I have a bug. It opens alright and i go in and it has the MoJang logo then it won do anything else. I wait a while, Nothing i click on it, it stops responding and it says Javas not responding and it never does. So i'm not sure whether its Java or Minecraft, HELP!!!!
Here's mine - not sure if this is a bug, or an infinitesimally low-chance glitch:
I have a portal hub, the main portal being located at (X,-46 Y,177). I have ventured some distance away, establishing a new camp, and while chopping wood discovered this at (X,584 Y,286). I DID NOT PLACE THIS PORTAL HERE!! (and I never add corners to my portals)
Notice that even though it is active, it is not glowing, but black. Stepping through the portal brings me to the nether at the place I should be if I went through the one far to the north!!!
Stepping back through indeed brings me back to my main portal at (X,-46 Y,177), causing me to have to journey the many miles (lol) back to the southern camp. Not a game-breaker, but certainly a head-scratcher!
BTW, I destroyed the damned thing and scavenged the obsidian... not sure if I want to make another one yet... I'm still kinda freaked out, man... I've seen things... terrible things!
The same thing happened to Coe, in Coe's Quest! It was around ep 60 or so I think, he discovered (or rather his viewers discovered) a portal like this one, out on a lake. He was trying to build a passage in the Nether from one base to another at the time. But that was when portals still were new, so it seems to be an older phenomenon. Still, really creepy man!
Well, I tried an experiment- I broke the portals down, and deleted the DIM-1 folder forcing a rebuild of the nether. I built a new portal at the southern camp, near where the mystery portal appeared. Entering rebuilt a fresh nether landscape (finally!), but re-entering to Earth did not put me back at the camp where I entered, but about a "half-mile" away down the shoreline. O.o
So then i journeyed back to the north camp, expecting to enter the portal there and be in a fresh location... however, entering teh original portal PLACED ME BACK AT THE SOUTHERN CAMP'S NETHER LOCATION.
Re-entering back to Earth of course brought me back to the new mis-located portal down the beach from the southern camp; the same exact error as before i deleted teh Nether, but IN REVERSE. Obviously, whatever pointer to nether locations stored in the level.dat got munged, but deleting the nether and destroying the portals and saving does NOT remove this info from the save, it only resets what your "ground-zero" point of entry is. This is seriously messed up, and it's obvious the difficulty it introduces trying to make a navigation network.
Quote from Obst »
Not listed yet in 1.4/1.3 bug report thread: Uneven distribution of all kinds of ore among the four different quadrants of the map
the whole engine will likely (and seriously MUST) be re-written from scratch. (maybe MC rel v2 will be in C++ using memory mapping and distributed processes for robust performance? :tongue.gif:)
Something else (related) I have noticed, although cannot quantitatively prove, is that terrain seemingly becomes increasingly mundane and featureless further to the northeast, while it becomes more intricate and "epic" to the southwest. So I follow Horace Greely's advice to "go west, young man", and keep to the south at the same time.
One other thing: I have been probing the game's performance on my setups... there has been much debate as to whether MC requires more from a GFX card, or from the mobo. It's well-known that MC eats up nearly a gig of system mem while chugging along, and most of your processor's overhead. Running it on my Celeron-D 3.3 GHz, 2GB ram, with a NVidia GeForce 8800GT (512MB), I have noticed that during play, the app averages around 25-40% GPU usage, while memory usage gradually climbs until it hits the roof, regardless of veiwport stress (how much world is visible in-frustum), at which point it chudders, chokes, and blackscreens.
Interestingly enough, if you start, and stay in one place (regardless of settings) the mem usage will rise as the terrain unfolds, maxing out at around 220MB or so in a lightly forested zone near water... this will plateau here if you stay put, but venturing out, causing the game to add terrain to be rendered, will add to the gfx RAM load until you hit your RAM's ceiling.
Unexpectedly, backing out of the game to the app's menu does NOT unload the RAM... indeed, the RAM retains this orphaned data until you actually quit the MC app.
Blackscreen (GL pipeline failure)on this setup, not surprisingly, happens with roughly 1/4 the frequency as when I run MC on the same system using a GeForce 6600GT with only 128MB video RAM.
I have severely overclocked AND underclocked the rig, with little to no change in behavior.
(Note: using only the mobo's integrated ATI Radeon Xpress 1100 with 32MB shared sys RAM yields perfectly unplayable results, < 1fps="" and="" 1="" minute="">
Knowing this it is easy to see why people with great systems have trouble, while many with average systems have few complaints... these types of snafu are very much influenced by the user's gameplay style, as well as how much RAM the gfx card has to work with. The "local miner" who never ventures far may never have but the occasional problem on a card such as mine, while even with a card having 1GB RAM the "wanderer" may experience frequent lag and blackscreen as the accumulated data crunk builds up. GPU speed or mobo spec has relatively little to do with it all. Indeed, while MC requires a buttload of system RAM and badwidth, it runs with little variance on a processor as slow as 1.8GHz, giving acceptable framerates on a heavy-duty gfx card
As mentioned, GPU stress=roughly 30% avg in-game, while it shoots up to 50%+ while paused in the options menu overlay! Also, this occurs in Fancy mode where the only difference seems to be teh fact that transparencies are rendered. Enabling alpha in textures should not cause such an undue load on the GPU.
Having done some GL application building, these symptoms are not unfamiliar, but they do point to some fundamental framework deficiencies in the rendering engine. All in all, the application is very poorly managing memory (I suspect buffer abuse in particular), and when there is no more bandwidth left, there will be an inevitable call to a GL function, which borks the pipeline, making blackscreen occur. There really needs to be a better culling method implemented, if there is one at all presently.