That's correct, AMD has announced that they support OpenCL as the GPGPU language. And OpenCL is an open standard as opposed to CUDA (NVidia proprietary).
Doc Searls (editor of Linux Journal) is working on such a donation system as part of his "Vendor Relationship Management" or VRM project at the Berkman Center at Harvard. The idea is to be able to make small voluntary donations to the software author, or more generally the creator of any piece of work. The goal is make this easy -- simple click of a button that says "donate $5" and put you in control of how much of your personal information (name, credit card details etc.) you want the recipient to know.
Great site! I would love to work on some of those projects (I have the necessary HW knowledge), but the commercial tools are horrendously expensive (e.g. VCS from Synopsys) Sure, there are a few open source tools (Icarus Verilog, gEDA etc.), but they are nowhere close to the commercial tools in terms of capabilities.
Besides, you need serious cash in order to get your chip fabricated! You have to have some big company pick up your design and fab them in volume. IIRC, some company is now selling chips (SoC) based on the OR1K design... but that is the only instance I know of a chip actually getting fabbed out of all the projects listed on OpenCores.
Nope, this is still the same old x86 architecture we all love. What they mean by a "new architecture" is that they are ditching the Pentium-IV micro-architecture and developing a new one based on the Pentium-M micro-architecture (itself loosely based on the Pentium-!!! design). As a result, the new chips promise to deliver higher performance at lower power levels.
For a detailed, albeit somewhat biased, story of Go, read Jerry Kaplan's book: "Startup : A Silicon Valley Adventure". It makes an interesting read, and in the book he even mentions that Microsoft was copying some of the interfaces and functionalities of PenPoint. Go apparently had discussions with Microsoft under NDA, and Jerry alleges that Microsoft stole their ideas and violated the NDA.
That is not true at all. The size of instructions is still 32-bits for all 64-bit RISC CPUs and the instruction size is the same in x86-64 as that of regular x86 (CISC). Now 64-bit CPUs give you the *option* to use 64-bit data/pointers, but you are free to continue using 32/16/8 bit data as you deem appropriate. Code/data size doesn't automatically increase.
Also, you may not necessarily be doing twice the work with a 64-bit CPU - it again depends on your compiler. In theory you *can* do two 32-bit operations in parallel using a 64-bit EX/FP unit if the ISA provides for packed operations (also known as SIMD instructions).
I agree that they are not being overly ambitious, and that is a *good thing*. But instead of doing AGP/PCI they should do PCI-e. The industry is pretty much standardizing around PCI-e, and AGP may not exist on most boards when this thing comes out.
1. It is an easier upgrade path for customers. I think Sun learnt that it is easier to sell its customers incremental upgrades than to sell them brand new designs. Remember that the market they sell to (telco, financial) absolutely despises having to test all their mission-critical applications on new, unproven hardware. So while the slow backplane is a performance limitation, many customers may prefer stability to cutting-edge performance.
2. Wait for the 'Zones' in Solaris 10... I've heard it is better than anything IBM & HP have to offer.
Exactly my thoughts! Can we include something in Slashcode that automatically filters the GNAA and goatse trolls? Perhaps as a user-configurable option.
Some Bayesian approach ought to do it... I wouldn't want jokes based on the "$PROJECT is dying - Netcraft confirms it!" troll to be filtered out!
As an Indian, and a big Simpsons fan, I can answer that question right here.
Apu represents the stereotypical image of an Indian. I personally find it quite amusing, others may have a different opinion. Of course, like any other stereotype, not everyone fits that description. What makes India so interesting is the different subcultures within the country (these differences are more pronounced than the subculture differences in the US), people skeak different languages in different states, different types of food and in some cases different types of clothes as well. So Apu cannot represent a typical Indian... there is no such thing as a typical Indian.
Apu & Manjula are indeed Indian names - not very common but not uncommon either. So are the names of all their kids... I never seem to remember them all. The last name however makes no sense! But, somebody obviously did some research in coming up with these characters.
Looks like MS doesn't want two different 64-bit x86 extensions. I'm pretty sure Intel has cross-licensing agreements with AMD that will allow it to use AMD's x86-64 extensions. 'Prescott' may already have it... may not be enabled though.
They can still maintain backwards compatibility if they think the market will demand it.
What prevents them from bolting on the current XBox processor (some P3 derivative IIRC) alongside the new IBM chip and having the hardware funnel the instructions to the appropriate processor depending on the version of the game being played? So an XBox1 game will run on the old processor whereas an XBox2 game can run on the new one - everything else can be shared (memory etc.) Again all this will depend on results from their "market surveys".
First, UltraSparc IV will be a Out-of-Order CPU. Any comparisson with the In-Order UltraSparc III ends here.
Read this article at ExtremeTech. It states that UltraSPARC IV will have 2 UltraSPARC III cores. If the UltraSPARC III are in-order, how can
UltraSPARC IV be out-of-order?
Besides how does it matter if the core is in-order or not? All I want is kick-ass app performance - I don't care if you do in-order, out-of-order or even get a group of chimps to run through the instructions.
AMD did not implement the full 64-bit virtual address (neither does Itanium2). The Opteron has *only* 48-bit virtual address and 40-bit physical address. That means it can address upto 256TB of virtual space and 1TB physical space.
Slashdot Mirror
More importantly, to speak of Raytheon's networking and security capabilities, they acquired BBN a few years ago.
Take a look at the Smartbook by Always Innovating: http://www.alwaysinnovating.com/products/smartbook.htm
It can run Ubuntu, Google Chrome or Android.
Yeah, I don't like this 'nose-longer'. I have trouble keeping it in place when I try to french kiss over the Internet using my robotic tongue.
AMD cards don't support CUDA.
That's correct, AMD has announced that they support OpenCL as the GPGPU language. And OpenCL is an open standard as opposed to CUDA (NVidia proprietary).
Yes, there is: OpenStreetMap. You can help make it better.
which one could be running wesnoth in the future?
Forget the "future". Maemo already runs Wesnoth, and quite well at that.
Doc Searls (editor of Linux Journal) is working on such a donation system as part of his "Vendor Relationship Management" or VRM project at the Berkman Center at Harvard. The idea is to be able to make small voluntary donations to the software author, or more generally the creator of any piece of work. The goal is make this easy -- simple click of a button that says "donate $5" and put you in control of how much of your personal information (name, credit card details etc.) you want the recipient to know.
Think of this conversation *after* they inhaled all that Helium! All jokes are more funny in He-induced voice.
Besides, you need serious cash in order to get your chip fabricated! You have to have some big company pick up your design and fab them in volume. IIRC, some company is now selling chips (SoC) based on the OR1K design
It is absurd that we need to have a "Read the Laws" campaign. Isn't it their job to read the bills?
Nope, this is still the same old x86 architecture we all love. What they mean by a "new architecture" is that they are ditching the Pentium-IV micro-architecture and developing a new one based on the Pentium-M micro-architecture (itself loosely based on the Pentium-!!! design). As a result, the new chips promise to deliver higher performance at lower power levels.
Amazon.com link to book: http://www.amazon.com/exec/obidos/tg/detail/-/0140 257314/qid=1120666767/sr=1-1/ref=sr_1_1/103-707629 3-4464656?v=glance&s=books/
Octave is a really nice piece of software! Also check out Scilab.
That is not true at all. The size of instructions is still 32-bits for all 64-bit RISC CPUs and the instruction size is the same in x86-64 as that of regular x86 (CISC). Now 64-bit CPUs give you the *option* to use 64-bit data/pointers, but you are free to continue using 32/16/8 bit data as you deem appropriate. Code/data size doesn't automatically increase.
Also, you may not necessarily be doing twice the work with a 64-bit CPU - it again depends on your compiler. In theory you *can* do two 32-bit operations in parallel using a 64-bit EX/FP unit if the ISA provides for packed operations (also known as SIMD instructions).
I agree that they are not being overly ambitious, and that is a *good thing*. But instead of doing AGP/PCI they should do PCI-e. The industry is pretty much standardizing around PCI-e, and AGP may not exist on most boards when this thing comes out.
What you say is absolutely true, but ...
... I've heard it is better than anything IBM & HP have to offer.
1. It is an easier upgrade path for customers. I think Sun learnt that it is easier to sell its customers incremental upgrades than to sell them brand new designs. Remember that the market they sell to (telco, financial) absolutely despises having to test all their mission-critical applications on new, unproven hardware. So while the slow backplane is a performance limitation, many customers may prefer stability to cutting-edge performance.
2. Wait for the 'Zones' in Solaris 10
Exactly my thoughts! Can we include something in Slashcode that automatically filters the GNAA and goatse trolls? Perhaps as a user-configurable option.
... I wouldn't want jokes based on the "$PROJECT is dying - Netcraft confirms it!" troll to be filtered out!
Some Bayesian approach ought to do it
As an Indian, and a big Simpsons fan, I can answer that question right here.
... there is no such thing as a typical Indian.
... I never seem to remember them all. The last name however makes no sense! But, somebody obviously did some research in coming up with these characters.
Apu represents the stereotypical image of an Indian. I personally find it quite amusing, others may have a different opinion. Of course, like any other stereotype, not everyone fits that description. What makes India so interesting is the different subcultures within the country (these differences are more pronounced than the subculture differences in the US), people skeak different languages in different states, different types of food and in some cases different types of clothes as well. So Apu cannot represent a typical Indian
Apu & Manjula are indeed Indian names - not very common but not uncommon either. So are the names of all their kids
Looks like MS doesn't want two different 64-bit x86 extensions. I'm pretty sure Intel has cross-licensing agreements with AMD that will allow it to use AMD's x86-64 extensions. 'Prescott' may already have it
They can still maintain backwards compatibility if they think the market will demand it.
What prevents them from bolting on the current XBox processor (some P3 derivative IIRC) alongside the new IBM chip and having the hardware funnel the instructions to the appropriate processor depending on the version of the game being played? So an XBox1 game will run on the old processor whereas an XBox2 game can run on the new one - everything else can be shared (memory etc.) Again all this will depend on results from their "market surveys".
A woman who recently had a baby claimed that her child was the cutest one on the planet!
Oh the irony!
Read this article at ExtremeTech. It states that UltraSPARC IV will have 2 UltraSPARC III cores. If the UltraSPARC III are in-order, how can UltraSPARC IV be out-of-order?
Besides how does it matter if the core is in-order or not? All I want is kick-ass app performance - I don't care if you do in-order, out-of-order or even get a group of chimps to run through the instructions.
Read the review ..
;)
AMD did not implement the full 64-bit virtual address (neither does Itanium2). The Opteron has *only* 48-bit virtual address and 40-bit physical address. That means it can address upto 256TB of virtual space and 1TB physical space.
And yeah, 256TB ought be enough for everyone
> because this DoubleClick exec happens to be a woman.
..."
Technically shouldn't the title be "Privacy Czarina"?
Yes I am nitpicking. I hate people who `nitpick' but start their posts with "Not to be nitpicking