There have been many articles lately about one company sueing another over patent infringement. The latest I read about was GraphOn sueing Google. This got me wondering if AMD had any case against Intel since Intel will now be using an integrated memory controller like AMD. Are the technologies different enough or is Intel just trying to copy AMD? I'm pretty sure Intel covered its ass before implementing the technology but you never know...
Oh, I didn't know there was already a lawsuit settlement. Also didn't know Intel already used an IMC in the past. Not sure about that 3rd comment. Thanks.
{Comment Edited by self for being overly sarcastic and not helpful to anyone}
A bit persnickety today...
Message edited by Scotteq on 08-19-2008 at 05:50:00 PM
------------------------------Which Chip? Well, it depends on which set of thieving b@stardz you choose to support: The ones who use insider trading to enrich themselves while running their company into the ground? Or the ones who illegally pay vendors to not support the first group?
Reply to Scotteq
in the 90's, amd often used the same socket intel was using for it's pentium processors!
that went to court and it was asumed you couldn't copyright a number, hence 286, 386 & 486, were made by many companies.
they then switched to Pentium which could be copy righted...
same with sockets, when they swiched to slot1 types.
AMD has no case regarding the IMC against Intel, or for just about any and all processor technologies for that matter
Back in the day, AMD licensed the 8088 ahnd 8086 processors so they could be a second source manufacturer of processors for IBM. Somewhere along the line AMD and Intel signed an agreement for Intel to use AMD's 64bit instruction set for processors. Specifically with the IMC, Intel was going to release the Timna line of processors back in 1999/2000 which had and IMC designed to work with RDRAM. Timna was cancelled due to rambus prices and the lawsuit between rambus and the rest of the world. These are just some examples of the technology that has been shared and/or traded between the two companies over the years, and it isn't likely to change any time soon.
At this stage in the game, Intel cannot live without AMD and AMD cannot live without Intel. They are two sides of the same coin.
Well since Intel has been working on a IMC since oh the 386 I would say no. There are certain technologies that get passed aroung and are common in the industry.
Its like in a car. All engines now use Direct Fuel Injection but no one sues anyone.
...and Yes, I'm kidding. I remembered the story, and thought people would think it's funny.
------------------------------Which Chip? Well, it depends on which set of thieving b@stardz you choose to support: The ones who use insider trading to enrich themselves while running their company into the ground? Or the ones who illegally pay vendors to not support the first group?
Reply to Scotteq
And Intel had integrated memory controllers on CPUs well before AMD... probably before AMD even existed.
At least educate yourself before you post such responses. They'd been working on a memory controller before AMD put out the K8. That IMC was for use with the Tinma System on a chip processor. That was meant to be used in the low end entry level or set top box market. But, AMD has existed since the late 70s, and had a second sourse cross license for the earliest part of their existence. See more below
in the 90's, amd often used the same socket intel was using for it's pentium processors!
Yeah, I remember the good old Socket 7 days. In fact thats the main reason Intel started to move away from the copyable socket patents. AMD, Cyrix and a few others had been producing processors that could work on any intel socket based board. AMD K6-2 being the last of the super 7 chips.
AMD has no case regarding the IMC against Intel, or for just about any and all processor technologies for that matter
Back in the day, AMD licensed the 8088 ahnd 8086 processors so they could be a second source manufacturer of processors for IBM. Somewhere along the line AMD and Intel signed an agreement for Intel to use AMD's 64bit instruction set for processors. Specifically with the IMC, Intel was going to release the Timna line of processors back in 1999/2000 which had and IMC designed to work with RDRAM. Timna was cancelled due to rambus prices and the lawsuit between rambus and the rest of the world. These are just some examples of the technology that has been shared and/or traded between the two companies over the years, and it isn't likely to change any time soon.
At this stage in the game, Intel cannot live without AMD and AMD cannot live without Intel. They are two sides of the same coin.
Mostly correct, under the current terms of the Cross license neither can sue the other over technology acquired during the last renewal of the license that regards to processor design. If the k8 already existed during the last renewal, then AMD can do Jack squat about them Using an IMC or moving to a HT like bus system.
Tinma's IMC problem was two fold, RDRam was too expensive, and the prices never dropped due to lack of competition for the original maker. And, Intel couldn't get the IMC to work properly with the Memory translator Hub that was needed to allow it to communicate with SDRam at the time, which would allow it to use much cheaper system memory. Actually even though Intel licensed x86-64 from AMD they never used it, they just re-engineered it for themselves and gave it the name IA64 so they wouldn't have to pay AMD royalties.
Geeez... You guys make this sh*t up as you go along?? x64 *IS* x86-64. Intel call it merely x64 (for short, and to try and break the AMD connection), and it stuck.
Itanium was a joint HP/Intel project. AMD had nothing to do with it. X86-64 and Itanium do not share any common instructions, and the programming model is also substantially different - an example would be x86, you have to explicitly code instructions to handle parallel computing, whereas in IA64 the compiler does that for you (increased parallellism was a primarly driver). Also, part of the basis for x86 is an emphasis on backwards compatibility. IA? Not so much...
The reason Itanium has suffered limited adoption rates is very simple: x86-64 (what Intel now calls x64) is a superset of the older x86 instruction set. Therefore an x86-64 processor can run x86 binaries as well as x64. This means you don't have to trash your current code base. Itanium is completely different and requires a complete rewrite because it cannot run your existing code. Therefore Itanium failed to achieve widespread acceptance for the simple reason that almost nobody wanted to rewrite everyting when they didn't absolutely have to.
------------------------------Which Chip? Well, it depends on which set of thieving b@stardz you choose to support: The ones who use insider trading to enrich themselves while running their company into the ground? Or the ones who illegally pay vendors to not support the first group?
Reply to Scotteq
Geeez... You guys make this sh*t up as you go along?? x64 *IS* x86-64. Intel call it merely x64 (for short, and to try and break the AMD connection), and it stuck.
Itanium was a joint HP/Intel project. AMD had nothing to do with it. X86-64 and Itanium do not share any common instructions, and the programming model is also substantially different - an example would be x86, you have to explicitly code instructions to handle parallel computing, whereas in IA64 the compiler does that for you (increased parallellism was a primarly driver). Also, part of the basis for x86 is an emphasis on backwards compatibility. IA? Not so much...
The reason Itanium has suffered limited adoption rates is very simple: x86-64 (what Intel now calls x64) is a superset of the older x86 instruction set. Therefore an x86-64 processor can run x86 binaries as well as x64. This means you don't have to trash your current code base. Itanium is completely different and requires a complete rewrite because it cannot run your existing code. Therefore Itanium failed to achieve widespread acceptance for the simple reason that almost nobody wanted to rewrite everyting when they didn't absolutely have to.
Yeah I corrected myself in the other thread. IA64 is the 64 bit architecture used for the Itanium that was done by Intel and HP. AMD's response was x86-64 later changed to AMD64, to allow better compatibility and future 64bit path. Intel adapted AMD64 to the Netburst architecture as EM64T, basically the same, but missing one of the multiply or divide I think it was operation commands.
Itanium could emulate x86 through software, but was crappy at it.
Message edited by Mathos on 08-20-2008 at 08:19:14 PM
------------------------------AMD Phenom X4 9850 Black Edition, ZeroTherm Nirvana 120 Premium CPU Cooler, MSI K9a2 Platinum bios 1.1b3 or P.0J, 4GB (2x2) Mushkin DDR2 1066 (pc8500) 5-5-5-15 2.05v RAM, Sapphire Toxic HD3870, Raidmax RX-700SS PSU, Seagate Barracuda 7200.10 320gb SATA2 X
Reply to Mathos
At least educate yourself before you post such responses. They'd been working on a memory controller before AMD put out the K8. That IMC was for use with the Tinma System on a chip processor. That was meant to be used in the low end entry level or set top box market. But, AMD has existed since the late 70s, and had a second sourse cross license for the earliest part of their existence.
The Intel 4004 had an integrated memory controller in 1972; RAMs were directly connected to the CPU with no need for external hardware to drive them. If you're correct about AMD being founded in the late 70s, then Intel had IMCs before AMD even existed.
The Intel 4004 had an integrated memory controller in 1972; RAMs were directly connected to the CPU with no need for external hardware to drive them. If you're correct about AMD being founded in the late 70s, then Intel had IMCs before AMD even existed.
If you can tell me, or show me on that die, where there is an integrated memory controller I'll go out of my way to find a way to send you a cookie, even if it's only virtual cookie on an ecard. There were no real memory controllers at the time, the closest thing was a DMA(direct memory access chip). Intel made their own memory chips then, in fact Intel started out making SRAM chips, which tends to be why their cache architecture is faster than AMD's.
If you can tell me, or show me on that die, where there is an integrated memory controller I'll go out of my way to find a way to send you a cookie, even if it's only virtual cookie on an ecard. There were no real memory controllers at the time, the closest thing was a DMA(direct memory access chip). Intel made their own memory chips then, in fact Intel started out making SRAM chips, which tends to be why their cache architecture is faster than AMD's.
That’s a base model they built varying models through their product cycle, look further if you must.
And Intel had integrated memory controllers on CPUs well before AMD... probably before AMD even existed.
Well, you were obviously born ten years after Intel was founded. AMD was the first to integrate a memory controller onto a cpu.
"If true–and that's a big if–then this project will be the first true integration of the on-die memory controller into an Intel processor, a step AMD took with its AMD64 technology, and one that has seen quite significant gains and advances in performance from NUMA configurations, allowing greater throughput with exclusively owned memory."
You are about to answer a thread that has been inactive for more than 6 months. If you still wish to proceed, please ensure that your posting is original and does not duplicate or overlap any prior responses to this thread.