• You've discovered RedGuides 📕 an EverQuest multi-boxing community 🛡️🧙🗡️. We want you to play several EQ characters at once, come join us and say hello! 👋
  • IS THIS SITE UGLY? Change the look. To dismiss this notice, click the X --->

Discussion - ChatGPT my first look (1 Viewer)

I have been feeding chatgpt macroquest stuff during beta for gpt4 data analyst here and there most of the year, its pretty up to date. Bing is a separate gpt4 model from what is being run on openai directly and not as trained or as advanced as the model being used by openai in alot of areas the MS model is completely seperate.

In GPT4 you can send it the actual file and don't have to work with just sending it code from the prompt. It is particularly handy for python stuff as well it can even build and run its own scripts in python. It still can screw things up pretty good too. Another problem is as more people start using it for more complex tasks the model is starting to lag and timeout a lot. They just took dalle3 and the more advanced data analyst stuff out of beta that created a big uptick in use, and they have also been making big upgrades to handle the demand. They need more upgrades.

Google bard is just ok, its a long ways off from openai.

The real thing to remember is that this is just a superpowered search engine/scraper combination. Its not really AI that's a buzz word that stuck what this really is behind the Ai mask is a giant ever growing database with advanced search functionality. It like any software is no better than those who programmed it. It contains their flaws and bias. Microsoft also now owns 49% interest in openai, with a 10 billion dollar investment it aims to use the tech to make bing the superior search engine in the world.

Other things to consider are the future in that this tech will change a lot of industries, and how we even work with the internet. Think about its massive potential for monetization, I would speculate that we will see a lot of the more powerful models gated by paid use, likely out of the scope of the recreational user that we have now. It will likely also be flooded with ads for publicly "Free" available models in the near future just like we saw in the evolution of search engines since the majority adoption of the internet unfortunately we have seen that type of "blow them up with spam ads" monetization method in a lot of great tech over the years as they became more widely used, while they have to pay the bills some how, I am really not looking forward to when this gets bogged down with ads.

My QKI Lua is written with GPT4 help its pretty capable if you know what to ask it.

gpt4.JPG

Lots of cool things coming as well. The new ui with myGPT's, More Plugins and stuff just launched about a week ago or so its a little buggy still. Also in beta now is "GPT Builder" This allows you to create and adjust specifics to your needs it adds personalized characterization to the GPT4 model, and can be fed your things more specific to what you want to do with the model, take a look here:


builder des.JPG
create gpt beta 2.JPG
create gpt beta 3.JPG
 
Motojeff said:
I have been feeding chatgpt macroquest stuff during beta for gpt4 data analyst here and there most of the year, its pretty up to date. Bing is a separate gpt4 model from what is being run on openai directly and not as trained or as advanced as the model being used by openai in alot of areas the MS model is completely seperate.

In GPT4 you can send it the actual file and don't have to work with just sending it code from the prompt. It is particularly handy for python stuff as well it can even build and run its own scripts in python. It still can screw things up pretty good too. Another problem is as more people start using it for more complex tasks the model is starting to lag and timeout a lot. They just took dalle3 and the more advanced data analyst stuff out of beta that created a big uptick in use, and they have also been making big upgrades to handle the demand. They need more upgrades.

Google bard is just ok, its a long ways off from openai.

The real thing to remember is that this is just a superpowered search engine/scraper combination. Its not really AI that's a buzz word that stuck what this really is behind the Ai mask is a giant ever growing database with advanced search functionality. It like any software is no better than those who programmed it. It contains their flaws and bias. Microsoft also now owns 49% interest in openai, with a 10 billion dollar investment it aims to use the tech to make bing the superior search engine in the world.

Other things to consider are the future in that this tech will change a lot of industries, and how we even work with the internet. Think about its massive potential for monetization, I would speculate that we will see a lot of the more powerful models gated by paid use, likely out of the scope of the recreational user that we have now. It will likely also be flooded with ads for publicly "Free" available models in the near future just like we saw in the evolution of search engines since the majority adoption of the internet unfortunately we have seen that type of "blow them up with spam ads" monetization method in a lot of great tech over the years as they became more widely used, while they have to pay the bills some how, I am really not looking forward to when this gets bogged down with ads.

My QKI Lua is written with GPT4 help its pretty capable if you know what to ask it.

View attachment 53277

Lots of cool things coming as well. The new ui with myGPT's, More Plugins and stuff just launched about a week ago or so its a little buggy still. Also in beta now is "GPT Builder" This allows you to create and adjust specifics to your needs it adds personalized characterization to the GPT4 model, and can be fed your things more specific to what you want to do with the model, take a look here:


View attachment 53278
View attachment 53280
View attachment 53281
Click to expand...
Part of the lag your seeing on GPT4 is they are letting it "roam" free on the live internet now. So while it should be able to give more up to date answers, and research to a better degree, that retrieving and crunching is eating into its CPU cycles.

also 100% agree its not AI, its just newer google (IE prompts are just google searches with flair).
 
I asked ChatGPT 4.0 to write a script I am wanting to try and implement on a beastlord. Didn't give it too much info, just the very basic and this is what it spit out:


Lua: 
Write me a lua script where if my everquest character is below 80% mana I till target myself and cast Focused Paragon
ChatGPT
Creating a Lua script for your specific need in EverQuest involves a few steps. The script will check your character's mana percentage and, if it's below 80%, will target your character and cast the spell "Focused Paragon". Here's a basic structure for such a script:

lua
Copy code
-- Function to get the current mana percentage
function GetManaPercentage()
    local currentMana = -- Add command to get current mana
    local maxMana = -- Add command to get max mana
    return (currentMana / maxMana) * 100
end

-- Function to cast Focused Paragon
function CastFocusedParagon()
    -- Add command to target self
    -- Add command to cast Focused Paragon
end

-- Main script execution
function Main()
    local manaPercentage = GetManaPercentage()

    if manaPercentage < 80 then
        CastFocusedParagon()
    end
end

-- Run the main function
Main()
In this script:

GetManaPercentage: This function needs to be filled with the correct command to fetch the current and maximum mana of your character. These values are used to calculate the mana percentage.

CastFocusedParagon: Here, you should insert the commands to target yourself and cast "Focused Paragon". These commands depend on how EverQuest's scripting interface allows you to interact with your character and abilities.

Main: This function checks if the mana is below 80%. If it is, it calls CastFocusedParagon.

Please note that you'll need to fill in the specific EverQuest commands for getting mana values, targeting, and spell casting, as this varies based on the game's API and scripting capabilities.
 
I just used Microsoft Copilot to create a C# program for me to de-duplicate my bazaar mule price files. I don't know Lua well (at all!) but I still dabble in C# so I gave Copilot the following request (including a misspelled word!):

1710799757803.png

The code it spit out compiled (VS2022) and ran with 1 minor error that Copilot then fixed for me. I had to manually deal with copying and pasting but the code did what I asked! I'm impressed!

C# console program to de-duplicate BZR INI files.: 
using System;
using System.IO;
using System.Linq;
using System.Collections.Generic;

class Program
{
    static void Main()
    {
        var files = new string[] {"file1.txt", "file2.txt", "file3.txt", "file4.txt", "file5.txt", "file6.txt"};
        var dict = new Dictionary<string, int>();

        foreach (var file in files)
        {
            var lines = File.ReadLines(file);
            foreach (var line in lines)
            {
                if (line.Contains('='))
                {
                    var parts = line.Split('=');
                    var key = parts[0];
                    var value = int.Parse(parts[1]);

                    if (dict.ContainsKey(key))
                    {
                        if (dict[key] < value)
                        {
                            dict[key] = value;
                        }
                    }
                    else
                    {
                        dict[key] = value;
                    }
                }
            }
        }

        using (var writer = new StreamWriter("output.txt"))
        {
            foreach (var pair in dict)
            {
                writer.WriteLine($"{pair.Key}={pair.Value}");
            }
        }

        Console.WriteLine("Done!");
    }
}
 
You must log in or register to reply here.
Discussion - ChatGPT my first look
Next unread thread 🦁

Users who are viewing this thread

Total: 2 (members: 0, guests: 2)
Back
Top