|
I'm enjoying SC:R thus far. I don't mind the new graphics in most scenarios. It seems the unit relations are pretty well-preserved. I'm not 100% happy with the coherence of the ladder, but I guess that's to be expected. Obviously the #1 complaint against Blizzard, AI, and other concerns--if they are to be addressed--is the question of whether the player base is consistent.
It's approaching the two week mark for play and I'm thinking about setting up some custom hotkeys. This was pretty common in the original game as well--as I remember remapping some of my keys--for instance ~ for probe production. This was quasi-illegal but considered fairly kosher. Obviously most people did not endorse auto-mining (in any form) or MBS (in any form) or max-select (in any form) etc. All that was frequently addressed by the same bans that would handle map hacks and resource juggling.
So I'll probably be setting up some custom hotkeys--something I'm glad the game supports. I wouldn't mind dual ladder system with all the SC2 features implemented, but this is something I don't expect to see for nostalgia's sake.
Has anyone else noticed some unusual glitches?
Z-glitch: Zealot does not attack (melee zealot was usually attacking).
Load Glitch: The biggest glitch currently. Settings are not preserved upon load. Scroll for both mouse and keyboard has to be set independently of each load.
Spatial glitches: Carrier/Goliath and Hydralisk/Zealot appear to be handled incorrectly as well as some common mining pathing problems.
Dragon AI: It seems the Dragoon is functioning correctly for its role in most matchups. This is a good sign--less Dragoon and Reaver glitching would be a positive. I can't say this has been arrived at with 100% consistency in each event but the Dragoon is functioning better.
|
I talked with some Blizzard agents from Lordaeron and it seems that theyweren't convinced we were facing difficulties with the load settings.
Is it confirmed that most othjer people are not receiving the load-glitches? Specifically are mouse settings, scroll settings etc saved? I assume this might be a simply endemic glitch other players are not experiencing
|
My settings fail to save say 1/3rd of the time, or they only partially save.
|
I dont know for the keys but the dragoons are acting the same as always. May be you got more used to it? I do play with low res graphics because protoss look too flashy in hd for my liking (and probes look like origami...may be just me) but i dont think it changes anything
|
The "Z-glitch" if I understand your situation correctly has been present since the dawn of time and is the same mechanic that causes stuck Dragoons, Marines, etc.
Haven't had any issues with loading whatsoever.
I haven't noticed any of these "spatial problems" that you are discussing either.
Honestly this reads more like a blog lol.
|
I stopped getting the settings load glitch months ago. I was under the impression it had been solved.
|
Yeah all my settings get saved and loaded correctly. But I haven't played too much recently so what do I know.
|
I only had my settings not load a couple of times, months ago. But I don't set my scroll speed for keyboard / mouse different from default. So there's no knowing wether that saves or not.
|
On February 20 2018 23:34 Tempest99 wrote:Z-glitch: Zealot does not attack (melee zealot was usually attacking).
Spatial glitches: Carrier/Goliath and Hydralisk/Zealot appear to be handled incorrectly as well as some common mining pathing problems.
Dragon AI: It seems the Dragoon is functioning correctly for its role in most matchups. This is a good sign--less Dragoon and Reaver glitching would be a positive. I can't say this has been arrived at with 100% consistency in each event but the Dragoon is functioning better. I doubt that there are even any changes. Game engine, scripted unit behaviour etc. are explicitly things that Blizzard did not touch and will not touch (except maybe for the attack freeze bug, which both the Zealot and Dragoon behaviours mentioned here seem to pertain to, but I am not even sure they went through with changing that, or succeeded in fixing it).
Particularly mining pathfinding problems are something that would require a completely remodeled pathfinding algorithm (basically getting rid of the primitive region-based long range pathfinding), not going to happen. If you want these bugs to actually get fixed, stop complaining about it to Blizzard as a general bug and better spend your energy filing bug reports for every single instance of it on every single map to both Blizzard and Afreeca, who still seem to mostly lack awareness of this and other serious map-related bugs. It's just a matter of finally establishing some proper quality criteria for maps and get some capable map makers to fix the maps.
|
Freakling, for those of us heathens not in the know can you explain what the mining pathfinding problems are? Like what do the workers actually do behavior wise, any common elements that cause it, and how significant the effect is?
Thank you!
|
On February 23 2018 04:22 L_Master wrote: Freakling, for those of us heathens not in the know can you explain what the mining pathfinding problems are? Like what do the workers actually do behavior wise, any common elements that cause it, and how significant the effect is?
Thank you! http://www.panschk.de/mappage/newsscript/viewarticle.php?newsid=35
There's a lot of information here; I think this should address some of your questions.
|
It's when a worker does not take a direct mining path from resource depot to resources or back. The cause is usually the resource depot being located in a pathfinding region that is not directly connected to the region the worker is moving/mining in. There are lot's of slightly different variations of this, but they all got to do with the local pathfinding region layout (a ton of other bugs and glitches, like vortices or cliff glitches are also directly linked to pathfinding regions, by the way). The effect can be, and usually is, very significant (particularly early game or in a ZvZ, read: whenever worker saturation is low and worker efficiency mostly depends on cycle duration): Even mild cases have been known to cause a ~10% in single saturation mineral mining rates (there are also some particular region layouts that actually improve worker pathing and thus mining rates from certain patches, though). Another well-known worker bug, workers migrating through gaps to the back of the mineral formations, is also often linked to this (when the minerals are located on a region with a node behind the mineral lines, workers consider the shortest path to it that way – coordinates being rounded up and regions being processed left to right are the reasons that this happens far more often with mineral formations on the right-hand side of resource depots than the left).
If you really want to get a feel for it, I suggest you toy around with ScmDraft a bit. The new ASL5 test maps are a good starting point, as they are not protected and already come with an extensive list of documented pathfinding-related bugs (among others).
EDIT: For any one who has never seen this before (which I assume are still lots of people): This is how the pathfinding regions and nodes grid looks like, which is calculated from terrain flag information of a map (terrain level and walkability) each time at game start (and now thankfully by ScmDraft too). The map is (2)Circulator0.920a from the current batch of ASL test maps: + Show Spoiler [show image] +Question to aspiring heathens: Find the mining bug and identify the affected resource patches (without in-game testing!) Bonus question: There are also two Ultra-vortices (spots where Ultralisks can get stuck on terrain) in the picture. Where?
Another example of interesting consequences of pathfinding regions (because I had the picture lying around already): + Show Spoiler [show image] +Question to aspiring heathens: What's going on here?
On February 23 2018 07:05 Jealous wrote:Show nested quote +On February 23 2018 04:22 L_Master wrote: Freakling, for those of us heathens not in the know can you explain what the mining pathfinding problems are? Like what do the workers actually do behavior wise, any common elements that cause it, and how significant the effect is?
Thank you! http://www.panschk.de/mappage/newsscript/viewarticle.php?newsid=35There's a lot of information here; I think this should address some of your questions. That info is actually very outdated, as it does not explicitly take pathfinding regions into consideration. We can replace replace most of the "starring at workers for hours on end" in the map debugging process with "undetermined amounts of time spent on starring at pathfinding regions" now. The section about gas balancing still contains some useful bits, though.
|
Fascinating stuff Freakling, thanks!
Picture 2 I got nothing, and I'm not sure what triggers ultra vortex either, but in picture one if I read correctly it sounds like the pathfinding could be suspect for workers at the top couple patches, due to the small rectangle adjacent right of the resource depot?
|
United States9652 Posts
fixing dragoon AI would litearlly break protoss.
|
On February 23 2018 11:30 FlaShFTW wrote: fixing dragoon AI would litearlly break protoss. Finding out what people actually mean by "fixing Dragoon AI" is the real challenge here.
|
|
|
|