NesuD
Diamond Member
On Thu, 07 September 2000, "Jeffrey Powers" wrote:
>
> Let me ask you a question. How can you guarantee a product if it goes from
> drawing board to mass production within 2 months? You barely can test
> motherboards in that time. Yet AMD and INTEL are putting out processors in
> that time.
>
> Just because an archetechure is 6 years old DOESN'T mean that outdated.
> Besides, the P6 is CORE ARCHETECTURE. It has kinda changed within the last
> few years. Now I am not saying that the INTEL chip is flawless, but in all
> reality, how STABLE is that chip? Before you answer that, make sure you can
> back up your sources.
>
> Oh, by the way. AMD ATHLON comes from a SEVENTH GENERATION CORE. Can you
> tell me when that core was built?
>
> Now, what I said from the article you mention is that if you build a chip
> and don't take the time to really work on it, it is going to become
> problematic. That goes for BOTH INTEL AND AMD.
>
> Bridgestone/Firestone wanted to make tires faster. In the process of working
> on more tires, they decided to not worry about safety, and used a
> substandard rubber for the protective layer.
>
> If you think that AMD is high and mighty because they support a 200 mHz bus,
> a dual ported 128 L1 split cache and an archetecture that includes a 3
> address calculation pipeline, then I only have one word for you. Sucker.
>
> Like I have said before, I am not for INTEL nor AMD. But I will not idly sit
> by and watch people cut down a processor simply because they had a problem.
> Yes, I run an INTEL on my system. But that was only because AMD had not
> proven themselves. There is a time that AMD could not be trusted. And in
> some aspects, I still don't trust them. They only care about speed.
>
> So you go ahead and drive your Firestone tires. I hope you do well.
>
My response,
It is obvious that you don't know much about the development process of a microchip. they don't go from the drawing board to mass production in 2 months. Amd has had 1.1 ghz production silicon for 6 months now. when a cpu core is manufactured it is fabricated on a large wafer of silicon. The size of the core determines how many cores can be produced from each wafer. The individual cores are then tested for to determine what speeds they can run at reliability they are then marked and packaged for those speeds and sold. sometimes a core can be coaxed to higher reliable speeds by increasing the core voltage and improving the cooling of the chip these are tricks used by the over clocking community. there is nothing wrong with intel and amd doing the same thing to attain faster speeds IF! they know the processor is reliable.
You are correct the p6 core is core architecture and there have been enhancements to it to keep it competetive But it is still the same basic core. 6 years is a long lifetime for an x86 based core.
You ask how stable is that chip the Athlon? back up my sources. Sir I have been running Athlon systems since they were introduced and have found it to be the most stable platform I have used yet. Check any of the major hardware sites (ex. WWW.anandtech.com Or www.tomshardware.com) and read through their reviews and you will find that athlon based systems are just as stable as the most stable intel systems.
And Yes I am aware that the Athlon is a seventh generation core that was officially released to the public in June of 1999 and had been in development for 2 years previous to that. So what was the point of that?
>
Actually the fact that the Athlon has a " a 200 mHz bus,
> a dual ported 128 L1 split cache and an archetecture that includes a 3
> address calculation pipeline," Well at a glance that wouldn't seem impressive wouldn't it since p6 core also includes a 3 address calculation pipeline. "However in the case of the P6, the decoders cannot accept 3 complex instructions and process them simultaneously. In the case of the Athlon, regardless of the nature of the instructions, the decoding units are capable of decoding three instructions, regardless of nature, simultaneously." "Another point of improvement the Athlon offers over Intel's P6 core (yes, you read that right, AMD is not only going for competitive performance but they are looking to surpass Intel) is the accuracy of the processor's branch tree prediction.We've all met individuals that are undoubtedly predictable in life, whether it's because of something they always say or do, our minds get used to expecting a certain type of behavior from them when placed in a specific set of circumstances. The same applies to the Athlon, whose 2048-entry branch prediction table takes on the task of storing commonly used operations and attempts to predict what the next operation to come down the pipeline will be. Not only is the table 4 times as large as Intel's branch prediction table, but in the event that the Athlon incorrectly "guesses" at what the next operation will be, the penalty isn't as great as that of the P6 core. This is simply because of the Athlon's shorter pipelines (10-stage integer and 15-stage FP pipelines vs 15-stage integer and ~30-stage FP pipelines for the P6 core) which make sure that the Athlon doesn't have to start from scratch after a missed prediction.- Anandtech's Athlon Review Aug 9,1999
http://www.anandtech.com/showdoc.html?i=1015
I have worked with a lot of amd and intel processors in the past 4 years and it has been my experience that in most cases any stability problems could be traced to flawed motherboard design or chipset design not the cpu itself. that goes for Intel and Amd.
I am sorry if I upset you All I was doing was pointing out that Amd has done a better job with the Athlon in the past year than Intel has with the p3.
>
> Let me ask you a question. How can you guarantee a product if it goes from
> drawing board to mass production within 2 months? You barely can test
> motherboards in that time. Yet AMD and INTEL are putting out processors in
> that time.
>
> Just because an archetechure is 6 years old DOESN'T mean that outdated.
> Besides, the P6 is CORE ARCHETECTURE. It has kinda changed within the last
> few years. Now I am not saying that the INTEL chip is flawless, but in all
> reality, how STABLE is that chip? Before you answer that, make sure you can
> back up your sources.
>
> Oh, by the way. AMD ATHLON comes from a SEVENTH GENERATION CORE. Can you
> tell me when that core was built?
>
> Now, what I said from the article you mention is that if you build a chip
> and don't take the time to really work on it, it is going to become
> problematic. That goes for BOTH INTEL AND AMD.
>
> Bridgestone/Firestone wanted to make tires faster. In the process of working
> on more tires, they decided to not worry about safety, and used a
> substandard rubber for the protective layer.
>
> If you think that AMD is high and mighty because they support a 200 mHz bus,
> a dual ported 128 L1 split cache and an archetecture that includes a 3
> address calculation pipeline, then I only have one word for you. Sucker.
>
> Like I have said before, I am not for INTEL nor AMD. But I will not idly sit
> by and watch people cut down a processor simply because they had a problem.
> Yes, I run an INTEL on my system. But that was only because AMD had not
> proven themselves. There is a time that AMD could not be trusted. And in
> some aspects, I still don't trust them. They only care about speed.
>
> So you go ahead and drive your Firestone tires. I hope you do well.
>
My response,
It is obvious that you don't know much about the development process of a microchip. they don't go from the drawing board to mass production in 2 months. Amd has had 1.1 ghz production silicon for 6 months now. when a cpu core is manufactured it is fabricated on a large wafer of silicon. The size of the core determines how many cores can be produced from each wafer. The individual cores are then tested for to determine what speeds they can run at reliability they are then marked and packaged for those speeds and sold. sometimes a core can be coaxed to higher reliable speeds by increasing the core voltage and improving the cooling of the chip these are tricks used by the over clocking community. there is nothing wrong with intel and amd doing the same thing to attain faster speeds IF! they know the processor is reliable.
You are correct the p6 core is core architecture and there have been enhancements to it to keep it competetive But it is still the same basic core. 6 years is a long lifetime for an x86 based core.
You ask how stable is that chip the Athlon? back up my sources. Sir I have been running Athlon systems since they were introduced and have found it to be the most stable platform I have used yet. Check any of the major hardware sites (ex. WWW.anandtech.com Or www.tomshardware.com) and read through their reviews and you will find that athlon based systems are just as stable as the most stable intel systems.
And Yes I am aware that the Athlon is a seventh generation core that was officially released to the public in June of 1999 and had been in development for 2 years previous to that. So what was the point of that?
>
Actually the fact that the Athlon has a " a 200 mHz bus,
> a dual ported 128 L1 split cache and an archetecture that includes a 3
> address calculation pipeline," Well at a glance that wouldn't seem impressive wouldn't it since p6 core also includes a 3 address calculation pipeline. "However in the case of the P6, the decoders cannot accept 3 complex instructions and process them simultaneously. In the case of the Athlon, regardless of the nature of the instructions, the decoding units are capable of decoding three instructions, regardless of nature, simultaneously." "Another point of improvement the Athlon offers over Intel's P6 core (yes, you read that right, AMD is not only going for competitive performance but they are looking to surpass Intel) is the accuracy of the processor's branch tree prediction.We've all met individuals that are undoubtedly predictable in life, whether it's because of something they always say or do, our minds get used to expecting a certain type of behavior from them when placed in a specific set of circumstances. The same applies to the Athlon, whose 2048-entry branch prediction table takes on the task of storing commonly used operations and attempts to predict what the next operation to come down the pipeline will be. Not only is the table 4 times as large as Intel's branch prediction table, but in the event that the Athlon incorrectly "guesses" at what the next operation will be, the penalty isn't as great as that of the P6 core. This is simply because of the Athlon's shorter pipelines (10-stage integer and 15-stage FP pipelines vs 15-stage integer and ~30-stage FP pipelines for the P6 core) which make sure that the Athlon doesn't have to start from scratch after a missed prediction.- Anandtech's Athlon Review Aug 9,1999
http://www.anandtech.com/showdoc.html?i=1015
I have worked with a lot of amd and intel processors in the past 4 years and it has been my experience that in most cases any stability problems could be traced to flawed motherboard design or chipset design not the cpu itself. that goes for Intel and Amd.
I am sorry if I upset you All I was doing was pointing out that Amd has done a better job with the Athlon in the past year than Intel has with the p3.
