The benchmark today is 3.0GHz Interlagos processors with 16 cores per socket, up to 4 sockets per motherboard, for a total of 64 AMD64 cores per node at 3GHz in 2U, or 1.5x better in blades. And of course GPGPU on top of that. Itanic may have some internal efficiencies, but it can't compete against that compute per RU on any benchmark I know of.
It's got some resilience, but frankly we've moved that feature to software. It's got some scalability, but FDR Infiniband is solving that problem too.
Itanic is toast. The clock doesn't ramp, and we knew that seven years ago - it was provable then, and it's proven now.
HP and Intel will bring out new chips for a few years because they promised to. It's a losing proposition now, but they promised and a deal's a deal. By eating some losses here HP and Intel gain some trust in the enterprise market where they have other products to sell, so it maths out financially.
Science isn't about winning or losing. So when the folks at OPERA had some interesting findings about superluminal neutrinos they said "Hey, guys, this is interesting - our timers show stuff moving at greater than c. Help us investigate: here's our methods and data.
But climatologists are all about preventing inspection of their methods and data to prevent political exploitation of it. So that's how it is.
The former are scientists. The latter are politicians. Accept your role and get to winning. If you think social change is required to save the Earth from your predicted apocalypse you have two options: 1) persuasion. 2) force. Since apparently you don't have force yet because you've not persuaded men with guns nor the people who lead them, persuasion of the common man is the tool you have to induce change. And at that you're failing miserably because you're unloveable. You could try not being a bunch of pricks. The science isn't about winning or losing, but you're not doing science, so that's OK.
You'll find that the serial publication of reference material goes back to the Library of Alexandria at least - but I have no doubt these folks hold patents that will prevent general use today. We've institutionalized being retarded. The progress we see now is retrograde, and accelerating.
Keep telling yourself that if it makes losing more comfortable. It's that the fossil fuel industry is out to get you, that the public is a mass of morons. It's not about scientific integrity, credibility, or not being a prick to people you're trying to convince. Stay with that. It's working Grrrrreat.
You guys can't even be coherent in your messaging, but can't fail at being condescending even as AC. You just don't get it. You've seized control of your little corner of academia and think that entitles you to rule the world. Well, it doesn't, and you're a fine example of TFA's point: a bunch of folk posting on blogs and well-placed news articles isn't going to sway the mass of people. That used to work, back in 2002. New age, new rules, new tools.
You could win if you weren't a bunch of pricks, if you understood how to do this in the current day. But I'm not going to help you.
Have you ever used it. Dont brag, I am an iPhone lover, but WP7 is good. Used all three platform now . No doubt WP7 is better. I am not a fanboy of anyone though. Try reading the reviews you will understand why eveyone is behind WP7. I think Microsoft has learned a lesson and turned around.
- A verbatim reply to one of my recent posts somewhere else. Hmm... 7 cents for that? They overpaid.
No mod points on hand today, so all I can give is kudos.
There's a dual-core HP AMD C-50 laptop out there on Black Friday special for $200. It's got 2GB and a 250GB HDD, and the usual stuff - even a webcam. If experience holds, that will be within 10% of the retail price in January. Over at Best Buy you can get a Lenovo one with a E-300 for $179. That's more than enough for most people - unless it's loaded down with crudware. I would juice it up with an SSD and RAM upgrade from Newegg, but it'll do.
I think the big part of the genius of the iPad was the timing: they struck just as the power of the ARM platform came to be "good enough for general use" - and had pre-built an ecosystem for the day. The new Tegra 3 Android tablets with 1080p HD should just fly off the shelves.
The PC is good enough. Now we need to work on making it thin and light and run all day on a battery.
Itanium predates AMD64 and was one way out of the 4GB limit. PAE and AMD64 were two others. That the clock would never ramp was provable in 2004, but by then it was too late. The promises were already made. Now we've got this immense chip that sucks watts, runs slow, but is very reliable. It will see two more revs before it goes into the dark - and a new enterprise CPU architecture will take its place.
HP and Intel don't want to support Itanium either. But if they start breaking their promises with these customers, their name is mud. Oracle doesn't seem to have a problem with that, but HP and Intel do.
Oracle agreed under contract to support this platform on their products. They got good valuable consideration for that. Now they don't want to hold up their end. Well that's too bad. A deal is a deal.
I have no pity for either HP or Intel on this one - they're taking a bath with Itanic, as some of us said they would 7 years ago and more. But at least they're not having to be sued to keep their promises.
In February Intel bout McAfee for about $7.68 billion. So in a sense no, Intel's not next. It's Intel's turn now. Considering this, the Window 8 on ARM stuff and some other things Intel's got to be just about sick of Microsoft right now.
AMD market cap today is $3.81B. I was already giving them a 30% premium over the existing market cap at $5B. Do you really think it would take a 160 percent premium to close the deal? That seems high. Though AMD's market cap does seem insanely low right now. I can't believe those guys got $8B for Skype, and AMD isn't half that on the market.
On the other hand for what Apple's paying for Intel chips Apple could just buy AMD and fix their supply chain problems. AMD could be had for about $5 billion today. Apple's moving about 16 million Macs a year. It wouldn't take too long for that to pay off. And 64-core Mac desktops would be pretty neat.
The Tegra 3 chip that's showing up in phones this spring and Transformer Prime tablet now is about 7.2 GFLOPs. That's more than enough to be top 10 in 1993. Current ARM architectures might go all the way up to fast enough to take that number one spot in reference sample designs now but they consume too much power to go in your pocket on retail shelves as yet. Maybe in a year or two.
Mali T658 and PowerVR are two to watch here. Mali is supposed to go up to 350 GFLOPs. It still amazes me that in 1993 that machine cost about $70 million in today's money and you can almost match it today for under $500.
Slashdot Mirror
The benchmark today is 3.0GHz Interlagos processors with 16 cores per socket, up to 4 sockets per motherboard, for a total of 64 AMD64 cores per node at 3GHz in 2U, or 1.5x better in blades. And of course GPGPU on top of that. Itanic may have some internal efficiencies, but it can't compete against that compute per RU on any benchmark I know of.
It's got some resilience, but frankly we've moved that feature to software. It's got some scalability, but FDR Infiniband is solving that problem too.
Itanic is toast. The clock doesn't ramp, and we knew that seven years ago - it was provable then, and it's proven now.
HP and Intel will bring out new chips for a few years because they promised to. It's a losing proposition now, but they promised and a deal's a deal. By eating some losses here HP and Intel gain some trust in the enterprise market where they have other products to sell, so it maths out financially.
Science isn't about winning or losing. So when the folks at OPERA had some interesting findings about superluminal neutrinos they said "Hey, guys, this is interesting - our timers show stuff moving at greater than c. Help us investigate: here's our methods and data.
But climatologists are all about preventing inspection of their methods and data to prevent political exploitation of it. So that's how it is.
The former are scientists. The latter are politicians. Accept your role and get to winning. If you think social change is required to save the Earth from your predicted apocalypse you have two options: 1) persuasion. 2) force. Since apparently you don't have force yet because you've not persuaded men with guns nor the people who lead them, persuasion of the common man is the tool you have to induce change. And at that you're failing miserably because you're unloveable. You could try not being a bunch of pricks. The science isn't about winning or losing, but you're not doing science, so that's OK.
It is pitch black. You are likely to be eaten by a grue.
We have that. It's called "Wikipedia".
Fiat money.
You'll find that the serial publication of reference material goes back to the Library of Alexandria at least - but I have no doubt these folks hold patents that will prevent general use today. We've institutionalized being retarded. The progress we see now is retrograde, and accelerating.
And that will put an end to evolution of this.
Of course, if you turn off your phone you can't check prices and reviews online, can you?
May as well just go online in the first place and avoid the pushy crowds.
Keep telling yourself that if it makes losing more comfortable. It's that the fossil fuel industry is out to get you, that the public is a mass of morons. It's not about scientific integrity, credibility, or not being a prick to people you're trying to convince. Stay with that. It's working Grrrrreat.
No, this is new. AMD has got some sizzle to announce in February. That wasn't in last week's stuff. Is there a steak behind the sizzle? We'll see.
You guys can't even be coherent in your messaging, but can't fail at being condescending even as AC. You just don't get it. You've seized control of your little corner of academia and think that entitles you to rule the world. Well, it doesn't, and you're a fine example of TFA's point: a bunch of folk posting on blogs and well-placed news articles isn't going to sway the mass of people. That used to work, back in 2002. New age, new rules, new tools.
You could win if you weren't a bunch of pricks, if you understood how to do this in the current day. But I'm not going to help you.
Thanks. Maybe I'll do a few more. It's a good idle amusement.
It's working for global warming. I mean climate change. No, enhanced weather variability. Wait, let me start again.
Have you ever used it. Dont brag, I am an iPhone lover, but WP7 is good. Used all three platform now . No doubt WP7 is better. I am not a fanboy of anyone though. Try reading the reviews you will understand why eveyone is behind WP7. I think Microsoft has learned a lesson and turned around.
- A verbatim reply to one of my recent posts somewhere else. Hmm... 7 cents for that? They overpaid.
</sarcasm>
You forgot to close your tag. You would think the AI folks could at least teach the bots how to html.
No mod points on hand today, so all I can give is kudos.
There's a dual-core HP AMD C-50 laptop out there on Black Friday special for $200. It's got 2GB and a 250GB HDD, and the usual stuff - even a webcam. If experience holds, that will be within 10% of the retail price in January. Over at Best Buy you can get a Lenovo one with a E-300 for $179. That's more than enough for most people - unless it's loaded down with crudware. I would juice it up with an SSD and RAM upgrade from Newegg, but it'll do.
I think the big part of the genius of the iPad was the timing: they struck just as the power of the ARM platform came to be "good enough for general use" - and had pre-built an ecosystem for the day. The new Tegra 3 Android tablets with 1080p HD should just fly off the shelves.
The PC is good enough. Now we need to work on making it thin and light and run all day on a battery.
Requiem Itanium: the clock won't ramp.
Itanium predates AMD64 and was one way out of the 4GB limit. PAE and AMD64 were two others. That the clock would never ramp was provable in 2004, but by then it was too late. The promises were already made. Now we've got this immense chip that sucks watts, runs slow, but is very reliable. It will see two more revs before it goes into the dark - and a new enterprise CPU architecture will take its place.
HP and Intel don't want to support Itanium either. But if they start breaking their promises with these customers, their name is mud. Oracle doesn't seem to have a problem with that, but HP and Intel do.
Oracle agreed under contract to support this platform on their products. They got good valuable consideration for that. Now they don't want to hold up their end. Well that's too bad. A deal is a deal.
I have no pity for either HP or Intel on this one - they're taking a bath with Itanic, as some of us said they would 7 years ago and more. But at least they're not having to be sued to keep their promises.
In February Intel bout McAfee for about $7.68 billion. So in a sense no, Intel's not next. It's Intel's turn now. Considering this, the Window 8 on ARM stuff and some other things Intel's got to be just about sick of Microsoft right now.
Whatever it is, AMD is up to something new, and will announce in February.
AMD market cap today is $3.81B. I was already giving them a 30% premium over the existing market cap at $5B. Do you really think it would take a 160 percent premium to close the deal? That seems high. Though AMD's market cap does seem insanely low right now. I can't believe those guys got $8B for Skype, and AMD isn't half that on the market.
On the other hand for what Apple's paying for Intel chips Apple could just buy AMD and fix their supply chain problems. AMD could be had for about $5 billion today. Apple's moving about 16 million Macs a year. It wouldn't take too long for that to pay off. And 64-core Mac desktops would be pretty neat.
The Tegra 3 chip that's showing up in phones this spring and Transformer Prime tablet now is about 7.2 GFLOPs. That's more than enough to be top 10 in 1993. Current ARM architectures might go all the way up to fast enough to take that number one spot in reference sample designs now but they consume too much power to go in your pocket on retail shelves as yet. Maybe in a year or two.
Mali T658 and PowerVR are two to watch here. Mali is supposed to go up to 350 GFLOPs. It still amazes me that in 1993 that machine cost about $70 million in today's money and you can almost match it today for under $500.
A picture of the Cray 1, for reference: Computer furniture.