Loading video...

Video Failed to Load

Go Home

Important to know that Postgres/MySQL on local SSD is *the* best way to get great performance. The reason it could get *even faster* boils down to abstraction boundaries. SSDs are built to be workload agnostic. They have their own controllers and garbage collectors meant to balance performance of many...

BenjDicken's profile picture

Ben Dicken

38,349 subscribers

79,406 views • 2 months ago •via X (Twitter)

EducationScience & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

“‘Red Rain’ is one of my favourite tracks from that record [‘So’]. It’s a good example of a band playing together with a lot of energy and again, I think from Daniel Lanois’ input, there’s a respect for the moment. So many times in studios you see people desperately trying to get the best possible sound, but they let the moment go because they want to get something right or fixed, or they know that this bit can be improved, but the moment is something instinctively that you feel and if you get that great performance everything else will bow to it. And you can fix a lot of things afterwards. But, if you get the best sound but you don’t get the great performance, you have nothing. And it’s so easy to forget that in the studio.” - PG
1:21

“‘Red Rain’ is one of my favourite tracks from that record [‘So’]. It’s a good example of a band playing together with a lot of energy and again, I think from Daniel Lanois’ input, there’s a respect for the moment. So many times in studios you see people desperately trying to get the best possible sound, but they let the moment go because they want to get something right or fixed, or they know that this bit can be improved, but the moment is something instinctively that you feel and if you get that great performance everything else will bow to it. And you can fix a lot of things afterwards. But, if you get the best sound but you don’t get the great performance, you have nothing. And it’s so easy to forget that in the studio.” - PG

Peter Gabriel

89,210 views • 2 years ago

Anthropic Technical Staff Jess Yan says the harness and the model cannot be pulled apart without giving up performance, though she flags her own bias. "I think I am quite biased but I also think that it is impossible to get the maximum possible performance without tying together the harness and the model." "Now the components of the harness and maybe the thickness of the harness will change over time as models get more and more capable." "However, when we test our models and when we are assessing their performance we always have to test it in conjunction with a harness." "And are we going to test it with all the different harnesses of the world? We're going to select the harnesses that we have built." "And so there is an aspect of the necessity of building models is that you have to be testing them with harnesses, and that sort of keeps them paired together."
0:46

Anthropic Technical Staff Jess Yan says the harness and the model cannot be pulled apart without giving up performance, though she flags her own bias. "I think I am quite biased but I also think that it is impossible to get the maximum possible performance without tying together the harness and the model." "Now the components of the harness and maybe the thickness of the harness will change over time as models get more and more capable." "However, when we test our models and when we are assessing their performance we always have to test it in conjunction with a harness." "And are we going to test it with all the different harnesses of the world? We're going to select the harnesses that we have built." "And so there is an aspect of the necessity of building models is that you have to be testing them with harnesses, and that sort of keeps them paired together."

Fireside Alpha

570,359 views • 3 days ago

Javascript, Ruby, and Go all use garbage collectors for memory management. But did you know your SSD has a garbage collector (GC) too? SSDs with high write/delete traffic are constantly being reorganized by the GC, which is built directly into the firmware running on the SSD controller. This is one of the reasons why sequential writes are still preferable to random, even if you aren't using a spinning disk.
0:37

Javascript, Ruby, and Go all use garbage collectors for memory management. But did you know your SSD has a garbage collector (GC) too? SSDs with high write/delete traffic are constantly being reorganized by the GC, which is built directly into the firmware running on the SSD controller. This is one of the reasons why sequential writes are still preferable to random, even if you aren't using a spinning disk.

Ben Dicken

29,599 views • 6 months ago

SITUATION EXPLAINED: Why is software the most important part of the AI hardware stack? We asked Anush Elangovan, VP of AI software at AMD "You got the data center, you got desktop, you got laptops, you got FPGAs. And in time, we have multiple generations of each. So if you map this whole terrain out, software is gonna have to map across so much hardware variation that this is kind of like the golden age of abstractions." "We wanna abstract it in a way that you almost have to automate it too. Because there's gonna be a combinatorial explosion of cases that we have to handle. But every one of them is gonna expect day zero support and performant models." "You're not doing performance optimization of one platform. You're doing performance optimization of N platforms across like a time machine. And when a new model drops, every combination is gonna be looking for it." "Because AI is gonna suck up every flop that exists. Any flop, any compute you have anywhere is something that AI could be using: your laptop doing local tokens for coding sessions, grepping files, bash commands, voice commands done locally for latency reasons."
4:16

SITUATION EXPLAINED: Why is software the most important part of the AI hardware stack? We asked Anush Elangovan, VP of AI software at AMD "You got the data center, you got desktop, you got laptops, you got FPGAs. And in time, we have multiple generations of each. So if you map this whole terrain out, software is gonna have to map across so much hardware variation that this is kind of like the golden age of abstractions." "We wanna abstract it in a way that you almost have to automate it too. Because there's gonna be a combinatorial explosion of cases that we have to handle. But every one of them is gonna expect day zero support and performant models." "You're not doing performance optimization of one platform. You're doing performance optimization of N platforms across like a time machine. And when a new model drops, every combination is gonna be looking for it." "Because AI is gonna suck up every flop that exists. Any flop, any compute you have anywhere is something that AI could be using: your laptop doing local tokens for coding sessions, grepping files, bash commands, voice commands done locally for latency reasons."

MTS

17,297 views • 28 days ago

FASTER for #Amiga OCS/ECS: I never imagined saying this in a million years, but I seem to be struggling to get the performance I need out of the blitter, and I don't even have all the stuff on screen yet! Anybody know a faster way of rendering bobs than the cookie cutter method?
1:58

FASTER for #Amiga OCS/ECS: I never imagined saying this in a million years, but I seem to be struggling to get the performance I need out of the blitter, and I don't even have all the stuff on screen yet! Anybody know a faster way of rendering bobs than the cookie cutter method?

Jonathan Thomas @atariste.bsky.social

13,897 views • 1 year ago

"We're going to have a safe, and one of the best World Cups in the history of this event. The truth is, though, with the Democrats defunding ICE & CBP for 115 days, and DHS for 76 days, it made it tough," says Secretary Markwayne Mullin on securing World Cup matches amid a Democrat shutdown. "When we are shut down, even though our men and women at Homeland Security... are willing to do their job for free—that's great. The problem is that it's our local law enforcement that we have to partner with, that we have to give grants to—they weren't able to be reimbursed, and not all of them can handle that... we're ready, but I feel positive that we could've been even in a better position if we weren't having to deal with the gamesmanship of the Democrat Party."
2:27

"We're going to have a safe, and one of the best World Cups in the history of this event. The truth is, though, with the Democrats defunding ICE & CBP for 115 days, and DHS for 76 days, it made it tough," says Secretary Markwayne Mullin on securing World Cup matches amid a Democrat shutdown. "When we are shut down, even though our men and women at Homeland Security... are willing to do their job for free—that's great. The problem is that it's our local law enforcement that we have to partner with, that we have to give grants to—they weren't able to be reimbursed, and not all of them can handle that... we're ready, but I feel positive that we could've been even in a better position if we weren't having to deal with the gamesmanship of the Democrat Party."

Rapid Response 47

111,397 views • 1 month ago

Tillman Holloway explains why your Bitcoin might actually be safer on an exchange than in cold storage "The simplest way to put it is that at the end of the day you're trusting someone. In the hardware wallet space we've ignored the fact that we're trusting hardware wallet providers and their firmware updates. That's now front and center for everybody to recognize, that these devices have a lot of complexity to them" "They can age out as well. If you've been in crypto long enough and had a hardware device, and you plug it in five years later, sometimes it doesn't even recognize the device. There's a path to get it back, but it's not for the layman, not for the average user. Cold storage is something that has to be managed, just like anything that has risk attached to it" "Risk has to be siloed so that it's not catastrophic, so it doesn't sweep across the whole board. We incur risk every day in everything, but the risk is siloed to one specific vertical. Maybe there's some disruption there, but the aggregate value represented by the entity can easily cover those losses"
1:39

Tillman Holloway explains why your Bitcoin might actually be safer on an exchange than in cold storage "The simplest way to put it is that at the end of the day you're trusting someone. In the hardware wallet space we've ignored the fact that we're trusting hardware wallet providers and their firmware updates. That's now front and center for everybody to recognize, that these devices have a lot of complexity to them" "They can age out as well. If you've been in crypto long enough and had a hardware device, and you plug it in five years later, sometimes it doesn't even recognize the device. There's a path to get it back, but it's not for the layman, not for the average user. Cold storage is something that has to be managed, just like anything that has risk attached to it" "Risk has to be siloed so that it's not catastrophic, so it doesn't sweep across the whole board. We incur risk every day in everything, but the risk is siloed to one specific vertical. Maybe there's some disruption there, but the aggregate value represented by the entity can easily cover those losses"

The Wolf Of All Streets

23,761 views • 10 hours ago

Q: Congresswoman, how are you feeling? Are you afraid? Rep. LaMonica McIver: Well, of course I am. I mean, I have the President of the United States and the Department of Justice trying to send me to jail for 17 years. That’s very frightening. I think that even the Department of Justice knows that this case is extremely bogus and that it makes no sense. But once again, the point of it all is cruelty. The process is the pain for them, and they enjoy that, and they’re watching me, go through that, you know, having to raise money for this case, that I’m not getting free dollars, you know, while they get to use your tax dollars to charge me. My family is stressed out about it. Obviously, I have a big belly bump in front of me, as you all can see. We should not be spending taxpayer dollars to prosecute members of Congress for them showing up to do their job.
0:58

Q: Congresswoman, how are you feeling? Are you afraid? Rep. LaMonica McIver: Well, of course I am. I mean, I have the President of the United States and the Department of Justice trying to send me to jail for 17 years. That’s very frightening. I think that even the Department of Justice knows that this case is extremely bogus and that it makes no sense. But once again, the point of it all is cruelty. The process is the pain for them, and they enjoy that, and they’re watching me, go through that, you know, having to raise money for this case, that I’m not getting free dollars, you know, while they get to use your tax dollars to charge me. My family is stressed out about it. Obviously, I have a big belly bump in front of me, as you all can see. We should not be spending taxpayer dollars to prosecute members of Congress for them showing up to do their job.

Headquarters

415,091 views • 1 month ago

This song no deserve nomination?! And you mentioned Laho? It’s like you people didn’t even take note of the name of the category, “Music Performance”. Abeg show me the performance for With You. This song is even too big to be categorized under “Africa”, this was a global performance and can stand with anything from even Beyoncé.
1:16

This song no deserve nomination?! And you mentioned Laho? It’s like you people didn’t even take note of the name of the category, “Music Performance”. Abeg show me the performance for With You. This song is even too big to be categorized under “Africa”, this was a global performance and can stand with anything from even Beyoncé.

Duru Bond

299,959 views • 6 months ago

🎥| “I think this is maybe just starting to happen, but I find it really interesting that sport psychology coaches are prevalent now throughout sport. They're a vital part of sport. It'd be interesting to see something similar in the music industry and especially people like us at young ages that were dealing with a lot of pressure at a lot a young age, but also to perform at a, you know, really good level. Um because often those kind of psych psychologists will be, you know, like um performance-based. Well, this is a performance-based job for sure. I do think something like that would have helped. It's only really in hindsight cuz like for my I can only speak on my own journey, but you're so young and it's so exciting and every day is so different. I didn't really have too much time to to really take it all in. So I think that having just remember having those conversations that would definitely help some kind of voice to chat to but not just in the name of like therapy also performance-based people that can bring out the best in you because you know I can remember that as an 18-year-old lad you literally get brought out thrown in the lies and go deal with it you know and you you have people are there to help you craft your skill but things like inner confidence you're just expected to have those things you know and that's something that definitely takes takes uh especially as a young person, it takes a bit of direction, I think.” Louis Tomlinson via Billboard.
1:23

🎥| “I think this is maybe just starting to happen, but I find it really interesting that sport psychology coaches are prevalent now throughout sport. They're a vital part of sport. It'd be interesting to see something similar in the music industry and especially people like us at young ages that were dealing with a lot of pressure at a lot a young age, but also to perform at a, you know, really good level. Um because often those kind of psych psychologists will be, you know, like um performance-based. Well, this is a performance-based job for sure. I do think something like that would have helped. It's only really in hindsight cuz like for my I can only speak on my own journey, but you're so young and it's so exciting and every day is so different. I didn't really have too much time to to really take it all in. So I think that having just remember having those conversations that would definitely help some kind of voice to chat to but not just in the name of like therapy also performance-based people that can bring out the best in you because you know I can remember that as an 18-year-old lad you literally get brought out thrown in the lies and go deal with it you know and you you have people are there to help you craft your skill but things like inner confidence you're just expected to have those things you know and that's something that definitely takes takes uh especially as a young person, it takes a bit of direction, I think.” Louis Tomlinson via Billboard.

World Tomlinson

21,362 views • 6 months ago

🚨If the US🇺🇸 thinks Venezuela🇻🇪 will be an easy target, it’s in for a BIG SURPRISE- Venezuela’s Former Deputy Foreign Minister for North America ‘As long as we have oil wealth and other mineral wealth, Venezuela is always going to be a target of the United States, of large multinational corporations, etc. I understand the United States has military superiority…but to think that Venezuela is an easy target, that Venezuela is someplace that you can come in and just take over, you’re up for a surprise. I think there’s a long history here of resistance, of morale to resist attacks. I think we have seen also the tragic sort of interventions that the United States has done even in recent years. I mean, you could go back to Vietnam and remember how tragic that was. But even Afghanistan, Iraq, all these more recent interventions have shown that if you don’t have the support of local populations, if there isn’t some sort of other arrangement, you’re not going to be successful. I mean, after 20 years of the disaster they caused in Afghanistan, they had to leave without accomplishing anything. That’s horrible for our people in the sense that, of course, if something like that comes to happen, and I really wish that it doesn’t come to happen — it’s going to be hurtful for the Venezuelan people. But I think they have to think twice about what really is at stake and how this is going to affect the stability of the United States. We’re not hundreds or thousands of miles away. We’re really close to the United States. And it could have a terrible effect for its economy, even for its own base. I mean, if you look at the base that supports President Trump right now, a lot of the criticism that is coming is precisely from its own base, who were promised that there would be no more forever wars.’ -Carlos Ron on the latest episode of Going Underground FULL INTERVIEW BELOW IN THE REPLIES👇
2:27

🚨If the US🇺🇸 thinks Venezuela🇻🇪 will be an easy target, it’s in for a BIG SURPRISE- Venezuela’s Former Deputy Foreign Minister for North America ‘As long as we have oil wealth and other mineral wealth, Venezuela is always going to be a target of the United States, of large multinational corporations, etc. I understand the United States has military superiority…but to think that Venezuela is an easy target, that Venezuela is someplace that you can come in and just take over, you’re up for a surprise. I think there’s a long history here of resistance, of morale to resist attacks. I think we have seen also the tragic sort of interventions that the United States has done even in recent years. I mean, you could go back to Vietnam and remember how tragic that was. But even Afghanistan, Iraq, all these more recent interventions have shown that if you don’t have the support of local populations, if there isn’t some sort of other arrangement, you’re not going to be successful. I mean, after 20 years of the disaster they caused in Afghanistan, they had to leave without accomplishing anything. That’s horrible for our people in the sense that, of course, if something like that comes to happen, and I really wish that it doesn’t come to happen — it’s going to be hurtful for the Venezuelan people. But I think they have to think twice about what really is at stake and how this is going to affect the stability of the United States. We’re not hundreds or thousands of miles away. We’re really close to the United States. And it could have a terrible effect for its economy, even for its own base. I mean, if you look at the base that supports President Trump right now, a lot of the criticism that is coming is precisely from its own base, who were promised that there would be no more forever wars.’ -Carlos Ron on the latest episode of Going Underground FULL INTERVIEW BELOW IN THE REPLIES👇

Going Underground

58,089 views • 7 months ago

‘IS GOD IS’ director Aleshea Harris discusses the themes behind the film: “I know the rules for being a Black woman and a Black girl, which is that the dictates of respectability are that you have to conduct yourself in a certain way, you have to hold things down, and you shouldn't show your anger. And I think that that's dehumanizing, which is part of the reason that I allowed these women the full measure of their humanity. Allowed them to get messy, allowed them their rage, which for me was restorative and healing, to say, 'No, actually, you have reasons to be angry, and you can express that anger. It's a perfectly human emotion.' So I knew that that line was there, and I was intentionally stepping over it.”
0:47

‘IS GOD IS’ director Aleshea Harris discusses the themes behind the film: “I know the rules for being a Black woman and a Black girl, which is that the dictates of respectability are that you have to conduct yourself in a certain way, you have to hold things down, and you shouldn't show your anger. And I think that that's dehumanizing, which is part of the reason that I allowed these women the full measure of their humanity. Allowed them to get messy, allowed them their rage, which for me was restorative and healing, to say, 'No, actually, you have reasons to be angry, and you can express that anger. It's a perfectly human emotion.' So I knew that that line was there, and I was intentionally stepping over it.”

Film Updates

178,039 views • 2 months ago

Rick Rubin tells Andrew Huberman how he deals with creative or writer’s block. He treats his work like a diary entry (and doesn’t worry about internal or external judgment): ➡️ “What's the cause of the block? The block is usually something that's either personal ("I'm not good enough") or it can be a confidence issue ("I don't have anything to say") or it could be...thinking about someone else ("nobody's going to like what I make"). Do you know what I'm saying? So, it's either fear of self-judgment or external judgment. If you're making something with a freedom of "this is something I'm making for myself for now", that is all [you have to do]. It is a diary entry. Everything I make is a diary entry. The beauty of a diary entry is that I can write my diary entry and you can't tell me that my diary entry wasn't good enough. Or that [the diary entry] is not what I experienced. Of course it's what I experienced: I'm writing a personal diary for myself and no one else can judge if it is my experience of my life. Everything we make can be that: a personal reflection of who we are in that moment of time. It doesn't have to be the greatest you could ever do. It doesn't have to have any expectation that it's going to change the world. It doesn't have to sell a certain number of copies for any reason. It doesn't have any of those things at all. It is "I'm making this thing for me and I want to do it to the best of my ability and to where I feel good about it". [The work] is honest of where I'm at and if you're living in this world of just being honest to where you're at, there's nothing blocking you. There are no blocks. The blocks are all based on dealing with a different force or a different perception that is made up.” ⬅️
2:34

Rick Rubin tells Andrew Huberman how he deals with creative or writer’s block. He treats his work like a diary entry (and doesn’t worry about internal or external judgment): ➡️ “What's the cause of the block? The block is usually something that's either personal ("I'm not good enough") or it can be a confidence issue ("I don't have anything to say") or it could be...thinking about someone else ("nobody's going to like what I make"). Do you know what I'm saying? So, it's either fear of self-judgment or external judgment. If you're making something with a freedom of "this is something I'm making for myself for now", that is all [you have to do]. It is a diary entry. Everything I make is a diary entry. The beauty of a diary entry is that I can write my diary entry and you can't tell me that my diary entry wasn't good enough. Or that [the diary entry] is not what I experienced. Of course it's what I experienced: I'm writing a personal diary for myself and no one else can judge if it is my experience of my life. Everything we make can be that: a personal reflection of who we are in that moment of time. It doesn't have to be the greatest you could ever do. It doesn't have to have any expectation that it's going to change the world. It doesn't have to sell a certain number of copies for any reason. It doesn't have any of those things at all. It is "I'm making this thing for me and I want to do it to the best of my ability and to where I feel good about it". [The work] is honest of where I'm at and if you're living in this world of just being honest to where you're at, there's nothing blocking you. There are no blocks. The blocks are all based on dealing with a different force or a different perception that is made up.” ⬅️

Trung Phan

1,619,840 views • 2 years ago

"One of the very confusing things about the models right now: how to reconcile the fact that they are doing so well on evals. And you look at the evals and you go, 'Those are pretty hard evals.' But the economic impact seems to be dramatically behind. There is [a possible] explanation. Back when people were doing pre-training, the question of what data to train on was answered, because that answer was everything. So you don't have to think if it's going to be this data or that data. When people do RL training, they say, 'Okay, we want to have this kind of RL training for this thing and that kind of RL training for that thing.' You say, 'Hey, I would love our model to do really well when we release it. I want the evals to look great. What would be RL training that could help on this task?' If you combine this with generalization of the models actually being inadequate, that has the potential to explain a lot of what we are seeing, this disconnect between eval performance and actual real-world performance"
2:21

"One of the very confusing things about the models right now: how to reconcile the fact that they are doing so well on evals. And you look at the evals and you go, 'Those are pretty hard evals.' But the economic impact seems to be dramatically behind. There is [a possible] explanation. Back when people were doing pre-training, the question of what data to train on was answered, because that answer was everything. So you don't have to think if it's going to be this data or that data. When people do RL training, they say, 'Okay, we want to have this kind of RL training for this thing and that kind of RL training for that thing.' You say, 'Hey, I would love our model to do really well when we release it. I want the evals to look great. What would be RL training that could help on this task?' If you combine this with generalization of the models actually being inadequate, that has the potential to explain a lot of what we are seeing, this disconnect between eval performance and actual real-world performance"

Dwarkesh Patel

502,249 views • 8 months ago

"of course i also want to be a mentor who can help them as much as i can on the parts that they need help. but more than that, even if it's not for a song, dance, or performance, i want to help them on other various areas and be remembered as a warm mentor" #방예담 #BANGYEDAM #BTN9DMediaLaunch vcr. philconcerts
0:32

"of course i also want to be a mentor who can help them as much as i can on the parts that they need help. but more than that, even if it's not for a song, dance, or performance, i want to help them on other various areas and be remembered as a warm mentor" #방예담 #BANGYEDAM #BTN9DMediaLaunch vcr. philconcerts

여우담 🦊

26,208 views • 1 year ago

Bam Adebayo GOES IN on the critics that says his 83 point game performance is unethical: "For the couch coaches, I mean, if you're in my shoes and you have, first of all, y'all are blaming me. You should be blaming the head coach. Get that first. I was not the one letting me go one-on-one the whole game until I had 70, and then you started to send a double. At that point, I got 70 with, like, what? nine minutes left to go in the game you think i'm not going for it like like and that's the thing that's crazy when they talk about the unethical part of the basketball i'm like if i have 70 points with 9 minutes to go Who would just be like, you know, coach, just take me out. Yeah, right. Anybody in my shoes with nine minutes left? Okay. A minute? All right. Nine? Yeah, I'm going for it. You can't be mad at that. If you are mad, I don't care because a lot of people, they're upset because if they did play, they never had a chance to get that close to chasing greatness. And then if you get that close to chasing greatness, that's the point of chasing it so you can surpass it. And some of the people have never played basketball. So like if you've been in the backyard and you and a couple of your homies have been playing 21 and you got 19. You're not going to get an easy look off. And four, they're going to talk about the free throws. It's not like I shoot 15 free throws a game. It's not like I average 10 free throws a game. You can watch the film. I was legitimately getting fouled every time. So I went to the free throw line."
1:51

Bam Adebayo GOES IN on the critics that says his 83 point game performance is unethical: "For the couch coaches, I mean, if you're in my shoes and you have, first of all, y'all are blaming me. You should be blaming the head coach. Get that first. I was not the one letting me go one-on-one the whole game until I had 70, and then you started to send a double. At that point, I got 70 with, like, what? nine minutes left to go in the game you think i'm not going for it like like and that's the thing that's crazy when they talk about the unethical part of the basketball i'm like if i have 70 points with 9 minutes to go Who would just be like, you know, coach, just take me out. Yeah, right. Anybody in my shoes with nine minutes left? Okay. A minute? All right. Nine? Yeah, I'm going for it. You can't be mad at that. If you are mad, I don't care because a lot of people, they're upset because if they did play, they never had a chance to get that close to chasing greatness. And then if you get that close to chasing greatness, that's the point of chasing it so you can surpass it. And some of the people have never played basketball. So like if you've been in the backyard and you and a couple of your homies have been playing 21 and you got 19. You're not going to get an easy look off. And four, they're going to talk about the free throws. It's not like I shoot 15 free throws a game. It's not like I average 10 free throws a game. You can watch the film. I was legitimately getting fouled every time. So I went to the free throw line."

Ahmed/The Ears/IG: BigBizTheGod 🇸🇴

373,876 views • 4 months ago

ELON MUSK: We believe the AI5 chip will be roughly comparable performance to an NVIDIA Blackwell, and at much less than 10% of the cost Transcription: I'm super hardcore on chips right now as you may be able to tell. I have chips on the brain. I dream about chips, Literally! Because in order to have a functional robot, you have to have a great AI chip. And it needs to be an inexpensive chip and it needs to be very power efficient So we think we believe the AI5 chip will be probably about a third of the power of say something like a Blackwell, an NVIDIA Blackwell, which is a great chip, for roughly comparable performance. And much less than 10% of the cost. This is a chip that is very much optimized for the Tesla AI software stack. So it's not meant to be a general purpose chip, it's meant to be an amazing chip for the Tesla AI software And I mean a couple of things that I think make... like how is Tesla able to achieve such an improvement? I think it is because we are specialized. We're not trying to... you know, NVIDIA has to serve the superset of all past and future customers. So all of their requirements, all of the software that they've written has to work, which is a very difficult problem. Whereas we just need to make it work for our software. And so we're able to simplify the chip dramatically And then we also, I think we're unique in this, but like we have an integer-based system. And integer operations are fundamentally more efficient than floating point operations. So we can do floating point, but the vast majority of our inference is done in integer. Which is, if you're familiar with sort of logic gates, the simplicity of integer... it's integer is much more power efficient, much more silicon efficient, but you have to, you actually have to train for integer inference, which everyone else is training for floating point. That's kind of like a niche technical detail, but it's actually very important. So, yeah, this is going to be a great chip So this chip will be made in basically in four places: TSMC Taiwan, Samsung Korea, TSMC Arizona, and TSMC Texas. And we already know what improvements to make for AI6. So I'm hopeful that we can within less than a year of AI5 starting production, we can actually transition in the same fab to AI6 and double all of the performance metrics
3:12

ELON MUSK: We believe the AI5 chip will be roughly comparable performance to an NVIDIA Blackwell, and at much less than 10% of the cost Transcription: I'm super hardcore on chips right now as you may be able to tell. I have chips on the brain. I dream about chips, Literally! Because in order to have a functional robot, you have to have a great AI chip. And it needs to be an inexpensive chip and it needs to be very power efficient So we think we believe the AI5 chip will be probably about a third of the power of say something like a Blackwell, an NVIDIA Blackwell, which is a great chip, for roughly comparable performance. And much less than 10% of the cost. This is a chip that is very much optimized for the Tesla AI software stack. So it's not meant to be a general purpose chip, it's meant to be an amazing chip for the Tesla AI software And I mean a couple of things that I think make... like how is Tesla able to achieve such an improvement? I think it is because we are specialized. We're not trying to... you know, NVIDIA has to serve the superset of all past and future customers. So all of their requirements, all of the software that they've written has to work, which is a very difficult problem. Whereas we just need to make it work for our software. And so we're able to simplify the chip dramatically And then we also, I think we're unique in this, but like we have an integer-based system. And integer operations are fundamentally more efficient than floating point operations. So we can do floating point, but the vast majority of our inference is done in integer. Which is, if you're familiar with sort of logic gates, the simplicity of integer... it's integer is much more power efficient, much more silicon efficient, but you have to, you actually have to train for integer inference, which everyone else is training for floating point. That's kind of like a niche technical detail, but it's actually very important. So, yeah, this is going to be a great chip So this chip will be made in basically in four places: TSMC Taiwan, Samsung Korea, TSMC Arizona, and TSMC Texas. And we already know what improvements to make for AI6. So I'm hopeful that we can within less than a year of AI5 starting production, we can actually transition in the same fab to AI6 and double all of the performance metrics

X Freeze

305,109 views • 9 months ago

Thought this was a very good (and balanced) explanation of xG from David Sumpter "If you're thinking about future performance, xG is a better way of predicting that than the goals you scored. But... "it is not a good idea to get too obsessed with the xG of the match that has just happened. "Once you get up to between 3 and 6 games, you DO want to know the xG. Once you get over half a season, real goals become the important thing. "Goals are reality, xG is a model." Premium Members can watch the full 'Build your own xG model Masterclass' via the link below Twelve Football 🔗
3:58

Thought this was a very good (and balanced) explanation of xG from David Sumpter "If you're thinking about future performance, xG is a better way of predicting that than the goals you scored. But... "it is not a good idea to get too obsessed with the xG of the match that has just happened. "Once you get up to between 3 and 6 games, you DO want to know the xG. Once you get over half a season, real goals become the important thing. "Goals are reality, xG is a model." Premium Members can watch the full 'Build your own xG model Masterclass' via the link below Twelve Football 🔗

Training Ground Guru

31,383 views • 4 months ago