2

I'm looking at some new laptops, and I'm seeing the AMD Neo CPUs at 1.5GHz, and some of the new Intels at 1.3-1.8GHz. The laptops they are on come well supplied in terms of RAM (up to 8GB) and Video Cards (ATI's with 2GB shared memory).

My question is, how do these laptops perform running heavy duity apps like Visual Studio, Photoshop, etc? I realize they wont run in full 3D settings, but how do games like WoW and Half-Life2 run on machines like these?

http://www.microcenter.com/single_product_results.phtml?product_id=0337298&utm_source=ACT_BYO&utm_medium=email&utm_campaign=E0897+eNews+20100824

and

http://www.newegg.com/Product/Product.aspx?Item=N82E16834157108

Bottom line: how do these lower clock-speeds effect using heavy duity apps and some games? I've read some reviews that claim these machines (and ones like them) play games with medium 3D settings without any issues at all.

1
  • Keep in mind that for gaming, most of the work is done by the GPU, not the CPU, so you could potentially play a game with high end graphics on a low-clocked CPU like one of the CULV processors and still get good performance if you have a dedicated GPU.
    – nhinkle
    Aug 24, 2010 at 17:30

5 Answers 5

4

Clock speed speed has been misleading for a long time now. Intel has had 3.2Ghz chips on the market since the P4 back in 2002/2003, and even at that time AMD's lower-clocked alternatives out-performed them. Unfortunately, there's not much to go on these days in terms of what the performance of a given chip really is without benchmarking.

But in the case of the AMD Neo family, the best comparison is to an Intel Atom processor used in the tiny netbooks. The Intel Atom is intended to give up some performance in exchange for better power use, a smaller chip size, and a lower price. In my experience, AMD's Neo alternative kept a little more performance (of course in exchange for power consumption). Some of the nicer Neo's perform as well as an older low-end Core 2 Duo, which, considering disk performance often matters than cpu these days, is still nothing to sneeze at. Pair that with one of the new low-power graphics chips from nVidia and you have a very portable and low-power system that still packs quite a wallop.

2

Its all about the architecture and bonus features.

Just like GPUs, CPUs have made it to a stage where looking at their specs you have no idea what their overall performance will be, the numbers on the box are fairly meaningless these days.

2
  • So you're saying that if there are many good reviews of it for doing some basic gaming and visual studio I should weigh that more than the Clock speed on the box?
    – Nate
    Aug 24, 2010 at 16:21
  • I don't know if that's exactly what (s)he's saying but I'll say it. You should weigh benchmarks more than the clock speed on the box.
    – Shinrai
    Aug 24, 2010 at 23:00
1

The Neo cpu is an AMD celeron analog.

The P6000 in your other example is a lower power core2 chip (25W vs 35W TDP), performance falls off as expected from the lower clock but you get better battery life in trade.

For intel's newer I3/5/7 mobile chips the base clocks of all the normal power ones are in the 2-2.5ghz range and are about 20% faster per clock than the core2's they're replacing. The i5/7 chips can also temporarily increase their speed by a few hundred mhz under heavy load.

The ultra low power chips are roughly half as fast but can boost even higher for short terms. They're also aimed as replacements for the ULV core 2's that ran in the 1-15ghz range. They're not intended for CPU intensive apps although for compiling turboboost can probably help a lot. Gaming less so because the high sustained loads will have the chips clock back down as they get hotter.

Intel doesn't currently have a direct replacement for the 25W TDP core2 chips like the P6000.

1
  • Neo is closer to Atom than Celeron. Turion is AMD's celeron clone. Aug 24, 2010 at 17:04
0

To answer your question: The deal is to conserve power, thus enabling it to run longer.

Can they run apps well? Well, depends solely on the app.

-2

Measuring a CPUs performance based on clock speed is like measuring a car's performance based on the flow of fuel through the fuel pump.

Examples:

I have seen a custom processor that can solve a problem that would take the fastest computer 20yrs to solve, in just a few hours. It runs at 100kHz (0.1Mhz, 0.0001GHz), but has a massively parallel architecture. (As do GPUs)

I have worked on another very simple processor that runs an over 100GHz, but can't do much just one add, one subtract, and a multiply per tick ever tick. (actually that is more than an x86 can do, but then most processors can do more than any x86 can do per clock tick).

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .