hauhauhauhauha CHATGPT LIES ...ROPE? lmfao??? i always sucked on this porra!!!!

But well, others stuff are true 

It says the same thing about everyone lol.

"achieved numerous victories in tournaments and competitions"

Pffft!

Quote from: TheWalrus on April 04, 2023, 06:16 PM

Quote from: TheKomodo on April 03, 2023, 01:40 PM
I once played a 2 hour Fort match, Korydex shared the video recently.

Only other scheme I've ever played for that length of time is a Warmer lol, I remember doing Warmers all night with Americans back in 2000, especially all nighters with TheFudo.

I've never even played a 1 hour hysteria match, and ive played some drawn out matches again turtle players

Top 3 longest hysteria rounds from 2 TRL seasons (3318 replays):
1. 0:56:22
2. 0:54:08
3. 0:46:15

Quote from: Korydex on April 04, 2023, 08:05 PM

Quote from: TheWalrus on April 04, 2023, 06:16 PM

Quote from: TheKomodo on April 03, 2023, 01:40 PM
I once played a 2 hour Fort match, Korydex shared the video recently.

Only other scheme I've ever played for that length of time is a Warmer lol, I remember doing Warmers all night with Americans back in 2000, especially all nighters with TheFudo.

I've never even played a 1 hour hysteria match, and ive played some drawn out matches again turtle players

Top 3 longest hysteria rounds from 2 TRL seasons (3318 replays):
1. 0:56:22
2. 0:54:08
3. 0:46:15

SPW a snail player, 2 of top 3!

chat gpt said thewalrus was a heavily armed british worm, and a reference to the beatles (it is)

it refused to answer leading questions about whether thewalrus was the worst player the worms community has ever seen, or if thewalrus was a cancer to the tus-wa community.

maybe we will never know

Lmaoo, dissapointed.

Number 10, top bants. WWP, as if

I wanted to see how the current GPT4-powered ChatGPT and Gemini Pro-powered Google Bard would perform, and these are the results:

Chatgpt is obviously working with data that has been implemented in him thru the net. Now going from this, when you read the answer of chatgpt, its like you're listening to the opinion of Komito 70% of TUS posts were made by Komito, so the easiest and quickest material and source he was probably using up was from Komo's forum comments about the "best players"

Quote from: Triad on January 28, 2024, 07:39 PM
I wanted to see how the current GPT4-powered ChatGPT and Gemini Pro-powered Google Bard would perform, and these are the results:

[attachment=1]

[attachment=2]

Pretty amazing, its right on about some things and clueless about others.  I wonder where it gets some random shit though, like Snoop?!  I like snoop but lol.  And the part about Lordhound being known for strategy gametypes when hes a ttrr god.  Fun read though, and dead on about mablak. 

Quote from: Triad on January 28, 2024, 07:39 PM
I wanted to see how the current GPT4-powered ChatGPT and Gemini Pro-powered Google Bard would perform, and these are the results:

[attachment=1]

[attachment=2]

Too easy question. Now ask it who is the worst player or maybe something about schemes

Quote from: Korydex on January 28, 2024, 08:28 PM
Too easy question. Now ask it who is the worst player or maybe something about schemes

Asked both ChatGPT and Bard about Hysteria and Darts. Bard got both right (altho there is some slight hallucinations, like I never saw explosive barrels on Darts lol), while ChatGPT completely hallucinated.

ChatGPT's answers:




Bard's answers:




Both models have access to internet, but Bard seems to know better when is the best time to use it.

Quote from: Korydex on January 28, 2024, 08:28 PM
Too easy question. Now ask it who is the worst player or maybe something about schemes

or ask about godmax and football predictions

Quote from: Triad on January 28, 2024, 09:40 PM
Asked both ChatGPT and Bard about Hysteria and Darts. Bard got both right (altho there is some slight hallucinations, like I never saw explosive barrels on Darts lol), while ChatGPT completely hallucinated.

ChatGPT's answers:
[attachment=1]

[attachment=2]

Bard's answers:
[attachment=3]

[attachment=4]

Both models have access to internet, but Bard seems to know better when is the best time to use it.

wow, Bards answers are insanely good. although now that services with language models have Internet access, the task is straight-forward.
but LLM inside ChatGPT still hallucinates thought out moust of the details, only knowing the nicknames and basic information