Teens (or just people) don't buy mp3 because they haven't the money, or they don't want to spend money for....there are more important things before mp3 & dvds:
- mobile connectivity
- garments
- drinks & food
- fuel and tickets for mobility
- periodicals & magazines
Digital content is just not too much important to share a high percentage of founds from the weekly family income or some extra bucks from daily working, expecially at the thief prices it's sold.
I mean: people all over world how many mp3 from P2P have donwloaded since the era of FTPs and P2Ps?
Billions?
Hundred billions?
Multiply them by 0.99$ (the medium price at wich you can buy legal musc) by one trillion: you get 1/6 the public debt of an european country, it is worth to spend this money in music?
I think no.
I've got about 10-15.000 illegal mp3 (lot of them are in full rar discography) and 60-70 xvid on my media, and -i give you a hint- i'm not teenager, but university student, simply i'm not so stupid to spend the value of a car in mp3s when i can take them for free.
If P2P one day will be completely erased, simply I'll stop listen music for something more useful.
This only happens if you don't use solutions used by AMD and IBM like SOI and DSL.
With the low-cost and low-quality 65nm intel process, the leakage con only get worse compared to 90nm. AMD from 130 to 90nm SOI+DSL reduced the leakage to less than 10% of total power consumption. Intel is at 35-40% or worse, depending on frequency.
Agreed.
Merom will also be less power hungry than yonah, because it can disable not only parts of cahce but entire logical units for power saving.
Another intel problem is leakage, idle power consumption is still too high.
How can they compare CPU power comsumption by TDP?!
This is nonsense.
An athlon FX57 merely goes over 60-70W, while a dual core P4 can reach a maximum power consumption over 150W.
humm i don't tink that a memory controller has so high variatons in power consumption.
But what i am trying to say is that a mobil CPU that runs at maximum power dissipating 35W drains a 66Wh battery (med-large one) in about one hour and 50min
So, i think that when you are speaking of Performance you should consider max power consumption, but when you are speaking of battery life you should consider minimal consumpion.
It is normal that if you want to play a 3dGame on a portable you don't pretend that the battery lasts 4 hours, and when you speak about battery performance you don't care too muck of the highest performance that the CPU can burst.
So, my idea is that The Register haven't understood well something:)
Ok, so you are saying that a north bridge dissipates 11W?
And a battery of 53Wh on a centrino how much does it last?
4,5Hour only powering the Northbridge?
A Pentium-M in full speedstep operation mode (eg: 600MHz and 0.8V and 3/4 of cache disabled) absorb from 1 to 2W, same for a VGA GPU with modern power saving features (eg a radeon M9 sorks at 80Mhz and 1.25V)
And the northbridge so should have 11W of power consumpiton?
Laptop is not a desktop and you forgot the first law of thermodinamycs:)
We were finally able to get our hands on a 90nm Athlon 64 3500+ this past weekend, and we've been testing it to see how it compares to the 130nm version. Since this 90nm Athlon 64 3500+ runs at the same clock speed as the 130nm Athlon 64 3500+, we were able to do a direct comparison between the chips running at 2.2GHz
The 2 CPU have same clockrates.
...remember that 130nm-K8 uses only SOI technology, instead 90nm-K8 uses SOI + LowK + SS.
sure, but the problem is that the 40-stage-long-pipeline of the P4 trashes twice many instructions than the 15-stage-pipe in a K8 core.
So the 1.5 clock factor in the P4 is useless... unless you have to fry eggs:D
I thought that strained silicon was part of the silicon crystal production, instead SOI was a post processing add to every single core (or wafer slice)
or not?
Yeah!
No more fat, no more nerd-fisionomy... you can connect to a server at 7:00 instead of doing your jogging session every day and you'll be more healthy than a triathlon performer
wow!
Slashdot Mirror
Teens (or just people) don't buy mp3 because they haven't the money, or they don't want to spend money for. ...there are more important things before mp3 & dvds:
- mobile connectivity
- garments
- drinks & food
- fuel and tickets for mobility
- periodicals & magazines
Digital content is just not too much important to share a high percentage of founds from the weekly family income or some extra bucks from daily working, expecially at the thief prices it's sold.
I mean: people all over world how many mp3 from P2P have donwloaded since the era of FTPs and P2Ps?
Billions?
Hundred billions?
Multiply them by 0.99$ (the medium price at wich you can buy legal musc) by one trillion: you get 1/6 the public debt of an european country, it is worth to spend this money in music?
I think no.
I've got about 10-15.000 illegal mp3 (lot of them are in full rar discography) and 60-70 xvid on my media, and -i give you a hint- i'm not teenager, but university student, simply i'm not so stupid to spend the value of a car in mp3s when i can take them for free.
If P2P one day will be completely erased, simply I'll stop listen music for something more useful.
This only happens if you don't use solutions used by AMD and IBM like SOI and DSL.
With the low-cost and low-quality 65nm intel process, the leakage con only get worse compared to 90nm. AMD from 130 to 90nm SOI+DSL reduced the leakage to less than 10% of total power consumption. Intel is at 35-40% or worse, depending on frequency.
Leakage power is static, and you see leakage consumpion in idle, not in full load.
BTW noone knows intel core duo power consumption specs, so i can't undesrtand how you can made a comparison with a manchester or a toledo.
Agreed. Merom will also be less power hungry than yonah, because it can disable not only parts of cahce but entire logical units for power saving. Another intel problem is leakage, idle power consumption is still too high.
How can they compare CPU power comsumption by TDP?! This is nonsense. An athlon FX57 merely goes over 60-70W, while a dual core P4 can reach a maximum power consumption over 150W.
...there will be a curses "Blue Screen Of Death" in replacement of kernel oops!, and segfaults?
humm i don't tink that a memory controller has so high variatons in power consumption.
:)
But what i am trying to say is that a mobil CPU that runs at maximum power dissipating 35W drains a 66Wh battery (med-large one) in about one hour and 50min
So, i think that when you are speaking of Performance you should consider max power consumption, but when you are speaking of battery life you should consider minimal consumpion. It is normal that if you want to play a 3dGame on a portable you don't pretend that the battery lasts 4 hours, and when you speak about battery performance you don't care too muck of the highest performance that the CPU can burst.
So, my idea is that The Register haven't understood well something
Ok, so you are saying that a north bridge dissipates 11W? And a battery of 53Wh on a centrino how much does it last? 4,5Hour only powering the Northbridge? A Pentium-M in full speedstep operation mode (eg: 600MHz and 0.8V and 3/4 of cache disabled) absorb from 1 to 2W, same for a VGA GPU with modern power saving features (eg a radeon M9 sorks at 80Mhz and 1.25V) And the northbridge so should have 11W of power consumpiton? Laptop is not a desktop and you forgot the first law of thermodinamycs :)
There are 3 kind of lies: lies, damned lies, and benchmarks.
Uh? 0.5W for the memory controller are enough?
Underclocked?
...remember that 130nm-K8 uses only SOI technology, instead 90nm-K8 uses SOI + LowK + SS.
We were finally able to get our hands on a 90nm Athlon 64 3500+ this past weekend, and we've been testing it to see how it compares to the 130nm version. Since this 90nm Athlon 64 3500+ runs at the same clock speed as the 130nm Athlon 64 3500+, we were able to do a direct comparison between the chips running at 2.2GHz
The 2 CPU have same clockrates.
sure, but the problem is that the 40-stage-long-pipeline of the P4 trashes twice many instructions than the 15-stage-pipe in a K8 core. So the 1.5 clock factor in the P4 is useless ... unless you have to fry eggs :D
I thought that strained silicon was part of the silicon crystal production, instead SOI was a post processing add to every single core (or wafer slice) or not?
Yeah! No more fat, no more nerd-fisionomy... you can connect to a server at 7:00 instead of doing your jogging session every day and you'll be more healthy than a triathlon performer wow!