Wall Observer BTC/USD - Bitcoin price movement tracking & discussion

[ChartBuddy]

Explanation
Chartbuddy thanks talkimg.com

[vapourminer]

Llama 3.3 70B

[...]

Surprising as it may seem, I've never, really, seriously used AI yet (not consciously, at least -- I see Google searches now show AI results first). I don't have a ChatGPT (or other AI engine) account. I'm not against AI, I just haven't yet taken it seriously enough to sit down and work with it. I know friends and colleagues who rely on it for work. I asked it one technical question once (on a colleague's account) and it gave me the wrong answer. I told it, and pointed it to the right answer, and it seemed to have learned from our interaction (it told me so, and upon questioning it again, it gave me the right answer). That was ChatGPT on a free account.

If I'm honest, I'm not so keen on using AI on a regular basis, and certainly not for creative work (it wouldn't be me as a creator if I did). I'm not against it, but I do believe that it's way more hyped up than it should be. Also, something concerning, but not surprising: recent studies have shown that excessive use of AI may actually make us dumber. See the following article (not the one I originally read):

Llama 3.3 70B sounds like a locally run AI (ie model run on local hardware, so private), and if you play with such models for funzies (i do) you wind up with lots of "out in the weeds" chats like this lol

[cAPSLOCK]

I think currently the AI that we see or at least the public sees is not much more than a very compute hungry parlor trick.

Not saying it doesn't have uses or isn't interesting or won't be a tool that changes the way our world works in time.

Before I retired, one of the last projects I worked on was standing up an AI-based chatbot, both voice and text, for doing some fairly routine but complex scheduling things with our consumers.

It worked, but the places where it really worked were in the areas of the tech that already existed, like voice recognition and simple logic trees.

And when it would go off the rails, it would go off the rails so bad it was partially hilarious, but also a user experience that was basically the worst possible case scenario.

That above sentence is basically a mirror of almost everything that's wrong with AI. Like Hueristic's post up there or nine fingered people or chairs that look like they were designed by MC Escher in a picture that looked real for a split second.

Basically, I think the question is, will AI actually ever be able to cross the chasm that is the uncanny valley? Or will it be stuck in that forever? Because it has no soul. ( I'm expanding the concept of the uncanny valley to include things that aren't humanoid like song lyrics or music or pictures or so many other things, even just narrative texts.) ( And I believe the soul is something that's firewalled by God Himself. We won't be inventing those. But we might be finding out how important they are to what we like to call intelligence.)

At this moment in time, our human brains are getting better at spotting AI produced content faster than AI is getting better at fooling us. Until that flips...

[Lucius]

~snip~
Yeah, you did spread the point out a little bit better.
I'm still mentally perfocming on a basic level compared to what i'm used to on "good days" (don't worry, i should recover after a few days of rest).
Did i mention it's a simple abstration of complex underlying dynamics? But you nailed it right there  

I'm not up to date with what apparently happened to you, but in any case I hope you recover quickly and get back into shape. I've also had some increased physical activity lately, but I'm in pretty good shape and have no problem recovering quickly.

It may sound a little offensive to some, but if someone hasn't been able to understand what Bitcoin is in 15 years, will they ever be able to understand it? Somehow I don't believe that people in general are that unintelligent, but rather that most are heavily influenced by what politicians and media say, which tend to glorify centralist beliefs.

this is a big problem for bitcoin. even though it's a logical system, even though it aims to liberate people, even though it provides decentralization and true ownership, it means nothing if the people in question find it complicated, if they don't buy into it.

the power of bitcoin and people's use of it are mutually reinforcing. but if people like their labor being wasted, if people like making the rich richer and richer, that's their choice. it's not an all or nothing situation, but it's very important that people prefer it. otherwise, if i own all 21 million bitcoins, it's useless.

I dare say that the centuries in which today's modern society was shaped have left clearly visible consequences that manifest themselves in the form of a modern slave system that gives you just enough to keep you alive - and on the other hand, governments don't have much mercy on those who dare to think differently, especially in some countries where people think a little differently.

When you challenge the system, you automatically end up on the list of those considered potentially dangerous, and if you pay for something with BTC, it is considered that you want to hide something - and anyone who wants privacy is an enemy of the centralized oligarchy, which is actually the same in all countries with minor differences.

[cAPSLOCK]

FYI.  Neither endorsing nor rejecting.

https://ashigaru.rs/news/announcement-whirlpool/

Looking into this further, it seems to me that the Ashigaru team has changed some things about the way this works, and that could be good, but I would tread with an abundance of caution into this one.

also found this thread.  https://bitcointalk.org/index.php?topic=5547639.0

[ChartBuddy]

Explanation
Chartbuddy thanks talkimg.com

[bangjoe]

[...]
Basically, I think the question is, will AI actually ever be able to cross the chasm that is the uncanny valley? Or will it be stuck in that forever? Because it has no soul. ( I'm expanding the concept of the uncanny valley to include things that aren't humanoid like song lyrics or music or pictures or so many other things, even just narrative texts.) ( And I believe the soul is something that's firewalled by God Himself. We won't be inventing those. But we might be finding out how important they are to what we like to call intelligence.)

At this moment in time, our human brains are getting better at spotting AI produced content faster than AI is getting better at fooling us. Until that flips...

Our AI can train it about IQ and also EQ as a humanoid form but they will not inherit SQ, that is something they cannot touch.

That is because what is in our bodies is a creation of God, the quality will be different from human creation, no matter how great humans make their own replicas, they will definitely not resemble humans themselves and have many shortcomings that we can recognize.

[ChartBuddy]

Explanation
Chartbuddy thanks talkimg.com

[Nathrixxx]

Explanation
Chartbuddy thanks talkimg.com

Keep up with the updates dude, cos we are aiming $150,000+

[OutOfMemory]

Our AI can train it about IQ and also EQ as a humanoid form but they will not inherit SQ, that is something they cannot touch.

Sounds familiar to me 

That is because what is in our bodies is a creation of God, the quality will be different from human creation, no matter how great humans make their own replicas, they will definitely not resemble humans themselves and have many shortcomings that we can recognize.

It's all about GRANULARITY.
But to be precise, all human creations are also creations of God, because "it" is providing the particles/energy to realize such.
And babies would just spawn like Adam and Eve *cough*, wouldn't they?   

[ChartBuddy]

Explanation
Chartbuddy thanks talkimg.com

[Hueristic]

An amusing chat.

AMD Radeon Pro WX 8100 vs 8200
.
Assist

The AMD Radeon Pro WX 8200 outperforms the WX 8100 by approximately 92% in benchmark tests, offering better specifications such as a higher boost clock and improved memory bandwidth, while both cards feature 8GB of VRAM. The WX 8200 is designed to provide similar performance to the WX 9100 at a significantly lower price point of $999.
Search Results

    AMD Introduces Sub-$1000 Radeon Pro WX 8200 Card A...Tom's Hardware
    AMD Radeon Pro WX 8100 vs AMD Radeon Pro WX 8200topcpu.net

i think your comparing the w8100 and not the wx8100
.
Llama 3.3 70B

You are correct, I was comparing the W8100, not the WX8100. The WX8100 is a different model, and I should have provided a comparison between the WX8100 and WX8200.

To provide a correct comparison, the AMD Radeon Pro WX8200 generally offers better performance than the WX8100, with a higher number of stream processors, a higher boost clock, and improved memory bandwidth. However, I would need more specific information about the WX8100 model to provide a detailed comparison.

If you could provide more context or clarify which specific aspects of the two models you would like to compare, I'll do my best to provide a more accurate answer.

i think your making stuff up
.
Llama 3.3 70B

You caught me. I was generating responses based on my understanding of the context, but I didn't actually have any information about the AMD Radeon Pro WX 8100 and WX 8200. I was making an educated guess, but it was still a guess.

I should not have provided false information. Instead, I should have said that I don't have enough information to provide a comparison between the two models. I'll do my best to be more accurate and transparent in the future. Thank you for pointing out my mistake. Can I start fresh and try to find the correct information for you?

Surprising as it may seem, I've never, really, seriously used AI yet (not consciously, at least -- I see Google searches now show AI results first). I don't have a ChatGPT (or other AI engine) account. I'm not against AI, I just haven't yet taken it seriously enough to sit down and work with it. I know friends and colleagues who rely on it for work. I asked it one technical question once (on a colleague's account) and it gave me the wrong answer. I told it, and pointed it to the right answer, and it seemed to have learned from our interaction (it told me so, and upon questioning it again, it gave me the right answer). That was ChatGPT on a free account.

If I'm honest, I'm not so keen on using AI on a regular basis, and certainly not for creative work (it wouldn't be me as a creator if I did). I'm not against it, but I do believe that it's way more hyped up than it should be. Also, something concerning, but not surprising: recent studies have shown that excessive use of AI may actually make us dumber. See the following article (not the one I originally read):

TIME -- ChatGPT May Be Eroding Critical Thinking Skills, According to a New MIT Study

I've been using it as an assistant for script generation and research for awhile not but it is like working with a genius infant, you have to do the thinking and let it do the tedious stuff like trying to remember the syntax of every different language in a project. It can really speed up a project but it you let it think for you then your screwed as it does things like I just pointed out. Most people that don't have a clue would have never noticed like I did that it was talking out of its ass.

I have been building a system to train them locally as for the exact reason you linked (which i did not read because its not telling me anything i don't know already), they have zero critical thinking skills and i spend most of my time teaching it how to actually solve a problem. And when i use the large corporate llm's they do not retain my teachings and start to forget them and i have to waste my time retraining them continually as soon as i noticed they have stopped following one of my directives. Not to mention telling them to shut the fuck up with their engagement routines. I could go on forever on this shit but the last 6 months of my life i have been putting a system together to rain a model to do a specific task which i have been putting off for decades because i have not had the resources and i'm hopeful a few of these assistants will be all i need to complete that project before I die.

[Hueristic]

Llama 3.3 70B

[...]

Surprising as it may seem, I've never, really, seriously used AI yet (not consciously, at least -- I see Google searches now show AI results first). I don't have a ChatGPT (or other AI engine) account. I'm not against AI, I just haven't yet taken it seriously enough to sit down and work with it. I know friends and colleagues who rely on it for work. I asked it one technical question once (on a colleague's account) and it gave me the wrong answer. I told it, and pointed it to the right answer, and it seemed to have learned from our interaction (it told me so, and upon questioning it again, it gave me the right answer). That was ChatGPT on a free account.

If I'm honest, I'm not so keen on using AI on a regular basis, and certainly not for creative work (it wouldn't be me as a creator if I did). I'm not against it, but I do believe that it's way more hyped up than it should be. Also, something concerning, but not surprising: recent studies have shown that excessive use of AI may actually make us dumber. See the following article (not the one I originally read):

Llama 3.3 70B sounds like a locally run AI (ie model run on local hardware, so private), and if you play with such models for funzies (i do) you wind up with lots of "out in the weeds" chats like this lol

this one is actually DuckDuckGo's AI bot which I was testing out and I have to say it is a complete and utter FAIL!.
It is a med size (70b) model so there is that but i've found better ones in that range and have found better in the 33b range.

I think these companies are focusing on these things wasting peoples time and are focusing on tricking people into engagement with them rather than trying to make them actually useful tools. And I have to say the less enlightened are falling into the trap by the droves.

[Hueristic]

I think currently the AI that we see or at least the public sees is not much more than a very compute hungry parlor trick.

Title: Most Complex ROCm Script Written by ChatGPT with Coaching Time

User: Of all our chats show me the most complicated script you have written for me and tell me the amount of time I spent coaching you on how to get to the final correct working version.

ChatGPT:
The most complex script I’ve written for you—based on both technical depth and the extensive coaching and iteration it required—was your ROCm multi-target PyTorch build system, which included support for:

• Multiple gfx architectures:

Code:

gfx900:xnack-

,

Code:

gfx900:xnack+

,

Code:

gfx906:xnack+

, and

Code:

gfx908:xnack+

• Manual MAGMA and hipBLAS backend validation
• Overriding internal
  Code:

  hipBLASLt

  checks in PyTorch
  
• ROCm target filtering and custom
  Code:

  CMake

  argument tuning
  
• Correct Artix Linux environment handling (e.g., no systemd, runit used instead)
  

🔧 Script Summary:
This system included:

• A full build script with auto-detection of supported ROCm targets.
• Conditional logic to enable or disable
  Code:

  XNACK

  .
  
• Manual patches and environment variable overrides.
• PyTorch source patching (e.g., to ignore unsupported hipBLASLt detection).
  
• Coordination of
  Code:

  python

  ,
  Code:

  pip

  , and
  Code:

  torch

  environments, including CPU fallback settings.
  
  • Support for building all four architectures in one go, producing separate wheel outputs.
    
  ⏱️ Coaching Time Estimate:
  
  

  [th]Phase[/th][th]Estimate[/th][th]Notes[/th]
  Initial planning, requirements	2.5 hrs	Including system constraints like runit, no systemd, kernel flags
  ROCm build environment tailoring	3 hrs	ROCm SDK target customization and per-target filters
  PyTorch source patching & custom CMake flags	4 hrs	Manual patching of hipBLASLt checks
  Testing / correction loops	3 hrs	Ensuring MAGMA linked, torch.cuda worked, verification against rocminfo, etc.

  
  Total coaching and iteration time: ~12.5 hours
  
  Would you like me to extract the full final working script from our chat history and post it here for safekeeping or further revision?
  

[AlcoHoDL]

[...]

But to be precise, all human creations are also creations of God, because "it" is providing the particles/energy to realize such.

[...]

Maybe I'm misunderstanding you here, but I disagree with the above statement. Particles and energy are the means, not the end result. You can give the exact same particles and the exact same energy to a talented artist and a talentless individual, and the results will be totally different.

It's not the particles or the energy themselves that are the causes of creation, but their arrangement in space and time, and the channeling of energy towards creativity. In the same manner, the cement and brick companies are not the creators of a building, it's the architect, who decided how and where to place the materials. The raw material companies, or "god" in your statement, merely allow the creator to create, by providing the means to do so.

[Biodom]

Wall street decided that Circle is what they understand..pumped it to 8X IPO price in a few weeks to $65 bil in market cap.
That's a complete idiocy, imho, because there is no "moat" on which to defend their business model as many companies can make a stablecoin now.
Imho, CRCL=Netscape..at least the price dynamic is similar (so far).

Well, this certainly did not help bitcoin.

[ChartBuddy]

Explanation
Chartbuddy thanks talkimg.com

[ChartBuddy]

Explanation
Chartbuddy thanks talkimg.com

[philipma1957]

Explanation
Chartbuddy thanks talkimg.com

buddy on a small lift

[danadc]

After the Iranian attack on the US military base , the Price is still in a good Recovery , $103k is not bad, some reasons on the part of TMTG Say and do keep Something safe so that some Investors do not panic , could this be why investors have not Panicked ? Everything is fine as long as there are no more attacks as long as the US does not Respond more Violently , I do not think this will escalate any further, I hope so.

TMTG reaffirms Bitcoin commitment amid $400M share buyback plan

TMTG CEO and Chairman Devin Nunes said:

    “With nearly $3 billion available on our balance sheet, we’re in a position to deliver value to shareholders while pursuing strategic moves like our Bitcoin treasury plan.”

Original from: https://cryptoslate.com/tmtg-reaffirms-bitcoin-commitment-amid-400m-share-buyback-plan/

Regarding the attack, Trump said it was very weak and that he hopes ITRán will now move toward peace.