AMD uses Infinity Fabric for the inter-chiplet communication, so those don't actually use PCIe lanes. But when there are two sockets, those will at least in the current implementations require the usage of PCIe lanes (which would essentially encapsulate the Infinity Fabric between the two sockets).
They should just call it USB 4.2 gen 1 for future-proofness so they don't need to rename it first from USB 4.0 to USB 4.1 gen 1 and then to USB 4.2 gen 1.
I had a minilease from Hertz, and it was strictly 30 days at a time. Had I renewed the lease at 29 days, for example, it would have cost me the daily rent instead, which would have been pretty much twice as the 30-days-at-a-time rate. Well, they would not have been upset about renewing / returning early, but they would have charged me A LOT more. (And they were very picky on the time, too - if I renewed at, say, 2:30 PM, then the next renewal had to be after 30 days, latest at 2:30 - and so I had to keep coming at the exact same time in order to not have the returning time shifting earlier and earlier).
You're wrong. Windows is a tradermark - see http://tsdr.uspto.gov/#caseNum.... And before saying that this is only related to the specific logo - no: "Standard Character Claim:
Yes. The mark consists of standard characters without claim to any particular font style, size, or color." And Microsoft does not claim "Microsoft Windows" to be a trademark, even unregistered: https://www.microsoft.com/en-u... (and they don't need to, since that is just two registered trademarks in a row).
Why cant one just cancel the subscription A when not finding enough stuff to watch there and only then paying for subscription B? Unlike with broadcast TV, one can actually have the whole library available at any time, and so I cannot really understand the point of having to have all of the services subscribed all of the time.
Yes, actually the use-it-everywhere-anytime greeting in Finnish, translatable to "hello", is "terve", which literally means "healthy". So I think there used to be the same kind of "How are you?" which was a real question, but then it got dumbed down to "How are you? I'm healthy" and finally just "Healthy" (whether or not you actually are). So apparently at some point there was a piece of small-talk in the language. Disclaimer: I'm not a linguist.
Also... "welcome" translates into Finnish as "tervetuloa", which is also "healthy-coming" - I'm not sure if that is a direct translation of some other language, or if it means "hello-come-in" or if there actually was the same meaning at some time - come in if you're healthy.
And yea, people rarely talk to strangers. I have a friend who said that it was fine to talk to strangers in English (because it is assumed a cultural difference), but when he speaks Finnish with strangers, people get weird (he's not a native speaker, but the reason is likely not the way he speaks Finnish, but just that people don't expect strangers - foreigners or not - to speak in Finnish with them).
https://en.wikipedia.org/wiki/List_of_countries_by_GDP_(PPP) - 2017 figures for GDP EU: 20.85 trillion USD, Germany: 4.15 trillion USD.
https://www.thelocal.de/20150924/what-the-vw-scandal-means-for-germanys-economy - Germany's car industry sales are 14% of Germany's GDP.
Certainly, car industry is very important for Germany. And also certainly, Germany's economy is very important for the EU. But to claim an equality between these is a hyperbole.
That's why Dieselgate was a powerful shock to the whole EU and a "friendly" reminder from president Obama that the US can turn off the EU at any given moment without Europeans being able to do anything about it
Also, a significant portion of the car sales does NOT go to the USA. I also don't understand the claim how a smaller economy can "turn off" a bigger economy without being affected by it themselves.
If a very efficient method for this was found, it would be worth so much more on Earth. The sugar could be used as a relatively high-density, stable, easy-to-transport energy storage - and if not viable directly, then for could be used for example through fermentation to alcohol as well (though I don't know how efficient that process is).
1. Ensure zero mistakes are made, by punishing people who make mistakes, and by other means, or
2. Create a culture where mistakes that are made are correctly reported and dealt with?
Hint: The first one is not possible, because perfection is impossible.
The mistakes should of course be minimized, and intentional incorrect actions, like sabotage, should of course be punished. Of course, you can argue that not reporting a mistake is an intentional incorrect action, but this could be avoided by choosing wisely how mistakes are dealt with. So paradoxally, trying to push for zero mistakes by punishment, you're actually 1) not necessarily reducing mistakes, and 2) additionally causing intentional incorrect actions, thus making the situation worse. If you're a manager, I hope you take some time to consider this.
And the push for getting rid of the time changes has been strongly pushed by Finland after the citizen initiative, which was signed by ~1.2 % of the Finnish population (not as much as almost 5% of the German population participating the EU level one, though, but still, quite likely not just lobbyists and people mobilized by them). And from what I've discussed with various people, most people have been against the clock shifts, whether or not they participated this poll.
I worked with a lot of Finish guys (Nokia, technomen) at the start of my career, even went to finland for a job interview in Jan 2001, but alas, it was not meant to be.
Interesting! I also worked at Nokia with some Spanish people, though they were in a different team and abroad, so I think I only had a chance to see them once, was around 2008 or 2009 maybe:)
What you say abot performance/watt in ultra large scale deployments (Like Amazon, Azure, facebook, Google) is 100% true, and well known, but my deployments are more large scale (Telcos). And also, if I took it to that level, the point would be lost to people who are still thinkig at the level of a gaming desktop, or a tower server for a 5 people company...
Okay, I didn't really know what your target audience is.
As per the security, AMD (and all the others) has similar issues with speculative execution and side channel attacks, some have more issues, some have less issues, is just that Intel gets the bulk of the publicity for the time being, due to their large size.
You're right, however, and the nature of the speculative execution is such that it is quite difficult to get performance from it without sacrificing some of the security. I saw an idea about resetting the state in case of a failed speculation, but that is quite difficult with the different cache levels etc. That said, there has been only a couple of Spectre-like issues affecting AMD CPUs, and the worst one, Meltdown, does not affect them due to differences (speculative execution in AMD products does not bypass privilege boundaries). Intel has been cutting corners for the sake of performance, and they're getting hit by it. So the size is not the only factor for Intel getting the publicity.
Thank you a lot for taking the time to write the reply, this was very informative! This seems to especially apply to data center kind of servers, and not necessarily so much often affect smaller scale servers (although in some cases may).
So there is indeed some truth behind the Intel's claim of their ecosystem, but it is not at least completely about compatibility (though to some extent also that, such as the transactional memory, VMM etc.), but also about stuff being optimized for Intel, and also the supply chain. But indeed highly depending on the use case.
I started to think about this more out-of-the-box and I realized that for very large scale implementations, performance-per-watt might also be an important factor. And I think this also needs benchmarking for the use-case. Adding that to the single-thread performance, it is not a trivial equation at all!
Of course, in the light of the constant security issues discovered with Intel processors, the risk and cost of data leak is also something to keep in mind.
PS: My native language is not English either, but Finnish, and I speak also German, Swedish, some Italian and a tiny bit of French:P Too bad I don't speak any Spanish, and with my level of Italian, it would not be comprehensible to me probably even in writing, even though for native speakers it is possible to an extent...
[--] there are many caveats beyond AMD's control that make it an uphill battle to use their chips.
I barely know anything about the server space, so I'd be interested in hearing what kind of issues there are. Could you shed some light on this? (Special care was tried to be taken to not sound like I'm asking because I doubt your statement - I genuinely am just interested to learn about this)
The chip manufacturers have known about the issues already for more than six months (and also many hardware manufacturers have been aware for quite some time), but they've just kept selling hardware they knew was 1) vulnerable and 2) soon about to become somewhat slower, or much slower (in case of Intel).
Perhaps they've been designing some new hardware based on this, but I don't think they're going to change the current ones, except for shipping with newer microcode (in case of CPUs) or patched software (in case of HW manufacturers). What I believe they will do instead is to launch new products that they then advertise being much faster than the previous one (as the old ones suddenly got slower in comparison), and with extra security (as the previous ones were vulnerable).
Exactly! I have sent Facebook feedback about this several times. But it always is either 1) some algorithm picking stuff for me, showing the same stuff over and over again or 2) the "most recent" which actually is 90% BS. Such as: somebody I don't even know recently commented/liked this post I've seen already five times. Somebody I know recently commented/liked a post/profile picture from somebody I don't even know. Somebody recently posted an ad/video I don't care about. Somebody recently commented on a news that I completely disagree with. Obviously, I'm not happy with either choice, but I finally gave up with the "most recent", as it is even more frustrating than the algorithm picking the top stories. But I wish that finally they are taking a step in the right direction, which is back to an older behaviour.
AMD actually beats intel in AES both in performance-per-watt, in performance-per-dollar and with most CPUs also in raw performance. See for example http://www.anandtech.com/show/... and http://techreport.com/review/3.... Note also that the only Intel offering in these beating the Ryzen 1800x in performance-per-CPU is the i7-6950x, which is more than three times the cost and almost 50% increase in TDP. That said, most of the benchmarks don't have the Intel server offerings included, so perhaps you can prove me wrong?
I think GP referred by "whole" milk exactly that, not whole milk. At least in Finland, the 3.5% milk is labeled as whole milk (or full milk - the word is ambiguous). And I think it is an EU wide thing, since before the EU, it was labelled as "consumption milk". The process for that might vary, but at least in Finland it is done by adding cream or skim milk to the "raw milk". But homogenization is achieved by forcing the milk through very small holes, not by using chemicals (https://en.wikipedia.org/wiki/Homogenization_(chemistry) ). That breaks the fat liposomes to so small units that they won't separate anymore. However, the homogenization does activate or accelerate some chemical processes, which then may alter the chemical composition of the milk as well. Though technically you could say that this is chemically treated fat, it might give the wrong impression of chemicals being used for achieving this.
All that said, I remember seeing some studies and debate that since the fat is broken down into small particles, it modifies how the milk behaves in digestion, and the homogenized milk doesn't create a protective layer on the intestine, which might make people more prone to allergies and digestive problems.
Yes, but in the context of history text books, this gives a wrong impression. Think about a physics text book where it says that atoms were discovered by me in the 1980s. As I learned about atoms during the '80s, technically that is true. But it certainly gives a wrong impression, and nobody in their right mind would print a textbook with such a text.
The whole point of what I am saying is what kind of impression the text books give to the students, not if such interpretation of the text exists that makes the text factually correct.
According to dictionary.com, to discover means "to see, get knowledge of, learn of, find, or find out; gain sight or knowledge of (something previously unseen or unknown): to discover America; to discover electricity." As if America was earlier unknown to the mankind. But Native Americans had been living there tens of thousands of years. (Incidentally, discovering America is used as an example, but there's not enough surrounding data to say in what sense the example is meant - unlike on the history lessons in school).
One could say that from the Europeans' point of view, it was a discovery. But for the history of the world, or the history of mankind, it wasn't a discovery any more than me seeing atoms and claiming that I've been discovering new particles. There is a difference between "Today I discovered a new continent beyond the Atlantic Ocean" (as in, I found something that had not been found earlier) and "Today I discovered that there's a continent beyond the Atlantic Ocean" (as in, I learned something that I didn't know before).
I can't believe your post has 0 points and the GP has +2. I've never understood really why it's taught in the schools that Columbus discovered America. But I guess it's just another case of history being written by the winners.
Slashdot Mirror
AMD uses Infinity Fabric for the inter-chiplet communication, so those don't actually use PCIe lanes. But when there are two sockets, those will at least in the current implementations require the usage of PCIe lanes (which would essentially encapsulate the Infinity Fabric between the two sockets).
They should just call it USB 4.2 gen 1 for future-proofness so they don't need to rename it first from USB 4.0 to USB 4.1 gen 1 and then to USB 4.2 gen 1.
I had a minilease from Hertz, and it was strictly 30 days at a time. Had I renewed the lease at 29 days, for example, it would have cost me the daily rent instead, which would have been pretty much twice as the 30-days-at-a-time rate. Well, they would not have been upset about renewing / returning early, but they would have charged me A LOT more. (And they were very picky on the time, too - if I renewed at, say, 2:30 PM, then the next renewal had to be after 30 days, latest at 2:30 - and so I had to keep coming at the exact same time in order to not have the returning time shifting earlier and earlier).
... for the already rich
Imagine someone saying that at the time of 8086, would kind of sound silly in retrospect, wouldn't it?
You're wrong. Windows is a tradermark - see http://tsdr.uspto.gov/#caseNum.... And before saying that this is only related to the specific logo - no: "Standard Character Claim: Yes. The mark consists of standard characters without claim to any particular font style, size, or color." And Microsoft does not claim "Microsoft Windows" to be a trademark, even unregistered: https://www.microsoft.com/en-u... (and they don't need to, since that is just two registered trademarks in a row).
Why cant one just cancel the subscription A when not finding enough stuff to watch there and only then paying for subscription B? Unlike with broadcast TV, one can actually have the whole library available at any time, and so I cannot really understand the point of having to have all of the services subscribed all of the time.
Yes, actually the use-it-everywhere-anytime greeting in Finnish, translatable to "hello", is "terve", which literally means "healthy". So I think there used to be the same kind of "How are you?" which was a real question, but then it got dumbed down to "How are you? I'm healthy" and finally just "Healthy" (whether or not you actually are). So apparently at some point there was a piece of small-talk in the language. Disclaimer: I'm not a linguist.
Also... "welcome" translates into Finnish as "tervetuloa", which is also "healthy-coming" - I'm not sure if that is a direct translation of some other language, or if it means "hello-come-in" or if there actually was the same meaning at some time - come in if you're healthy.
The bus waiting is exactly the same in Finland:
https://www.mtv.fi/uutiset/kotimaa/artikkeli/aamuruuhka-nain-pidetaan-turvavali-bussipysakilla-katso-kuva/1845498#gs.0CMrYBc
And yea, people rarely talk to strangers. I have a friend who said that it was fine to talk to strangers in English (because it is assumed a cultural difference), but when he speaks Finnish with strangers, people get weird (he's not a native speaker, but the reason is likely not the way he speaks Finnish, but just that people don't expect strangers - foreigners or not - to speak in Finnish with them).
So the comparison fails - and none of those three points apply to DNT. It would have been better for example to compare DNT to homeopathy instead.
Europe's economy IS German economy.
https://en.wikipedia.org/wiki/List_of_countries_by_GDP_(PPP) - 2017 figures for GDP EU: 20.85 trillion USD, Germany: 4.15 trillion USD.
https://www.thelocal.de/20150924/what-the-vw-scandal-means-for-germanys-economy - Germany's car industry sales are 14% of Germany's GDP.
Certainly, car industry is very important for Germany. And also certainly, Germany's economy is very important for the EU. But to claim an equality between these is a hyperbole.
That's why Dieselgate was a powerful shock to the whole EU and a "friendly" reminder from president Obama that the US can turn off the EU at any given moment without Europeans being able to do anything about it
Also, a significant portion of the car sales does NOT go to the USA. I also don't understand the claim how a smaller economy can "turn off" a bigger economy without being affected by it themselves.
Yes, exactly - that's why I wrote that "if not viable directly". But there is at least some research also on glucose fuel cells.
If a very efficient method for this was found, it would be worth so much more on Earth. The sugar could be used as a relatively high-density, stable, easy-to-transport energy storage - and if not viable directly, then for could be used for example through fermentation to alcohol as well (though I don't know how efficient that process is).
Which one do you think is better:
1. Ensure zero mistakes are made, by punishing people who make mistakes, and by other means, or
2. Create a culture where mistakes that are made are correctly reported and dealt with?
Hint: The first one is not possible, because perfection is impossible.
The mistakes should of course be minimized, and intentional incorrect actions, like sabotage, should of course be punished. Of course, you can argue that not reporting a mistake is an intentional incorrect action, but this could be avoided by choosing wisely how mistakes are dealt with. So paradoxally, trying to push for zero mistakes by punishment, you're actually 1) not necessarily reducing mistakes, and 2) additionally causing intentional incorrect actions, thus making the situation worse. If you're a manager, I hope you take some time to consider this.
And the push for getting rid of the time changes has been strongly pushed by Finland after the citizen initiative, which was signed by ~1.2 % of the Finnish population (not as much as almost 5% of the German population participating the EU level one, though, but still, quite likely not just lobbyists and people mobilized by them). And from what I've discussed with various people, most people have been against the clock shifts, whether or not they participated this poll.
I worked with a lot of Finish guys (Nokia, technomen) at the start of my career, even went to finland for a job interview in Jan 2001, but alas, it was not meant to be.
Interesting! I also worked at Nokia with some Spanish people, though they were in a different team and abroad, so I think I only had a chance to see them once, was around 2008 or 2009 maybe :)
What you say abot performance/watt in ultra large scale deployments (Like Amazon, Azure, facebook, Google) is 100% true, and well known, but my deployments are more large scale (Telcos). And also, if I took it to that level, the point would be lost to people who are still thinkig at the level of a gaming desktop, or a tower server for a 5 people company...
Okay, I didn't really know what your target audience is.
As per the security, AMD (and all the others) has similar issues with speculative execution and side channel attacks, some have more issues, some have less issues, is just that Intel gets the bulk of the publicity for the time being, due to their large size.
You're right, however, and the nature of the speculative execution is such that it is quite difficult to get performance from it without sacrificing some of the security. I saw an idea about resetting the state in case of a failed speculation, but that is quite difficult with the different cache levels etc. That said, there has been only a couple of Spectre-like issues affecting AMD CPUs, and the worst one, Meltdown, does not affect them due to differences (speculative execution in AMD products does not bypass privilege boundaries). Intel has been cutting corners for the sake of performance, and they're getting hit by it. So the size is not the only factor for Intel getting the publicity.
Thank you a lot for taking the time to write the reply, this was very informative! This seems to especially apply to data center kind of servers, and not necessarily so much often affect smaller scale servers (although in some cases may).
So there is indeed some truth behind the Intel's claim of their ecosystem, but it is not at least completely about compatibility (though to some extent also that, such as the transactional memory, VMM etc.), but also about stuff being optimized for Intel, and also the supply chain. But indeed highly depending on the use case.
I started to think about this more out-of-the-box and I realized that for very large scale implementations, performance-per-watt might also be an important factor. And I think this also needs benchmarking for the use-case. Adding that to the single-thread performance, it is not a trivial equation at all!
Of course, in the light of the constant security issues discovered with Intel processors, the risk and cost of data leak is also something to keep in mind.
PS: My native language is not English either, but Finnish, and I speak also German, Swedish, some Italian and a tiny bit of French :P Too bad I don't speak any Spanish, and with my level of Italian, it would not be comprehensible to me probably even in writing, even though for native speakers it is possible to an extent...
[--] there are many caveats beyond AMD's control that make it an uphill battle to use their chips.
I barely know anything about the server space, so I'd be interested in hearing what kind of issues there are. Could you shed some light on this? (Special care was tried to be taken to not sound like I'm asking because I doubt your statement - I genuinely am just interested to learn about this)
The chip manufacturers have known about the issues already for more than six months (and also many hardware manufacturers have been aware for quite some time), but they've just kept selling hardware they knew was 1) vulnerable and 2) soon about to become somewhat slower, or much slower (in case of Intel).
Perhaps they've been designing some new hardware based on this, but I don't think they're going to change the current ones, except for shipping with newer microcode (in case of CPUs) or patched software (in case of HW manufacturers). What I believe they will do instead is to launch new products that they then advertise being much faster than the previous one (as the old ones suddenly got slower in comparison), and with extra security (as the previous ones were vulnerable).
Exactly! I have sent Facebook feedback about this several times. But it always is either 1) some algorithm picking stuff for me, showing the same stuff over and over again or 2) the "most recent" which actually is 90% BS. Such as: somebody I don't even know recently commented/liked this post I've seen already five times. Somebody I know recently commented/liked a post/profile picture from somebody I don't even know. Somebody recently posted an ad/video I don't care about. Somebody recently commented on a news that I completely disagree with. Obviously, I'm not happy with either choice, but I finally gave up with the "most recent", as it is even more frustrating than the algorithm picking the top stories. But I wish that finally they are taking a step in the right direction, which is back to an older behaviour.
AMD actually beats intel in AES both in performance-per-watt, in performance-per-dollar and with most CPUs also in raw performance. See for example http://www.anandtech.com/show/... and http://techreport.com/review/3.... Note also that the only Intel offering in these beating the Ryzen 1800x in performance-per-CPU is the i7-6950x, which is more than three times the cost and almost 50% increase in TDP. That said, most of the benchmarks don't have the Intel server offerings included, so perhaps you can prove me wrong?
I think GP referred by "whole" milk exactly that, not whole milk. At least in Finland, the 3.5% milk is labeled as whole milk (or full milk - the word is ambiguous). And I think it is an EU wide thing, since before the EU, it was labelled as "consumption milk". The process for that might vary, but at least in Finland it is done by adding cream or skim milk to the "raw milk". But homogenization is achieved by forcing the milk through very small holes, not by using chemicals (https://en.wikipedia.org/wiki/Homogenization_(chemistry) ). That breaks the fat liposomes to so small units that they won't separate anymore. However, the homogenization does activate or accelerate some chemical processes, which then may alter the chemical composition of the milk as well. Though technically you could say that this is chemically treated fat, it might give the wrong impression of chemicals being used for achieving this.
All that said, I remember seeing some studies and debate that since the fat is broken down into small particles, it modifies how the milk behaves in digestion, and the homogenized milk doesn't create a protective layer on the intestine, which might make people more prone to allergies and digestive problems.
Yes, but in the context of history text books, this gives a wrong impression. Think about a physics text book where it says that atoms were discovered by me in the 1980s. As I learned about atoms during the '80s, technically that is true. But it certainly gives a wrong impression, and nobody in their right mind would print a textbook with such a text.
The whole point of what I am saying is what kind of impression the text books give to the students, not if such interpretation of the text exists that makes the text factually correct.
According to dictionary.com, to discover means "to see, get knowledge of, learn of, find, or find out; gain sight or knowledge of (something previously unseen or unknown): to discover America; to discover electricity." As if America was earlier unknown to the mankind. But Native Americans had been living there tens of thousands of years. (Incidentally, discovering America is used as an example, but there's not enough surrounding data to say in what sense the example is meant - unlike on the history lessons in school).
One could say that from the Europeans' point of view, it was a discovery. But for the history of the world, or the history of mankind, it wasn't a discovery any more than me seeing atoms and claiming that I've been discovering new particles. There is a difference between "Today I discovered a new continent beyond the Atlantic Ocean" (as in, I found something that had not been found earlier) and "Today I discovered that there's a continent beyond the Atlantic Ocean" (as in, I learned something that I didn't know before).
I can't believe your post has 0 points and the GP has +2. I've never understood really why it's taught in the schools that Columbus discovered America. But I guess it's just another case of history being written by the winners.