• 0

    posted a message on Custom Ore Generation: First Revival
    Did some more examination on the current server setup and it seems like the biomes/chunks that have vanilla ore clusters ALSO have COG distributions. I've been coming across areas with both clusters and noticed COG veins scattered through them too. Some areas or biomes like Highlands "meadow" have only COG distributions and nothing else. Strange!

    Edit: Further updates, and here's what my config looks like:

    COG Config
    oreFreq=0.25
    oreSize=0.5
    deferred range=0
    vanillaOreGen=false
    Layered Veins for: coal, iron, gold, redstone, diamond, lapis, emerald, IC2 copper, IC2 tin, IC2 lead
    Disabled: IC2 Uranium, SimpleOres in its entirety

    IC2 Config
    All ores disabled in IC2

    SimpleOres Config
    All ores set to 0 spawn rate

    RESULTS: Actually pretty good at the moment, still looking for any biome or random chunk conflicts. Vanilla generation seems to be less problematic or not really present in the areas I've looked while flying around. I should note I also temporarily disabled Railcraft to help diagnose some tickrate spike problems, so next test will be with Railcraft back in and no other changes.
    Posted in: Minecraft Mods
  • 0

    posted a message on Custom Ore Generation: First Revival
    Quote from noogenesis

    Would you please help me understand these results? The first screenshot looks like only copper and tin are generating, or what are those blocks? Are you sure IC2 ore generation was disabled? The second screenshot looks like one would expect, as well. Should that be the case? I'm pretty sure that if you're seeing missed chunks that it's the age-old deferred generation bug, and is completely biome-independent. The variation biomes are just different parameterizations of the existing types. The parameters should have no effect on COG. I will try to debug this this weekend. Thanks for looking into this.

    Sorry, should've been more specific there :)

    The first screenshot showing only copper and tin does in fact have ALL IC2 ore generation disabled. That's why it's kind of strange; the vanilla ore gen option is set to false, IC2 ore gen is disabled in the IC2 config, and it's for whatever reason still generating IC2 ores using vanilla ore generation clusters. Fiddling with it is probably clouding the waters in regards to the main problem of missing/skipping chunks and/or biomes so it's probably best to ignore that part of my rambling for now.

    The second screenshot is exactly what we're expecting: a chunk with vanilla ore generation that COG is skipping. The thing is, this ONLY occurs in specific biomes and not random chunks. It's 100% repeatable with Cold Taiga (or possibly Taiga variants in general, will have to test more) while many other biomes including Highlands ones work perfectly fine with COG vein distributions. I've tested having Highlands installed or removed and the same results come up, Taiga types revert to vanilla clustered ore and setting the deferred range to 0 doesn't change how the ore generation works in the slightest. Kind of crazy! I figure if it's the deferred generation bug you're describing, changing the range setting would have some impact at least.

    *edit* I should also add that in most of these cases, the affected biomes are more than 1 chunk in size. I was using biome chunk size 1 in Highlands to maximize exposure to different biomes and find ones causing problems for testing!

    Hope that's a bit more clear. It's definitely a weird issue, let me know anything I can do to try and help pin down what's going on. I may also outright strip my mods out to just COG and experiment to eliminate any other variables just in case. The only three things I have which touch ore are Railcraft, Simple Ores, and IC2. Removing any or all of them doesn't seem to change the results though.

    Let me know how I can help and I'll do what I can.
    Posted in: Minecraft Mods
  • 0

    posted a message on Custom Ore Generation: First Revival
    Quote from noogenesis

    Note that setting the deferred population range to zero (in debug options) should avoid this, with the caveat that some vanilla generation will persist (in addition to the COG generation). Anyway, it would be helpful to know whether you still see these issues even after disabling deferred generation.

    Will do, I'll post up some results when I get to testing it this afternoon. I don't mind some residual vanilla oregen clusters dotting the terrain, so long as the bulk of the ore is in COG distributions :)

    It does seem totally repeatable in certain vanilla biomes. Cold Taiga is the one I've seen the most, but I've also encountered vanilla Forest biomes exhibiting the crazy cluster ore gen problem too. Interesting that it seems to apply the vanilla cluster distribution but at an insanely high frequency for no apparent reason. Gotta love bugs, eh?

    Edit: Results are... interesting. This is with vanilla ore gen disabled in COG, deferred generation range 0, and IC2 ores disabled in IC2.cfg. If vanilla oregen is left on, there is no change in the affected biome(s). So it does seem to be totally skipping the Cold Taiga and possibly other biomes for unknown reasons. Cold Taiga is a variation biome in 1.7 (or something like that, I need to read up more) so could that be a factor? (Further edit: seems like other variation biomes like Mega Taiga do it too.)
    Posted in: Minecraft Mods
  • 0

    posted a message on Custom Ore Generation: First Revival
    Quote from Lonemind_


    I could be wrong but it looks like the ores that are generating strangely are limited to a single chunk. It also appears as though they are generating in a vanilla fashion. Perhaps that one chunk is an unsupported biome? First thought that comes to mind is a Taint biome from Thaumcraft. This is all conjecture but perhaps it may help. :)

    Good thinking, but I should've mentioned that in the Highlands config I specifically set the biome size setting to 1 to maximize my exposure to other biomes while testing. That would probably account for the small area.

    In any case I'm not using Thaumcraft or anything that I can think of which would throw unsupported chunks out barring any major screwups in my configuration. The weird ore gen is repeatable with the Cold Taiga variation biome, so I'm going to keep poking around to see if any others pop up or if it's indeed some sort of chunk issue.

    Edit: Cold Taiga biome causes weirdness even on non-Highlands worldgen mode. Doesn't spawn any ore with COG frequency globally set to 0, but any amount beyond that (even 0.05) will cause insanity on the ore spawns. Tried with Highlands uninstalled, same result. "Forest" biome also seems to do crazy things, so I'm not sure what all biomes are acting funny combined with COG.
    Posted in: Minecraft Mods
  • 0

    posted a message on Custom Ore Generation: First Revival
    It's been a while since I set up a server but once again I've gotten talked into it...

    For some reason Cold Taiga (vanilla biome variation) is causing some seriously weird ore generation for me on 1.7.2. All the other biomes seem fine as far as I can tell. Pictured here iron is red, IC2 tin is white, IC2 copper is orange and black is coal. I'm also using Highlands for the moment and haven't seen any odd ore distributions in the Highlands biomes, only the Cold Taiga biome specifically.

    By the way, great work bringing COG back to life. This was the first mod I hunted down when I started putting together a server for the bunch of goobers that talked me into playing Minecraft again. Looking forward to full SimpleOres support as well in the future.
    Posted in: Minecraft Mods
  • 0

    posted a message on Mob Spawning Observations and Tests - updated 8/28/13
    Wow, major stuff going on over here. I didn't get the chance to read every post in the past 3 pages or all the OP updates but I've passed the link along. There are now 3 of us working on getting the mob configs and all that sorted out instead of just one bedraggled Rhaz :)

    I had started working with JAS just before vanishing due to work, so we'll be getting back to that hopefully over the weekend.
    Posted in: Discussion
  • 0

    posted a message on Mob Spawning Observations and Tests - updated 8/28/13
    Woo, a few days out of the loop and I have a page and a half to catch up on!

    Quick note on the DRPG mobs: I can say with complete certainty that most of them DO in fact despawn. What I might be seeing with the "deep cave" mobs is that without any above ground mobs able to spawn in daylight the spawner decides something is better than nothing and drops stuff underground. The town is at around 58 elevation so it's within the sphere of spawning range in all directions, allowing stuff to spawn down in the depths despite no one actually being nearby.

    @Gurrok: Great news! I've been fiddling with weights a lot more lately myself. Hopefully that works out well and some of us can replicate the results!
    Posted in: Discussion
  • 0

    posted a message on Mob Spawning Observations and Tests - updated 8/28/13
    I've gotten considerably better results in the past day or two but sadly had to completely omit DivineRPG's overworld mobs because I believe sirdave is correct in saying they're non-despawning. With the Divine mobs (only caveclops and enthralled dramcryx) inserted into the surrounding biomes, spawns all but ceased in less than 40 minutes of testing with only two players on the server. Without those and using some tweaked spawn weights with a larger cap we saw spawns continue to work normally for several hours before I had to call it a night. I'll be leaving the config as is this evening for a longer test run.

    At one point I thought I saw spawns taper off, but as soon as I moved away from the swamp I was mining under they fired back up. No new chunks were generated, all the area had been previously explored on the surface. None the less it does seem like staying in one spot for longer than 45 minutes to an hour almost forces you to move to a new biome to kick the spawns back in.

    If I don't find better solutions to this I may just take the CMS only route myself. I think the reason I avoided Highlands initially is that CMS doesn't seem to spawn anything in the Highlands biomes, but I could be wrong there. There was a reason I went with ExtraBiomesXL initially and I seem to recall it involving mob spawns since Mo'Creatures was the first thing I installed on the server. If CMS works okay with Highlands I'll give it a shot though.
    Posted in: Discussion
  • 0

    posted a message on Mob Spawning Observations and Tests - updated 8/28/13
    I wonder what else causes the so-called leaking world issue... the excessive lag sounds VERY similar to what I experienced those couple evenings on my server where tickrates went crazy and I had to just restart the server to clear it up.

    @Gurrok: Did you also import the lizard as creature instead of monster, or try both ways? I've messed with the overrides a bit mostly with ambient vs. creature though. It seems to have an impact, but it's hard to quantify or describe the overall global spawning trends to say exactly what it's doing. I think I have some mismatches in my config where generally not hostile mobs are listed as creature or the opposite, I'll have to double check and make sure to watch for that and correct any that seem to not be spawning as intended.
    Posted in: Discussion
  • 0

    posted a message on Mob Spawning Observations and Tests - updated 8/28/13
    @Gurrok: I could swear there's something fishy going on in the spawner because I've seen some anomalies like that too. We've had all of us in a small area around our town and had spawns slowly get crippled, though no creatures or monsters seemed to be spawning, nor were any save sometimes a handful visible on the minimap/overhead map. Sometimes the ForgeEssentals butcher command is helpful but seems to trigger spawns all over the place once some of the cap is freed up. I have also witnessed running a /butcher <radius> PASSIVE command cause MONSTERS to spawn immediately once the spawn cap is freed up by removing animals. I really don't know what's going on with that.

    I won't be able to do any real testing for a few days due to being insanely busy with my businesses here but I'm working on a list of things to test and coming slowly to the conclusion I might have some mod conflicts causing some "ghosts in the machine" as it were. If things are too much of a pain to deal with I'll just ride out the bugs and quirks as best I can until it's time to make the jump to 1.5 and start with a fresh client and server jar etc.
    Posted in: Discussion
  • 0

    posted a message on Mob Spawning Observations and Tests - updated 8/28/13
    Damn shame they broke the DivineRPG downloads so badly. That's recent news with the latest release, the prior releases had perfectly functional downloads :huh: Sadly it does seem like some of the frequent posters on their official thread are rather snippy, but in their defense I guess they have hordes of rabid mod fanboys swarming them constantly with the same questions or common user error problems. The perils of customer service, as it were.

    @sirdave - it seems we have very similar ideas on how the global animal and mob populations should work in theory, but sadly I think you're correct in saying we won't see anything very in-depth for a very long time. I would not be at all against using CMS in the future if it can give better results than I'm seeing now. I had far more issues with CMS though relating to modded biomes and just far too many monsters all over the place regardless of spawn weights. Perhaps due to my inexperience with mob spawning in general and the settings available I didn't fully understand the settings. I can always revisit CMS and the regular Mo'Creatures configs though, just to see exactly what sort of differences I get in practical use.
    Posted in: Discussion
  • 0

    posted a message on Mob Spawning Observations and Tests - updated 8/28/13
    @Crudedragos: Yes that's essentially what I was getting at, "short circuiting" the properties of mobs using a separate mod entirely. I'm not too surprised that from a practical standpoint a coremod might be the main or only way to accomplish it, though. My logic was that having one central mod able to override or configure ANY mob mod will avoid having to pester every modder to expose their config settings on a per-mod basis, as well as not having to juggle every mod's configs separately since they'd all be in one place that way. Alas, it seems like it's not especially doable at least for the time being, and it's a huge pain either way.

    @Mohawky: Replace ALL the spawn code! :lol:

    I've had good luck with Ruins and Highlands so far, so that's another of its perks I suppose. I do really like some of the biomes though, the Sahel and Outback are really interesting twists blending grassland and desert somewhat which is a void I felt needed filled with some new biomes. I don't know if I'll go too overboard adding any more biome mods simply for the sake of per-biome mob spawns, and I guess there are only so many variants on biomes you can throw out there before they don't give a whole lot more variety versus the time spent configuring them.

    An 18 hour stretch petering down on spawns isn't too bad, definitely progress compared to what we started with not that long ago! I'll be generally content to have at least 12 hours of continuous uptime before spawns get scarce. I'm almost happy with the creature/animal spawns right now, it's the hostiles that seem to be hitting a wall. Or at least they're hitting the wall sooner, I suppose.
    Posted in: Discussion
  • 1

    posted a message on Mo' Creatures - v12.0.0 for Minecraft 1.12.1!! Now Opensource!!
    Quote from kawaiimiyo

    A switch would seem slightly out of place on a farm, what about a wooden fence, stone fence, or fence gate? Wooden fences are used in mineshafts, so don't know if that would work, but stone fences or gates do not appear naturally,[...]


    My immediate thought was fence. They're completely at home on most farm setups, and though fence does appear in mineshafts no vanilla animals spawn there. I don't think there would be any concern with vanilla animals spawning near fences that appear underground. Considering that fences and walls are "pens" they could serve nicely in this regard, taking them all into account would let folks choose what they want to use or mix and match.

    I also like the technical solution posed above. To be frank, it's how spawning should work in the first place in my opinion. Pen in and feed your animals to make them stick around for the long haul, but let the game clear out the spawns otherwise to avoid the kind of slow bottleneck we've discussed at length in many threads already.
    Posted in: Minecraft Mods
  • 0

    posted a message on Mob Spawning Observations and Tests - updated 8/28/13
    @Mohawky - I also used EBXL at first but didn't like the extra blocks it added and thought a lot of the biomes were really an eyesore. When I got tired of trudging through "extreme _____" biomes and we decided to migrate to a new world I replaced it with Highlands and REALLY like most of the biomes. I mean, volcanic islands, how do you say no to that? :)

    It seems like one of the Divine dimensions has a "Wraith" while Mo'Creatures has the "wraith". The only difference is the case, which isn't taken into account anywhere but the formal name of the entity I guess.

    I went ahead and migrated my "ambient" critters (sans bats, they're ambient by default anyway) to the "creature" category and saw immediate changes in spawning behavior globally. I may need to increase the cap if these mobs are now correctly counting against the right category. I can definitely confirm that regardless of the MSC override, those mobs WILL be considered "creature" because Rei's Minimap will show them as white dots on the radar.

    With the omission of vanilla animals in the spawn config (except the "highlands" biome as noted) I'm getting pretty solid results on animal spawns. I have to take an off day or weekend and run the server a long stretch of 6+ hours straight to see how long it takes to bottleneck. Hostile mobs seemed to be doing well but there is a distinct lack of spawns in certain biomes which I need to investigate. What good is a jungle with dangerous spawns if they never show up, right?

    Last two days I had instances where the server tickrate plummeted (ticks reading 100-160ms per tick which is insane) leading to server crashes. Not sure of the cause yet but both cases involved lots of long distance exploring and chunk generating by at least 1 player.
    Posted in: Discussion
  • 0

    posted a message on Mob Spawning Observations and Tests - updated 8/28/13
    I agree that for the moment disabling the Highlands conflicted biomes is the way to go, but given that at least two such biomes exist in the current world I guess I'll have to keep on top of pruning down vanilla animals :) At least this way there's some source for them, and the two spots are a fair distance from our home area as well.

    Since we know from Crudedragos' posts that the ambient mobs I've imported do not internally count as "ambient" in regards to the spawn caps, I may undo that change and add them to the "animals" category. This would let me spread out spawn weights considerably and provide more mobs to sparse biomes like deserts where scorpions can flood the place easily.

    If either of you guys (re: Mohawky and sirdave) drop in DivineRPG could you test the Mo'Creatures wraith mob and let me know if it spawns for you? It seems like Divine has a "wraith" as well that might be causing neither entity to spawn. It's not a huge issue because I can always use the flamewraith instead, I'm just wondering if it's that mod combo doing it or my config.
    Posted in: Discussion
  • To post a comment, please .