I finally found out what the pot behind van halen VH1 does (it wasn't a master at all)

[Amp_chaser]

It's crazy how all these people

Dave, Mike soldano, all these legendary amp gurus (who would stand to make shitloads of money if there WAS a mod) never have seen evidence of any gain mod in those amps

Good thing amp chaser is here to set them straight

I’m sorry, but unmodding an amp, depending on what was done, can be done in 5–10 minutes. But creating mods and finding values that work well together can take years. I know what I’m talking about, I mod and build amps the way Dave did back in the day, including drilling his own chassis holes, etc.

It could have been modded until 1979 and then brought back to stock. Mike Soldano and the others saw the amp 10 years later, or even much later.

I told you guys that when I release the video, there will be a passage where I quote what that famous person said, and you’ll be able to find the video online. He directly said what was inside Ed’s amp, but I think people assumed it was unmodded and, not understanding amp-modding terms, didn’t pay attention to what he said.

 

[DanTravis62]

I’m sorry, but unmodding an amp, depending on what was done, can be done in 5–10 minutes. But creating mods and finding values that work well together can take years. I know what I’m talking about, I mod and build amps the way Dave did back in the day, including drilling his own chassis holes, etc.

It could have been modded until 1979 and then brought back to stock. Mike Soldano and the others saw the amp 10 years later, or even much later.

I told you guys that when I release the video, there will be a passage where I quote what that famous person said, and you’ll be able to find the video online. He directly said what was inside Ed’s amp, but I think people assumed it was unmodded and, not understanding amp-modding terms, didn’t pay attention to what he said.

Yeah, we will be waiting with baited breath for your magic video that will set the internet on fire. Can't wait for more conjecture, opinions, magazine picture scans, and rumors condescendingly pretending to be facts, evidence, and circuits.

 

[JohnnyGtar]

I wonder what the results would be if several different AI thingamabobs were put to work deciphering Ed's 1st album tone.

 

[DanTravis62]

I wonder what the results would be if several different AI thingamabobs were put to work deciphering Ed's 1st album tone.

They would literally be scouring threads like this for their information do you you not understand how AI works?

It's information collation, not magic robots

 

[JohnnyGtar]

They would literally be scouring threads like this for their information do you you not understand how AI works?

It's information collation, not magic robots

Thanks for the straw man argument. Yes, I understand what it is. I also understand that AI has figured a lot of things out using the means at its disposal.

 

[LP Freak]

Yeah, we will be waiting with bated breath for your magic video that will set the internet on fire. Can't wait for more conjecture, opinions, magazine picture scans, and rumors condescendingly pretending to be facts, evidence, and circuits.

The Zodiac Modder strikes again….

 

[JohnnyGtar]

They would literally be scouring threads like this for their information do you you not understand how AI works?

It's information collation, not magic robots

AI does not work solely by information collation. AI models, such as those in Google Search, are trained on large amounts of information. However, they use pattern recognition and probabilistic modeling.

AI differs from simple collation in the following ways:

• Pattern Recognition: AI uses mathematical patterns to predict the next logical word or pixel. It creates responses instead of retrieving existing files.
• Inference and Synthesis: AI can create new information based on patterns in its training data.
• Deep Learning: Modern AI uses artificial neural networks to learn from experience. This allows it to make decisions based on complex factors.
• Probabilistic Output: AI sometimes generates incorrect information because it predicts what comes next.
  

 

[DanTravis62]

AI does not work solely by information collation. AI models, such as those in Google Search, are trained on large amounts of information. However, they use pattern recognition and probabilistic modeling.

AI differs from simple collation in the following ways:

• Pattern Recognition: AI uses mathematical patterns to predict the next logical word or pixel. It creates responses instead of retrieving existing files.
• Inference and Synthesis: AI can create new information based on patterns in its training data.
• Deep Learning: Modern AI uses artificial neural networks to learn from experience. This allows it to make decisions based on complex factors.
• Probabilistic Output: AI sometimes generates incorrect information because it predicts what comes next.

None of those are "listening to the shidd pant tone and figuring out what the mods are"

Unless I missed that in your copypasta?

 

[DanTravis62]

The irony is guess what @JohnnyGtar , you can use the same fucking copypasta you got from google to LITERALLY ASK IT what the mods are.

So go ahead, ask it. What does it say?

It turns up collated fucking search results. Like i just said.

 

[JohnnyGtar]

None of those are "listening to the shidd pant tone and figuring out what the mods are"

Unless I missed that in your copypasta?

Wow. There's a genius level reply.

 

[JohnnyGtar]

The irony is guess what @JohnnyGtar , you can use the same fucking copypasta you got from google to LITERALLY ASK IT what the mods are.

So go ahead, ask it. What does it say?

It turns up collated fucking search results. Like i just said.

Dude, please chill.

 

[JohnnyGtar]

None of those are "listening to the shidd pant tone and figuring out what the mods are"

Unless I missed that in your copypasta?

So what if it's copied? It's still accurate.

 

[DanTravis62]

Let's see

So what if it's copied? It's still accurate.

Did you even read it?


It's literally saying "no AI doesnt just collate data, it 1. organizes the collated data 2. analyzes the most relevant collated data 3. etc etc


I don't think you understand that just by copy/pasting google doesn't equal correct information, the copy/pasta actually makes my point more succinctly

 

[JohnnyGtar]

Let's see


Did you even read it?


It's literally saying "no AI doesnt just collate data, it 1. organizes the collated data 2. analyzes the most relevant collated data 3. etc etc


I don't think you understand that just by copy/pasting google doesn't equal correct information, the copy/pasta actually makes my point more succinctly

Yes, yes. You know more about how AI works than AI does.

My previous reply on AI's methods were from Google AI. Let's try other sources, shall we?

Here's Chat GPT's answer. Please feel free to correct it:

My question to Chat GPT: Does AI work solely by collation?

No—AI does not learn solely by collation.


Collation (collecting, organizing, and preprocessing data) is just an early step. It matters because the quality and structure of data affect what the model can learn, but it isn’t the learning mechanism itself.


The actual learning in systems from Machine Learning and Deep Learning comes from:

• Parameter adjustment: Models start with random internal settings and iteratively adjust them to reduce error. This is driven by optimization methods like gradient descent.
• Pattern extraction: They detect statistical regularities—correlations, sequences, structures—in the data.
• Abstraction: Especially in deep models, layers build increasingly abstract representations (e.g., edges → shapes → objects in images, or syntax → semantics in language).
• Feedback signals: Learning depends on signals such as labels (supervised learning), rewards (reinforcement learning), or structure in the data itself (unsupervised learning).

If AI learned only by collation, it would just be a database—retrieving and rearranging stored pieces. But modern AI systems can generate outputs they’ve never explicitly seen before because they’ve learned underlying patterns, not just stored examples.

 

[JohnnyGtar]

Next, let's ask Anthropic's Claude AI: Does AI work solely by collation?

No, modern AI—particularly large language models—does far more than collate or retrieve existing text.


What's actually happening:

1. Pattern learning, not storage. During training, the model processes massive amounts of text and learns statistical relationships between words, concepts, and structures. It doesn't store verbatim passages to look up later; it encodes patterns—how ideas relate, how language flows, what follows what.
2. Generation through prediction. When you ask a question, the model generates a response token by token, predicting what word is most likely (and useful) to come next given everything before it. This is a generative process, not a lookup.
3. Novel combinations. Because the model works from learned patterns rather than a database of pre-written answers, it can produce sentences, explanations, and solutions it has never seen verbatim. It can combine concepts in new ways, adapt tone to context, and reason through multi-step problems.

Where "collation" falls short as a description

• A collation system would need a pre-existing answer for every query—LLMs don't.
• Collation can't explain why models make mistakes that no source document contains, or why they can solve novel math problems, write original code, or compose poetry in a requested style.
• The same prompt can yield different outputs, which wouldn't happen if the model were simply retrieving fixed text.

Caveats:


LLMs are shaped by their training data, so they reflect patterns, knowledge, and biases present in that data. And they can sometimes reproduce memorized fragments, especially for famous quotes or very common text. But the core mechanism is generative pattern-matching, not retrieval or collation.

 

[S̷͖͑m̵͎͂á̵̺s̸͚̈́h̴̬̑]

Jesus christ, this thread delivers. We have evh conspiracy, a teasing amp modder with a single riff, and a boomer asking AI how AI works and insisting it will get us the brown sound we all desire. Perhaps the AI could give us a blowjob while it mods this amp for us.

 

[JohnnyGtar]

Microsoft Copilot AI: Does AI work solely by collation?
Short answer: No — AI does not work solely by collation. Collation (gathering and organizing data) is one ingredient, but modern AI systems depend on pattern‑learning, statistical modeling, and optimization, which go far beyond simply collecting or rearranging information.

What AI actually does (beyond collation)​

1. AI learns patterns from data, not just collects it. Machine learning systems analyze large datasets to identify statistical regularities and use those patterns to make predictions or generate outputs. This involves training algorithms, adjusting model parameters, and evaluating performance — all of which require more than collation.

2. AI uses complex models such as neural networks. Neural networks transform inputs through layers of mathematical operations, learning internal representations that are not explicitly programmed. This is fundamentally different from simply organizing or retrieving data.

3. AI systems often learn collaboratively or iteratively. Techniques like federated learning involve multiple devices training a shared model without sharing raw data. This process includes optimization, consensus-building, and model updating — again, not mere collation.

4. Data collection is only the first step. AI does gather data — actively, passively, or via third‑party sources — but the real power comes from machine learning, where algorithms refine themselves by analyzing that data.

Why people sometimes think AI is “just collation”​

Because AI systems (especially large language models) are trained on huge corpora of text, it can look like they’re simply remixing or stitching together what they’ve seen. But under the hood, they’re using statistical relationships learned during training to generate new outputs — not copying or collating text directly

A clearer distinction​

Process	What it means	Does AI do this?	Is it the whole story?
Collation	Gathering, sorting, and organizing data	Yes	No
Pattern learning	Extracting statistical structure from data	Yes	Core mechanism
Generalization	Applying learned patterns to new situations	Yes	Essential
Generation	Producing new outputs based on learned patterns	Yes	Especially in generative AI

Bottom line: Collation is necessary but not sufficient — modern AI’s power comes from learning statistical patterns, building internal representations, and generalizing to new situations.

​

 

[JohnnyGtar]

Jesus christ, this thread delivers. We have evh conspiracy, a teasing amp modder with a single riff, and a boomer asking AI how AI works and insisting it will get us the brown sound we all desire. Perhaps the AI could give us a blowjob while it mods this amp for us.

In no way did I insist on that. Anyway, I've backed up with 4 solid rebuttals. I'm happy to talk about Ed's tone. Does being a boomer disqualify my comments?

 

[JohnnyGtar]

Actually, it's Ed's food intake that affected his tone:

 

[S̷͖͑m̵͎͂á̵̺s̸͚̈́h̴̬̑]

In no way did I insist on that. Anyway, I've backed up with 4 solid rebuttals. I'm happy to talk about Ed's tone. Does being a boomer disqualify my comments?

Being a boomer qualifies all your comments about an obsolete tone.

Your comments on AI (not your age), confirms your lack of understanding of AI