I plugged the same question into Bing ChatGPT 4, and got this response:
AI in Real Use
- Thread starter GravitySucks
- Start date
Fermion
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
Try this one. I did not get a good answer. It seems totally wrong but maybe my initial conditions were not defined. I am sort of thinking also that ChatGPT is wrong and a traditional firearm cartridge DOES work on the moon even though it says it doesn't. Gunpowder carries its own oxidizer.
"If I fired a 0.220 swift cartridge on the moon at an angle of 10 degrees from the horizon, how far would it go?"
I got "Calculating the trajectory of a projectile fired on the Moon involves considering factors such as initial velocity, launch angle, and the Moon's gravitational field. However, it's important to note that using conventional firearms or cartridges on the Moon is not currently feasible due to the lack of atmosphere and the absence of oxygen, which is essential for combustion in traditional firearms.
Assuming hypothetical conditions where a conventional firearm could operate on the Moon, we can use basic projectile motion equations. However, keep in mind that these calculations would not accurately represent real-world conditions on the Moon.
*blah blah formula blah*
So, under these hypothetical conditions, the projectile fired from a 0.220 Swift cartridge on the Moon at an angle of 10 degrees would travel approximately 0.0102 meters (or 1.02 centimeters) before hitting the lunar surface. Keep in mind that these calculations are based on simplified assumptions and do not consider real-world limitations or the absence of an atmosphere on the Moon.
"If I fired a 0.220 swift cartridge on the moon at an angle of 10 degrees from the horizon, how far would it go?"
I got "Calculating the trajectory of a projectile fired on the Moon involves considering factors such as initial velocity, launch angle, and the Moon's gravitational field. However, it's important to note that using conventional firearms or cartridges on the Moon is not currently feasible due to the lack of atmosphere and the absence of oxygen, which is essential for combustion in traditional firearms.
Assuming hypothetical conditions where a conventional firearm could operate on the Moon, we can use basic projectile motion equations. However, keep in mind that these calculations would not accurately represent real-world conditions on the Moon.
*blah blah formula blah*
So, under these hypothetical conditions, the projectile fired from a 0.220 Swift cartridge on the Moon at an angle of 10 degrees would travel approximately 0.0102 meters (or 1.02 centimeters) before hitting the lunar surface. Keep in mind that these calculations are based on simplified assumptions and do not consider real-world limitations or the absence of an atmosphere on the Moon.
I entered that and got this:
Attachments
Well, I asked it about the resistance of the 50 ft osmium cylinder, and it came back with:
The resistance of a 50 foot long cylinder of osmium with a diameter of 0.081 inches is 0.0114 ohms.
ERD50
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
That's impressive that it breaks down the language and gets right to an answer.
But I was curious how this compares to good old Google search. So I entered that exact phrase, and I generally got useful links to tables, calculators etc. So not a spoon-fed answer, but sure gets you close.
But more interesting, was the links to discussions on forums or things like reddit, quora, etc.
OK, Smarty-ChatGPT - who said it was 50/60 Hz AC, or DC, or 2.5 GHz? Some humans had tore the question apart when asked in one of those forums. State all the conditions (frequency, capacitance, etc), don't make assumptions.
Heck, it would be easy to assume that for the above question that the intent was to get the voltage drop in a house circuit, which would be double for the round-trip. I know that wasn't the question, but maybe the round trip number should have been mentioned?
So again, if the person asking doesn't really know how to figure this out, they may not know what to ask to get the right answer.
Hey, this is a little like an oft-repeated phrase here - once you know enough to evaluate a for-hire Financial Advisor, you know enough to do it yourself (for the investment aspect at least).
Again, I'm not knocking on ChatGPT, I'm super-impressed, and it will be super useful for many things. Like any tool, the limitations and application suitability must be understood by the user, or Garbage in - Garbage out.
An old-time poster on a technical forum I follow often said something like "A question well asked, will often provide the answer." IOW, if you put enough thought into everything behind the question, and how to state it clearly and take all the factors into consideration, you just might come to the answer on your own. Or at least, really help someone give you a clear, correct answer.
I think I also mentioned earlier, when people talked about AI being able to write code - I said that you need to give it precise definition of what you want the code to do. Often, that's the hard part, and the coding becomes fairly straightforward. So people will still need to define the conditions in detail, or GIGO.
-ERD50
