Samsung 960 PRO specs in at 3500MB/sec an that's not just marketing wank. I'll see if I can dig up the screenshot of mine far exceeding that (to my own surprise) if it will shut you the fuck up. The ability to maintain those speeds when using encryption falls on the other components of the system, as the drive itself is not (and should not be) aware of the encryption in the first place; Ryzen and anything put out by Intel for the last few generations will be able to handle encryption at well above those rates through dedicated silicon on the CPU die, so that was a silly goal post to erect in the first place.
I will, however, post a correction to my earlier claims: The 960 PRO was released in October of 2016, so it hasn't actually been out for over 2 years as I had stated. I thought it had.
Anyone know who makes these SSDs for Apple? Because you know damn well they aren't making them themselves and Foxconn wouldn't know where to start. If I had to guess, they're 960 PROs in Apple's custom configuration, which requires rearranging some data paths and making things slightly less efficient, thus the just under 3000MB/sec speeds on parts I've personally seen run much faster.
I said "it seems" because that's how I interpreted it. Beyond that, I wouldn't very well say I lost the argument given that you were barking up the wrong tree in the first place. Oh and how hurtful, some random anonymous asshole on the internet thinks I'm a loser. I'm so hurt.
So they've been doing what iOS started doing in September 2015 since at least sometime in 2014 and... it seems you're implying that Android is copying Apple when Android actually had the feature first. Yes, I know things.
Apparently, though, you don't know how to follow the context of a conversation; for example, the "THIS" was in response to the comment I was... well... responding to. That is, I was agreeing with ebrandsberg about it being a battery size issue -- low power mode had nothing at all to do with the "THIS." portion of my reply.
Plus, all of my Android phones for the past few years have come with a battery extender mode that does what Apple is doing -- but at my option and at any time I choose. If I never enable that function, I just start seeing less battery life over time, but the phone still works because the power management circuitry and battery aren't under-engineered.
I don't know why a non-programmer would be on Slashdot
Because it's "News for Nerds", not "Peniswaving for Programmers". I remember when ours (programming) used to be an open and welcoming community. You are what is wrong with it today.
Discussing a form of addiction in a discussion about addiction is hardly off-topic. It's, perhaps, just slightly enough removed from some very specific definition of the topic that people who don't see the problem will scream "off-topic" as an excuse to not have to hear it; but those people are just sticking their heads in the sand.
The only people I've ever heard say smartphone addiction isn't a problem have said it while glaring at me over the screen of their smartphone. The only people I've ever seen say it have been posting from their smartphone. I've not once had someone set their phone down and explain to me that they were answering a work email or helping a friend work through something, then pick it back up and carry on; no, the addicts can't take their eyes off the screen long enough to defend themselves.
Not that there's much defense against that car they're about to step in front of, anyway.
That, of course, in no was lessens what's being talked about here; a whole lot of gaming is done on smartphones, you see.
You said that it doesn't even matter if AI has to be taught.
Try taking that statement in the context of the question I was answering. You asked who was smarter and I pointed out that it doesn't matter who's smarter as long as we're the ones who have to do the teaching; that in no way means that it doesn't matter that we're the ones doing the teaching. In fact, it places all of the importance on us doing the teaching which, in turn, places importance on the AI doing the learning.
What you were actually replying to just then was an entirely different point: it is a given that the AI would have to be taught. Of course, since it's also a given that a human would also have to be taught, you see the same thing on both sides of the equation, ergo no, it actually doesn't matter and you can remove it from the equation altogether. That's a rational reduction, the same as removing "+ 3" from "N + 3 = 5 + 3".
Who does the teaching is what matters, not that both sides must learn. As long as we're the teachers, we can prevent any AI from becoming smarter than us if we so choose.
So which human could identify a rifle on the moment of their birth ?
None, of course. Both the AI and the human will, hopefully, grow smarter as they're taught, of course; so the real question is whether a given AI or a given human will fare better given identical inputs over time. You seem to want to reduce things to absurd levels for some reason.
It's really not important that humans and AIs both start at 0, it's more or less a given, nor is it important that one or the other might fare better with better inputs; again, that should be expected. What really matters with regard to who's "smarter" is, given identical starting points consistently identical inputs over a period of time, which of them realizes the greatest improvement in their abilities.
Something tells me our current generation of AIs could give you a run for your money in that test.
I'm not the one who originally made the point, perhaps they'll have some prepared talking points. The real answer, though, is whoever actually properly identifies the thing being looked at more often. The real real answer is that it doesn't matter as long as we have to teach the AI in the first place.
The AI that wrote the headline knew it was a story about another AI seeing one thing and thinking it was another. Since it, itself, saw the turtle as a rifle, it had to think of what another AI might mistake a rifle for and it came up with helicopter.
Ah, actually we were both wrong. We both fell victim to bad math; mine too simple and yours too complex. I realized my error half way through reading your post.
A battery that can deliver 1W for 2.28hr would, as you correctly stated, lat 137 minutes (136.8, but I digress). At 10x the power draw, the battery life would actually be 1/10th as long, or 13.7 minutes.
I shouldn't do math with a head cold... and I dare say you were closer to right the first time (3.7min under) than I was (9.1min over). In actual practice, the battery won't be allowed to run flat, so 10 minutes is probably spot on.
Or buying a phone that takes a fairly standard and readily available battery. Yes, phones do still exist that use the same battery design Nokia was using in the mid '00s. Those can be had, new and from reputable manufacturers, for around $10.
Additional errata: 1900mAh at 1.2V is 2.28Wh, the battery would take 22.8 minutes ti run flat at 10W. Not that it affects your point in any way but, if you're gonna post corrections...
As long as microwave isn't running a phone-home-for-activation OS for which the manufacturer has already announced the end of support, no, I don't really care where the clock is as the microwave will still work in 2 years.
That's a fair bit different than suggesting someone migrate to a platform that is already EOL, has an already deteriorating ecosystem, and will be unsupported and unable to activate or install apps in two years time. Especially when you're making that suggestion to someone who just told you they're looking to upgrade from their current 9 year old phone; clearly, they don't want to have to migrate again right away.
Slashdot Mirror
Samsung 960 PRO specs in at 3500MB/sec an that's not just marketing wank. I'll see if I can dig up the screenshot of mine far exceeding that (to my own surprise) if it will shut you the fuck up. The ability to maintain those speeds when using encryption falls on the other components of the system, as the drive itself is not (and should not be) aware of the encryption in the first place; Ryzen and anything put out by Intel for the last few generations will be able to handle encryption at well above those rates through dedicated silicon on the CPU die, so that was a silly goal post to erect in the first place.
I will, however, post a correction to my earlier claims: The 960 PRO was released in October of 2016, so it hasn't actually been out for over 2 years as I had stated. I thought it had.
Anyone know who makes these SSDs for Apple? Because you know damn well they aren't making them themselves and Foxconn wouldn't know where to start. If I had to guess, they're 960 PROs in Apple's custom configuration, which requires rearranging some data paths and making things slightly less efficient, thus the just under 3000MB/sec speeds on parts I've personally seen run much faster.
Does YOUR SSD do writes on a fully encrypted partition at 2996MB/sec
Mine's actually closer to 4000 and it's been on the market for over 2 years, actually. Apple's not anywhere near cutting-edge in that regard.
I said "it seems" because that's how I interpreted it. Beyond that, I wouldn't very well say I lost the argument given that you were barking up the wrong tree in the first place. Oh and how hurtful, some random anonymous asshole on the internet thinks I'm a loser. I'm so hurt.
So they've been doing what iOS started doing in September 2015 since at least sometime in 2014 and... it seems you're implying that Android is copying Apple when Android actually had the feature first. Yes, I know things.
Apparently, though, you don't know how to follow the context of a conversation; for example, the "THIS" was in response to the comment I was... well... responding to. That is, I was agreeing with ebrandsberg about it being a battery size issue -- low power mode had nothing at all to do with the "THIS." portion of my reply.
That's why it was a separate paragraph.
THIS.
Plus, all of my Android phones for the past few years have come with a battery extender mode that does what Apple is doing -- but at my option and at any time I choose. If I never enable that function, I just start seeing less battery life over time, but the phone still works because the power management circuitry and battery aren't under-engineered.
Because that would be much less fun for those who run the show.
No, he shamed skinny ones.
I don't know why a non-programmer would be on Slashdot
Because it's "News for Nerds", not "Peniswaving for Programmers". I remember when ours (programming) used to be an open and welcoming community. You are what is wrong with it today.
Discussing a form of addiction in a discussion about addiction is hardly off-topic. It's, perhaps, just slightly enough removed from some very specific definition of the topic that people who don't see the problem will scream "off-topic" as an excuse to not have to hear it; but those people are just sticking their heads in the sand.
The only people I've ever heard say smartphone addiction isn't a problem have said it while glaring at me over the screen of their smartphone. The only people I've ever seen say it have been posting from their smartphone. I've not once had someone set their phone down and explain to me that they were answering a work email or helping a friend work through something, then pick it back up and carry on; no, the addicts can't take their eyes off the screen long enough to defend themselves.
Not that there's much defense against that car they're about to step in front of, anyway.
That, of course, in no was lessens what's being talked about here; a whole lot of gaming is done on smartphones, you see.
I propose that Bos Taurus Fecal Disorder be added to the DSM, as well. Too many people simply can not pull themselves away from bullshit.
Hell, I'll present myself as the first case study, even, since I can't seem to leave this site.
So, then, is he wrong?
You said that it doesn't even matter if AI has to be taught.
Try taking that statement in the context of the question I was answering. You asked who was smarter and I pointed out that it doesn't matter who's smarter as long as we're the ones who have to do the teaching; that in no way means that it doesn't matter that we're the ones doing the teaching. In fact, it places all of the importance on us doing the teaching which, in turn, places importance on the AI doing the learning.
What you were actually replying to just then was an entirely different point: it is a given that the AI would have to be taught. Of course, since it's also a given that a human would also have to be taught, you see the same thing on both sides of the equation, ergo no, it actually doesn't matter and you can remove it from the equation altogether. That's a rational reduction, the same as removing "+ 3" from "N + 3 = 5 + 3".
Who does the teaching is what matters, not that both sides must learn. As long as we're the teachers, we can prevent any AI from becoming smarter than us if we so choose.
So which human could identify a rifle on the moment of their birth ?
None, of course. Both the AI and the human will, hopefully, grow smarter as they're taught, of course; so the real question is whether a given AI or a given human will fare better given identical inputs over time. You seem to want to reduce things to absurd levels for some reason.
It's really not important that humans and AIs both start at 0, it's more or less a given, nor is it important that one or the other might fare better with better inputs; again, that should be expected. What really matters with regard to who's "smarter" is, given identical starting points consistently identical inputs over a period of time, which of them realizes the greatest improvement in their abilities.
Something tells me our current generation of AIs could give you a run for your money in that test.
I'm not the one who originally made the point, perhaps they'll have some prepared talking points. The real answer, though, is whoever actually properly identifies the thing being looked at more often. The real real answer is that it doesn't matter as long as we have to teach the AI in the first place.
But it doesn't output answers that aren't on its list... did you not follow the conversation?
The AI that wrote the headline knew it was a story about another AI seeing one thing and thinking it was another. Since it, itself, saw the turtle as a rifle, it had to think of what another AI might mistake a rifle for and it came up with helicopter.
When I look at the dress, I only see white/gold, so I failed your test despite not being an AI.
He said an AI, not every AI. Right now, while not every person can do what is prescribed above, some people can but no AI can.
Browsing with "Ads Disabled" checked.
What, don't you have that option?
Ah, actually we were both wrong. We both fell victim to bad math; mine too simple and yours too complex. I realized my error half way through reading your post.
A battery that can deliver 1W for 2.28hr would, as you correctly stated, lat 137 minutes (136.8, but I digress). At 10x the power draw, the battery life would actually be 1/10th as long, or 13.7 minutes.
I shouldn't do math with a head cold... and I dare say you were closer to right the first time (3.7min under) than I was (9.1min over). In actual practice, the battery won't be allowed to run flat, so 10 minutes is probably spot on.
Revert back to the older iOS version, and it'll perform just like it did when it was new.
Absent the battery-age-based throttling being discussed here, perhaps...
Or buying a phone that takes a fairly standard and readily available battery. Yes, phones do still exist that use the same battery design Nokia was using in the mid '00s. Those can be had, new and from reputable manufacturers, for around $10.
Additional errata: 1900mAh at 1.2V is 2.28Wh, the battery would take 22.8 minutes ti run flat at 10W. Not that it affects your point in any way but, if you're gonna post corrections...
As long as microwave isn't running a phone-home-for-activation OS for which the manufacturer has already announced the end of support, no, I don't really care where the clock is as the microwave will still work in 2 years.
That's a fair bit different than suggesting someone migrate to a platform that is already EOL, has an already deteriorating ecosystem, and will be unsupported and unable to activate or install apps in two years time. Especially when you're making that suggestion to someone who just told you they're looking to upgrade from their current 9 year old phone; clearly, they don't want to have to migrate again right away.
So I wouldn't recommend buying a new one - or spending ANY money on one - but if you have one with 10 on it or are handed a clean one for free...
That's more or less my point. This entire discussion is predicated on a user looking for a platform to switch to; that is, getting a new phone.
Indeed, we saw exactly this with Metro apps on Windows and I fear that's where Apple is taking the Mac if they do this.
Time will tell, and I'll be ecstatic if I'm wrong.