Paul Otellini comments/ Intel Dual core better then AMD'S
- Thread starter clarkey01
- Start date
Markfw
Moderator Emeritus, Elite Member
- May 16, 2002
- 27,282
- 16,123
- 136
Intel's shared cache is a slight performance edge in some instances, whereas amd's seperate cache architecture is a large edge in some instances. Overall, amd has a much better cache and memory architecture. If that is what intel is trying to brag about, then they are in really bad shape. If I were them I'd really be focusing on their thermal advantages (assuming dual cores are made using the dothen-style processors)
HAHAHAHAHAHAHAHAH
Hiillllarious
Yeah, hypertransport and incredible scalability and less power and less heat blow chunks. LOL
Performance wise we shall see but I have a supsicion that Intel is going to be spanked
Intel has dual core P4's and Xeon's for awhile? Just like they had X86-64 on their roadmaps for awhile also.
Intel says they could have made 4.0 Ghrtz but didn't. What the heck does that mean? LOL
I have Intel and AMD computers but Intel seems to be dreaming but market dominance will allow that. I hope their dual core is awesome but beating the AMD dual core that was designed from the beginning to be dualcore. I don't think so. They believe their external memory controller will work better than the integrated controller. HMMMM. Probably not.
Hiillllarious
Yeah, hypertransport and incredible scalability and less power and less heat blow chunks. LOL
Performance wise we shall see but I have a supsicion that Intel is going to be spanked
Intel has dual core P4's and Xeon's for awhile? Just like they had X86-64 on their roadmaps for awhile also.
Intel says they could have made 4.0 Ghrtz but didn't. What the heck does that mean? LOL
I have Intel and AMD computers but Intel seems to be dreaming but market dominance will allow that. I hope their dual core is awesome but beating the AMD dual core that was designed from the beginning to be dualcore. I don't think so. They believe their external memory controller will work better than the integrated controller. HMMMM. Probably not.
For some reason I have a soft spot for Paul Otellini , Dr barret was more for the "Ghz= More".
Intel seem to be more vocal then AMD, I wish hector would get of his arse and say something.
Intel seem to be more vocal then AMD, I wish hector would get of his arse and say something.
"Intel says they could have made 4.0 Ghrtz but didn't. What the heck does that mean? LOL "
They could have but it would of taken another stepping for prescott and I have no idea what the power consumtion on that would be , im sure a 3.6 E uses 115 watts.
They could have but it would of taken another stepping for prescott and I have no idea what the power consumtion on that would be , im sure a 3.6 E uses 115 watts.
From the Inquirer article: Intel speaking
But that didn't mean Intel wasn't capable of producing a 4GHz Pentium 4. It was. But, Otellini told Intel's worldwide staff, it was pulled because the company intended to get back to a 90 per cent confidence level on both its schedules and its road maps.
Intel doesn't have infinite resources and had to make choices.
He said it was more important for Intel to move to 2MB cache processors and to get to dual core chips faster. Intel would have to have done another stepping for the Prescott core to get to 4GHz and 4.2GHz and that would have been counterproductive "in the grand scheme of things".
Intel is now marketing performance to include features and not just gigahertz, he said."µ
What a load! So now instead of another core revision they are going to pile on extremely expensive cache and marketing performance other than megahertz on a platform that is designed completely on megahurtz frequency increase? How much are the 2meg Prescott's going to be 3.6 for 999.00. They are using the market dominance to try and stave off an AMD that is gaining respect in the IT world. It will be interesting but I found this article to be an amazing arrogance and revisionism. (I have both Intel and AMD machines and like them both)
LTC8K6
Lifer
- Mar 10, 2004
- 28,520
- 1,576
- 126
Looking at the D0 Prescotts, it's pretty clear that Intel could have done a 4Ghz part.
Anand just meant that AMD was ahead time wise didn't he? Not necessarily technology wise.
Isn't this the sort of thing that all company reps say about their products vs the competition? Why is this such a big deal?
We will know which solution is better soon enough.
Anand just meant that AMD was ahead time wise didn't he? Not necessarily technology wise.
Isn't this the sort of thing that all company reps say about their products vs the competition? Why is this such a big deal?
We will know which solution is better soon enough.
- Aug 22, 2001
- 32,072
- 32,599
- 146
Markfw
Moderator Emeritus, Elite Member
- May 16, 2002
- 27,282
- 16,123
- 136
- Aug 22, 2001
- 32,072
- 32,599
- 146
Originally posted by: LTC8K6
Why are you guys trying to say that the power used by the whole computer is the power used by the cpu alone?
The election is over, no need for any more propaganda.
Now logically it follows that if Anand did a good job isolating the load to the CPU subsystem that the increased power draw for each system was created by the CPU's power demands. Now since Anand is a bad cat and all that I trust his testing methodology, so I need only check the chart to see the 560 is drawing considerably more power than the FX55, and the difference between the 530 and 3500+ is :shocked:
"They could have but it would of taken another stepping for prescott and I have no idea what the power consumtion on that would be , im sure a 3.6 E uses 115 watts"
I didnt ask markfw900
I didnt ask markfw900
Learn to read.
Measuring CPU usage using system power draw is like measuring your blood pressure at the doctor's office. Nowhere near accurate and only useful because it's consistent.
Also, if I remember correctly, virtually all the power use from a CPU goes into thermal waste. So much so that CPUs can be pretty well modeled as resistors.
Full system draw is 210 W at full, 124 at idle. The difference is 86W vs 28W for the 90nm Athlon64. The 115W figure is pretty close to actual power use after factoring in other system components.
Back to the main topic, if the story is true, then Intel may be trying to essentially expand HyperThreading into two fully independent pipelines. You'll lose HT per core, so each core would act like a non-HT P4 with less arbitration logic.
Just speculation on my part.
Zebo
Elite Member
- Jul 29, 2001
- 39,398
- 19
- 81
Thats cause they throttle when they get too hot equal less performance.
The 3.6 presshot uses 150W under ful bore accroding to intels own specs !!! (not TDP which is really BSP)
http://www.cpuheat.wz.cz/html/IntelPowerConsumption.htm
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 23K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter