So for the many of you who don't know EverQuest is an MMORPG originally developed by Sony online entertainment also know as (SOE). It has been developed on by some other studios, and is currently under the wonderful guidance of daybreak games. Of course, this is not the point while EverQuest was doing awesome they decided to make an EverQuest2. Which even if I think is trash it did better than EverQuest next and landmark. This was most likely because of the development of World of Warcraft (WOW). In short, when they had big competition they decided to split their player base.
So that's very interesting but you may be wondering what this has to do with Minecraft. Don't worry this thread is not a joke about the infamous Minecraft 2. So as the game that looks like it could become better than Minecraft if development continues the way it does comes nearer should we be worried about Mojang developing Minecraft dungeons? In case you were wondering the game I am referring to is Hytale.
I don't get all the fuss over Hytale... We've just seen a machinima trailer and it looks like a retexturized Minecraft with 4 or 5 new mobs and a view distance of 16 chunks that will bring even the mightiest PC to its knees... Plus, knowing the corporate mindset of Hypixel, it will be a server-based game without the gazillion custom options Minecraft offers.
a view distance of 16 chunks that will bring even the mightiest PC to its knees...
Minecraft is terribly unoptimized, and only getting worse with each update; for comparison, the developer of Fastcraft was able to get 256 chunk render distance and performance exceeding that of Bedrock even as that entire area was being simulated (part of the better performance of Bedrock is because it "cheats" a lot):
In my own tests FC2 could easily outperform everything else, including the W10 edition (MCPE/Bedrock).
The config allows increasing the view distance limit up to 256. Performance will likely tank once the game runs out of VRAM.
All rendering happens with full detail and for a view distance setting of d FC2 will render (d*2+1)2 chunks, some bugs around post-initial area and very large distances aside. The whole view distance is being simulated (ticked). MCPE doesn't do either.
My test environments are only a Nvidia GTX 780 and an Intel HD 2000 [note that Mojang recommends a GTX 700 series GPU and HD 4000 as a minimum for the current version], both with Intel quad core CPUs, your experience on other platforms may vary and is of elevated interest.
I've made order-of-magnitude optimizations myself, including far offsetting the performance impact of new features (TMCWv5, in development, currently has over 300 additions and changes yet runs far better than the current release version, never mind vanilla 1.13+, and even 1.6.4). This includes optimizations that enable increasing graphics quality without decreasing performance; in fact, I got clouds to render twice as fast on Fancy as on Fast (I've often seen clouds blamed for lag), and reduced the time taken to render leaves on Fancy by more than 75%:
This was measuring the time needed to render a leaf block on Fancy graphics (the changed rendering actually had no effect on steady-state FPS as this time counts to the time needed to re-render a chunk section during a chunk update; since they do take time out of the time allotted per frame this increases dynamic FPS, as is the case when moving around). The times shown correspond to a total time of 11.4 and 2.59 milliseconds per 4096 blocks (worst-case per chunk update); if this were the only time spent per frame the maximum FPS would be 87.7 and 386.3 respectively:
17:56:58 - Took 2780 nanoseconds to render leaf block (vanilla)
17:56:58 - Took 2795 nanoseconds to render leaf block (vanilla)
17:56:59 - Took 2775 nanoseconds to render leaf block (vanilla)
17:57:00 - Took 626 nanoseconds to render leaf block (modded)
17:57:00 - Took 645 nanoseconds to render leaf block (modded)
17:57:01 - Took 625 nanoseconds to render leaf block (modded)
These are the times for rendering Fancy clouds, which is probably the single biggest optimization that I've made (the times in vanilla don't seem to support clouds having a major impact on performance, at least not on my hardware. A bigger factor may be that vanilla re-draws clouds every frame by directly issuing draw commands to the GPU instead of using a display list, and these times do not include that time):
22:12:23 - Took 188486 nanoseconds to render vanilla clouds
22:12:29 - Took 188199 nanoseconds to render vanilla clouds
22:12:34 - Took 188571 nanoseconds to render vanilla clouds
22:12:50 - Took 16207 nanoseconds to render optimized clouds
22:12:55 - Took 17033 nanoseconds to render optimized clouds
22:13:00 - Took 16701 nanoseconds to render optimized clouds
This was measuring the time it took the mob spawning code to run in a Superflat world with no mobs able to spawn (worse-case). The main change that I did was reduce the per-chunk spawn rate by a factor of 4 (the game still makes so many attempts to spawn mobs that there was no noticeable effect on the ability to maintain the mob cap even when moving quickly; the 75% reduction in mob farm efficiency is not an issue as I don't even like them and have nerfed them in other ways. Bedrock has such bad mob spawn rates in part because it reduces the chance per chunk per tick to only 1/200, just twice as often as passive mob spawning on Java, though even that is enough to maintain the mob cap if you don't move much; much of the rest is due to the very low mob density it sets per 9x9 chunk area):
16:02:49 - Took 633764 nanoseconds per spawn cycle (vanilla)
16:03:39 - Took 633854 nanoseconds per spawn cycle (vanilla)
16:04:29 - Took 634061 nanoseconds per spawn cycle (vanilla)
16:05:19 - Took 122141 nanoseconds per spawn cycle (modded)
16:06:09 - Took 129139 nanoseconds per spawn cycle (modded)
16:06:59 - Took 124512 nanoseconds per spawn cycle (modded)
These are the times taken to generate caves, in microseconds per chunk, over successive optimizations (the first is based on TMCWv4 code while I've since added many more types of caves - and still significantly sped up the code):
Changing Java's (slow) Random to custom full 64 bit RNG and other optimizations:
initializeTime: 210; validTime: 320; generationTime: 1730; totalTime: 2262
initializeTime: 244; validTime: 406; generationTime: 1778; totalTime: 2429
initializeTime: 269; validTime: 387; generationTime: 1577; totalTime: 2235
initializeTime: 270; validTime: 398; generationTime: 1813; totalTime: 2482
Latest timings, including after adding many new types of caves and increasing cave volume to nearly twice that of vanilla, which is about as fast as the "generationTime" (other timings are specific to TMCW) and not including ravines (vanilla generates them separately from caves while I generate both at once):
initializeTime: 143; validTime: 223; generationTime: 1059; totalTime: 1426
initializeTime: 153; validTime: 224; generationTime: 1157; totalTime: 1534
initializeTime: 144; validTime: 231; generationTime: 1171; totalTime: 1547
initializeTime: 149; validTime: 223; generationTime: 1109; totalTime: 1482
In general, every time I've added some new feature to TMCWv5 I've more than offset the performance impact and otherwise make it as fast/lightweight as practical; this sort of thinking is what Mojang should be doing (they seem to think that that if their code is fast enough on their system then everybody else should be able to run it, and/or everybody regularly upgrades).
Apples and Oranges. Games like EQ survive on retaining players who pay recurring subscription. If I already bought MC, I am not really paying anything else unless I am willing to fork for Realms. But who on Java MC actually USES Realms?
So if company makes 'Minecraft 2' and gets me to buy it, it's hardly "splitting the playerbase".
The EQ case is also portrayed rather poorly. I played EQ since beta, with a couple extensive breaks here and there, and am quite active now with 3 subbed accounts.
EQ2 did not really 'split' the playerbase. I was quite actively raiding with my guild in 2004 when EQ2 and WoW went live, and I did not really notice much of a population drop. EQ was, and still is a rather brutal game by most modern MMO standards. Pretty much any significant progress requires grouping and larger scale effort, with significant dedication and time investment. Latest expansions such as Gates of Discord and Omens of War were heavily endga.e raiding focused. But at the time there was a growing demand for a more casual friendly game where people could easily progress solo if so inclined. EQ2 aimed to be that sort of game and was to draw the casual crowd who found EQ too intense, rather than pillage players from EQ. Of course, World of Warcraft happened, nailing the casual vibe better, so EQ2 never got that much traction.
IMHO, EQ2 was the right decision. Overhauling EQ to be more casual oriented would alienate the core invested playerbase and not guarantee attracting the casuals, especially given the upcoming competition.
Not familiar with EQ (beyond a few articles), but MC does appear to be balkinizing it's player commuinty:
⊕ the java / non-java division is undoubtable with the two versions being mutually incompatable not just electronically, but in terms of mechanics
⊕ the pre-1.9 / post-1.9 diivision (occasioned by a change to basic combat mechanics) is a bit less absolute, but has strongly entrenched advocates in both camps
⊕ recently I found a report (apparently confirmed by one of the denizens of the server) that SciCraft is still using 1.12.0 as the changes through 1.12.2 were too minor and 1.13.x too laggy to update. (This was in a discussion of the preferred/recommended version to use for technical play showcase/let's plays that seemed to come to something of a consensus that 1.12 or 1.12.2 was best suited to the purpose)
Adding in the to-be-expected divisions based on equipment quailty [not system type] & playstyle (builders vs explorers vs redstoners or PvE vs PvP) only increases the number of (often fairly insular) groups.
scorrp10's final sentence "Overhauling EQ to be more casual oriented would alienate the core invested playerbase and not guarantee attracting the casuals, especially given the upcoming competition."
might be taken as advice why explicitly catering to at least two of the clusters of groups [FRPG adherents and SandboxBuilders] could be a wise decision.
MC dungeons (which seems to have dropped off the radar) appeared to be intended as the FRPG fork of the game.
As with many other situations, two products (each optimized to a particular purpose) may well be more successful than a single product that attempts to be all things to all people (and fails to satisfy any as well as its competitors).
Rollback Post to RevisionRollBack
"Why does everything have to be so stoopid?" Harvey Pekar (from American Splendor)
WARNING: I have an extemely "grindy" playstyle; YMMV — if this doesn't seem fun to you, mine what you can from it & bin the rest.
So for the many of you who don't know EverQuest is an MMORPG originally developed by Sony online entertainment also know as (SOE). It has been developed on by some other studios, and is currently under the wonderful guidance of daybreak games. Of course, this is not the point while EverQuest was doing awesome they decided to make an EverQuest2. Which even if I think is trash it did better than EverQuest next and landmark. This was most likely because of the development of World of Warcraft (WOW). In short, when they had big competition they decided to split their player base.
So that's very interesting but you may be wondering what this has to do with Minecraft. Don't worry this thread is not a joke about the infamous Minecraft 2. So as the game that looks like it could become better than Minecraft if development continues the way it does comes nearer should we be worried about Mojang developing Minecraft dungeons? In case you were wondering the game I am referring to is Hytale.
Minecraft/Minecraft Dungeons
-Official site
-Dungeons
-Twitter
-FaceBook
-Wiki
-Youtube
-Mojang
EverQuest/EverQuest2/Daybreak
-daybreak
-EverQuest
-EverQuest2
-FaceBook
-Twitter
-FaceBook(EQ2)
-Twitter(EQ2)
-Youtube
I don't get all the fuss over Hytale... We've just seen a machinima trailer and it looks like a retexturized Minecraft with 4 or 5 new mobs and a view distance of 16 chunks that will bring even the mightiest PC to its knees... Plus, knowing the corporate mindset of Hypixel, it will be a server-based game without the gazillion custom options Minecraft offers.
Minecraft is terribly unoptimized, and only getting worse with each update; for comparison, the developer of Fastcraft was able to get 256 chunk render distance and performance exceeding that of Bedrock even as that entire area was being simulated (part of the better performance of Bedrock is because it "cheats" a lot):
I've made order-of-magnitude optimizations myself, including far offsetting the performance impact of new features (TMCWv5, in development, currently has over 300 additions and changes yet runs far better than the current release version, never mind vanilla 1.13+, and even 1.6.4). This includes optimizations that enable increasing graphics quality without decreasing performance; in fact, I got clouds to render twice as fast on Fancy as on Fast (I've often seen clouds blamed for lag), and reduced the time taken to render leaves on Fancy by more than 75%:
In general, every time I've added some new feature to TMCWv5 I've more than offset the performance impact and otherwise make it as fast/lightweight as practical; this sort of thinking is what Mojang should be doing (they seem to think that that if their code is fast enough on their system then everybody else should be able to run it, and/or everybody regularly upgrades).
TheMasterCaver's First World - possibly the most caved-out world in Minecraft history - includes world download.
TheMasterCaver's World - my own version of Minecraft largely based on my views of how the game should have evolved since 1.6.4.
Why do I still play in 1.6.4?
Apples and Oranges. Games like EQ survive on retaining players who pay recurring subscription. If I already bought MC, I am not really paying anything else unless I am willing to fork for Realms. But who on Java MC actually USES Realms?
So if company makes 'Minecraft 2' and gets me to buy it, it's hardly "splitting the playerbase".
The EQ case is also portrayed rather poorly. I played EQ since beta, with a couple extensive breaks here and there, and am quite active now with 3 subbed accounts.
EQ2 did not really 'split' the playerbase. I was quite actively raiding with my guild in 2004 when EQ2 and WoW went live, and I did not really notice much of a population drop. EQ was, and still is a rather brutal game by most modern MMO standards. Pretty much any significant progress requires grouping and larger scale effort, with significant dedication and time investment. Latest expansions such as Gates of Discord and Omens of War were heavily endga.e raiding focused. But at the time there was a growing demand for a more casual friendly game where people could easily progress solo if so inclined. EQ2 aimed to be that sort of game and was to draw the casual crowd who found EQ too intense, rather than pillage players from EQ. Of course, World of Warcraft happened, nailing the casual vibe better, so EQ2 never got that much traction.
IMHO, EQ2 was the right decision. Overhauling EQ to be more casual oriented would alienate the core invested playerbase and not guarantee attracting the casuals, especially given the upcoming competition.
Not familiar with EQ (beyond a few articles), but MC does appear to be balkinizing it's player commuinty:
⊕ the java / non-java division is undoubtable with the two versions being mutually incompatable not just electronically, but in terms of mechanics
⊕ the pre-1.9 / post-1.9 diivision (occasioned by a change to basic combat mechanics) is a bit less absolute, but has strongly entrenched advocates in both camps
⊕ recently I found a report (apparently confirmed by one of the denizens of the server) that SciCraft is still using 1.12.0 as the changes through 1.12.2 were too minor and 1.13.x too laggy to update. (This was in a discussion of the preferred/recommended version to use for technical play showcase/let's plays that seemed to come to something of a consensus that 1.12 or 1.12.2 was best suited to the purpose)
Adding in the to-be-expected divisions based on equipment quailty [not system type] & playstyle (builders vs explorers vs redstoners or PvE vs PvP) only increases the number of (often fairly insular) groups.
scorrp10's final sentence "Overhauling EQ to be more casual oriented would alienate the core invested playerbase and not guarantee attracting the casuals, especially given the upcoming competition."
might be taken as advice why explicitly catering to at least two of the clusters of groups [FRPG adherents and SandboxBuilders] could be a wise decision.
MC dungeons (which seems to have dropped off the radar) appeared to be intended as the FRPG fork of the game.
As with many other situations, two products (each optimized to a particular purpose) may well be more successful than a single product that attempts to be all things to all people (and fails to satisfy any as well as its competitors).