Forum Replies Created
-
Posts
-
“Crafting Solver” Feedback
– First attempt took 23 secs to calculate and since I’m assuming I didn’t have the abilities to create 100% Quality, resulted in a successful creation, but at 1% Quality. I’d love to see an algorithm that prioritizes highest HQ with successful craft.
– Second attempt I tried something low level and I kept getting a “Step 1. Recommended skill: (Nothing)” error
– Third attempt I re-tried attempt #2 (took 6 secs) and was successfully able to get a HQ item. However due to its very low level I capped out on Quality very early and I spent half the time doing Observe, Tricks of the Trade. If it’s possible to recognize and stop at 100% Quality, it would optimize the experience
– Surprised that some of the very basic cross-class crafting skills were not available (such as… Steady Hand II, Byregot’s Blessing, Manipulation, Careful Synthesis II, and Comfort Zone). Maybe a set of checkbox of what cross-skills you currently have would be useful.
– Overall good first attempt, but still some work to do, before I’d trust it over my basic set of crafting macros.Very interesting. Could you please give us more details about what was happening in First and Second attempts?
The behaviour you describe is unexpected (unless it was at 51+, for which we have very little data so far).- What was your crafter’s level and what was the recipe’s level?
- A screenshot from the Monitor tab would be extremely helpful (for First and Second attempts).
- What was the exact sequence of actions in the First attempt? Miqobot always tries to maximize quality first, so how did it turn out to be 1% quality?
- Did you press Stop between First and Second attempts? Or did you change recipes on the fly?
- How long did it take Miqobot to solve the Second attempt? Was it the same 6 seconds as in Third attempt?
And of course, we’ll be definitely making improvements along the way. We are already working on changing strategy at 100% Quality.
As for adding more skills, the main issue here is that it takes a lot of time to adjust the mathematical model for each of them. For example, Comfort Zone will add a whole new dimension to solving algorithm, and it will result in x10 time required to solve (230 sec instead of 23 sec). Manipulation would make time x3, Steady Hand II would make time x5, and so on. We are working on optimizations to keep this reasonable, but it will take a while. And that’s why we are releasing early – to communicate and find the best solution together 🙂
Fantastic beta. Loving the “Gather by Name” feature and I’m putting the other beta features through a gambit of tests. Keep up the great work!
-
“Gather By Name” Feedback
– Needs an easier way to change priority (like drag/drop or up/down buttons)
– Enhancement: Would love there to be a pre-populated list that you could pick from Search Filter + (Gathering Type > Name) or Search + (Zone > Name)
– Enhancement: Zone Detection to help narrow down what you can gather in a specific zone
– Enhancement: A count of what’s been gathered during a sesssionThank you for the feedback!
We’ll look into your suggestions and see whether they can be pushed into the full version without delaying it much 🙂We also hope that adding a Gathering Presets list would help with quick population of Gather by Name list. And it would also allow sharing. But you’re right, a Search feature would be a great addition.
Yes, it will be possible with the release of Scenario Scripting.
If you want to know more details, please read our Gathering Development section. Under Advanced Scenarios, you can see that we have planned Aethersand and Favor farming.Indeed, we have overlooked F1-F12 keys when we implemented macro feature. At the moment, only letters and numbers supported. But this is an easy adjustment, so we are going to implement it for the upcoming version of Miqobot.
When using icons from hotbars, there’s a different algorithm involved which takes keys directly from your ingame keybinds. So this is just an issue with macro parser.
Thank you for this detailed explanation 🙂
This is well described in terms of feature request, so we can easily implement it right away.This is exactly what we needed. Thank you very much!
We have reproduced the issue, it indeed happens when you run the game in background, and when the mouse cursor passes over the game window (even if covered by another window), it interrupts any keypresses. These lines were the key to understanding this issue.
(8:46:40 AM) Sorry, couldn't move cursor with key '[{NUMPAD6}, 0x00]' :( (8:46:43 AM) Sorry, couldn't move cursor with key '[{NUMPAD6}, 0x00]' :(We will make adjustments to Miqobot’s AI, and this issue will be fixed in the upcoming version. Thank you for providing such a detailed report 🙂
Can you please confirm that Miqobot detects collectability rating correctly?
Collectability value appears in the Output Area each time Miqobot makes a decision whether to keep or discard the fish. Also, the Output Area is the first place we recommend to look for clues when something weird happens.It seems like a rare bug caused by selecting a wrong button in Yes-No window, and we’ll be making adjustments based on your feedback. If there’s any other valuable information in the Output Area, please tell us 🙂
We have achieved an almost stable version of a Crafting Solver, but it supports only a few skills so far. Even not in full glory, it can still do a lot better in some cases when compared to macro crafting. We are planning to release it as a beta feature in the upcoming version of Miqobot, so you will be able to try it out and provide feedback.
A full list of supported skills will be revealed in patch notes for Miqobot v1.2.4.
We have implemented the basic AI for Minigames and carried out several experiments. The results give an average of ~250 MGP per minute, but this is only for Out on a Limb. Finer Miner seems impossible to solve within 4 or 5 attempts, and with 6 attempts the reward is very low.
We still think about implementing both Minigames and let Miqobot switch between them to look less like a bot. Of course, it will be made available as a setting for you to decide.
Yes, there is a way.
You can use any text editor with JSON formatting. For example, this one: http://jsonviewer.stack.hu/
Racing waypoints are stored in a single JSON line. Copy and paste that line into the editor window and press Format. Make any adjustments you want, then press Remove White Space. Copy and paste the resulting line back into your presets file, and it’s ready. Import this into Miqobot as usual, and she would never tell the difference.I think we should post a guide about editing JSON in our Help section 🙂
All right 🙂
Mochi will design a list of sections, and we’ll try to optimize our database for them. Just one question before we do that.You wouldn’t get lost among many sections, would you?
I’m a little concerned, because we’ve seen other forums making 20+ sections and in the end 19 of them stayed empty most of the time. At the moment all discussion threads are grouped together, so it’s easy to see what’s going on. Do you think it will be the same if we add more sections?Also, if you could provide more information, we would be able to reproduce your environment closer and see whether we should adjust the underlying math.
- What stats does your chocobo have?
- What class do you participate in?
- Does your chocobo have any undistributed stats?
- Do you use one of the basic presets? Or have you recorded your own waypoints? If you have your own, would you please export them to text file and upload here?
- What is your average ping to server?
- What is your average FPS?
- Does walking in snake lines happen all the time? Or like 50% of the time?
- Does it make Miqobot fall into traps often?
- What place does Miqobot usually take in the end?
- If you disable rails and waypoints on 3D Radar, does Miqobot perform better?
Thank you 🙂
Let’s note that steering in Chocobo Racing involves a lot of randomness. A partial list of factors would be:
- Chocobo stats
- Weather
- Debuffs applied
- Network latency
- FPS
- CPU Load
Without cheats it’s impossible to follow waypoints with 100% precision. And Miqobot doesn’t use cheats, only basic controls. But she does her best trying to accommodate to your environment considering all the factors involved.
Walking in snake lines is inevitable.
So the main question would be: Do you hate Miqobot for doing that?
