Do you sometimes dislike the way technology is headed?

[Issac Zachary]

42 minutes ago, htimsenyawed said:

I still eyeball the Sennheiser with a wire for sitting at my desk, I loved my earbuds and was blown away by how good they were compared to what I had used before, and one day when I can spend big money on headphones I might get me another pair. But if you want to do a decent upgrade, here is one, two and three Sennheiser options that won't break the bank and will still be noticeably better than the Skullcandy.

I wonder if I were going to break the bank a little if these would be worth it: Amazon.com: 7Hz x Crinacle Salnotes Dioko HiFi 14.6mm Planar Diaphragm Driver in Ear Earphone IEMs with CNC Aviation-Grade Aluminum Case, Detachable 4-Core Litz OCC Copper Cable for Audiophile Musician Studio : Electronics.

2 minutes ago, Donut417 said:

I disagree. My Sennheiser HD4.40 sounds a shit load better even on Bluetooth than my Corsair Void gaming headset.

But do they cost the same?

 

I did say that the wireless $200 Beats Powerbeats Pros sound slightly better than the wired $10 Skullcandys I have.

[Donut417]

1 minute ago, Issac Zachary said:

But do they cost the same?

The Void's were $63 when I bought them a couple years back. The Sennheiser HD4.40 were about $80 USD when I bought them 5 or so years ago. So not much of a difference.

[Issac Zachary]

4 hours ago, htimsenyawed said:

Cost isn't indictivate of quality.

Not always, but I still think there is a certain amount truth to the phrase "you get what you pay for."

 

There's also a difference in tastes that can make one person hate one product and love another while someone else has the complete opposite opinion of those products.

 

Seeing Linus love the Drop Panda's makes me want to buy them (although I'm not sure if they are still available and they cost like $400):

Drop + THX Panda Wireless Headphone | Bluetooth Headphones Audiophile

This does bring up another problem with tech, and that's not being able to try things. How can I tell if the Drop THX Pandas would sound good to me? How can I tell if they'd sound $400 good to me? How can I tell if I'd like or dislike Airpods? What about if I'd like or dislike using an iPhone? I hate the phone I have now and wish I would have known more about it before buying it, but there was no 1 week trial period that I was aware of.

[Issac Zachary]

41 minutes ago, htimsenyawed said:

It's true that you get what you pay for, but some branding does push prices higher than they need to be - like those Corsair headphones the other user was talking about. Paying for the name Corsair and that it's 'gaming' orientated.

Agreed. That also explains part of it.

42 minutes ago, htimsenyawed said:

but the science behind a wire being better than Bluetooth is not one of those.

I could be wrong, but it seemed that before Bluetooth 5 there was a noticeable bandwidth limit, which of course affects quality. Generally speaking wires (or fibers) have always been able to transfer more data when comparing similar generations and price points of technology.

 

Of course digital also has the ability to be completely lossless. Analog is virually impossible to be completely lossless. They never made a phone with an XLR connector that I'm aware of.

 

45 minutes ago, htimsenyawed said:

As for iPhone - you can either go by an Apple store or a wireless retail store (which will also have more than just iPhone set up) for you to mess around with and see if you like it short term.

When you live in a small town like me there are no wireless retail stores for hundreds of miles. The next trip I have to the big city won't afford me any time to go look at phones or other items.

 

Which is odd when I remember back when there was a Radio Shack here in my little town and you could get all sorts of electronics right at a local store that's within walking distance from my house. Now I order capacitors on Digikey and sometimes end up with the wrong ones.

[Donut417]

2 hours ago, htimsenyawed said:

- like those Corsair headphones the other user was talking about. Paying for the name Corsair and that it's 'gaming' orientated.

That's not why I bought them. I needed a headset that worked. See Bluetooth and Windows is not always a good experience otherwise Id just use the Bluetooth headset that I already had. My friend recommended these as he has the wireless version.

[Issac Zachary]

1 minute ago, Donut417 said:

That's not why I bought them.

Ok, I'm being nit-picky. But the reason why we buy products doesn't always have to coincide with the reasons given when they're advertised.

 

Just because I buy a Toyota Corolla GR because I simply prefer a stick shift with zero interest in driving it fast like a sports car doesn't mean it isn't advertized as a fast sports car. But just because all I want is a Corolla with a manual transmission doesn't make it cost $22,000 instead of $37,000.

[smcoakley]

On 12/9/2023 at 8:19 AM, htimsenyawed said:

A wire will always sound better than Bluetooth.

Ehh... kinda... it's more complicated than that. It's about what bottlenecks are harming the sound quality, and artificial filters improving it.

 

A wire will do very little to degrade sound quality (unless the quality of your wire sucks), whereas Bluetooth can and will degrade sound quality, anywhere from "hardly perceptible" to "very obvious".

 

The earphone drivers have a big impact on sound quality, regardless of how audio got to them. So for this reason, wired earphones with crappy drivers will probably sound way worse than Bluetooth earphones with "average" Bluetooth codec support but very high quality drivers.

 

Additionally, sound quality isn't a fully objective thing, there's also a degree to where different EQs can sound more pleasing to different people. So digital earphones that have onboard processing can add additional filters to the audio to make it seem better or more pleasing to the listener (even though strictly you can't restore good quality audio to a degraded audio source). It just so happens that Bluetooth earphones have a battery and onboard processors anyway and are thus more likely to support such additional audio processing, while processors are unnecessary on analog wired earphones and probably not going to be added.

 

So yeah, generally, expensive wired earphones will usually sound better than expensive Bluetooth earphones. I can agree to that generalization. And cheap wired earphones very often sound better than cheap Bluetooth ones. But these are generalizations.

On 12/9/2023 at 3:28 AM, Issac Zachary said:

The biggest pain: The batteries in the Beats

This seems to be a problem with that specific product or maybe your specific unit. I have multiple Bluetooth earbuds that easily give me 12+ hours of listening time.

 

On 12/9/2023 at 3:28 AM, Issac Zachary said:

What was more of a pain was when someone talked to me as I'm used to pulling an earphone out and letting it hang off where I have it tied around the top button of my shirt. But with the Beats I would pull one out and then wasn't sure where to put it. Again, maybe they just need some getting used to.

Some of the nicer or newer earbuds have transparency modes -- basically the opposite of active noise cancellation. Just tap on the bud and it will pipe in audio from the outside world into your ear so that you can talk to people or whatever. Then tap again to go back to isolation.

[smcoakley]

On 12/9/2023 at 3:29 PM, Issac Zachary said:

Of course digital also has the ability to be completely lossless. Analog is virually impossible to be completely lossless. They never made a phone with an XLR connector that I'm aware of.

Technically yeah, analog is lossy to a very small degree. But this is irrelevant, because if your drivers are analog (they are), then you have to use a DAC somewhere inside your earphone to drive the drivers. So you can keep everything lossless digital as far as possible, but when it gets to the driver you have to convert back to lossy analog anyway.

[Polderviking]

The thing that worries me most is AI and what it will mean for the work force. Specifically the low end of the ladder. Jobs that would be fairly easy to automate away in the next ten, fifteen, twenty or so years.
But in hindsight, that is not even necesarily a problem I have with AI itself, as much as a problem with the lack of capacity of gouvernments to deal with the fallout in a timely manner.

 

People like to say there are labour shortages everywhere and subsequently stop talking about it, but that is one economic mood swing away of drasticly changeing and then when the caca hits the fan everybody is going pretend nobody saw it coming.

[Kisai]

On 12/13/2023 at 5:01 AM, Polderviking said:

The thing that worries me most is AI and what it will mean for the work force. Specifically the low end of the ladder. Jobs that would be fairly easy to automate away in the next ten, fifteen, twenty or so years.
But in hindsight, that is not even necesarily a problem I have with AI itself, as much as a problem with the lack of capacity of gouvernments to deal with the fallout in a timely manner.

 

People like to say there are labour shortages everywhere and subsequently stop talking about it, but that is one economic mood swing away of drasticly changeing and then when the caca hits the fan everybody is going pretend nobody saw it coming.

AI's impact is very over-stated.

 

In creative industries (writing, artwork, music, film, television, video games, food), generative AI is utter garbage, and will not replace anyone. Anyone who believes they can cut the artist out of the creation of a creative work, and have something people will want to buy, at any price, is a fool. Generative AI has not "learned" to be creative, at best it has achieved the "infinite monkey theorem". Without being prompted, it does nothing. When prompted, it just auto-completes using a statistical model of what the next letter, word, music, note, etc is. 

 

We are not anywhere close to an AI that actually knows, or can infer an answer to any question. You know the classic trope of "book smarts, street smarts?" The AI achieves a high score on book smarts, because it has effectively memorized things, but has zero in the street smarts, and when presented with any question that requires it to solve based on experience, it will always fail. It can not learn from experience, only other people's instructions. If those instructions are wrong because statistically people get it wrong, or people commonly repeat misinformation, then yes, the AI is going to learn wrong information.

 

This is why GPT sucks at math. It doesn't understand a math question, because it can not statistically memorize the answer to every math question. That's not how math works. That's also why you have to be careful and double-check the results. It's not good enough to have a 7% on report card any more than a 86%. Given a properly formatted math question, it should only have one, correct answer.

 

Generative AI also can't generate anything that can be copyrighted, because the training materials may contain copyrighted sources. So if you use Generative AI to write a page of a novel, there is a 100% chance it will plagiarize entire sentences. A few days ago, when the Mistral LLM was published, I went and tried an AI accuracy battle website (I forget what it's called) and I asked it to describe "climate change" and the entire opening paragraph was more than 50% identical, word-for-word. Two different LLM's. That tells me both models ingested the same source. Were they both correct? Yes. But the format of the output was different, one wrote it like you would write an essay for school, and the other one wrote it like a Quora answer in bulleted points.

 

I don't see any possibility of generative AI replacing creative workers. People are going to try, just like people like to pass off procedurally generated music and video games as being innovative. It's not. You can only fool people with randomly generated garbage until they get frustrated with the lack of consistency and accuracy.

 

Now, as for where it will be useful, and we've seen some of this with with generative AI, but mostly I see it being good as an assistive technology. So let's say you are losing your ability to draw due to carpel tunnel or arthritis, yes, I could see image-to-image type of generative AI that has been trained on your previous finished works able to "autocomplete" from your line drawing. Unfortunately that's not the direction a lot of generative AI is aimed at. Generative AI is frequently aimed at trying to replace the artist, and passing off "good enough" with no consistency. Kinda hard to make consistently good works that don't look/sound/read like previous stuff attributed to you right?

 

Likewise for music and voice generation. I can see it being useful to retaining your voice of a specific age by training a model on your younger self, but it's still your self that is providing the input. You can't just drop a billboard top-100 into it and claim you made a cover (this is what people are presently doing, and the result is terrible.) No two covers will sound like the same singer, because ultimately it's an advanced auto-tune. It's not "Singing".

 

Where "AI"'s are helpful presently are "second opinion" situations. Because the AI could be trained to recognize things in data that a human would not, because a human can't do it any level of accuracy because it doesn't have infinite time. So analyzing DNA, X-rays, MRI's, etc for problems that might be smaller than what the human doctor will notice. But you can't say "AI says you have cancer" without having a human actually go double check that. Wouldn't it suck to be diagnosed with a medical condition by an AI, be treated for it (eg surgical intervention) and then turns out the AI was wrong? Welp, medical malpractice insurance is gonna have a field day with doctors who don't know how to use AI, just like how people don't know Wikipedia isn't a source for writing school work.

 

We're pretty much in a situation where AI should be helping, not replacing humans. If anything, outsourcers should be the ones to be afraid, because their jobs were always replaceable. When a business outsources their core service or product, then their product or service will be duplicated, cheaper, off-the-clock by the outsourcer. Data will be exfiltrated to a third party, and then suddenly all these copy-cat services and products that look just like yours show up. Funny how many times I've seen this with kickstarters. Where some copycat shows up before the official product gets released. AI is only going to make this easier. So my prediction is that South/South-East Asia is going to feel the most pain. Physical products won't need to be made by humans if the AI can just be fed a photo or video of some clothing they want to clone, it will produce a pattern, and then the robots will knit or 3d-print clothing that is of the same quality as a hand-made version of the product. Presently this is done rather poorly, by humans. Imagine being able to create products, tailored to fit you perfectly, rather than mass-produced fast fashion that poorly fits.

 

[Touch My Hamm]

AI needs some controls or companies will quickly replace as much as possible to make a quick $$

I hate how AI is being pushed when in reality its simply the same chat/audio bots we have had for years. Companies are quickly trying to replace humans with bots to save money to then sell products to the same out of work People...We could be heading to another Industrial revolution, where we need to find work for people while there is a transition where alot of jobs are replaced by automation. To me as someone who actively researches AI and how to maximize its use for IT this is something that is coming sooner then people think and is one of the biggest issues facing the next oncoming years. Before simply replacing positions we need to find new places and new facets for the labour force to work.

 

 

Bring back physical media

Digital ownership is a mess. Every game now isnt yours and is simply a rented extension that you have no control over. Gaming is at a sorry state where big devs are more inclined to milk you for every penny rather then making a fun game. Convenience of not having to go to a store and have a physical medium has made it so people dont own most games they buy currently. I still buy physical media as it feels like collecting now where its not simply a fun thing to won but also an investment as the game/movie/show/etc could be taken off or changed and I would have an original version. Think my old disney VHS's.

 

 

[Polderviking]

2 hours ago, Kisai said:

AI's impact is very over-stated.

 

In creative industries (writing, artwork, music, film, television, video games, food), generative AI is utter garbage, and will not replace anyone. Anyone who believes they can cut the artist out of the creation of a creative work, and have something people will want to buy, at any price, is a fool. Generative AI has not "learned" to be creative, at best it has achieved the "infinite monkey theorem". Without being prompted, it does nothing. When prompted, it just auto-completes using a statistical model of what the next letter, word, music, note, etc is. 

 

We are not anywhere close to an AI that actually knows, or can infer an answer to any question. You know the classic trope of "book smarts, street smarts?" The AI achieves a high score on book smarts, because it has effectively memorized things, but has zero in the street smarts, and when presented with any question that requires it to solve based on experience, it will always fail. It can not learn from experience, only other people's instructions. If those instructions are wrong because statistically people get it wrong, or people commonly repeat misinformation, then yes, the AI is going to learn wrong information.

 

This is why GPT sucks at math. It doesn't understand a math question, because it can not statistically memorize the answer to every math question. That's not how math works. That's also why you have to be careful and double-check the results. It's not good enough to have a 7% on report card any more than a 86%. Given a properly formatted math question, it should only have one, correct answer.

 

Generative AI also can't generate anything that can be copyrighted, because the training materials may contain copyrighted sources. So if you use Generative AI to write a page of a novel, there is a 100% chance it will plagiarize entire sentences. A few days ago, when the Mistral LLM was published, I went and tried an AI accuracy battle website (I forget what it's called) and I asked it to describe "climate change" and the entire opening paragraph was more than 50% identical, word-for-word. Two different LLM's. That tells me both models ingested the same source. Were they both correct? Yes. But the format of the output was different, one wrote it like you would write an essay for school, and the other one wrote it like a Quora answer in bulleted points.

 

I don't see any possibility of generative AI replacing creative workers. People are going to try, just like people like to pass off procedurally generated music and video games as being innovative. It's not. You can only fool people with randomly generated garbage until they get frustrated with the lack of consistency and accuracy.

 

Now, as for where it will be useful, and we've seen some of this with with generative AI, but mostly I see it being good as an assistive technology. So let's say you are losing your ability to draw due to carpel tunnel or arthritis, yes, I could see image-to-image type of generative AI that has been trained on your previous finished works able to "autocomplete" from your line drawing. Unfortunately that's not the direction a lot of generative AI is aimed at. Generative AI is frequently aimed at trying to replace the artist, and passing off "good enough" with no consistency. Kinda hard to make consistently good works that don't look/sound/read like previous stuff attributed to you right?

 

Likewise for music and voice generation. I can see it being useful to retaining your voice of a specific age by training a model on your younger self, but it's still your self that is providing the input. You can't just drop a billboard top-100 into it and claim you made a cover (this is what people are presently doing, and the result is terrible.) No two covers will sound like the same singer, because ultimately it's an advanced auto-tune. It's not "Singing".

 

Where "AI"'s are helpful presently are "second opinion" situations. Because the AI could be trained to recognize things in data that a human would not, because a human can't do it any level of accuracy because it doesn't have infinite time. So analyzing DNA, X-rays, MRI's, etc for problems that might be smaller than what the human doctor will notice. But you can't say "AI says you have cancer" without having a human actually go double check that. Wouldn't it suck to be diagnosed with a medical condition by an AI, be treated for it (eg surgical intervention) and then turns out the AI was wrong? Welp, medical malpractice insurance is gonna have a field day with doctors who don't know how to use AI, just like how people don't know Wikipedia isn't a source for writing school work.

 

We're pretty much in a situation where AI should be helping, not replacing humans. If anything, outsourcers should be the ones to be afraid, because their jobs were always replaceable. When a business outsources their core service or product, then their product or service will be duplicated, cheaper, off-the-clock by the outsourcer. Data will be exfiltrated to a third party, and then suddenly all these copy-cat services and products that look just like yours show up. Funny how many times I've seen this with kickstarters. Where some copycat shows up before the official product gets released. AI is only going to make this easier. So my prediction is that South/South-East Asia is going to feel the most pain. Physical products won't need to be made by humans if the AI can just be fed a photo or video of some clothing they want to clone, it will produce a pattern, and then the robots will knit or 3d-print clothing that is of the same quality as a hand-made version of the product. Presently this is done rather poorly, by humans. Imagine being able to create products, tailored to fit you perfectly, rather than mass-produced fast fashion that poorly fits.

 

I appreciate sharing your view on this so extensively, in regards to it taking away jobs. And I fully agree that it should work as a tool. (And mostly does I guess? I use it as such anyway.)
But I do think you're disregarding the fact all these technologies are still very much new a little bit easily.
It feels to me at least we are still very much at the start of where this is going.

[cooky560]

Generative AI is something it's impossible to judge at this stage, as it's growing at an alarming pace. Right now I agree with sentiments shared here, but will it be so limited in 10 years? It's going to be interesting to see what happens. It'll either be gone as a fad by then, or be having a massive impact on our lives.

 

 

[Issac Zachary]

On 12/11/2023 at 8:39 AM, smcoakley said:

This seems to be a problem with that specific product or maybe your specific unit. I have multiple Bluetooth earbuds that easily give me 12+ hours of listening time.

Ok, but essentially that means that $200 earphones are essentially a throw away item.

 

 "My $200 earphones aren't working."

"Well just go throw those away and buy another pair."

[cooky560]

3 hours ago, Issac Zachary said:

Ok, but essentially that means that $200 earphones are essentially a throw away item.

 

 "My $200 earphones aren't working."

"Well just go throw those away and buy another pair."

This is the biggest concern with modern products, I have hi-fi from the 70s that still works, kitchen appliances from the 60s that still work, and yet I'm more worried that my new gaming headphones are what'll need replacing next

[Brookz]

On 12/14/2023 at 7:34 PM, Touch My Hamm said:

AI needs some controls or companies will quickly replace as much as possible to make a quick $$

I hate how AI is being pushed when in reality its simply the same chat/audio bots we have had for years. Companies are quickly trying to replace humans with bots to save money to then sell products to the same out of work People...We could be heading to another Industrial revolution, where we need to find work for people while there is a transition where alot of jobs are replaced by automation. To me as someone who actively researches AI and how to maximize its use for IT this is something that is coming sooner then people think and is one of the biggest issues facing the next oncoming years. Before simply replacing positions we need to find new places and new facets for the labour force to work.

 

 

Bring back physical media

Digital ownership is a mess. Every game now isnt yours and is simply a rented extension that you have no control over. Gaming is at a sorry state where big devs are more inclined to milk you for every penny rather then making a fun game. Convenience of not having to go to a store and have a physical medium has made it so people dont own most games they buy currently. I still buy physical media as it feels like collecting now where its not simply a fun thing to won but also an investment as the game/movie/show/etc could be taken off or changed and I would have an original version. Think my old disney VHS's.

 

 

Sadly game industry got to this state exactly 'cause it was more convinient(and cheaper) to get a digital copy compared to physical. Thinking from disctributor point of view: you can keep prices the same without needing to make physical copies or you can lower the price for the same $ you would've spent on physical copy creation. It's a win-win for them and customer doesn't need to go and pickup the copy/pay for delivery, etc. It's kinda lame, but I see why both sides kinda agreed that it's the best option(and you still get to milk some extra $ from special editions with maps/memorabilia from the game, but on a smaller scale for actual fan that collect stuff).
VHS era was really dope tho, still got like 2x boxes of VHS in my basement I switched from collecting VHS to I think posters and then to archery stuff eventually(Reminds me I gotta pick up my new bow from local sports store)

[Issac Zachary]

All I know is that when the internet goes out for several days or weeks that after a while I can't access any of my Steam games.

[Recharged]

Faking everything is becoming annoying.