Several pictures show just how snug an Eee PC 901 can be stored away inside the protective shell of a hollowed out book, and could inspire those to use the idea as an anti-theft device, considering how much, when the book is closed, it would be difficult to guess that it wasn’t actually a book at all!

This brings to mind the idea that if one wanted to push the concept further, there are various software configurations available for viewing ebooks and flipping the display on the Eee PCs to mimic a book standing vertically, so you could be reading an ebook on your Eee PC while it is housed and resting in your actual hollowed out book.

So check out the project and let us know of any other similar modifications other Eee Users have come up with.

Clove Technology, a UK supplier of Eee PCs and accessories, has begun offering a cheap alternative to the (as yet unseen) ASUS extended battery. These are made by the Chinese firm Cameron Sino, and are available in 4-cell 5200mAh and 6600mAh 6-cell variants.

Given the uncertainty over the ASUS UK 900 Eee PC battery upgrade offer, it seemed a good idea to revisit these batteries, and perform some comparative performance tests against the ASUS batteries that come with the 701 and 900 Eee PC variants.

Appearance.

First off, let’s consider the appearance. All the batteries here are suitable for fitting to white Eee PCs. The CS (Cameron Sino) 5200mAh battery has been sprayed with a pearlescent finish that is very close to, but not quite the same colour as, the standard ASUS Pearl White batteries. The difference is quite difficult to capture with a digital camera.

See what I mean? The CS is on the right, but this shot exaggerates the differences somewhat. In practice, you barely notice the difference; it is slightly darker in shade. The CS 6600 mAh battery is a different matter, though, as it is finished in matte white plastic, unpainted, so it is very similar to the finish of the Eee PC 700 (2G) Surf models. Fitted to the 900, it doesn’t really matter that much; all you are missing out on is a bit of pearlescent sheen.

I’ve tried to take a better shot in daylight of all three here (CS 6600mAh on the top, CS5200 bottom left, ASUS 4400 bottom right).

The CS batteries are, of course, available for the Galaxy Black Eee PCs, but I’m not able to compare their colour against a black Eee PC as I don’t have one to hand.

Update : eeeuser forum member GrahamK posted some excellent images here.

Physical dimensions and fit

The CS 5200mAh battery, as has been commented on elsewhere, is a little bit tight fitting, because it is marginally bigger in size when compared to the equivalent ASUS battery. No such problems with the larger 6600mAh CS battery, though; it seems to fit much more easily into the Eee PC’s battery slot. However, because it is a six cell battery, the manufacturers have had to make a decision as to where to put the extra two cells, and they have elected to extend the battery backwards.

There it is, sticking out by around 1cm. The other option would have been to mount the extra cells underneath the battery pack, thus raising the back of the machine. What the CS battery does, therefore, is not alter the height of the rear of the keyboard, but in return you are unable to open the screen as far as you would have done had you opted for the other design. This next picture demonstrates how far back you can push the screen, before the CS battery gets in the way.

Weight-wise there’s not much difference between the three 4-cell batteries – they all come in around 200g on my (admittedly not too accurate) kitchen scales. Understandably, the 6-Cell battery weighs an extra 75g.

Performance

Before I state how much battery life each of the batteries on test yielded, I’ll explain my testing method.

All testing was performed with a 20Gb Linux 900 upgraded to 2Gb of memory. The test comprised of fully charging the battery under test, then (from within a shell window) launching a shell script (click on the link to see it – I posted it in the UK battery thread last week) that writes a timestamp and some battery information into a file, then repeatedly touches the file with a 10 second delay. The Linux “touch” command simply updates the “last altered” timestamp on the file. Then, with the wi-fi turned on and associated with an access point, a compressed AVI file was run in the foreground using SM Player on repeat. The machine was then left alone to run down its battery (at which point it would shutdown automatically). The next time the machine was booted, the file was examined and the in-file timestamp compared with the file’s last modified timestamp (using the command “ls -l <file>”) to derive a run time duration for the battery. This was done for the ASUS 4400mAh battery (supplied with the UK 900), the ASUS 5200 mAh battery (supplied with the UK 701), the CS 5200mAh battery and the CS 6600MAh battery.

The results, then:-

• ASUS 4400mAh : 1 hr 58 minutes
• ASUS 5200mAh : 2 hrs 34 minutes
• Cameron Sino 5200mAh : 2hrs 17 minutes
• Cameron Sino 6600mAh : 4hrs 37 minutes

For some reason the 6600mAh battery is giving much better performance than might otherwise be expected, considering it only has an extra 1400mAh capacity over its smaller sibling. I re-ran the test, thinking that I’d made an error of some sort, and again it yielded 4hrs 39 minutes. Impressive.

Some other things I’ve observed whilst using the 6600 mAh battery:

• With wi-fi off, and watching uncompressed video stream files from a WD Passport LifeDrive (320Gb) USB powered hard drive, I usually manage 1hr 20 mins before getting the first low power warning with the ASUS 5200 mAh battery. With the CS 6600 mAh battery, I can do this for twice the time and still have 60% battery life.
• The CS 6600mAh battery will take around 3 hours 45 mins to recharge from flat.
• Running the standard test without wifi turned on will give you over 5 hours of usage with the 6600mAh battery.

Here’s something else that is interesting. If you divide the rated power of the battery by the number of minutes of life, you get an approximation of the average number of mAh that the life of the battery is reduced by every minute. For all the batteries under test, apart from the CS 6600mAh model, this figure is around 35mAh. The figure for the 6600mAh battery is 23.8, much lower. This is not because the Eee PC is drawing less power from the battery, it is because the battery has more capacity than is stated on its label. Working backwards from a drain of 35mAh / minute for 4 hrs 37 mins (277 minutes) gives you a theoretical capacity of 9695mAh, which is a real surprise.

Please note that I’m not asserting it really does have this capacity. We observe in the test that it is giving uenexpectedly long life, so I’ve used some simple math to try to quantify what that extra capacity might be.

A note about “jump starting”.

Some reviewers have noted that the Cameron Sino batteries suffer from the so called “Jump Starting” problem. This is where you change to a CS battery, but the machine refuses to boot up until you plug it into the mains adapter. Then, on unplugging, the battery continues as normal.

I’m pleased to say that this has not occurred once with these particular batteries.

Prices.

There are many places to get these batteries from, but most of them are overseas eBay sellers. My recommendation would be to leave them and order from a reputable UK source. Thankfully, Clove are offering them at reasonable prices (£29.38 inc. VAT for the 5200mAh variant, and £41.12 inc. VAT for the 6600mAh) and as they have a reputation for top notch after sales service you can buy with confidence.

Conclusion.

I have to recommend the Clove Technology / Cameron Sino 6600mAh battery on performance alone. Sure, it sticks out a little at the back of the machine and interferes with the screen’s normal opening angle, but it lasted much longer than I had any right to expect, and in normal use the extra bulk isn’t such a big deal. That’s why I put my money where my mouth is, and bought one.

So, with this battery, the Eee PC 900 delivers on its promise of portability. Just imagine how much more mileage you’ll get on a 701…

Final note : My thanks to Clove Technology for supplying the two Cameron Sino batteries for testing.

Typical of the responses our UK users are getting from ASUS UK is this one, which eeeuser member loopz posted on the battery discussion thread. loopz says it was received from ASUS customer service in the UK.

Dear Customer,

Like I promised I would give you an update as soon as I got one.
The Exchange for the battery is postponed until further notice.
I am so sorry for this late call but I only heard 2 minutes ago about it.
The minute I know something else I will let you know.

With Kind Regards
Glady’s

It seems to me that, with the competition (for example, the MSI Wind) hot on the heels of the ASUS EEEPC 901, ASUS could have handled the UK battery situation far batter. Potential buyers will surely be asking themselves if this is the sort of after sales service they deserve after forking out their hard earned money on a new UMPC – and they will also no doubt be considering the alternatives.

The ASUS press release quoted by Mobile Computer Magazine’s web page starts off with the bold statement “ ASUS delivers unique customer service“.

UK 900 buyers must be finding the customer service offered by ASUS UK truly “unique” indeed.

Effectively, ASUS are to provide an upgrade path. This follows a concerted effort by eeeuser.com forum members, led by user shade360, to lobby ASUS for a battery replacement. The downside is that there is a nominal cost involved (£10 + VAT) and the original 4400 mAh battery has to be returned, but at least UK 900 owners can now opt to enjoy the full 900 experience, as the replacement batteries will be the proper 5800mAh rated items.

According to the press release, “ASUS will release information regarding the detailed procedure in a few days and anticipate the battery purchase service commencing from 10th June 2008.”

Read the full discussion on our forum here.

The Eee PC 900 was unveiled for the Asia Pacific market during late April and for the US market on May 12. But Asustek is said to have stopped taking any more Eee PC 900 orders as the Eee PC 901 is set to launch on June 3, the sources said.

It’s not really specified whether or not production will resume or if the device will be discontinued.

What will happen to users who are still on a backorder list for the device? What about the battery situation? It appears as though the 900 was simply a 1-2 month transitional device in the eyes of ASUS. What do you think about this?

[edit: originally stated discontinued, but perhaps it's just a pause in production]

While other parts of the world already can get their hands on Eee PC 900 devices, it has not yet been introduced to the US market. Now, it looks like it is getting closer to a release: Amazon.com has both Eee PC 900 versions listed for pre-order through themselves (and not a 3rd party retailer): the 20G model that comes with Linux and the 12G model that comes with Windows XP can both be found on the website. Of course, they feature the new 8.9″ screen with a resolution of 1024×600.

[thanks jqw3]

Reviews for the new ASUS EEE PC 900 UMPC are coming in thick and fast, but none of them seem to go into the sort of depth that the forum members here at eeeuser.com would like. With that in mind, and with my new 900 sitting on the dining room table waiting to be unboxed, I decided to attempt to answer some of the more technical questions that people have been asking on the forums, and to compare the 900 with the 701 machine I already own.

About this particular EeePC.
This machine is an official UK Eee PC 900 model, supplied by Clove Technology (www.clovetechnology.co.uk), running the Xandros Linux operating system.

UPDATE : A note about the performance measurements.
Since writing this review, I found out that setting the 900 to Power Saving mode has no effect if the machine is connected to the mains supply. To get the BIOS to actually do this means pulling the plug and running it on battery power alone. So when I saw no difference in performance between Performance and Power Saving modes in my review, I was being mislead – it was all effectively in Performance mode, since the machine was plugged into the mains. I’ve since added some more results in real Power Saving mode, and they seem to indicate little or no gains over the 701 (obviously the Performance mode does give a real boost, though).

Unboxing the EeePC 900

Before we get to the nitty gritty, let’s take a very quick look at what is in the package.

The box is exactly the same size as the 701’s box. What you find on unpacking is the machine itself, the power supply with separate cable to connect to the mains socket, a 4400mAh battery (more on this later), the Quick Guide and full User Guide, a pair of DVDs (one is the restore disk, the other the support disk) and a carrying pouch. The pouch is not made of neoprene like the 701’s pouch; it is a soft foam-padded material which looks like it will protect the machine well, but pick up a lot of dirt on the way. It looks like light grey colour will not help in this regard – I would have preferred it if ASUS had stuck to black.

Physical dimensions

The 900 is a little bigger than the 700 series.

Looking from above, with the 900 on the right, you can immediately see it is a little deeper than the 700. The difference is around 7mm. They exactly are the same width, and, at the highest point (where the hinge is), they are also identical. However, due to the speaker repositioning and a slightly thicker screen, the front edge of the 900 stands proud of the 701 by around 3mm. As you can imagine, this is not much of a problem, and the 900 can be fitted into a Brando or PDair leather folio style case with no effort. I was unable to accurately measure the relative weights of the two machines, but the difference is in the region of 50 grammes (the 900 is slightly heavier), but you are not going to notice this if you are used to carrying a 701around.

External connections

The Eee PC 900 is almost identical to the 701. On the right hand side there are the MMC memory card slot, two USB 2 ports, a VGA port for an external monitor and an attachment for a Kensington lock (which, like the 700, cannot be used at the same time as the VGA port). On the rear is the power socket, which is the same size as the 701’s, so if you have both machines, take care not to accidentally plug the wrong adapter into either of the machines. On the left hand side is the Ethernet port, a third USB2 port and microphone and headphone sockets. The microphone socket on the 701 is said to be stereo, and also to support line-in levels for digital capture from an audio source, so it is likely that the 900 can do this as well.

The only real difference is the modem plug, which is missing on the 900. There is no rubber bung as are fitted to the 701’s; the new case fills the hole.

Boot times

The 900 sat in the kitchen for a couple of hours having its initial charge. One the orange charge light had gone off, it was time to see how fast it booted. Before doing this test, I went through the first time start wizard, and found myself in the now familiar Easy Mode interface. In order to compare like-with-like, I re-imaged the 701, so it would be as it comes from the factory. This would give me a reasonable comparison, as previously the 701 was running the Advanced Mode desktop, and loaded several kernel modules at startup that are not in the ASUS Xandros Linux distribution – all things that adversely affect boot times.

To time the boot up sequence I used a wristwatch. The time measurement began when the power button was pressed and ended when the Easy Mode desktop finished displaying itself.

Boot times were not quite as expected. The 701 boots in 26 seconds. The 900 boot time was 30 seconds in Power Saving mode, and 26 seconds in Performance mode (these settings are available in the 900’s BIOS). I timed shutdown as well, from the moment the shutdown confirmation dialog box’s OK button was clicked, to the power light going out. Once again, the 701 beat the 900, but only by a second. The timings were 701 – 5 seconds, 900 – 6 seconds.

Screen Quality

The 900’s screen had no dead pixels, but looked a little grainy. This is caused by the anti-glare membrane, which is matte to cut down on reflections. The 701 has this film on its display, too, but – perhaps because it is lower resolution – the graininess isn’t so obvious.

The 900 screen is bright and clear, but has a slightly washed out look to it, which you notice even more if you compare it against the 701’s panel. I also noticed, when setting up my Advanced Mode desktop, that the colour balance on the 900’s panel is different to the 700’s. Whites look kind of green tinged, whereas on the 701 panel they are very white. When setting the window manager colour settings, I copied colours exactly as they were on the 701 (by setting RGB values in the colour selector), but they looked different on the 900. I also found it a little difficult to get good contrasts – it was necessary to tilt the screen back quite a way.

The 900’s brightness control seems to operate over a slightly greater range than the 701 – meaning it can go darker and brighter, and this is very handy in very bright or very dark environments.

Keyboard comparison

The 900’s keyboard is exactly the same as the 701’s keyboard. It’s a little cramped, but has good response. I find myself making quite a few mistakes when typing long pieces of text, but I have improved quite a bit since first trying out the 701. For a machine this size, it remains one of the best you can get, so prospective purchasers should not worry that it is under par.

Touchpad

One of the main improvements made to the 900 concern the touchpad. It is quite a bit bigger than the one found on the 701, and there is a good reason for this. It is a dual touch device, that supports various “gestures”, which have been demonstrated on various web sites already.

Pointing, tapping

The 900’s touchpad works very much like the 701’s for moving the mouse pointer about and tapping. It does seem a bit less responsive, though, and I have spent a little time adjusting the sensitivity setting in the Touchpad setup program (on the Settings tab). When you wish to move the pointer minutely, it is very difficult. On the 701 touchpad, this can be achieved by rolling your fingertip slightly, but on the 900’s pad, doing this has little effect. It can be frustrating at times. Tapping is more or less the same as the 701, except that the 900’s touchpad does not get so sensitive when it gets hot, so cutting down on accidental taps. Single and double taps are, of course, supported.

Clicker bar

The 900 has a smart new clicker bar under the touchpad, just like the 701 does. This time it sports a bevelled edge that matches the bevel on the front edge of the machine, plus a trough that suggests there are two buttons rather than one. It seems likely that ASUS designed this to avoid the (momentary) confusion that some new users of the 701 faced. In use the clicker bar operates as right and left mouse clicks, and, like the 701, is stiff and difficult to use. This isn’t helped by the width of it, as you often find yourself trying to click it somewhere between the centre and the edge, which takes much more effort than if you click at the edge. I would have preferred the bar to have been the same size as the one on the 701. The 900’s click is also sharper sounding; more pronounced somehow, but I put this down to the machine being new.

For those rare occasions when you need to do a centre button click (that’s a text Paste operation in most Linux applications), you can press both sides simultaneously, but you need to use both thumbs to get enough pressure.

Scrolling

Gone is the 701’s “hit and miss” right hand side scrolling edge. To perform a vertical scroll, you touch the pad with two fingers and drag up or down depending on the desired direction of scroll. This works well, and is a big improvement on the 701’s single finger scroll function. The only problem is that you often find yourself scrolling too far, especially if the application in question cannot keep up with the speed of your fingers. If you do go too fast, then stop, the application continues scrolling, because it (or the touchpad driver) has buffered up the scroll messages, and it won’t stop until it has acted on all of them. What this means is you have to be a little careful with this feature.

Side scrolling is also implemented, again with two fingers. To do this, slide your fingers to the left or right. Different applications seem to interpret this action in different ways. For example, Firefox treats a left side scroll as “go back one page” and a right scroll as “go forward one page”. Again, though, the feature is plagued by the buffering problem of the vertical scroll, and I found a single left hand swipe took me all the way back to the very first page I visited in the Firefox session. This is very frustrating indeed, and I have to hope that this feature will be improved upon in the future.

Zoom

You can perform a zoom gesture on the 900’s touchpad by touching it with two fingers and then dragging your fingers apart. Similarly, you zoom out by bringing your fingers together. This has no effect in Firefox under Xandros. According to the User Gide, zooming only works in the Open Office Writer / Impress, Adobe Reader and Photo Manager applications.

Touchpad settings

The Touchpad settings dialog box on the 701 (under the Settings tab) has been extended for the 900’s features. There is a new “Two finger” tab that replace the 701’s “Edge” tab. There are settings for:-

• Two finger zoom (on / off) – “Two finger” tab
• Two finger vertical scroll (on / off and speed) – “Scrolling” tab
• Two finger horizontal scroll (on/off and speed) – “Scrolling” tab

The circular scrolling feature of the 701 touchpad has been disabled.

I found no setup tool that would allow me to assign operations to the various gestures you could perform on the 900 touchpad, so it seems that for now, Xandros users are stuck with what they are given. ASUS have missed a real opportunity to innovate here in not giving Xandros Linux users a similar interface to that enjoyed by XP users (as demonstrated to good effect on various other web sites).

We can only hope that a future software update will address this.

Loudspeakers

Because the screen is so much bigger on the 900, there is no wide bezel around it in which to mount the little stereo loudspeakers. ASUS moved them to a point under the front edge of the machine, and the effects of this are evident when you compare the sound quality with that of the 701. Whilst the maximum volume you can get out of a 900 is similar to that which you get from a 701, the 900’s sound seems to have a little more body but slightly less clarity than the 701’s. This is only to be expected, as the speakers are facing downwards, pointing at the desk surface the machine is sitting on and it is this that most affects it. If you’re fond of perching it on your lap, you’re probably not going to hear anything at all – so headphones are the order of the day, just as they are with the 701.

For the occasional system notification sound the 900’s speakers are just fine.

Microphone / Camera

The microphone has been moved away from the underside of the machine (where it is on the 701) up to the top of the screen, next to the new 1.3 megapixel camera. This is the most sensible place for it, and recordings made with it have more clarity as a result. The camera is a big improvement on the 700’s camera, and there are a pair of images taken by using the built-in Webcam application below. Let’s begin with a shot from the 701’s webcam. This is a picture showing the “well organised” shelves in my study.

With the 900’s webcam, things become a little clearer.

Both images are displayed here at 100%, so you can see the difference the 900’s camera makes. The poor colour balance is caused by the lighting conditions in my study. When filming a moving image, such as your face going into and out of shot, as it would if you were having a video conversation with someone, there’s not much to choose between the 701 and the 900 webcam speed – both tend to blur movement to a certain degree. However, for web conferencing (which is the intended use of the device) this is not really an issue.

Internal storage

The 701 Eee PC was criticised for its lack of on-board storage, having only 4GB in total. With Xandros installed, this left just 1.3GB for user data and extra applications. Not much, so it is no surprise that some of the earliest hardware modifications added in-built USB pendrives.

With the 900, there is much more space to breathe. The Linux variant gets a full 20GB of SSD, split between 4GB on the motherboard and a 16GB mini-pcie card which is accessable under the memory bay door (more about this later). Again, though, ASUS have chosen an odd partitioning scheme in order to protect the operating system files from being inadvertently damaged by the user, and this operates broadly in the same way as the 701, with various physical partitions hidden by an implementation of unionfs. All user files are now located on the 16GB drive, where /home is mounted. Using the ASUS Disk Utility (under the Easy Mode “Settings” tab), we can see how much free space is available on each of the drives.

This looks pretty good so far. The extra 16 gigs (14 after partitioning, seemingly) goes a long way to justify the 900’s price increase over the 701 (if we leave the obvious benefit of the screen to one side). One thing that is odd is that the free space on the internal 4Gb drive is nearly the same on the 900 as it is on the 701, and this implies to me that there has not been much added to the software lineup on the 900.

There is good news in the performance stakes, though. For the 900, ASUS have dropped the Silicon Motion chipset used on the 701’s motherboard for one made by Phison, and this yields some improvements in read / write speed.

To demonstrate and measure this, I first created a file of random bytes using the command (from the Terminal application – press CTRL-ALT-T to open one).

openssl rand -out 1Meg 1048576

The last number is important, as it tells the command to create a file that is exactly 1Mb in size (enter “man rand” to find out more about this command). Having created the test file, I wrote a short shell script to copy it to another file then delete the new file, and do this in a loop 1024 times:

#!/bin/bash

#--- Test the ssd write speed

count=0

while [ $count -lt 1024 ]; do
  cp 1Meg 1Meg.cp
  rm 1Meg.cp
  let count=count+1
done

This has the effect of reading and writing a total of 1Gb of data from the disk where the test file is located. I ran this test using the time command, so I could gather some rough timings for the SSD performance. On the 701, it was execcuted in the /home/user directory, because that resides on the internal 4Gb disk. For the 900, I ran it once in /home/user (which is mounted on the 16Gb SSD card) and once as root in /root (which resides on the 4Gb drive on the motherboard). The timings are interesting.

The speed of the 900’s SSD is appears faster that that of the 701.

Update : I since found that with the mains adapter plugged in, the machine stays in Performance mode all the time, irrespective of the Bios setting. A retest indicates similar performance to the 701 in Power Saving mode.

Update 2 : It’s been rightly suggested that because my test script doesn’t have a “sync” after each disk operation, then the timings won’t be too accurate, as all the I/O will all happen in cache.. Hmmm, I really should have thought of that. I updated the script with a sync command after the cp and rm commands and retimed it.

On the 16 Gb drive we get this:

real : 7m 21.968s
user: 2.920s
sys :8.38s

That’s about 0.431 sec / Mb (or 2.32 Mb / second).

…and on the internal SSD (/root):-

real : 3m 42.098s
user : 3.2s
sys : 9.11s

..a more respectable .217 sec / Mb – 4.6Mb (megabytes) per second. This tends to support the theory that the internal SSD is faster, but remember, my little script is not a particularly scientific test.

The battery

Since this is a proper “official” UK spec Eee PC 900, it is supplied with a woefully inadequate 4400mAh battery. This is a major disappointment, because early review machines were sent out with a 5800mAh battery, so this is what we have come to expect. The batteries are reputed to be in short supply following a fire at one of ASUS’s battery supplier’s factories, and so ASUS chose to supply some regions like the UK with an inferior unit. This is odd, considering the 900 is the flagship of the Eee PC range, and the lesser 701 has a 5200mAh battery. We in the UK will be looking toward ASUS to provide a free replacement as soon as better cells become available (as they have reputedly done in other markets).

Unfortunately I have not had enough time to perform formal battery testing with the 900, but it seems to last around 2 hours before needing a recharge. This can be extended by switching the wireless card off or setting the operating mode to Power Saving.

It was rather telling when, testing the 900 and 701 side by side, doing the same things to each in turn, the 900 announced that it was going to shut itself down in 3 minutes because the battery was nearly dead – and the battery monitor on the 701 was still showing 60%. This was after a full charge on both machines.

Fortunately, though, the 701 and 900 batteries appear to be interchangeable, despite the slight voltage difference. The 701 battery is rated at 7.2 volts, whereas the 900 battery comes in at 7.4 volts, thus improving its actual energy storage capacity a little. Despite this, when swapped, both machines were happy to start up using the other’s battery, so I imagine I’ll be using the 701’s bigger capacity battery with the 900 for the time being.

Charger

The charger is one more thing that ASUS have changed. Rather than having a mobile phone style “wall wart” device, where the transformer is integrated into the wall plug as per the 701, for the 900 ASUS introduced a more conventional “brick”, design which has a figure of eight plug in the side to which is attached a short lead that in turn is plugged into the wall. The specifications are a more conventional 12v 3A as can be seen in the picture. Yet this charger is very, very small in consideration with its power output. I measured the “brick” at 85 x 34 x 25mm (LxWxD), which is actually quite alot smaller than the 701’s charging unit (even if we take the wall plug part of it out of the equation). The 900’s power supply is actually easier to pack into a small bag, because the wall plug lead is separate and so can be fitted into an odd spot in the bag. Total cable length is comparable to the 701’s rather generous 3m trailing lead (the 900’s cable is around 30cm shorter, with the transformer box around 70cm away from the wall socket).

Due to the increased power output of the 900 charger, it should be able to charge the battery much quicker than the 701’s charger, and support more plugged in USB devices as well. For comparison, the 701’s charger is rated at an unusual 9.5v at 2.315A, which has made it difficult finding alternative power supplies. It is a great relief that the 900’s power supply is so much more conventional in output (12v / 3A).

The other addition of note is the inclusion on the 900’s power supply of a nice bright blue LED, which comes on whenever it is connected to the mains.

Wireless

It is a great pity that the Eee PC 900 is still fitted with the Atheros 5700 wi-fi card that the 701 comes with. The reason I say this is that the chipset it uses is not properly supported by the Open Source community, so it can be difficult getting it to work with other Linux distributions like Ubuntu. There is a solution, but it involves using a driver that is supplied partially as a binary to which there is no freely available source,m and it has to be manually installed.

That said, the 900’s wireless card works pretty much as it does on the 701 under Xandros with one notable exception – range. The 900 does not seem to have the same range as the 701. I tested this rather unscientifically by measuring the signal strength of my wireless router given by the “Wireless Networks” utility under the Internet tab of the Easy Mode desktop of each machine and walking round the house. At various locations, I put both machines down and clicked the dialog box’s refresh button a few times, noting what effect the new location had on the signal strength. This cannot be considered a scientific test, but as a comparative test, it raises some concerns.

What this test is showing us is either an alteration of the driver configuration between the machines, or that the WiFi antenna in the 900 is different to that of the 701. I suspect the latter, as there is less space in the 900’s screen surround, so it seems likely it has a smaller and less efficient antenna. Whatever the reason, it is clear that the 900 has an inferior WiFi range when compared against the 701.

Heat and fan noise

The 900 seems to run a little cooler than the 701, but the fan comes on much sooner. When it does it is very noticeable. Gone is the whisper-quiet 701 fan – now what we have is something altogether more noisy. I have not had a chance to install the well known eee.ko kernel module on the 900 that provides a means of controlling the fan, but I hope that, providing it is compatible, it should be possible to at least alter the “fan on” threshold.

Potential owners should not be unduly alarmed by this. The fan seems noisy to me because I am used to the 701’s fan. You still need to be in a quiet place to be able to hear it.

As a side note, 701 owners may have noticed that their batteries get quite warm in use. This appears to be mostly caused by direct heat transfer from the machine to the battery. The 900 has an extra piece of plastic glued to the bottom of the battery bay and I speculate that this may be there to provide a little extra insulation for the battery. If so, it’s a good idea, as the Lithium-Ion cells found inside the 900’s battery pack will deteriorate faster when warm.

Memory

The 900 comes with a full 1Gb of memory, twice what you get with the 701. That is enough for most people who are using the Xandros Linux operating system, but it’s always nice to have a bit more. Upgrading the memory in the 900 couldn’t be easier. First, lay the machine down with the bottom facing upwards. Then remove the “Eee” sticker that is covering the right hand memory bay door screw. This isn’t a warranty sticker as was found on the early 701s, and it can be removed and replaced without damage if you are careful.

Now, taking a small cross head screwdriver, remove the two memory bay screws. Use a finger nail to remove the memory bay door. This is what you will see:

The top compartment contains the 16Gb SSD drive plugged into the mini PCI-E socket on the motherboard, but in the bottom compartment resides the object of our interest. It’s an A-Data DDR2-400 (200 pin) SO-DIMM module, which is the same specification as the memory fitted in the 701. Use the screwdriver to lever the clips on either side of the memory card, and it will pop up. Carefully lift it out, and replace it with a 2 gig DDR-2 SO-DIMM module (I used a DDR2-667 module which was previously in my 701), then push the module down until the clips locate with a click. Replace the memory bay door and carefully tighten the little screws up.

As you can see, the 900’s BIOS recognises the memory, as it would do on the 701 BIOS. But the great news is…

…yes, you guessed it – the stock Xandros kernel does, too. On the 701, the kernel only recognises a maximum of 1Gb of RAM, making it necessary to compile and install a different kernel (without the hardcoded 1Gb limit), and this is a problem for novice users. Well, at least on the 900, it isn’t a problem any more.

Since I’d scavenged the 2Gb memory stick from my 701, I put the A-Data card from the 900 in its place. Naturally, it worked perfectly.

Graphics performance

For this test I have opted to use the program “glxgears” which is a 3D demonstration program that animates a set of three interlocking cogs in a window that is 300×300 pixels. By passing the (undocumented) argument -printfps, you can get it to print the number of frames the graphics card is drawing per second as the animation progresses. I should point out here that glxgears is far from an exhaustive test of the capabilities of the graphics chipset, and as such is not regarded as a particularly good benchmark; however, it is easy to install and many people know what it is.

The results of running this program are interesting, especially when compared to the same results obtained on a 701.

There is a clear performance improvement here. The 900 is supposed to have the same on-board hardware as the 701 (apart from the SSD controller chip), yet here it is, returning nearly twice the performance in a simple 3D rendering test, and even in Power Saving mode, there is a small improvement.

Perhaps a better test would be to play Penguin Racer, the 3D game which is supplied with both machines. It is accessible from the Games icon of the “Play” tab of the Easy Mode desktop.

The graphics in this 3D game on the 900 are visibly much smoother than on the 701, and that is with a higher resolution display.

Update : I since found that with the mains adapter plugged in, the machine stays in Performance mode all the time, irrespective of the Bios setting. A retest indicates 560fps in Power Saving mode (the table was updated to reflect this).

The multimedia experience

Right now you’re probably wondering how the 900 performs when playing a DVD or a video file, compared to the 701. Well, the answer is that, like the 701, you will need to install the libdvdcss.so library from the debian-multimedia repository to do this (there are clear instructions on the eeeuser.com wiki), and you do need an external DVD drive, of course.

For this test I used the LG GSA E50-N “Slimdrive”, which is a full featured DVD writer that can be powered from a single USB port.

After installing libdvdcss, the 900’s in-built media player, smplayer, worked flawlessly. There is one tweak you need to do to get the best from it, and that is to set it up to use the “xv” video driver (that’s found under Options -> Preferences -> “General” tab -> “Output Drivers – Video”). This gives a better picture with less tearing. Picture quality was fine and I notice with this new screen that the blacks really are a deep black, so it must have a good contrast ratio.

The built-in video playing application can handle a wide variety of video formats, including those downloaded from a digital video recorder (for example, the Humax PVR-9200T). Again, picture quality was very good, although I noticed some tearing (as I do on the 701 with these .ts files). I think this is caused by the video player setup, but I have not been able to experiment with it as yet.

Advanced mode desktop

The advanced mode desktop can easily be installed by following the instructions on the eeeuser.com wiki. I used the instructions entitled “Enable Advanced Desktop Mode – The Easy Way”, because the (even easier) technique, which uses a shell script called “pimpmyeee.sh” is said to not be 100% compatible with the 900 (it was written for the 701).

However, following the indicated instruction set, I now have a KDE desktop on my 900, and so far it has worked flawlessly.

A quick Ubuntu test

It’s now time to have a go with another Linux distribution. I’m going to boot up Ubuntu 8.04 “Hardy Heron” on the 900, to see what it looks like on the bigger screen.

That seems fine. The Atheros wireless card, as expected, does not work properly and the acpi keys aren’t all there either, but the display is very crisp and there are various fixes to the minor incompatibilities we are faced with here. See the eeeusr.com wiki for more details.

The geeky tidbits.

This information comes from the Eee PC 900’s operating system and as such is only going to be of any interest to the Linux die-hards out there. Please let me know if I missed anything out!

cat /proc/devices

Character devices:

  1 mem

  2 pty

  3 ttyp

  4 /dev/vc/0

  4 tty

  5 /dev/tty

  5 /dev/console

  5 /dev/ptmx

  7 vcs

 10 misc

 13 input

 14 sound

 29 fb

 81 video4linux

116 alsa

128 ptm

136 pts

180 usb

189 usb_device

226 drm

254 usb_endpoint

Block devices:

  1 ramdisk

  7 loop

  8 sd

 11 sr

 65 sd

 66 sd

 67 sd

 68 sd

 69 sd

 70 sd

 71 sd

128 sd

129 sd

130 sd

131 sd

132 sd

133 sd

134 sd

135 sd

Output of lsmod command

Module                  Size  Used by

uvcvideo               49288  0

compat_ioctl32           576  1 uvcvideo

videodev               23552  1 uvcvideo

v4l1_compat            12868  2 uvcvideo,videodev

v4l2_common            11200  2 uvcvideo,videodev

usb_storage            22212  0

libusual                5968  1 usb_storage

wlan_scan_sta           9728  1

ath_pci                74084  0

ath_rate_atheros       24768  1 ath_pci

wlan                  144280  4 wlan_scan_sta,ath_pci,ath_rate_atheros

ath_dfs                25476  1 ath_pci

ath_hal               139968  3 ath_pci,ath_rate_atheros

pciehp                 31172  0

pci_hotplug             9672  1 pciehp

atl2                   23064  0

ehci_hcd               25420  0

uhci_hcd               18636  0

usbhid                 13444  0

usbcore                91992  7 uvcvideo,usb_storage,libusual,ehci_hcd,uhci_hcd,usbhid

fuse                   34516  0

asus_acpi               6240  0

button                  5648  0

processor              19820  0

battery                 7940  0

ac                      3524  0

autofs4                15876  0

sr_mod                 13284  0

cdrom                  30624  1 sr_mod

snd_hda_intel          14168  0

snd_hda_codec         174280  1 snd_hda_intel

snd_pcm                52872  2 snd_hda_intel,snd_hda_codec

snd_timer              15300  1 snd_pcm

snd                    32964  4 snd_hda_intel,snd_hda_codec,snd_pcm,snd_timer

soundcore               3744  1 snd

snd_page_alloc          6472  2 snd_hda_intel,snd_pcm

genrtc                  6028  0

Output of mount command

rootfs on / type rootfs (rw)

/dev/sda1 on / type ext2 (ro)

none on / type aufs (rw,xino=/.aufs.xino,br:/=rw:/=ro)

proc on /proc type proc (rw)

sysfs on /sys type sysfs (rw)

devpts on /dev/pts type devpts (rw)

tmpfs on /dev/shm type tmpfs (rw)

tmpfs on /tmp type tmpfs (rw)

/dev/sdb1 on /home type ext3 (rw,noatime,data=ordered)

usbfs on /proc/bus/usb type usbfs (rw)

Output of cat /proc/partitions

major minor  #blocks  name

   8     0    3940272 sda

   8     1    2409718 sda1

   8     2    1510110 sda2

   8     3       8032 sda3

   8     4       8032 sda4

   8    16   15761088 sdb

   8    17   15759733 sdb1

The Conclusion

It’s certain that the Eee PC 900 represents a significant step up from the 701. The price difference is an additional £110, which brings the cost to £330 in the UK. For that kind of money, you can easily get a more capable laptop PC with a 15″ screen, dual core processor and built-in DVD drive.

However, it is the portability that makes people want to buy the Eee PC, and in the case of the 701, the low price tag. The question is, has ASUS improved the product enough to justify the increased price?

I think they have. Consider this : if you had a 701 and upgraded it with a 16Gb USB stick and extra memory, you’d be well into the extra £110, and that is without including the bigger screen. If portability is your thing, the 900 is for you. There is also evidence of genuine improvements in performance, too, and this is something you cannot modify a 701 to emulate.

The only major flaw with the 900 is the provision in the UK market of such a low capacity battery on what is essentially a brand leading product. This is a real disappointment to the many existing Eee PC users who expected, based on initial reviews, a far better battery (to be absolutely fair, ASUS never advertised the battery capacity, but they did send the initial reviewers Eee PC 900s equiped with 5800mAh batteries, thus raising expectations). It seems that in this instance, enthusiastic early adopters in the UK at least will be penalised.

The issue of the battery really overshadows the other problems I mentioned and for me at least, taints what should have been a very good experience. Buying another battery of a more appropriate capacity will add at least £60 to the retail price, and for many I suspect this will be a step too far.

However, that said, I think it a worthy successor to the 701and a good upgrade – especially if you keep the 701’s battery!

You can read the related post here.

But ASUS has stepped up to the plate to give this story a good ending: they are going to replace the 4400mAh batteries of the early adopters with the better 5800mAh ones. Good work, ASUS!