Best laptop for a programmer?

[shef]

What's the best laptop out there for someone who writes a lot of code?

Having really snappy fast performance can enhance a programmer's productivity a lot. I need a system with a fast disk and a fast processor (and yes, I lot of what I do is multithreaded, so dual core would help). Really good integer performance matters.

What I don't need: fancy video, speakers, or other things that a gamer might need. Although support for dual monitors would be nice.

This is basically a desktop replacement. I take it home at night to keep working. I travel now and then, but performance is more important than weight.

Any suggestions?

 

[Goran]

For programming anything with a nice big screen and high resolution. Top notch dual core pentium m processor and at least 2 gigs of ram. For gaming you only need a fast GPU like 7800GTX go

 

[killernotebooks]

Why don't you wait a couple months for the Turion X2? Then you will have 64 bit performance and future upgradability and also dual core that you wanted? I would aso recommend a system with a video card capable to running Vista since any system you get now is going to last you for several years.

 

[DougalJ]

Killernotebooks, while I respect your opinion, I cannot agree with the decision to purchase a turion laptop or wait for a turion X2.

It has been shown from benchmarks that the advanced form of the P6 architecture used in the pentium M 'Core' CPUs has anywhere from 1.6 to 2.0 times the IPC of netburst. By contrast, the AMD K8 architecture used for Turion64 has been shown to perform at approximately 1.5 to 1.8 times the IPC of netburst.

Given this, we can examine the maximum core clock speed frequencies. The fastest clocked single core pentium M is the 780, clocked at 2.26 Ghz, the turions are limited to a 2.2 Ghz clock speed at most. While this only leaves a 60 Mhz difference between the two chips, the greater IPC of the Pentium M whisks it into the lead by as much as 10%.

In addition, while the Turion64s theoretically Should consume less power than the combination of Pentium M and external memory controller, it has been shown in practice that the pentium m actually uses less power, this is likely attributable to the ability of the Pentium M to dynamically deactivate its L2 cache to save power.

In argument for dual core, they arent significantly more expensive, dont use significantly more power, and arent clocked significantly slower, as a result, I see very little reason to purchase a single core processor for either laptops or desktops anymore.

Also, your term "64 bit performance" I find a bit flawed. 64 bit processors are not automatically faster because they are 64 bit, this is very evident For intel's 64 bit netburst CPUs when compared to the (now very few) 32 bit netburst variants. 64 bit processor almost exclusive benefit is that of addressable memory space. There are certain situations where double-precision operations may be sped up by 64 bit CPUs, but these are rare. The sole reason 64bit is useful now is that it can address more than 4 Gbyte of memory.

Given all of that, for desktop systems I would ABSOLUTELY use AMD CPUs, I myself have two AMD desktop systems. But for laptops, I cannot reccomend AMDs turion line of processors.

(The mention of a middle/low end Graphics system is probably a good idea, laptops with 6100/6150/6200/6400/X300 or equivalent will be plenty for your purposes, and will help offload some of the graphics acceleration off of the CPU)

 

[killernotebooks]

That argument is well thought out and presented. This is my rebuttal, not a personal attack (although my writing style tends to come off that way).

It's clear that NGMA chips will be much better than Netburst-based chips, AMD's Richard said. But he's not convinced that the improvements will be enough to overcome what he called a 15 percent performance advantage enjoyed by AMD's chips today.

While synthetically your argument may be true numerically; I have used Intel single and dual core as well as Turion processors and X2 desktops. There is absolutely NO WAY you can compare a 2.26 Pentium M to a 2.2 GHz Turion. It isn’t even remotely close, but hey, use them both and you be the judge... let's move on to this core duo.

Talk about a marketing phenomena... core due it is just that… marketing. I tried a 1.66 core duo and it doesn’t cut the muster. I have no idea how much money Intel is unloading on marketing, but if I were them I would be freaking praying that Conroe and Merom live up to all the initial claims and that AMD doesn't make any new advances by then to throw more dirt in the hole Intel is enjoying at this point.

Why would the notebook arena be any different then the desktop arena? The X2 dual core desktop humiliates the Intel class dual cores. Intel makes all sorts of wild claims about AMD's technology not stacking up to their own, and downplaying the features AMD processors support, but when Intel's R&D finally catches up they start implementing the things AMD is doing that were previously stated as junk.

NGMA chips can process more instructions per clock cycle than their predecessors, take advantage of larger amounts of cache memory, and route instructions more intelligently through the central processing unit (CPU). This will allow Intel to retake the performance lead currently held by the AMD64 architecture without resorting to techonology similar to AMD's integrated memory controller or point-to-point interconnects, said Mooly Eden, general manager of Intel's mobile platforms group.

So Intel's own general manager of mobile platforms admits that AMD has the performance lead... interesting. Reading that statement it appears as if the onboard memory controller that Intel has previously said makes "insignifcant performance gains" in reality is putting Intel to the tap and the funny thing is Intel cvan't get their integrated memory controller to work!

Intel has been hesitant to embrace the integrated memory controller since the failure of its last attempt to use such a design. Integrating the memory controller allows that vital gateway between the CPU and the memory to run at the speed of the processor, whisking data into the processor at a high rate of speed. But it also forces the processor to be designed specifically for a certain type of memory, which doomed Intel's Timna processor when its integrated memory controller was designed for Rambus' short-lived RDRAM standard.

Intel executives, such as Chief Technology Officer Justin Rattner, have talked about pursuing integrated memory controllers for chips toward the end of the decade, but the company has not shared any specific plans.

Intel's talking about a lot of things that are "going to be, but sounds like right now they are tapping out. What's going to happen when AMD moves to DDR2?

Why if Intel is so great didn't we have 64 bit computing 3 years ago? AMD had it but Intel couldn't figure it out... so they downplayed it. They stagnated the market because no one wanted to move forward without (at the time) the dominant market share player. Why did a company with so much market share and so much money for R&D get taken downtown in the desktop market ... then the server market ...

The Opteron chip has lifted AMD's share of the x86 server processor market from virtually zero a few years ago to 14.3 percent in the fourth quarter of 2005, according to IDC.

then the notebook market? If Intel is so great why are they hemorraging market share?

Strike three for Intel
By Michael Kanellos and Tom Krazit
Staff Writer, CNET News.com
Published: March 3, 2006, 1:20 PM PST

Intel warned on Friday that its revenue for the first quarter would come in at between $8.7 billion and $9.1 billion, roughly $500 million lower than estimates the company issued in January. The Santa Clara, Calif.-based chipmaker cited a weak market and a "slight" market share loss.

AMD's surge can be seen most strongly in the U.S. retail market, which accounts for about 9 percent of global PC shipments. In the first seven weeks of 2006, AMD's share in desktops in that area climbed to 81.5 percent, while Intel's has slid to 18.5 percent, Baker said. That's almost a complete reversal of their typical relative positions.

Intel is getting to be like Microsoft... big claims then delays and striped down versions of what they initially presented. The really annoying thing about Intel is instead of doing it right they rush to market and then change their stuff a million times, change the name of everything 5 million times and hype their stuff and throw marketing and advertising dollars out 10 million times more than AMD.

How long has AMD made socket 939 last? How long have they made DDR memory last? When they design something it is desighned right so they don't burn their customer base

AMD will introduce support for DDR2 (double data rate 2) memory along with a new socket technology called AM2 in the second half of the year. That will allow system builders to drop quad-core processors into the same chipsets for upcoming dual-core chips, he said.

Why don't you speculate on how much more advanced the computer industry would be right now had Intel not been a bloated pig stonewalling the industry?

Now everyone is entitled to their own opinion, and I am not going to try to change your mind. I am going to ask one question... Of everything you said above... how fast will the core-duo Intel be when everyone is using 64 bit O.S. & programs? and I will leave you with this quote,

As might be expected, Advanced Micro Devices doesn't think much of Intel's performance claims for its upcoming Conroe and Merom products.

"It's driven by the fact that they can't talk about their current products, because everybody knows their current products aren't very good,"