soviras

Forum Replies Created

Viewing 15 posts - 61 through 75 (of 123 total)
  • Author
    Posts
  • in reply to: Endwalker Miqo Roadmap #33808
    soviras
    soviras
    Participant
    0

    With new update, how does it impact Miqo recovery? Of course, it’s not known right now, but question is meant for after patch 6.01 release.

    in reply to: Endwalker Scenario plans #33764
    soviras
    soviras
    Participant
    7+

    I’m planning to make a fully automated ocean fishing scenario that runs infinitely and buys items of choice. No plans for any 6.x specific stuff yet, but that may change.

    in reply to: Endwalker Miqo Roadmap #33745
    soviras
    soviras
    Participant
    16+

    Am I weird? I check every hour to see if update happened… I can’t wait for the return.

    in reply to: Multiple CPU core use #33522
    soviras
    soviras
    Participant
    0

    All settings is basically user error, honestly.

    And I’m in penta melded gear, literally BiS for crafters, so uh… Stats? I got ’em. :v

    Doubt. You can be penta melded, and still not have good stats, or far lower CP. Did you follow a guide for the melds, or just slap random ones on? Or did you calculate yourself exactly how much you need of craftmanship before it no longer saves steps, which changes based on recipe? How much CP? 554, like one of the guides that’s popular tells you? Which is great for normal crafts, but 660 CP is better for consistency with expert recipes, and thats a big difference in the amount of steps. More steps means more chances for beneficial RNG.
    Saying “literally BiS” is saying you don’t know what BiS is for crafting, because there is no single solution, BiS for one recipe is not BiS for another.

    The reason I prefer all settings is simple, I want optimal efficiency in terms of success rate, and everything specializes in that. I don’t want failures, and even if the failure rate is reduced by a small amount, I feel it’s worth it. Using a feature that the bot has, even if unneeded, isn’t an error, it’s a choice.

    in reply to: Multiple CPU core use #33515
    soviras
    soviras
    Participant
    0

    And I’m saying you must be doing something wrong, because if you have a 5800H and you’re getting “Almost an hour” and I’m sitting pretty here with my 2600X and getting at max two minutes per solver, there’s no other way to see it.

    If you only have 2 minutes, you don’t have the full settings, nor the full stats.

    in reply to: Multiple CPU core use #33512
    soviras
    soviras
    Participant
    0

    Some complex things, already take an hour.

    On this PC, there are 15 other cores available

    Considering this, I’d say threading was the least of your worries.

    In my puny Ryzen 5 2600X, it takes at most (if I really want to push it) 1-2mins on an expert recipe. So if you’re getting an HOUR on some random thing, then look inwards to fix the issue, cuz no one else has that issue from what I’ve seen.

    I have a Ryzen 7 5800h, the problem is not that it’s weak. It works perfectly fine on high end benchmarking. The problem is that it has 16 threads, and that makes it take a significant time, as only a fraction of its capability is used. When going for expert crafts, with all settings active, it takes an incredible amount of time, far more than it should, because the single thread, about 6% of the CPU, is capped. Even with less settings, it takes 3-4 times the estimated time. Stats also matter a lot. I get almost guaranteed maxed out expert recipes, and have max CP, with food active. More steps with more CP results in longer calculation. Max CP + food + syrup + specialist, and the estimated time goes to over 10 minutes. Once it’s done, it fails to max out about 1 in 10, and complete fails around 1-2%. Just takes ages to get there. It’s not some random thing, it’s expert recipes with stats that are good enough to cap the quality rather consistently, and even the estimated time is higher than 1-2 min, closer to 12 min.

    in reply to: more command dungeons #33476
    soviras
    soviras
    Participant
    0

    For my needs, it would be sufficient, even if not efficient. A lot of farming is unsynced, and for that, optimal efficiency is not needed, and so the ability to do this would provide a fast way to do this when most mechanics aren’t as important anymore.

    Want drops for glamour/desynth? Unsync is perfect.
    Want materials that drop? Unsync is perfect.
    Want light farm? Unsync is perfect.
    Want farm for minion, mount, barding or orchestrion? Unsync is perfect.
    All these things do not need to be synced, and simple AI would cover it when unsynced.
    For example, I already made scenario to do The Vault unsynced, and it clears even without dodging, even if performance is horrible. However, with basic AI, performance would improve a lot.

    Also you mention Trust dungeons not being possible with a simple system, because it does not do mechanics, and in some cases, that is true… But, important to note is that with some classes, is already possible to skip most mechanics, especially healer. Combat Assist and standing still let me clear all fights in the last dungeon as SCH (Not in Matoya relict though, there it failed at water boss). I just ignore mechanics, and clear it.

    in reply to: more command dungeons #33474
    soviras
    soviras
    Participant
    0

    Of course, additional tools for making custom combat scenarios are already planned.
    But they will not be able to perform dungeon mechanics until we manually implement each of them in Miqobot source code. Without additional algorithms, Miqobot will simply stand in acid puddles and AoE omens. So this feature will not allow you to farm any dungeon out of the box.

    You can find more information in this thread:
    <span class=”d4pbbc-underline” style=”text-decoration: underline;”>Open scenario scripting engine for squadrons/combat</span>

    Would it be possible to detect generic damage from an attack by name, or look for text in similar way to what ACT does, and then have it move to a different combat waypoint? It seems that may be shortcut, over implementing all mechanics specifically? In some cases, it may not be enough, but for many, it seems it could work?
    Like, if detecting an attack called “Acid Puddle”, move to waypoint at least X distance away. In scenario, that could be like dodgeField(Acid Puddle,2)?
    Or is there something that prevents miqo from detecting things like ACT detects?

    in reply to: more command dungeons #33463
    soviras
    soviras
    Participant
    0

    I would hope to get tools for making our own dungeon scenario at one point. I think, many basic attacks are detected, and most dungeons do not have unique special mechanics, so basic tools could let us make most dungeons into scenario, even if fail rate may be higher at times… It is a lot of work, so I prefer that developer focus on bigger features, and let user make rest.

    in reply to: Crafting Suggestion #33450
    soviras
    soviras
    Participant
    0

    That’s already possible with the scenario system.

    in reply to: Catnip Error #33323
    soviras
    soviras
    Participant
    0

    Same issue as here, so probably answer there also applies here.

    in reply to: Diadem Mining with boom #24200
    soviras
    soviras
    Participant
    1+

    You have 8 diferent sets for Diadem mining, which is the viable one?

    It’s a single scenario. It switches grids on it’s own.

    in reply to: Diadem Mining with boom #24165
    soviras
    soviras
    Participant
    0

    I have no issues with hi-cordials if I want to use them, so I can’t help with that… Did you save the gathering preset after ticking it? That’s my only guess.

    It should exit and reenter on it’s own as long as you don’t move the mouse with the window active (if it isn’t minimized it breaks the interactions due to how the game works) and the default keybinds are used. If you changed the keybinds, you will need to adjust them in chapter 3 so that it accepts the exit thing and interacts to enter gain properly.

    in reply to: (Phase 2) Diadem Full #23908
    soviras
    soviras
    Participant
    0

    What Scenario do I attach this to? I still get the blackout.

    (4:16:47 AM) Sorry, can’t move. I had a blackout and I’m still hiding in the corner!
    (4:16:48 AM) Blackout while travelling… Good night! <3
    (4:16:48 AM) (NOTE) Blackout switch is triggered.
    (4:16:48 AM) (NOTE) Please save the message log for analysis and restart Miqobot.

    That’s not for a scenario, it’s just a grid.
    I have a scenario + grids combo in the works for miner though, but it will be a bit before I’m satisfied with it running properly.

    in reply to: Grin Request Diadem Tier 3 #23901
    soviras
    soviras
    Participant
    5+

    While I haven’t had time to work on grids, I did spend a little bit in Diadem. Not only do we have new weather conditions to deal with, but the actual rotations for the nodes is actually reverse from Phase 2. Mining now is best counterclockwise and Botany is Clockwise. So it’s gonna take a lot more work to get things back in order.

    I’m working on a grid myself at the moment, which seems to be working mostly fine for mining already… But won’t make a botanist one til after the rankings… I should be able to release it tomorrow, if testing goes well.

Viewing 15 posts - 61 through 75 (of 123 total)