First dual-core PC's to ship on Monday, Intel says

[DAPUNISHER]

The high-end workstation and server sector has a higher profit margin, I agree with AMD that this provides an opportunity to capture some more share there. I really believe that finally starting to have a presence in these areas is more important right now than trying to take more market share in the desktop sector. They couldn't produce enough chips to supply 50% of the desktop market if what I read is accurate, so why not maximize profits on what you can produce?

 

[aka1nas]

Originally posted by: keysplayr2003

Originally posted by: mamisano

Originally posted by: keysplayr2003
My friend just handed me 475.00 and said, "Please build me the best PC you can for this much money. Just the tower no monitor, keyboard or mouse." So I got the AMD Sempron 2800+ and an ASUS KT400 mobo, RAM, HDD, DVD burner, floppy, case, PSU. I feel dirty. J/K. I could have gone Celery, but the Sempron packs more punch clock for clock and dollar for dollar. So, I do have a use for AMD. I am glad they are around.

Wow, for $475 he could have had an Sempron 2800 (A64 Core), ECS Nforce-4, 1GB PC3200, 80GB SATA, DL-DVD/RW, Nvidia 6600 PCIe

Well, I only spent 402.00 of his 475.00. He was with me when we bought the parts so he was happy to save money. And why on earth would you even consider buying an ECS motherboard for a FRIEND? I got him an ASUS bro. Nuff said. Friends don't let friends use cheap arse mobo's. He used the remaining 73.00 to buy an AGP FX5200 128MB 128 bit card and a SB Live Value. He's not a big gamer, but he does want to use 2 monitors. So the FX was fine.

I think I did rather well with the money he had to spend. 64 bit processor not needed so why spend the extra cash? hehehe... ECS....

Friends don't let friends buy VIA chipset-based motherboards either

 

[Keysplayr]

Originally posted by: aka1nas

Originally posted by: keysplayr2003

Originally posted by: mamisano

Originally posted by: keysplayr2003
My friend just handed me 475.00 and said, "Please build me the best PC you can for this much money. Just the tower no monitor, keyboard or mouse." So I got the AMD Sempron 2800+ and an ASUS KT400 mobo, RAM, HDD, DVD burner, floppy, case, PSU. I feel dirty. J/K. I could have gone Celery, but the Sempron packs more punch clock for clock and dollar for dollar. So, I do have a use for AMD. I am glad they are around.

Wow, for $475 he could have had an Sempron 2800 (A64 Core), ECS Nforce-4, 1GB PC3200, 80GB SATA, DL-DVD/RW, Nvidia 6600 PCIe

Well, I only spent 402.00 of his 475.00. He was with me when we bought the parts so he was happy to save money. And why on earth would you even consider buying an ECS motherboard for a FRIEND? I got him an ASUS bro. Nuff said. Friends don't let friends use cheap arse mobo's. He used the remaining 73.00 to buy an AGP FX5200 128MB 128 bit card and a SB Live Value. He's not a big gamer, but he does want to use 2 monitors. So the FX was fine.

I think I did rather well with the money he had to spend. 64 bit processor not needed so why spend the extra cash? hehehe... ECS....

Friends don't let friends buy VIA chipset-based motherboards either

If it's good enough for ASUS, then I'm ok with it. VIA KT400 chipset is a very good chipset. So I had minimal issues with it. But ECS mobos are a no no even with the very best of chipsets on it.

 

[Keysplayr]

Originally posted by: DAPUNISHER
The high-end workstation and server sector has a higher profit margin, I agree with AMD that this provides an opportunity to capture some more share there. I really believe that finally starting to have a presence in these areas is more important right now than trying to take more market share in the desktop sector. They couldn't produce enough chips to supply 50% of the desktop market if what I read is accurate, so why not maximize profits on what you can produce?

I wish I knew what it actually took to make a CPU. Not so much the design or architecture, but the actual production. Does it really take that long to make a CPU?
Dunno. I can understand shortage of supply from a shortage of materials. But saying they can not produce them in quantity has me stumped. Maybe if I knew more about the process (I will still continue to survive if I don't) then I could understand. If the reason AMD went for the enterprise market first is because they cant' keep up with quantity for the desktop market, then that is not called good marketing, but should be called "The only thing we could do". Dunno.

 

[EndGame]

Originally posted by: keysplayr2003

Originally posted by: DAPUNISHER
The high-end workstation and server sector has a higher profit margin, I agree with AMD that this provides an opportunity to capture some more share there. I really believe that finally starting to have a presence in these areas is more important right now than trying to take more market share in the desktop sector. They couldn't produce enough chips to supply 50% of the desktop market if what I read is accurate, so why not maximize profits on what you can produce?

I wish I knew what it actually took to make a CPU. Not so much the design or architecture, but the actual production. Does it really take that long to make a CPU?
Dunno. I can understand shortage of supply from a shortage of materials. But saying they can not produce them in quantity has me stumped. Maybe if I knew more about the process (I will still continue to survive if I don't) then I could understand. If the reason AMD went for the enterprise market first is because they cant' keep up with quantity for the desktop market, then that is not called good marketing, but should be called "The only thing we could do". Dunno.

Still going to be dang hard to do considering Dell has a large % of the business/corporate systems and that you have been able to order a dual system from Dell since Wed.....

 

[bunnyfubbles]

whoever can offer me a $300 (although idealy around $200) or less dual core is the winner in my book, everything else is worthless bragging rights, the EE didn't save the P4, it certainly won't save the PD

 

[DAPUNISHER]

Originally posted by: keysplayr2003

Originally posted by: DAPUNISHER
The high-end workstation and server sector has a higher profit margin, I agree with AMD that this provides an opportunity to capture some more share there. I really believe that finally starting to have a presence in these areas is more important right now than trying to take more market share in the desktop sector. They couldn't produce enough chips to supply 50% of the desktop market if what I read is accurate, so why not maximize profits on what you can produce?

I wish I knew what it actually took to make a CPU. Not so much the design or architecture, but the actual production. Does it really take that long to make a CPU?
Dunno. I can understand shortage of supply from a shortage of materials. But saying they can not produce them in quantity has me stumped. Maybe if I knew more about the process (I will still continue to survive if I don't) then I could understand. If the reason AMD went for the enterprise market first is because they cant' keep up with quantity for the desktop market, then that is not called good marketing, but should be called "The only thing we could do". Dunno.

There were approx. 189,000,000 total PC sales worldwide for 2004 AMD says they did 30 million in fab30 mostly on 130nm and will be able to do 50 million on 90nm, and 100 million total when fab36 comes on-line man I feel dirty linking to inq

 

[mamisano]

I really can't wait for Dual Core Athlon so I can have better performance running VMWare 5

 

[Lonyo]

Somehow I think AMD won this one.
PD EE? That's not a mainstream chip, it's a very niche market.
Opteron DC? That's where the focus should be anyway IMO. Xeons are lagging behind after this (and they only just managed to catch up).
PD seems like a stop-gap to me, based on the design (ie: lets stick 2 cores on a CPU, rather than making a real dual core design, ie: Pressler).
Not exactly sure how the AMD dual core's have been done (have they just thrown 2 cores on a chip? I don't think they have, but I can't remember what I have read).

I will only really care either way when one or other company has REAL chips around, ones that people will actually be able to afford while keeping all their bodily organs. Opterons are server chips, EE's are stupid chips

 

[Sentential]

Originally posted by: stevty2889

Originally posted by: AnandThenMan

Originally posted by: stevty2889
Yeah, SOI is being used for 65nm, it has much less current leakage than the prescott, so it should run much cooler at the same clock speeds, and be able to reach higher ones.

How can you know this? You know how a new process will behave months before it is even released? A smaller process tends to introduce more leakage issues not less. The older .13 Intel chips run cooler and actually faster clock for clock.

Because the process is already running, and should be ramping up around June..the problem with prescott is they did the process shrink, without making any improvements to help control the leakage, with 65nm they made the correction they should have done with prescott..SOI...

Still, they ARENT going to be using any form of SOI. Last I heard they were using some sort of black diamond low-k process for 65nm (dont quote me on that tho). Either way one thing is for DAMN sure its not SOI.

In addition SOI doesnt control leakage. ALl SOI does is lace the copper interconnects between the various levels and help to act as a superconductor thus reducing vcore requirements. It does nto STOP leakage it just helps reduce the vcore.

 

[AnandThenMan]

AMD has another advantage in dual core as well. The arcitecture is designed to have more than 2 cores. We will see 4 core Opterons as soon as the fab technology makes it feasable. Intel has no hope of doing a 4+ core Pentium at this point. Unless Intel copies AMD's hypertransport design (among other things) doing more than 2 cores will be severely bandwidth limited.

 

[stevty2889]

Originally posted by: Sentential

Originally posted by: stevty2889

Originally posted by: AnandThenMan

Originally posted by: stevty2889
Yeah, SOI is being used for 65nm, it has much less current leakage than the prescott, so it should run much cooler at the same clock speeds, and be able to reach higher ones.

How can you know this? You know how a new process will behave months before it is even released? A smaller process tends to introduce more leakage issues not less. The older .13 Intel chips run cooler and actually faster clock for clock.

Because the process is already running, and should be ramping up around June..the problem with prescott is they did the process shrink, without making any improvements to help control the leakage, with 65nm they made the correction they should have done with prescott..SOI...

Still, they ARENT going to be using any form of SOI. Last I heard they were using some sort of black diamond low-k process for 65nm (dont quote me on that tho). Either way one thing is for DAMN sure its not SOI.

In addition SOI doesnt control leakage. ALl SOI does is lace the copper interconnects between the various levels and help to act as a superconductor thus reducing vcore requirements. It does nto STOP leakage it just helps reduce the vcore.

I stand corrected, it is the low-k dielectric thats reducing the power consumption. They are using strained silicon, and I got that mixed up with SOI. They don't realy give us many details, they just want us to keep the wafers movin..

 

[sxr7171]

Originally posted by: Duvie
And Intel lands a stunningblow to the morons of AMD......

So basically they may even beat the opterons to the market...total fools at AMD......

Unfortunatley it is the EE or extremely expensive version which I would never want even if I had the cash.....BUt all in all it is the victories Intel loves so much, and AMD so graciously loves to give them....

I really would like to see some get out there so we can get a good idea on the cooling needs, the overclocking ability, and the power needs....

Actually: http://forums.anandtech.com/messageview.aspx?catid=28&threadid=1567754

 

[Keysplayr]

Originally posted by: mamisano
I really can't wait for Dual Core Athlon so I can have better performance running VMWare 5

How long do you think you will have to wait?

 

[mamisano]

Originally posted by: keysplayr2003

Originally posted by: mamisano
I really can't wait for Dual Core Athlon so I can have better performance running VMWare 5

How long do you think you will have to wait?

Hmmm, I figure June/July.

 

[Keysplayr]

Originally posted by: mamisano

Originally posted by: keysplayr2003

Originally posted by: mamisano
I really can't wait for Dual Core Athlon so I can have better performance running VMWare 5

How long do you think you will have to wait?

Hmmm, I figure June/July.

Not too bad.

 

[EndGame]

Originally posted by: AnandThenMan
AMD has another advantage in dual core as well. The arcitecture is designed to have more than 2 cores. We will see 4 core Opterons as soon as the fab technology makes it feasable. Intel has no hope of doing a 4+ core Pentium at this point. Unless Intel copies AMD's hypertransport design (among other things) doing more than 2 cores will be severely bandwidth limited.

The "quad cores" are already in the rodmap for both AMD and Intel. AMD has the K9 scheduled for Q4 '07 and Intel has the Whitefield scheduled for Q3 '07.....those are the first quad core chips for either at this point, but yes, they both have them.....

 

[Sentential]

Originally posted by: AnandThenMan
AMD has another advantage in dual core as well. The arcitecture is designed to have more than 2 cores. We will see 4 core Opterons as soon as the fab technology makes it feasable. Intel has no hope of doing a 4+ core Pentium at this point. Unless Intel copies AMD's hypertransport design (among other things) doing more than 2 cores will be severely bandwidth limited.

Doesnt matter, Q1 2007 will begin the reign of Pentium 5 and Conroe. Thats when PentiumM chips will finally move to desktop. Conroe should be a beast. Last I heard they were tossing the idea of making a Quad-Core Dothan @ .45nm process with 16MB of cache *WITH* a new form of hyperthreading. Thus making 8 logical threaded CPUs.

Its going to be awefully hard to compete with a 3ghz Conroe that dissapates less than 70W (which it is scheduled to do)

 

[EndGame]

Originally posted by: Sentential

Originally posted by: AnandThenMan
AMD has another advantage in dual core as well. The arcitecture is designed to have more than 2 cores. We will see 4 core Opterons as soon as the fab technology makes it feasable. Intel has no hope of doing a 4+ core Pentium at this point. Unless Intel copies AMD's hypertransport design (among other things) doing more than 2 cores will be severely bandwidth limited.

Doesnt matter, Q1 2007 will begin the reign of Pentium 5 and Conroe. Thats when PentiumM chips will finally move to desktop. Conroe should be a beast. Last I heard they were tossing the idea of making a Quad-Core Dothan @ .45nm process with 16MB of cache *WITH* a new form of hyperthreading. Thus making 8 logical threaded CPUs.

Its going to be awefully hard to compete with a 3ghz Conroe that dissapates less than 70W (which it is scheduled to do)

^^^That would be the Whitefield........

 

[Sentential]

Originally posted by: EndGame

Originally posted by: Sentential

Originally posted by: AnandThenMan
AMD has another advantage in dual core as well. The arcitecture is designed to have more than 2 cores. We will see 4 core Opterons as soon as the fab technology makes it feasable. Intel has no hope of doing a 4+ core Pentium at this point. Unless Intel copies AMD's hypertransport design (among other things) doing more than 2 cores will be severely bandwidth limited.

Doesnt matter, Q1 2007 will begin the reign of Pentium 5 and Conroe. Thats when PentiumM chips will finally move to desktop. Conroe should be a beast. Last I heard they were tossing the idea of making a Quad-Core Dothan @ .45nm process with 16MB of cache *WITH* a new form of hyperthreading. Thus making 8 logical threaded CPUs.

Its going to be awefully hard to compete with a 3ghz Conroe that dissapates less than 70W (which it is scheduled to do)

^^^That would be the Whitefield........

Ah, guess they changed the name....coulda sworn it was still Conroe

 

[AnandThenMan]

LOL here we go again with wild speculation. The P5 will rule this and that blah blah in 2007. Good thing you can see into the future.

 

[stevty2889]

Originally posted by: AnandThenMan
LOL here we go again with wild speculation. The P5 will rule this and that blah blah in 2007. Good thing you can see into the future.

There is nothing wrong with a little speculation, or don't you like to look forward to the future? Besides, it's not that wild of speculation when the 45nm proccess is already being qualified, and Intel already has some details and names for future chips well before they are ready to come out...like Morom for example, not sure if it will be on 65nm or 45nm, but it's supposed to be dual core Pentium-M type chip, that will get some FPU enhancments to help it compete in it's weaker areas, and it will have 4mb of L2 cache, and will be used with what Intel is calling it's "Entertainment PC" platform. So speculating on that info..sounds like it's going to be for some form of gaming PC...and why not? Do you think Intel just sits around and ignores what people want, and don't realize that AMD has the gaming lead, and the price to performance lead in many area's?

 

[EndGame]

Originally posted by: AnandThenMan
LOL here we go again with wild speculation. The P5 will rule this and that blah blah in 2007. Good thing you can see into the future.

To be honest, the only one "speculating" was you in your first post. What we posted was fact from both AMD & Intel......LOL!

 

[AnandThenMan]

Speculate all you want. But stating that, "Q1 2007 will begin the reign of Pentium 5" is not speculation but a statement of fact. Perhaps you are unclear of the difference.

 

[Cooler]

Are they even going to name it Pentium 5 that is redundant. I think they might start going with Pentium with model numbers .