Pentium II power is strong but not long

In this issue, GCN concludes its two-part look at more than 30 portable computers. Last
issue, the GCN Lab reported on 11 Pentium MMX notebooks and one AMD K6 notebook. This
issue the lab examines a dozen 233- and 266-MHz Pentium II units. Read about 300-MHz
models in the Test Drive on Page 1.


To keep a Pentium II notebook computer alive on a cross-country plane trip, you must
either fly a supersonic jet or carry extra batteries. None of the dozen portables in this
comparison could survive more than 3' hours on a charge, even under the most conservative
power settings.


But users who buy notebooks with 233- or 266-MHz Pentium II processors likely care more
about processing power than electrical power. Pentium II power in a notebook? Yes, really.


Because notebooks run at far lower voltages than desktop PCs, they suffer from an
inherent electrical disadvantage. Even so, the portable Pentium II processors in this
comparison achieved up to 95 percent of the speed of desktop processor equivalents on the
GCN Lab’s GCNdex32TM benchmark suite.


Most users, however, notice little difference between the two types of Pentium IIs.
What they pay attention to is notebook weight and battery life.


That’s why the IBM Corp. ThinkPad 600 and the Dell Computer Corp. Latitude CPi
earned Reviewer’s Choice designations. Each demonstrated superior features and
performance with reasonably low weight and acceptable battery life.


As in last issue’s comparison, this dozen notebooks—plus the five in the Test
Drive on Page 1—just do not have much staying power. Before investing in a Pentium II
notebook, consider whether you truly need this processor.


You do under one of three conditions:


If you’ll mainly check e-mail, write with a word processor or work in other basic
applications, a Pentium MMX notebook will suffice.


To be fair, it is not solely the Pentium II processor that curtails battery life.


Most manufacturers regard Pentium II notebooks as the top of the heap, so they load
them up with the largest possible displays and hard drives and other power-hungry
components.


Let’s dispel a few myths about mobile computing:


Compaq Computer Corp. boasts that its Armada 7800 has, among other things, an
integrated AC adapter. But the unit alone weighs 8 pounds, 13 ounces. IBM’s ThinkPad
600 does not have an integrated adapter, yet its luggable weight with an adapter and all
the other extras is 7 pounds, 6 ounces—about 1 1/2 pounds less than the Armada by
itself. Which would you rather lug around?


Only five notebooks in this comparison had acceptable battery life: IBM’s ThinkPad
600 and 770ED, Dell’s Latitude CPi, Hewlett-Packard Co.’s OmniBook 4100 and
Toshiba America Information Systems’ Satellite 4005CDS. The ThinkPad 770ED was heavy
at almost 8' pounds. The Satellite 4005CDS had audio problems and a passive-matrix
display that held it down in overall ratings.


My favorite, the IBM ThinkPad 600, weighed less than 6 pounds. For the most part, IBM
did not compromise on features except for one item: a lithium-ion battery that lacked a
smart LED readout to show the cell charge.


Thin, light and well-designed, the ThinkPad 600 earned top marks and was superior to
other ThinkPads I’ve used. Its dual-spindle design makes a third spindle accessible
via a cable to the bay.


Hard-drive and CD-ROM access speeds were a little slower than for other notebooks in
this review. Overall, benchmark performance put the ThinkPad 600 right in the middle of
the dozen.


Even though the Dell Latitude CPi D266XT scored slightly low on most benchmarks, it is
a sophisticated, mature unit worthy of any mobile computer user. A 61'4-pound weight and
the longest battery life among the dozen made the Latitude stand out.


Many notebook computers have an energy gauge in a tiny window in a corner. The
Latitude’s gauge fills the screen so it is easy to see battery status. And the BIOS
layer was easy to reach by a keystroke to adjust power consumption and other settings.


The Latitude’s stereo speakers trumpet crisply—the best of the bunch. Priced
at $2,686, the Latitude easily earned the only Bang for the Buck designation among the
30-plus notebooks examined.


The IBM ThinkPad 770ED would have scored a little better had it not weighed 81'4
pounds. Other vendors managed to accommodate most of the 770ED’s features in packages
weighing less than 8 pounds.


One feature in the 770ED topped the others: a DVD-ROM drive. It was cool to have but
not essential for most government users. IBM’s drive worked for the DVD games I
tested, but it does not have the appropriate software drivers to play DVD movies.


Moving from the heaviest to almost the lightest unit tested, NEC Computer Systems
Division’s Versa SX would have made the top three except for its inefficient battery
use. Perhaps the 14.1-inch display shortens its battery life.


The Versa has an attractive silver-gray chassis and is the thinnest of the bunch at
about 1 ' inches. Its CD-ROM drive also works the fastest. Overall, the Versa performed
strongly on benchmarks.


Hewlett-Packard’s OmniBook 4100 also performed well for a 233-MHz Pentium II. I
liked the dual input devices. The presence of an eraser-tip pointer as well as a touchpad
should satisfy almost any user except for those who prefer trackballs.


The OmniBook 4100 would be especially good at government sites that let multiple users
share notebooks.


The OmniBook did have one quirk: The curved chassis makes it difficult to insert a
floppy disk or CD-ROM.


Compaq’s Armada 1700 and Armada 7800 both weighed too much at 81'2 pounds or
more, and both had short battery life. The 7800 cost about $2,100 more than the 1700. For
the extra money, the 7800 has a 7.6G hard drive, 14.1-inch display and better overall
performance.


Although Toshiba’s Satellite 4005CDS has a passive-matrix display as mentioned
above, its major problem turned out to be sound. Sometimes it worked, sometimes it
didn’t. Otherwise the Satellite performed fine.


IBM’s ThinkPad 380XD is a little bulky and pricey for a bargain system. Moreover,
the 380XD always thought it was plugged into wall current even when it was running from
battery. I never knew how long the 380XD would last before dying.


Micron Electronics Inc.’s GoBook2 officially weighed the least, at about 5 pounds,
but I could barely get it to operate for 90 minutes. Furthermore, unlike all the other
notebooks in this comparison, the GoBook2 only has one spindle working—the hard
drive.


Gateway Inc.’s Solo 2500LS gained a pound over its predecessor, the Solo 2200
examined last year [GCN, Sept. 8, 1997, Page 33]. The 2500LS has only a 12.1-inch display,
so that didn’t boost the weight. Perhaps it has a heavier nickel-metal hydride
battery. But the Solo 2500LS is the only one among the dozen to incorporate two Universal
Serial Bus ports, and it performed particularly well on the 3-D benchmark.


Dell’s Inspiron 3200 D233XT performed pretty well on the 3-D benchmark and
generally well elsewhere. Unfortunately, its 2G hard drive is oddly partitioned with the
c: drive at 1.5G and a d: drive at 495M.


Despite the presence of the 32-bit Windows 98 operating system, Dell formatted the hard
drive using the 16-bit Windows File Allocation Table, which still could have recognized
the full 2G. 


The GCN Lab invited notebook computer manufacturers to send a selection of distinctive
portables for evaluation, and 14 companies sent us 30 units.


After checking them for shipping damage, the lab staff categorized them by processor
and speed, weighed them and examined components. We powered up each system with AC
current, installed an operating system if necessary and looked for conflicts.


The lab then loaded test applications and files and ran them in the same sequence:
Symantec Norton Utilities 3.0 for Windows 9x with appropriate updates, GCNdex32TM
benchmark suite, year 2000 readiness tests, multimedia files and an OpenGL 3-D video
benchmark.


Sometimes other applications were also loaded and run afterward.


During testing, the lab staff disabled certain components of the hardware, operating
system or test applications if they adversely affected benchmark performance.


To simulate typical use patterns for battery life, we carried the notebooks out of the
lab for a period of time.


After each notebook returned to the lab, the lab staff turned it off and attached it to
an AC outlet to charge overnight before running the maximum-drainage battery test.


We examined and recorded the notebook’s features in a testing database, attached
the unit to the GCN Lab network, and downloaded and printed the results of all tests. When
testing was complete, the lab wiped the hard drive before returning the unit to the
manufacturer.


For further details about the testing methodology, visit the GCN Lab’s Web site at
http://www.gcn.com/gcnlab and click on “More
about the GCN Lab.”  

GCN Lab assistant Donovan Campbell contributed to this report.

inside gcn

  • HPE SGI 8600

    New supercomputers headed to DOD

Reader Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above

More from 1105 Public Sector Media Group