Pentium-M Notebook Put To The Test

BedivereW writes "Tom's Hardware has an interesting review of the first Intel Pentium-M (codenamed Banias) notebook. There are a few pieces of information missing, like heat production, but on the whole it is a good review. Intel appears to be moving in the correct direction." I'm looking forward to seeing more info on this one - seems to be the x86 response to the PowerBook series.

Comments

Wireless choice?

[xWeston]

It is weird to me that Intel plans to have an 802.11a/b solution in these notebooks as a standard. It seems that going with 802.11g would be a better choice considering it works with B and is fast as A but has better range (some speculation here about speed etc since it is not as well tested).

These notebooks arent planned to come out for a while, and considering there are 802.11g-draft products already available for purchase it seems that they would also go this route.

Anybody know why they may not be considering this? Possibly it is in the works, who knows.

Re:Wireless choice?

[dWhisper]

Major manufacturers are still reluctant to put in b solutions, and a solutions have all but vanished for laptops (line of sight requirements). The g standard hasn't been finalized for wide market production. As it is, most wireless is still a pain to deal with, especially the Intel products.

Wouldn't this reduce compatability?

[amigaluvr]

This sounds a little odd. Combining cpu and lan and some other things all on the one chip. It's suspiciously like lock-in

This means you don't get a chance to upgrade without completely changing your system. ie you are locked-in to one solution, the one commercial vendors want.

Now technically it's possible to add for example 802.11g to this, but why would any manufacturer bother when there is already 802.11b, and likely a "pentium-Mg" or somesuch, which will have it.

You will be forced to upgrade to an entirely new machine to get just one feature you need.

Perhaps they are following Apple's lead more than it seams there.

Not only the lock-in effect, but with all these features in one chip means you can't for example repair your 802.11b if it goes down. You'll need to replace a whole new processor

x86 response to the PowerBook...?

[djupedal]

I would have expected the x86 response to the PowerBook series to be something more along the lines of "......um, we will...ah...ummm...we have this new....ummmm.....what do we have, now? .....?".

After all, PB's are all about form and function, and since Intel doesn't make a laptop, all the function in the world won't help if the form sucks.

Or was the PB bit just an attempt to start another war? Afterall, why compare a processor to a complete product...guess I don't get it.

Re:x86 response to the PowerBook...?

[Golias]

After all, PB's are all about form and function, and since Intel doesn't make a laptop, all the function in the world won't help if the form sucks.

Or was the PB bit just an attempt to start another war? Afterall, why compare a processor to a complete product...guess I don't get it.

I think I get it.

In the desktop market, Intel and AMD machines have Macintosh machines pretty handilly beat, in terms of raw performance/price. You can easilly find a $600 PC which makes the $1000 eMac look downright sluggish. Apple is forced to add value to their computers in other ways in order to sell them.

The CPU which is Apple's achilies tendon in the desktop market is actually their strength in the laptop market. iBooks and Powerbooks match up pretty well with Intel laptops in terms of performance/price, and also feature much longer battery life, and less heat (which translates into being able to build much tighter designs without down-chipping the CPU.)

If Intel's new chip allows high-speed processing at cooler temperatures and with less power consumption, it means that companies like Sony and Toshiba have a much better chance at making comparable products to the Powerbook line.

Where is AMD?

[Phigrin]

It seems as though, with each Intel announcement, that AMD is not quite there as much as it used to be. Although notebooks have never been AMD's strong point because of greater heat production.

Unfair practice?

[tanveer1979]

IMHO, wont this be unfair. There are small manufacturers in WLAN area who provide wireless cards for notebooks. In such a case intel can easily produce the whole package and very much capture the notebook wireless market bu keeping operating margins low initially. After competition is eliminated then they can put any pricing to their advantage. Its like intel manufacturing PCs with intel motherboards and only those PCs with intel motherboards will be allowed to be named as Genuine intel.

Its always good to have peripherals separate from chip. Another danger is that intel can have a set of proprietry registers and pipelines which are not disclosed however the WLAN card from intel will be able to use it to give better performance!. Not really good i think. Its okay to have own WLAN cards but the way intel is going about is not really ethical

Re:What did we need Pentium IV for?

[Anonymous Coward]

What's funny is that the "new" processor is based on the PIII, and could have been released years ago. And it's not so much funny as sad - we'll now still be stuck with the big, slow, space-heater PIV instead of a cheap, ballanced, quiet desktop processor that could have been the standard two years ago.

Re:Intel admitting that clock speed isn't all that

[Anonymous Coward]

I fail to understand how because Intel has a couple of products that are more IPC heavy than Clock speed heavy that makes 'further proof' that there is a Megahertz myth?. Megahertz *alone* doesn't make a processor faster, but sometimes Megahertz IS the only way to speed up certain types of tasks.

Intel engineered the P4 for overall performance. It's clearly spanking Athlon now, and it may be the chip that kills AMD in the end (what would happen if AMD had another round of 4-6 quarterly losses?), it would be very ironic if Megahertz is what killed them literally :-)

For notebook performance however, it seems that higher IPC = less overall power consumption (although P-M has several other advantages over P4-M besides IPC tweaks), so Intel is creating a chip catered to that market.

I'm really suprised to see the # of Slashdot readers here that can't seem to grasp that performance is IPC x Megahertz. They just shout "holy cow, Megahertz Myth!!!", well, let's take the reverse approach. I'll hold an Itanium 1 and a P4 in each hand, and yell "IPC Myth!!!", why?, because the Itanium 1 was crushed in Integer applications by the P4, so it *must* be the megahertz that won. Don't you see the ridiculousness of this argument?.

Let's just give Intel kudos on designing an advanced low power x86 core, that will hopefully serve as a wakeup call for the other vendors (ex Apple, Transmeta, etc) that it's time they update their designs significantly :-).

Long live the Banias!

Why is Apple always ahead of the curve?

[afantee]

Apple just introduced 802.11g, bluetooth and Firewire 800 in one stroke, but Intel and the rest of Wintel box makers are still pushing the awkward and more expensive b & a combination. Let's face it, even the name is confusing - version a is a few years older than version b - wtf.

Similarly, when Apple dropped the floppy drive and added USB and Firewire 4 or 5 years ago, people were laughing at them. Now even Dell are moving away from the floppy.

As another example, USB is a nice and simple low bandwidth technology for connecting mice and printers, and Firewire is ideal for high bandwidth. But oh no, things have to be more complicated, so Intel must chip in and muddle the water with the theoretically faster but practically slower 480 Mbps USB2 than Firewire 400. Now Apple has Firewire 800 and potentially 1600 and 3200 soon, are we going to see USB3 and USB4?

Hrm, intel does have SOME control...

[PCBman!]

For the most part that's true, but it also comes down to the question of what platform runs your $20k per seat license workstation software fastest AND most reliably. In which case, you might decide an entire platform and just go with the standard IT OEM buy. In this case, Intel does have some nice control because they sell the platform, not just the processor. After all, if you're an OEM and you want to sell intel chips, do you offer them on VIA or Sis chipset motherboards if you're, say, Dell or even IBM?

Now if my software ran in 10 minutes on an AMD box vs 30 minutes on Intel box and they asked what they could do to improve my performance, I'd ask them to direct me to an OEM building around their reference systems--after all, the performance is already there, I just want reliability, and no company should be sending out crappy machines as reference boxes.

Power consumption miscalculation in the article?

[bandannarama]

I think the authors got their fractions reversed. AFAICS the data indicates a significant increase in battery life, not just 20 minutes. From the article:

"One thing becomes immediately apparent: the Pentium-M system runs longer. In order to estimate approximately how much longer it runs compared to the Pentium 4-M notebook, however, the batter capacity must be considered as well. The capacity of the battery in the Pentium-M notebook is 48.8 Wh, while that of the Pentium 4-M battery is about 59 Wh.

If you then take the operating time of both platforms with respect to battery capacity, figuring in a factor of an imaginary battery with a capacity of 10 Wh for example, then the Pentium-M system gives you an operating time of about 20 minutes longer."

The chart immediately above this quote shows the Pentium-M lasting about 30% to 50% longer than the older Pentium-4-M, depending on the test. On a smaller battery. So the conclusion that it only gets you about 20 minutes is far too conservative.

On the other hand, I'm suspicious of any such dramatic increases in battery life. It would probably be fair to deduct points for the smaller screen size on the Pentium-M because it should draw less power than the 15-incher on the P-4-M.