoft even caved and let multiple games run in DX12 mode in Windows 7.
    Windows 7 completely leaves support this year. It's not a current operating system. If you're still on it, upgrade.

    Vulkan? No such limitation to start with. Works on anything newer than XP, including Linux and . If you're still on XP, you're not playing Warhammer to start with.

    Core count? Same deal. It doesn't matter if 50% of Steam's users still have 4 cores. DX12 or Vulkan would only be "bad" for people with 2 cores, which is arguable because DX11 already has high overhead for threading with the inactive state. People who do only have two cores, can't play Warhammer for **** already. They need to upgrade.

    DX12, or Vulkan, is a performance improvement, not some kind of nightmare scenario for the guy with a five year old computer. They're actually the ones that get the most benefit, because without it they can't even feed the newest video cards.
    Optimizing with the highest of high-ends in mind (which is what OP suggests) never brings "the most benefits" to people with mid-low range hardware. You need to account for everybody to a certain extent, and that can be complicated depending on the engine and the different configurations.

    My only real point is that CA is not going to spend money for engine alterations if the end result of that is gating out from the game even just 1% of its current user-base, because that would be monetarily counter-productive, especially on an ongoing product that uses past releases as DLC for the newer ones. If they will actually go through with the DX12 implementation for WH3, it will NEVER be with the objective of simultaneously dropping DX11, because there are just too many current and potential customers that would be cut-off for no real sane monetary reason.

    PS: Microsoft has stated that it will allow the use of DX12 on Windows 7 only on a very select number of specific games, which probably means private agreements between companies and Microsoft, which probably means paying Microsoft for the privilege. And I doubt CA is going to do that, just to counteract the damage they would themselves create to their product (already at their expenses) for no real good reason.
  • Deep_echo_soundDeep_echo_sound Registered Users Posts: 601
    Sadly, some people are just do not understand what DX12 or Vulcan are about.
    Words or videos are just transparent to them.
    Many fans of some corporations, at the beginning of the DX12 and Vulcan era, complained about them: "we do not need this because it will mostly help weak CPUs, we have 120 fps already!". Now low spec users think new APIs are only for the top multi-core processors. People are just so...
    Sometimes I understand developers. You try to make the game better and then some users come and complain about it.
    Just like this: "Ahhh, you made a new patch, and it broke my mods!! No more new patches and DLCs forever, because I want to finish my pervert-modded campaign!"
    At the same time when most users want faster DLC and patches with a quality.
    Some not understanding and toxic egocentric people that are against improvements must be ignored.
  • GoatforceGoatforce Registered Users Posts: 5,845

    Sadly, some people are just do not understand what DX12 or Vulcan are about.
    Words or videos are just transparent to them.
    Many fans of some corporations, at the beginning of the DX12 and Vulcan era, complained about them: "we do not need this because it will mostly help weak CPUs, we have 120 fps already!". Now low spec users think new APIs are only for the top multi-core processors. People are just so...
    Sometimes I understand developers. You try to make the game better and then some users come and complain about it.
    Just like this: "Ahhh, you made a new patch, and it broke my mods!! No more new patches and DLCs forever, because I want to finish my pervert-modded campaign!"
    At the same time when most users want faster DLC and patches with a quality.
    Some not understanding and toxic egocentric people that are against improvements must be ignored.

    If you are replying to someone, I advise you to use the "quote" button under their post, or @ them, it makes things a lot clearer.
  • psychoakpsychoak Registered Users Posts: 3,337

    psychoak said:

    Better multi-core optimization would definitely be welcome, and it's probably not a daunting task on the technical and compatibility side if CA decides to put the time and resources into it.


    On the other hand, OP's do-or-die stance is completely unjustified. 16 threads will NOT be the norm in 2 years, it will be the high-end of the spectrum. Same goes for DX 12 only support, which would leave a lot of people (aka customers, aka sales) behind.

    According to the lastes Steam Survey, about 1/3 of the users still don't have a DX12+Windows 10 setup, and the overwheling majority of people have a CPU with 4 cores or less: https://store.steampowered.com/hwsurvey/Steam-Hardware-Software-Survey-Welcome-to-Steam?platform=pc

    And keep in mind that this is a voluntary survey, which probably means most people who even just take the time to answer it are already from the "top" strata of the gaming market, and therefore these data represent a more "enthusiast" slice of the total gaming population, not the real average Steam user.


    TL;DR Optimization should keep progressing over time, as it has been doing so far, but a sudden and drastic jump in technology would be unwarranted for and probably more damaging than anything else for a good portion of the user base.

    Switching to a modern API that actually functions with modern processors, isn't even remotely drastic.

    DX12 has been put into Windows 8, and Microsoft even caved and let multiple games run in DX12 mode in Windows 7.
    Windows 7 completely leaves support this year. It's not a current operating system. If you're still on it, upgrade.

    Vulkan? No such limitation to start with. Works on anything newer than XP, including Linux and . If you're still on XP, you're not playing Warhammer to start with.

    Core count? Same deal. It doesn't matter if 50% of Steam's users still have 4 cores. DX12 or Vulkan would only be "bad" for people with 2 cores, which is arguable because DX11 already has high overhead for threading with the inactive state. People who do only have two cores, can't play Warhammer for **** already. They need to upgrade.

    DX12, or Vulkan, is a performance improvement, not some kind of nightmare scenario for the guy with a five year old computer. They're actually the ones that get the most benefit, because without it they can't even feed the newest video cards.
    Optimizing with the highest of high-ends in mind (which is what OP suggests) never brings "the most benefits" to people with mid-low range hardware. You need to account for everybody to a certain extent, and that can be complicated depending on the engine and the different configurations.

    My only real point is that CA is not going to spend money for engine alterations if the end result of that is gating out from the game even just 1% of its current user-base, because that would be monetarily counter-productive, especially on an ongoing product that uses past releases as DLC for the newer ones. If they will actually go through with the DX12 implementation for WH3, it will NEVER be with the objective of simultaneously dropping DX11, because there are just too many current and potential customers that would be cut-off for no real sane monetary reason.
    Because having a crappy game experience because your turn times are 3 minutes long, and your minimum fps in typical game play dips down to around 20, is totally not costing them sales...

    I'd suggest you reread the OP. He's not arguing that they need to optimize for monster PC's. He's pointing out the obvious, that you always have one core pegged at 100%, and your performance goes to **** in a big battle, no matter how much money you spend.

    This franchise is massively in need of substantial improvements to parallelism, in the form of either DX12 or Vulkan, to solve the draw call cluster **** that is graphics performance, and in the form of a heavily threaded campaign map AI that doesn't spend longer than a bleeding player does, making moves. We could have utterly massive maps, and stupendously huge battles several times their current limits, and people with modern computers would have better performance than they do now. It's an atrocity.
  • ITA_Vae_VictisITA_Vae_Victis Senior Member ItalyRegistered Users Posts: 1,657
    edited August 2019
    psychoak said:

    I'd suggest you reread the OP. He's not arguing that they need to optimize for monster PC's. He's pointing out the obvious, that you always have one core pegged at 100%, and your performance goes to **** in a big battle, no matter how much money you spend.

    Comment removed.


    What OP said:

    "Second thing, may be even more important in some cases, - DX12 or Vulcan exclusive game. No more "windows 7" and DX11 in 2020-2021 gigantic strategy game."

    "And, at least future game 3, with a new or really modernized engine with much-needed new APIs exclusivity."


    To which I replied (twice) with the two following points:

    - Good/better API & Multicore optimization = GOOD, I hope CA does it
    - DX12 / Vulcan exclusive engine for WH3 = NOT HAPPENING because it LOSES THEM CUSTOMERS, THEREFORE MONEY (Current money - optimization money - less sales money = less money than Current money = not worth it)



    Are we all clear now?
    Post edited by dge1 on
  • Aram_theheadAram_thehead Registered Users Posts: 1,005
    Having a game as big as this that still runs on a single core, like games that were developed in the 2000s, is so sad.
  • Deep_echo_soundDeep_echo_sound Registered Users Posts: 601
    edited August 2019
    Customers for this game, and many other AAA-games upgrade their operating systems and hardware.
    By middle-late 2019 already be on "windows 7" is a joke. Not having graphics card from at least 2014 and wanting to play the latest and greatest is a joke. And we are talking about future! 2020, 2021, you know.
    This is just a standard for few years. Real P.C. gaming - "Win 10 64 bit" O.S. Period.

    Some people in 2013 still were on "windows XP". And not that small part, but mostly budget or backward. Industry should follow them? Games released in 2015 must have been for "XP"? We will orient all our future products on 10 and more years old systems?

    I propose to make "MS DOS" variant too!!! "Warcraft 2" was on that system? Yes! Strategy? Yes! Elves, orcs, humans, undead, dwars, etc.? Yes! "MS DOS" then seems like a great choice!!! Every computer for the last 25 years can run "Warhammer 3" on it. Well, on 0.002 fps but more player base! Profit!
  • psychoakpsychoak Registered Users Posts: 3,337
    edited August 2019

    psychoak said:

    I'd suggest you reread the OP. He's not arguing that they need to optimize for monster PC's. He's pointing out the obvious, that you always have one core pegged at 100%, and your performance goes to **** in a big battle, no matter how much money you spend.

    Comment removed.


    What OP said:

    "Second thing, may be even more important in some cases, - DX12 or Vulcan exclusive game. No more "windows 7" and DX11 in 2020-2021 gigantic strategy game."

    "And, at least future game 3, with a new or really modernized engine with much-needed new APIs exclusivity."


    To which I replied (twice) with the two following points:

    - Good/better API & Multicore optimization = GOOD, I hope CA does it
    - DX12 / Vulcan exclusive engine for WH3 = NOT HAPPENING because it LOSES THEM CUSTOMERS, THEREFORE MONEY (Current money - optimization money - less sales money = less money than Current money = not worth it)



    Are we all clear now?
    I'm sure everyone is clear you're arguing a false choice, so maybe you should stop?

    By the time Warhammer III comes out, even DX12 will be compatible with every version of Windows currently functional.

    Vulkan already is. It's fully supported for Windows 7. It works on every OS that DX11 work on, and a few that it doesn't.

    Contrary to shrinking their customer base, it expands their customer base. As I previously stated, low power, thermally restricted, and just plain old processors, see larger benefits from the new API's in real world gaming performance, than newer ones do, because even a quad core is twice as many cores as DX11 can functionally use. Then you add Linux, and Mac, which add up to twice as many people as Windows 8.

    Only if you get crazy big, does that nice new 9900K see a bigger benefit over the six year old quad core. Which would be nice. Better to build Warhammer III to scale with technology, than set artificial limitations based around several years old, low end hardware. That's what small unit sizes and no large armies settings are for.

    Which they can still do if they run two API's, but two API's is more expensive to work with than one. While subtracting variables, you kinda left that one out.
    Post edited by dge1 on
  • MystraMystra Registered Users Posts: 1
    Regards!
    Excuse my english
    I am a fan of Warhammer and total war and seeing the first post I agree with the user that the engine must be renewed, all those who speak of a new pc cost etc etc, what we cannot pretend is that we have been 100 years With the same engine and charging the same for each game edition, this already looks like EA's fifa every year to go through the box for what seems like a simple update, I have already thought that fifa should charge an annual fee or monthly and period and the Total war engine I think it has already taken out money to renew itself as it should be since users have already been paying for the same but different content for years, that if many things are added and many others are improved, but To me this seems more like an update than a new game, and what they say in the first post is quite true, it needs a facelift, I play like King Arthur from Neo Core they looked very good at the time compared to total wars.
    Greetings and let's hope that twist of the screw to close a great trilogy!
Sign In or Register to comment.