I don't mean to start a "AMD is better than Intel" or vice versa war. I
just am thinking of buying a new laptop and I first want to decide
which processor.

Why I am looking at Turion 64 X2
* 64 bit processor
* Different (My mother already has a Centrino Duo)

Why I am looking at Intel Centrino Duo
* Faster Processor
* Cooler Processor (That's Tempreture not Status)
* Better Battery Life


My main question is how much help is a 64bit processor. I have seen AMD
videos that show a game on a 32bit processor and one on a 64bit
processor, with much higher detail on the 64bit, but I suspect that is
a bit of a lie. My brother talks about how he is one of the few people
to use the PowerMac G5's 64bit processor through some low level
operations with 64bit integers and such. Is it true that buying a
Turion 64 will only actually have the benefit of me being able to say
it has a 64bit processor? Will Vista make a difference?

Also one final point, is there any company to recommend. I was looking
at a compaq and noticed that there seemed to be a lot of friction
between my finger and the touchpad, same for the HP, is that normal.
Toshiba seems good, but that finishes my processor decision right
there, Intel. Also I had a Toshiba but it's battery seems almost
completly gone after 1.5 years, but that is probably normal. Sony isn't
bad, but a bit expensive.

Last point, I think I will need around 1GB of RAM, is that about right?
Are buisness models more reliable and should I consider them?

Thanks,
Ben Pazolli

"bpazolli" <bpazolli******.com> wrote in message
news:1162097199.033647.93190@f16g2000cwb.googlegro ups.com...
>I don't mean to start a "AMD is better than Intel" or vice versa war. I
> just am thinking of buying a new laptop and I first want to decide
> which processor.
>
> Why I am looking at Turion 64 X2
> * 64 bit processor
> * Different (My mother already has a Centrino Duo)
>
> Why I am looking at Intel Centrino Duo
> * Faster Processor
> * Cooler Processor (That's Tempreture not Status)
> * Better Battery Life
>
>
> My main question is how much help is a 64bit processor. I have seen AMD
> videos that show a game on a 32bit processor and one on a 64bit
> processor, with much higher detail on the 64bit, but I suspect that is
> a bit of a lie. My brother talks about how he is one of the few people
> to use the PowerMac G5's 64bit processor through some low level
> operations with 64bit integers and such. Is it true that buying a
> Turion 64 will only actually have the benefit of me being able to say
> it has a 64bit processor? Will Vista make a difference?
>
> Also one final point, is there any company to recommend. I was looking
> at a compaq and noticed that there seemed to be a lot of friction
> between my finger and the touchpad, same for the HP, is that normal.
> Toshiba seems good, but that finishes my processor decision right
> there, Intel. Also I had a Toshiba but it's battery seems almost
> completly gone after 1.5 years, but that is probably normal. Sony isn't
> bad, but a bit expensive.
>
> Last point, I think I will need around 1GB of RAM, is that about right?
> Are buisness models more reliable and should I consider them?
>
> Thanks,
> Ben Pazolli
>

The processor is only part of the equation.

On laptops, Intel lags behind AMD in the Front Side Bus, which is the bus
that the processor uses to communicate with RAM and the PCI express and/or
video. Intel uses variants of a 400 MHz FSB, using Clock multipliers" to
get supposed "faster" buses, but they are still much slower than the nominal
1.6GHz Hypertransport Bus AMD uses.

AMD has on die memory controller, which communicates with the ram at the CPU
base frequency, again much faster than Intel, which uses the FSB or the L2
Cache.


While Intel Core 2 Duo chips may be "faster" in core frequency, a top of the
line AMD Turion64 X2 will outperform them because of the differences in the
motherboard architecture. The AMD can access data in RAM, HDD Buffers, or
the PCI express bus much faster than Intel can, and it has significantly
less latency because of the bandwidth and speed that the Hypertransport bus
offers over the Intel.

FWIW, Intel has announced that it will adopt the on-die memory controller
and Hypertransport bus at some time in the future, but meanwhile AMD
continues to offer the best performance per dollar ratio.

Another serious consideration is heat...AMD Turion64 X2 processors produce
less heat than do the Intel equivalents, although Intel spends huge amounts
of money trying to make everyone think otherwise.

While Intel has made significant strides with the Core 2 Duo, it still
cannot compete in the notebook/laptop arena because Intel continues to build
their systems using archaic technology. Hopefully this will change soon.

My best recommendation: buy an AMD system and use the money saved to buy
accessories or upgrades. You'll have a computer that will perform as good
as or better than any Intel in the same price range, it will run cooler, and
it will be true 64 bit dual core (instead of two single cores fused together
like Intel).

I hope this helps, and I hope it does not cause a firestorm from the
hardcore Intel zealots...they both offer good choices, but in the laptop
arena AMD still has the best bang for the buck.

Honu

I don't think it's productive to try and factor in numerous low-level
features like that.

Instead, compare benchmarks on complete laptops including form factor,
performance on relevant applications, and battery life.

"timeOday" <timeOday-UNSPAM@theknack.net> wrote in message
news:f7ednXMhPIxneNnYnZ2dnUVZ_q6dnZ2d@comcast.com. ..
>I don't think it's productive to try and factor in numerous low-level
>features like that.
>
> Instead, compare benchmarks on complete laptops including form factor,
> performance on relevant applications, and battery life.

Those "numerous low level factors" are what makes a computer fast and
efficient and worth the money or a dog. They are what determines your so
called "performance on relative applications" and battery life.

Honu

Hertz_Donut wrote:

>
> I hope this helps, and I hope it does not cause a firestorm from the
> hardcore Intel zealots...they both offer good choices, but in the laptop
> arena AMD still has the best bang for the buck.
>
> Honu

So you are implying you are just in the opposite fence...the side of
the AMD maddogs?

Hertz_Donut wrote:
> "timeOday" <timeOday-UNSPAM@theknack.net> wrote in message
> news:f7ednXMhPIxneNnYnZ2dnUVZ_q6dnZ2d@comcast.com. ..
>
>>I don't think it's productive to try and factor in numerous low-level
>>features like that.
>>
>>Instead, compare benchmarks on complete laptops including form factor,
>>performance on relevant applications, and battery life.
>
>
> Those "numerous low level factors" are what makes a computer fast and
> efficient and worth the money or a dog. They are what determines your so
> called "performance on relative applications" and battery life.
>
> Honu
>
>

Of course. But the only way to know how they'll all work together and
make a significant difference for you is to benchmark. You don't want
to wind up buying a slower computer just because it has some cool
feature, without which it would be even slower.

"timeOday" <timeOday-UNSPAM@theknack.net> wrote in message
news:RIadneZ4IayN7tjYnZ2dnUVZ_tudnZ2d@comcast.com. ..
> Hertz_Donut wrote:
>> "timeOday" <timeOday-UNSPAM@theknack.net> wrote in message
>> news:f7ednXMhPIxneNnYnZ2dnUVZ_q6dnZ2d@comcast.com. ..
>>
>>>I don't think it's productive to try and factor in numerous low-level
>>>features like that.
>>>
>>>Instead, compare benchmarks on complete laptops including form factor,
>>>performance on relevant applications, and battery life.
>>
>>
>> Those "numerous low level factors" are what makes a computer fast and
>> efficient and worth the money or a dog. They are what determines your so
>> called "performance on relative applications" and battery life.
>>
>> Honu
>>
>>
>
> Of course. But the only way to know how they'll all work together and
> make a significant difference for you is to benchmark. You don't want to
> wind up buying a slower computer just because it has some cool feature,
> without which it would be even slower.

When you understand the concept of what I am talking about, then perhaps we
can resume this thread. I suggest you educate yourself on HyperTransport
and the other things I discussed. They do not cause the system to be
slower. They are not "cool features", but are enhancements that yield
significant improvements in data throughput and benchmarking speed.

Buying a computer based solely on the speed of the processor is foolish.
Many of the fastest processors available are attached to motherboards and
chipsets that cannot take advantage of the power of the chip.

Honu

Hertz_Donut wrote:

>>Of course. But the only way to know how they'll all work together and
>>make a significant difference for you is to benchmark. You don't want to
>>wind up buying a slower computer just because it has some cool feature,
>>without which it would be even slower.
>
>
> When you understand the concept of what I am talking about, then perhaps we
> can resume this thread. I suggest you educate yourself on HyperTransport
> and the other things I discussed. They do not cause the system to be
> slower.

You obviously didn't even read what I said, or don't understand it.

"timeOday" <timeOday-UNSPAM@theknack.net> wrote in message
news:Qa2dnYpEfIqIXtvYnZ2dnUVZ_vydnZ2d@comcast.com. ..
> Hertz_Donut wrote:
>
>>>Of course. But the only way to know how they'll all work together and
>>>make a significant difference for you is to benchmark. You don't want to
>>>wind up buying a slower computer just because it has some cool feature,
>>>without which it would be even slower.
>>
>>
>> When you understand the concept of what I am talking about, then perhaps
>> we can resume this thread. I suggest you educate yourself on
>> HyperTransport and the other things I discussed. They do not cause the
>> system to be slower.
>
> You obviously didn't even read what I said, or don't understand it.

Your claim that the on-die memory controller, direct connect architecture,
and HyperTransport Bus are just "cool features" instead of intrinsic to the
speed and efficiency of the A64 architecture demonstrates that you don't
understand.
If you have a 500 horsepower engine attached to a one speed transmission, it
won't be very effective at all. It is the same with the motherboard
architecture.
Benchmarks are only good if the benchmarks measure what *you* use the
computer for. Most benchmarks are synthetic, and don't give a useable
reference for how a particular configuration will do unless you are using if
for the things that were benchmarked.

I have used systems that did poorly using standard benchmarking software,
but performed admirably as a terminal or a data server. It depends on many
things. As it is, too much emphasis is placed on benchmarks, particularly
those that Intel uses, which are nearly completely synthetic and have little
bearing on how the system will perform in the "real world".

My concern with your response is that you seem to think that CPU speed and a
set of benchmarks are the bible to buy a system by...nothing can be farther
from the truth.

Honu

Hertz_Donut wrote:
> "timeOday" <timeOday-UNSPAM@theknack.net> wrote in message
> news:Qa2dnYpEfIqIXtvYnZ2dnUVZ_vydnZ2d@comcast.com. ..
>
>>Hertz_Donut wrote:
>>
>>
>>>>Of course. But the only way to know how they'll all work together and
>>>>make a significant difference for you is to benchmark. You don't want to
>>>>wind up buying a slower computer just because it has some cool feature,
>>>>without which it would be even slower.
>>>
>>>
>>>When you understand the concept of what I am talking about, then perhaps
>>>we can resume this thread. I suggest you educate yourself on
>>>HyperTransport and the other things I discussed. They do not cause the
>>>system to be slower.
>>
>>You obviously didn't even read what I said, or don't understand it.
>
>
> Your claim that the on-die memory controller, direct connect architecture,
> and HyperTransport Bus are just "cool features" instead of intrinsic to the
> speed and efficiency of the A64 architecture demonstrates that you don't
> understand.

Except I didn't say that. They're good features and make the chips
faster than they would otherwise be. The question, though, is whether
the resulting system is faster than the competition, which has a few
tricks up its own sleeves. There are a large number of factors in favor
of each side, and nobody but nobody can tell you exactly how a given
design will work until they put it into a complete system and try it out.

> If you have a 500 horsepower engine attached to a one speed transmission, it
> won't be very effective at all. It is the same with the motherboard
> architecture.
> Benchmarks are only good if the benchmarks measure what *you* use the
> computer for.

I did say "performance on relevant applications."

Most benchmarks are synthetic, and don't give a useable
> reference for how a particular configuration will do unless you are using if
> for the things that were benchmarked.
>
> I have used systems that did poorly using standard benchmarking software,
> but performed admirably as a terminal or a data server. It depends on many
> things. As it is, too much emphasis is placed on benchmarks, particularly
> those that Intel uses, which are nearly completely synthetic and have little
> bearing on how the system will perform in the "real world".

Agreed, you can't rely on benchmarks from either manufacturer. But I
still disagree that benchmarks aren't the bible. They *are* the bible.
The only way to know how well a system works is to try it, and that's
what a benchmark is. But as I said in the first place, you also have to
include other factors aside from speed such as form factor and battery
life, if those matter to you. And, yes, you have to be careful not to
focus on performance benchmarks that don't matter to you (including all
narrow, synthetic benchmarks).