GuruPlug Server and JTAG interface

The GuruPlug Server Plus that I ordered from GlobalScale Technologies in February finally arrived around the middle of May, about two weeks later than anticipated.

These ARM-based Linux computers draw a mere 5-6 Watt of power at idle according to my WattChecker Plus, yet provide as much port connectivity as a regular notebook or desktop PC. It’s a full-featured Linux server, provided your application matches the storage available (or or you add enough storage) and integer performance comparable to a Pentium III 800 MHz is adequate for your purpose.

Here’s the specification of the machine:

  • 1.2 Ghz Marvell Kirkwood ARM CPU
  • 512 MB of DDR2 RAM
  • 512 MB of NAND flash
  • two gigabit Ethernet ports
  • 802.11b/g WLAN
  • Bluetooth
  • two hi-speed USB (480 Mbps) ports
  • eSATA port
  • microSDHC slot
  • Debian GNU/Linux 5.0 (2.6.32 kernel)

Normally the GuruPlug boots Linux off its NAND flash (see log below), but it can be reconfigured to load an image from a microsSD / microSDHC card (4 / 8 / 16 GB) and it can also use USB or eSATA disk drives. To reconfigure the boot loader, one needs to access a serial port via a JTAG interface, one of which came bundled with my order but is also available separately. To access this serial port one needs a terminal program such as PuTTY for Windows, a micro-USB cable and drivers for the USB-connected serial port.

If you look at the GuruPlug from the top there will be two tiny sockets on the right, a narrow one with 4 pins for the serial interface and a wider one for the flash interface. Connect these to the JTAG module. Plug the micro-USB cable into the socket at the opposite end of the JTAG module, but don’t connect it to the PC just yet.

Download the FTDI driver .zip file from here into a folder and extract its contents. Modify the two files FTDIBUS.INF and FTDIPORT.INF as described here. (EDIT): Download the FTDI driver archive file from here and save it in a folder. Extract the archive within the archive and unpack it into a folder.

When you finally connect the JTAG module to the PC via the micro-USB cable, it will start the plug and play device detection for it. Windows will not find a matching driver for the two ports (“SheevaPlug JTAGKey FT2232D B”) and so you’ll need to manually chose the location where it may find the driver and INF files (the ones you extracted and edited as above).

If anything goes wrong with the device installation, you can always delete the unrecognized devices in the device manager of Windows and disconnect and reconnect the USB cable to have another try.

Configure the virtual serial port (COM10 in my case) in device manager for 115200 bps. Download and install PuTTY. Then create a PuTTY profile for a connection to the virtual serial port. When you power-cycle the GuruPlug you should see messages come up. Here is a sample of an uninterrupted boot process:

U-Boot 2009.11-rc1-00602-g28a9c08-dirty (Feb 09 2010 – 18:15:21)
Marvell-Plug2L

SoC: Kirkwood 88F6281_A0
DRAM: 512 MB
NAND: 512 MiB
In: serial
Out: serial
Err: serial
Net: egiga0, egiga1
88E1121 Initialized on egiga0
88E1121 Initialized on egiga1
Hit any key to stop autoboot: 0
*** ERROR: `ipaddr’ not set
ping failed; host 192.168.2.1 is not alive
No link on egiga1
*** ERROR: `ipaddr’ not set
ping failed; host 192.168.2.1 is not alive
(Re)start USB…
USB: Register 10011 NbrPorts 1
USB EHCI 1.00
scanning bus for devices… 3 USB Device(s) found
scanning bus for storage devices… Device NOT ready
Request Sense returned 02 3A 00
1 Storage Device(s) found

NAND read: device 0 offset 0x100000, size 0x400000
4194304 bytes read: OK
## Booting kernel from Legacy Image at 06400000 …
Image Name: Linux-2.6.32-00007-g56678ec
Image Type: ARM Linux Kernel Image (uncompressed)
Data Size: 2789756 Bytes = 2.7 MB
Load Address: 00008000
Entry Point: 00008000
Verifying Checksum … OK
Loading Kernel Image … OK
OK

Starting kernel …

Uncompressing Linux…………………………………………………..
…………………………………………………………………………..
…………………………….. done, booting the kernel.
Linux version 2.6.32-00007-g56678ec (root@msi-linux-build.marvell.com) (gcc version 4.1.2 20070925 (Red Hat 4.1.2-33.fa1)) #1 PREEMPT Mon Feb 8 03:49:55 PST 2010
CPU: Feroceon 88FR131 [56251311] revision 1 (ARMv5TE), cr=00053977
CPU: VIVT data cache, VIVT instruction cache
Machine: Marvell Plug2L Reference Board
Memory policy: ECC disabled, Data cache writeback
Built 1 zonelists in Zone order, mobility grouping on. Total pages: 130048
Kernel command line: console=ttyS0,115200 ubi.mtd=2 root=ubi0:rootfs rootfstype=ubifs
PID hash table entries: 2048 (order: 1, 8192 bytes)
Dentry cache hash table entries: 65536 (order: 6, 262144 bytes)
Inode-cache hash table entries: 32768 (order: 5, 131072 bytes)
Memory: 256MB 256MB = 512MB total
Memory: 513024KB available (5144K code, 1034K data, 148K init, 0K highmem)
SLUB: Genslabs=11, HWalign=32, Order=0-3, MinObjects=0, CPUs=1, Nodes=1
Hierarchical RCU implementation.
NR_IRQS:114
Console: colour dummy device 80×30
Calibrating delay loop… 1192.75 BogoMIPS (lpj=5963776)
Mount-cache hash table entries: 512
CPU: Testing write buffer coherency: ok
NET: Registered protocol family 16
Kirkwood: MV88F6281-A1, TCLK=200000000.
Feroceon L2: Cache support initialised.
bio: create slab at 0
vgaarb: loaded
SCSI subsystem initialized
usbcore: registered new interface driver usbfs
usbcore: registered new interface driver hub
usbcore: registered new device driver usb
cfg80211: Using static regulatory domain info
cfg80211: Regulatory domain: US
(start_freq – end_freq @ bandwidth), (max_antenna_gain, max_eirp)
(2402000 KHz – 2472000 KHz @ 40000 KHz), (600 mBi, 2700 mBm)
(5170000 KHz – 5190000 KHz @ 40000 KHz), (600 mBi, 2300 mBm)
(5190000 KHz – 5210000 KHz @ 40000 KHz), (600 mBi, 2300 mBm)
(5210000 KHz – 5230000 KHz @ 40000 KHz), (600 mBi, 2300 mBm)
(5230000 KHz – 5330000 KHz @ 40000 KHz), (600 mBi, 2300 mBm)
(5735000 KHz – 5835000 KHz @ 40000 KHz), (600 mBi, 3000 mBm)
cfg80211: Calling CRDA for country: US
Switching to clocksource orion_clocksource
NET: Registered protocol family 2
IP route cache hash table entries: 4096 (order: 2, 16384 bytes)
TCP established hash table entries: 16384 (order: 5, 131072 bytes)
TCP bind hash table entries: 16384 (order: 4, 65536 bytes)
TCP: Hash tables configured (established 16384 bind 16384)
TCP reno registered
NET: Registered protocol family 1
RPC: Registered udp transport module.
RPC: Registered tcp transport module.
RPC: Registered tcp NFSv4.1 backchannel transport module.
JFFS2 version 2.2. (NAND) ร‚ยฉ 2001-2006 Red Hat, Inc.
JFS: nTxBlock = 4010, nTxLock = 32080
msgmni has been set to 1002
alg: No test for stdrng (krng)
io scheduler noop registered
io scheduler anticipatory registered
io scheduler deadline registered
io scheduler cfq registered (default)
Serial: 8250/16550 driver, 2 ports, IRQ sharing disabled
serial8250.0: ttyS0 at MMIO 0xf1012000 (irq = 33) is a 16550A
console [ttyS0] enabled
brd: module loaded
loop: module loaded
sata_mv sata_mv.0: version 1.28
sata_mv sata_mv.0: slots 32 ports 1
scsi0 : sata_mv
ata1: SATA max UDMA/133 irq 21
NAND device: Manufacturer ID: 0xec, Chip ID: 0xdc (Samsung NAND 512MiB 3,3V 8-bit)
Scanning device for bad blocks
Bad eraseblock 1265 at 0x000009e20000
Creating 3 MTD partitions on “orion_nand”:
0x000000000000-0x000000100000 : “u-boot”
0x000000100000-0x000000500000 : “uImage”
0x000000500000-0x000020000000 : “root”
UBI: attaching mtd2 to ubi0
UBI: physical eraseblock size: 131072 bytes (128 KiB)
UBI: logical eraseblock size: 129024 bytes
UBI: smallest flash I/O unit: 2048
UBI: sub-page size: 512
UBI: VID header offset: 512 (aligned 512)
UBI: data offset: 2048
ata1: SATA link down (SStatus 0 SControl F300)
UBI warning: ubi_eba_init_scan: cannot reserve enough PEBs for bad PEB handling, reserved 39, need 40
UBI: attached mtd2 to ubi0
UBI: MTD device name: “root”
UBI: MTD device size: 507 MiB
UBI: number of good PEBs: 4055
UBI: number of bad PEBs: 1
UBI: max. allowed volumes: 128
UBI: wear-leveling threshold: 4096
UBI: number of internal volumes: 1
UBI: number of user volumes: 1
UBI: available PEBs: 0
UBI: total number of reserved PEBs: 4055
UBI: number of PEBs reserved for bad PEB handling: 39
UBI: max/mean erase counter: 2/0
UBI: image sequence number: 0
UBI: background thread “ubi_bgt0d” started, PID 454
MV-643xx 10/100/1000 ethernet driver version 1.4
mv643xx_eth smi: probed
net eth0: port 0 with MAC address 00:50:43:01:5c:56
net eth1: port 0 with MAC address 00:50:43:01:5c:57
ehci_hcd: USB 2.0 ‘Enhanced’ Host Controller (EHCI) Driver
orion-ehci orion-ehci.0: Marvell Orion EHCI
orion-ehci orion-ehci.0: new USB bus registered, assigned bus number 1
orion-ehci orion-ehci.0: irq 19, io mem 0xf1050000
orion-ehci orion-ehci.0: USB 2.0 started, EHCI 1.00
usb usb1: configuration #1 chosen from 1 choice
hub 1-0:1.0: USB hub found
hub 1-0:1.0: 1 port detected
Initializing USB Mass Storage driver…
usbcore: registered new interface driver usb-storage
USB Mass Storage support registered.
usbcore: registered new interface driver ums-datafab
usbcore: registered new interface driver ums-freecom
usbcore: registered new interface driver ums-jumpshot
usbcore: registered new interface driver ums-sddr09
usbcore: registered new interface driver ums-sddr55
mice: PS/2 mouse device common for all mice
rtc-mv rtc-mv: rtc core: registered rtc-mv as rtc0
i2c /dev entries driver
cpuidle: using governor ladder
cpuidle: using governor menu
sdhci: Secure Digital Host Controller Interface driver
sdhci: Copyright(c) Pierre Ossman
mmc0: mvsdio driver initialized, lacking card detect (fall back to polling)
Registered led device: plug2l:red:health
Registered led device: plug2l:green:health
Registered led device: plug2l:red:wmode
Registered led device: plug2l:green:wmode
mv_xor_shared mv_xor_shared.0: Marvell shared XOR driver
mv_xor_shared mv_xor_shared.1: Marvell shared XOR driver
mmc0: new high speed SDIO card at address 0001
mv_xor mv_xor.0: Marvell XOR: ( xor cpy )
mv_xor mv_xor.1: Marvell XOR: ( xor fill cpy )
mv_xor mv_xor.2: Marvell XOR: ( xor cpy )
mv_xor mv_xor.3: Marvell XOR: ( xor fill cpy )
usbcore: registered new interface driver usbhid
usbhid: v2.6:USB HID core driver
oprofile: using timer interrupt.
TCP cubic registered
NET: Registered protocol family 17
lib80211: common routines for IEEE802.11 drivers
rtc-mv rtc-mv: setting system clock to 2009-08-08 08:10:20 UTC (1249719020)
usb 1-1: new high speed USB device using orion-ehci and address 2
UBIFS: mounted UBI device 0, volume 0, name “rootfs”
UBIFS: file system size: 516225024 bytes (504126 KiB, 492 MiB, 4001 LEBs)
UBIFS: journal size: 9033728 bytes (8822 KiB, 8 MiB, 71 LEBs)
UBIFS: media format: w4/r0 (latest is w4/r0)
UBIFS: default compressor: zlib
UBIFS: reserved for root: 0 bytes (0 KiB)
VFS: Mounted root (ubifs filesystem) on device 0:13.
Freeing init memory: 148K
usb 1-1: configuration #1 chosen from 1 choice
hub 1-1:1.0: USB hub found
hub 1-1:1.0: 4 ports detected
INIT: version 2.86 booting
usb 1-1.1: new high speed USB device using orion-ehci and address 3
usb 1-1.1: configuration #1 chosen from 1 choice
scsi1 : SCSI emulation for USB Mass Storage devices
Starting the hotplug events dispatcher: udevd.
Synthesizing the initial hotplug events…done.
Waiting for /dev to be fully populated…Bluetooth: Core ver 2.15
NET: Registered protocol family 31
Bluetooth: HCI device and connection manager initialized
Bluetooth: HCI socket layer initialized
libertas_sdio: Libertas SDIO driver
libertas_sdio: Copyright Pierre Ossman
libertas_sdio mmc0:0001:1: firmware: requesting sd8688_helper.bin
libertas: can’t load helper firmware
libertas: failed to load helper firmware
libertas_sdio: probe of mmc0:0001:1 failed with error -2
Bluetooth: vendor=0x2df, device=0x9105, class=255, fn=2
btmrvl_sdio mmc0:0001:2: firmware: requesting sd8688_helper.bin
btmrvl_sdio_download_helper: request_firmware(helper) failed, error code = -2
btmrvl_sdio_download_fw: Failed to download helper!
btmrvl_sdio_probe: Downloading firmware failed!
done.
Setting the system clock.
Activating swap…done.
Setting the system clock.
scsi 1:0:0:0: Direct-Access Generic STORAGE DEVICE 9909 PQ: 0 ANSI: 0
sd 1:0:0:0: Attached scsi generic sg0 type 0
sd 1:0:0:0: [sda] Attached SCSI removable disk
scsi 1:0:0:1: Direct-Access Generic STORAGE DEVICE 9909 PQ: 0 ANSI: 0
sd 1:0:0:1: Attached scsi generic sg1 type 0
sd 1:0:0:1: [sdb] 7954432 512-byte logical blocks: (4.07 GB/3.79 GiB)
sd 1:0:0:1: [sdb] Write Protect is off
sd 1:0:0:1: [sdb] Assuming drive cache: write through
sd 1:0:0:1: [sdb] Assuming drive cache: write through
sdb: sdb1
sd 1:0:0:1: [sdb] Assuming drive cache: write through
sd 1:0:0:1: [sdb] Attached SCSI removable disk
Cleaning up ifupdown….
Loading kernel modules…done.
Checking file systems…fsck 1.41.3 (12-Oct-2008)
done.
Setting kernel variables (/etc/sysctl.conf)…done.
Mounting local filesystems…done.
Activating swapfile swap…done.
Setting up networking….
Configuring network interfaces…done.
Starting portmap daemon….
Setting console screen modes and fonts.
cannot (un)set powersave mode
Setting up ALSA…done (none loaded).
INIT: Entering runlevel: 2
Starting enhanced syslogd: rsyslogd.
Starting system message bus: dbus.
Starting OpenBSD Secure Shell server: sshdNET: Registered protocol family 10
.
Starting MySQL database server: mysqld.
Checking for corrupt, not cleanly closed and upgrade needing tables..
Starting MTA: exim4.
ALERT: exim paniclog /var/log/exim4/paniclog has non-zero size, mail system possibly broken failed!
Starting Network Interface Plugging Daemon:ADDRCONF(NETDEV_UP): eth0: link is not ready
eth0.
Starting web server: lighttpd.
Starting internet superserver: inetd.
Starting Samba daemons: nmbdeth0: link up, 100 Mb/s, full duplex, flow control disabled
ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
smbd.
Starting file alteration monitor: FAM.
Starting Hardware abstraction layer: hald.
Starting periodic command scheduler: crond.
Sat Aug 8 08:08:00 UTC 2009
uap_probe: vendor=0x02DF device=0x9104 class=0 function=1
uap_sdio mmc0:0001:1: firmware: requesting mrvl/helper_sd.bin
uap_sdio mmc0:0001:1: firmware: requesting mrvl/sd8688_ap.bin
UAP FW is active
ADDRCONF(NETDEV_UP): uap0: link is not ready
SSID setting successful
BSS started!
ip_tables: (C) 2000-2006 Netfilter Core Team
nf_conntrack version 0.5.0 (8022 buckets, 32088 max)
CONFIG_NF_CT_ACCT is deprecated and will be removed soon. Please use
nf_conntrack.acct=1 kernel parameter, acct=1 nf_conntrack module option or
sysctl net.netfilter.nf_conntrack_acct=1 to enable it.
Starting very small DHCP server: udhcpd (v0.9.9-pre) started
udhcpd.
Starting DNS forwarder and DHCP server: dnsmasq.
Starting bluetooth: bluetoothdBluetooth: L2CAP ver 2.14
Bluetooth: L2CAP socket layer initialized
Bluetooth: RFCOMM TTY layer initialized
Bluetooth: RFCOMM socket layer initialized
Bluetooth: RFCOMM ver 1.11
.
Bluetooth: BNEP (Ethernet Emulation) ver 1.3
Bluetooth: vendor=0x2df, device=0x9105, class=255, fn=2
Bridge firewalling registered
Bluetooth: SCO (Voice Link) ver 0.6
Bluetooth: SCO socket layer initialized
Agent registered

Debian GNU/Linux 5.0 sheevaplug-debian ttyS0

sheevaplug-debian login:

The initial admin password for user “root” is “nosoup4u”. Change this as soon as you can to something different.

By default the Plug acts as a broadband access point. Join the GuruPlug WLAN from a laptop and you’ll get an IP address in the 192.168.1.100-192.168.1.200 range. You can access a simple website at htpp://192.168.1.100:80 to learn more about the GuruPlug settings.

http://plugcomputer.org/ is a great resource for finding more information about setting up your GuruPlug.

RsyncServer not started because pid file rsyncd.pid exists

I had a scary experience today because one of my Windows PCs “blue screened”, i.e. it crashed. While the “Blue Screen of Death” was fairly common in Windows 95 and 98, it is unusual to see it on any healthy machine running Windows 2000, XP, 2003 or 2008 Server, Vista or Windows 7.

Since the CPU fan of the machine had been running faster than normal recently, I had already suspected it of dust buildup in the CPU heat sink and when I opened up the machine, my suspicion was confirmed. I pulled out some dust and vacuumed the machine before starting it up again.

The machine booted normally and the fan was quiet, but I got alerts from another machine that it could not connect to the Rsync server of the restarted machine. I use an Rsync service for Windows to synchronize data updates between various machines. The service was shown with startup mode “Automatic”, which means it should start whenever Windows is booted, but it was not showing as “Started”. Looking into the log file at “C:\Program Files\ICW\rsyncd.log” I found several of these error messages, one each for every restart I had performed after the crash:

2010/05/08 15:37:45 [2128] rsync: failed to create pid file rsyncd.pid: File exists (17)
2010/05/08 15:37:45 [2128] rsync error: error in file IO (code 11) at clientserver.c(985) [receiver=3.0.6]
2010/05/08 15:52:32 [2120] rsync: failed to create pid file rsyncd.pid: File exists (17)
2010/05/08 15:52:32 [2120] rsync error: error in file IO (code 11) at clientserver.c(985) [receiver=3.0.6]
2010/05/08 16:11:13 [2644] rsync: failed to create pid file rsyncd.pid: File exists (17)
2010/05/08 16:11:13 [2644] rsync error: error in file IO (code 11) at clientserver.c(985) [receiver=3.0.6]

Probably the sudden crash had prevented the Rsync service from deleting its pid (process ID) file, as it normally does when it is shut down orderly.

After I deleted “C:\Program Files\ICW\rsyncd.pid” I was able to start the service and the other machine could connect to it and synchronize with it properly.

Hopefully the machine will keep running reliably again now that the heat sink is clean again. I will keep checking it for dust buildup at least monthly now that the summer months are approaching.

GuruPlug Server Plus, a $129 Linux server based on ARM

GuruPlug Server Plus

Over the years, Intel and its microprocessors have become a household name, recognized by millions of PC users. However, far more people own computers based on ARM processors without ever having heard of that CPU – it’s their mobile phones. More than a billion mobile phones are sold worldwide every year, each one more powerful than an office computer that used to run Windows 95/98 not too long ago. Whatever the brand or model, almost all of them (98%) use at least one CPU based on the ARM architecture. Also, if you own a car navigation system or a broadband router that connects your PC to a DSL or cable modem, it probably uses an ARM chip. Do your kids own a Gameboy Advance, a Nintendo DS, do you listen to music on an iPod? ARM CPUs are common to all of these, as well as the iPhone, the Blackberry and other mobile phones.

From the start, ARM was meant to compete head-on with Intel CPUs in desktop computers, but for many years there really was no direct competition between the two. The issue was mostly software, and as that may change, we’ll increasingly see ARM and Intel compete directly in the same markets. This is bad news for Intel (and AMD) as per-CPU revenue has been far higher in their markets than for makers of ARM chips. As Intel is trying to get a bigger slice of the exploding handheld market with Atom while ARM is taking a bite out of Intel-dominated markets, this gap in pricing is set to erode.

A short history of ARM

ARM has a colourful history. The first ARM chips were developed by Acorn Computers Ltd in Cambridge/UK, who in the early 1980s had successfully launched the 6502-based BBC Micro home/educational computer, an 8-bit machine. I first visited Acorn in the summer of 1984 as a young software engineer working at Digital Research UK. Under an OEM contract I ported Concurrent DOS to a prototype Acorn ABC 310, a 6502-based machine with an 8 MHz 80286 second processor equipped with 1 MB of memory. That 286 machine was one of several projects meant to take Acorn into the more lucrative business market, using a range of 16-bit and 32-bit machines to compete against the likes of the new IBM PC/XT, PC/AT and the 8086-based ACT Apricot that was very successful in the UK. Acorn was well aware of the limitations of the 6502, which was only an 8-bit processor that could not address more than 64 KB of memory. Acorn then decided to replace the 6502 with a simple but efficient 32-bit RISC design, the Acorn RISC Machine (ARM) which came out in 1986.

As it turned out, machines based on ARM never took much market share from the new IBM-compatible PCs that were gradually taking over the desktop market. So instead of building entire computers, or making chips and selling them like Intel does, the developers of ARM decided to license the design to anyone wishing to integrate it into their own custom designs. That’s how ARM became both ubiquitous and nameless. One reason for the lack of success on the desktop was the huge library of software written for MS-DOS and later the various MS Windows versions, which only worked on Intel-compatible CPUs. Despite that, the ARM design eventually became successful beyond anybody’s wildest dreams in mobile and embedded markets where low power usage and low cost were crucial. Mobile phones and other handheld devices and appliances had no need to run desktop applications written for MS Windows. Subsequently, ARM as a hardware platform became so successful that Microsoft ported its Windows CE operating system to ARM (besides x86, MIPS and SuperH as other supported platforms). Even Intel was involved for 8 years with StrongARM and XScale, which it sold to Marvell in 2006.

The marriage of ARM and Open Source

Now it looks like ARM is returning to its roots, with the announcement of ARM-based netbooks and with ARM-based Linux servers. Until recently notebooks and servers were mostly Intel domains, but open source and portable Linux is bridging the gap between Intel and ARM hardware, leveling the playing field for applications. In early 2009 Marvell launched the SheevaPlug, a $99 self-contained ARM and Linux-based mini-server about the size of a notebook power brick. A year later, Globalscale Technologies announced the GuruPlug range of low cost, energy saving computers as the successor to the SheevaPlug; both SheevaPlug and GuruPlug are based on a Marvell chip set, with similar specs and sizes. The GuruPlug is 95mm (L) x 65mm (W) x 48.5 mm (H) and draws under 5 Watts. Its built-in universal power supply handles 100-240V, 50/60Hz for worldwide use with only a plug adapter. The GuruPlug is supposed to start shipping in April and I have two on order ๐Ÿ™‚

The most interesting of the three GuruPlug Server models in my opinion is the “Plus” model, which sells for $129, $30 more than the GuruPlug Server “Standard” model or the older SheevaPlug base model. All GuruPlug Server models sport a 1.2 Ghz Marvell Kirkwood ARM CPU, 512 MB of DDR2 RAM, 512 MB of NAND flash, 802.11b/g WLAN and Bluetooth. While the Standard model offers one gigabit Ethernet port and one hi-speed USB (480 Mbps) port, the Plus has two of each and also offers an eSATA port (3 Gbps) plus an external micro-SD slot.

Dual network interfaces make the Plus very suitable as a router (one port for the WAN, one for the LAN). Just add a 4-port hub and you have a direct replacement for a regular broadband router, but with the ability to also hook up USB or eSATA hard disks for Network Attached Storage (NAS) or as a web server or media server, or plug in a USB-printer to be shared over the LAN. Or you could run a VPN, or a mail server, or a spam filter – or all of these combined. Basically anything that works on a Linux server should work in this tiny box, as long as there is enough storage for files on either the internal flash or RAM or the micro-SD or a USB-stick or an external USB or eSATA hard disk.

A third model, the GuruPlug Display for $179 has also been announced, but (as of 2010-02-21) can not be ordered yet. This one will come with 3 USB ports and an HDMI video connector for hooking up an external monitor, so you could use it with a keyboard and mouse (via USB or Bluetooth) for web browsing.

RCA Airnergy looks like a hoax

Gizmodo reported about a Gadget shown at CES 2010 that supposedly harvests energy from a wireless hotspot. The “RCA Airnergy WiFi Hotspot Power Harvester” consists of a small battery, a USB connector and some circuitry that is supposed to convert wireless signals into DC power to top up the battery. The gadget can then be used to recharge or power any device that can draw power from a USB port, such as a cell phone or iPod.

A claim was made in a Youtube video on the Gizmodo site that the gadget will charge a Blackberry mobile phone from 30% to fully charged in 90 minutes. That may well be true, if the internal battery of the gadget starts off fully charged and is big enough. The big question is, how much energy can this wireless harvester actually draw out of thin air to replenish its internal battery, if any?

The whole thing reminds me of the hoax of the Japanese “car that runs on water” demonstrated in June 2008 by now apparently defunct company Genepax Co. Ltd. (their website went offline the following year). That car turned out to have a set of lead-acid batteries that — fully charged — could have powered the car some distance even if the proprietary fuel cell announced by Genepax was completely dysfunctional. In any case, the quoted power output of the fuel cell of only 300W was completely inadequate to power a car, meaning the batteries (the real power source) would have had to be recharged from a wall socket before too long.

Likewise, the amount of power available from a WiFi hotspot is nowhere near enough to run a computer or mobile phone. Take my cheap Samsung mobile phone with a 880 mAh 3.7 V Li ion battery (a battery capacity of 3.2 Wh) that I normally need to charge every other day or so. 3.2 Wh over 48 h works out as about 67 mW, which is not that much. However, the maximum power at which a wireless access point may transmit under FCC regulations without needing a broadcast license is a mere 100 mW. Even if the “Hotspot Harvester” could convert 2/3 of the radio energy into usable DC power, it would have to suck up 100% of all energy radiated by the access point, which would have to broadcast at full blast all the time instead of just when there is traffic, just to keep my cell phone charged.

In reality, there is no way the harvester can grab 100% or 10% or even 1% of the energy from the hotspot, which radiates wireless signals in all directions. The gadget can only harvest the small fraction of the airwaves that cross its antenna, which is only a few centimetres by a few centimetres in size, while the hotspot may be metres or tens of metres away. The numbers simply don’t add up.

What that device is then is just a glorified spare battery that will need to be recharged by plugging it into a wall socket or the USB port of a mains-powered computer. The “energy harvesting” function can make no meaningful contribution to the battery charge – unless maybe you happen to put it inside a microwave oven and radiate it with 1000W of power (boys, don’t try this at home! ๐Ÿ˜‰ ).

The sad thing is how many websites and blogs have given free publicity to these claims, without doing the math to check if they make any sense at all.

NTFS disk corruption on frequent file creates and deletes

I recently upgraded a secondary hard disk in a Windows machine from a Seagate Barracuda 7200.11 1 TB drive to a Western Digital Caviar Green WD15EADS 1.5 TB drive. Mirror imaging the data off the old drive to the new drive using the Linux dd utility from an Ubuntu live CD went very smoothly and completed in a little over three hours, but I subsequently hit a snag when I tried to resize the partition on the larger drive.

The image copy of the 1 TB NTFS partition resulted in a 1 TB partition followed by 500 GB of unallocated space, which I wanted to add to the free space of the NTFS partition. That should be no problem for Linux gparted, but only if the partition to be resized is internally consistent. If it is not, one will have to use CHKDSK /F under Windows to fix it first.

C:\>chkdsk m: /f
CHKDSK is verifying files (stage 1 of 3)...
0 percent complete. (0 of 21942736 file records processed)
Deleted corrupt attribute list entry
with type code 144 in file 2300.
Deleted corrupt attribute list entry
with type code 176 in file 2300.
Deleting corrupt attribute record (144, $I30)
from file record segment 2300.
Deleting corrupt attribute record (176, $I30)
from file record segment 2300.
Deleted corrupt attribute list entry
with type code 144 in file 2332.
Deleted corrupt attribute list entry
with type code 176 in file 2332.
Deleting corrupt attribute record (144, $I30)
from file record segment 2332.
Deleting corrupt attribute record (176, $I30)
from file record segment 2332.
Deleted corrupt attribute list entry
with type code 144 in file 2342.
Deleted corrupt attribute list entry
with type code 176 in file 2342.
Deleting corrupt attribute record (144, $I30)
from file record segment 2342.
Deleting corrupt attribute record (176, $I30)
from file record segment 2342.
21942736 file records processed.
File verification completed.
41475 large file records processed.
0 bad file records processed.
1552 EA records processed.
0 reparse records processed.
CHKDSK is verifying indexes (stage 2 of 3)...
10 percent complete. (4118 of 81648013 index entries processed)
Correcting error in index $I30 for file 2300.
10 percent complete. (5149 of 81648013 index entries processed)
Correcting error in index $I30 for file 2332.
Correcting error in index $I30 for file 2342.
34 percent complete. (23316097 of 81648013 index entries processed)
Correcting a minor error in file 2300.
Correcting a minor error in file 2332.
Correcting a minor error in file 2342.
81648013 index entries processed.
Index verification completed.
CHKDSK is recovering lost files.
997 unindexed files processed.
997 unindexed files processed.
CHKDSK is verifying security descriptors (stage 3 of 3)...
21942736 security descriptors processed.
Security descriptor verification completed.
Inserting data attribute into file 2300.
99 percent complete. (1 of 939015 data files processed)
Inserting data attribute into file 2332.
Inserting data attribute into file 2342.
939015 data files processed.
Correcting errors in the master file table's (MFT) BITMAP attribute.
Correcting errors in the Volume Bitmap.
Windows has made corrections to the file system.

1465135996 KB total disk space.
871385828 KB in 18183533 files.
7651336 KB in 939018 indexes.
0 KB in bad sectors.
22057856 KB in use by the system.
65536 KB occupied by the log file.
564040976 KB available on disk.

4096 bytes in each allocation unit.
366283999 total allocation units on disk.
141010244 allocation units available on disk.

Googling for these error messages, I found a Knowledgebase article by Microsoft about disk corruption on Windows NT that could occur with frequent file creates and deletes (for which they had a fix) and a posting by a Windows XP user who also frequently creates and deletes files.

The three files reported by CHKDSK were three folders on my hard disk. CHKDSK converted them into zero length files. After recognizing this, I deleted these files and recreated the original directories. Since then my application has been working again.

As it so happens, these three directories are three places on my computer where the largest number of files is created and deleted in rapid and random succession (there’s one process each creating them and one process each asynchronously consuming them), just like in the articles that I googled.

If anyone else comes across a similar problem with NTFS with similar symptoms, I would be glad to hear from them.

Update 2009-11-20:
The three corrupted folders that were turned into zero length files by ChkDsk did not actually disappear completely. Instead their contents was moved to a hidden folder tree. Note the above lines:

CHKDSK is recovering lost files.
997 unindexed files processed.

I looked on the hard disk for hidden files or folder from the command prompt:

M:\> dir /ah m:\found.*
 Volume in drive M is wd15
 Volume Serial Number is E861-DD52

 Directory of m:\

2009-11-18  17:39    <DIR>          found.000
               0 File(s)              0 bytes
               1 Dir(s)  576,809,738,240 bytes free

This hidden folder contains the three lost directories:

M:\>dir m:\found.000
 Volume in drive M is wd15
 Volume Serial Number is E861-DD52

 Directory of m:\found.000

2009-11-20  12:06    <DIR>          dir0000.chk
2009-11-18  17:39    <DIR>          dir0001.chk
2009-11-18  17:39    <DIR>          dir0002.chk
               0 File(s)              0 bytes
               3 Dir(s)  576,804,810,752 bytes free

Looking at the contents of each of these folders I could soon tell which was which and move those files back to the original path.

Marvell MC85 with NdisWrapper on Ubuntu 9.10

Last week I updated my Gateway M-6750 from Ubuntu 9.04 to Ubuntu 9.10 (“Karmic Koala”) and managed to get myself into a right mess, as I lost access to both my wired and wireless internet connections. I then burnt an Ubuntu 9.10 live CD from an ISO image dowloaded with uTorrent and reinstalled from there. That got the wired connection working, but the wireless was still gone. Before I fixed that problem I upgraded my notebook hard disk drive, complete with dual-boot Vista and Ubuntu partitions, to a new 500 GB drive (see yesterday’s blog post).

The Marvell MC85 doesn’t have a native Ubuntu driver yet. Therefore you have to use a Windows XP driver for it as described here. You can download the Netgear wn311t_setup_4_1.exe driver set and extract its content on either XP or Wine. You’ll end up with NetMW14x.inf, netmw143.sys and netmw145.sys. Only NetMW14x.inf and netmw145.sys are actually needed.

See my instructions in my earlier blog post for the initial steps (look for a section labeled “Update, 2008-03-18”).

Installing the driver with ndiswrapper does almost everything. The magic ingredient that was missing when I tried to get it working this time was to ensure that ndiswrapper loads every time, before using the Network Manager or when booting up. Without ndiswrapper the Windows driver won’t be loaded and Ubuntu simply won’t see the wireless adapter. It will show up as unclaimed on this command:

lshw -C network

*-network UNCLAIMED
description: Ethernet controller
product: Marvell Technology Group Ltd.
vendor: Marvell Technology Group Ltd.
physical id: 0
bus info: pci@0000:02:00.0
version: 03
width: 32 bits
clock: 33MHz
capabilities: pm msi pciexpress bus_master cap_list
configuration: latency=0
resources: memory:f6000000-f600ffff memory:f4000000-f400ffff
*-network
description: Ethernet interface
product: RTL8101E/RTL8102E PCI Express Fast Ethernet controller
vendor: Realtek Semiconductor Co., Ltd.
physical id: 0
bus info: pci@0000:06:00.0
logical name: eth0
version: 01
serial: 00:e0:b8:XX:XX:XX
size: 10MB/s
capacity: 100MB/s
width: 64 bits
clock: 33MHz
capabilities: pm vpd msi pciexpress bus_master cap_list rom ethernet physical tp mii 10bt 10bt-fd 100bt 100bt-fd autonegotiation
configuration: autonegotiation=on broadcast=yes driver=r8169 driverversion=2.3LK-NAPI duplex=half latency=0 link=no multicast=yes port=MII speed=10MB/s
resources: irq:28 ioport:4000(size=256) memory:fa200000-fa200fff memory:c0000000-c001ffff(prefetchable)

Ubuntu uses file /etc/modules to enumerate extra drivers and kernel modules that need to be loaded at boot time. Therefore I needed to add a line in /etc/modules using gedit:

gksudo gedit /etc/modules

Add

ndiswrapper

in a line of its own at the bottom of the file and save it, then restart your machine. After the reboot you should be able to use the Network Manager to manage wireless connections, including looking for available networks and configuring security parameters (for WPA, WPA2 / RSN) to be able to connect.

I found many other threads and blog postings that discussed manually editing configuration files and configuring wpa_supplicant, but none of that is required if you just configure Ndiswrapper for NetMW14x.inf, tell NDiswrapper about the device ID and then ensure that Ndiswrapper is always laoded. Here is the lshw output with Ndiswrapper having claimed the wireless adapter:

*-network
description: Wireless interface
product: Marvell Technology Group Ltd.
vendor: Marvell Technology Group Ltd.
physical id: 0
bus info: pci@0000:02:00.0
logical name: wlan0
version: 03
serial: 00:16:44:XX:XX:XX
width: 32 bits
clock: 33MHz
capabilities: pm msi pciexpress bus_master cap_list ethernet physical wireless
configuration: broadcast=yes driver=ndiswrapper+netmw14x driverversion=1.55+NETGEAR,10/04/2006, 2.1.4.3 ip=192.168.42.122 latency=0 link=yes multicast=yes wireless=IEEE 802.11g
resources: irq:16 memory:f6000000-f600ffff memory:f4000000-f400ffff
*-network
description: Ethernet interface
product: RTL8101E/RTL8102E PCI Express Fast Ethernet controller
vendor: Realtek Semiconductor Co., Ltd.
physical id: 0
bus info: pci@0000:06:00.0
logical name: eth0
version: 01
serial: 00:e0:b8:XX:XX:XX
size: 10MB/s
capacity: 100MB/s
width: 64 bits
clock: 33MHz
capabilities: pm vpd msi pciexpress bus_master cap_list rom ethernet physical tp mii 10bt 10bt-fd 100bt 100bt-fd autonegotiation
configuration: autonegotiation=on broadcast=yes driver=r8169 driverversion=2.3LK-NAPI duplex=half latency=0 link=no multicast=yes port=MII speed=10MB/s
resources: irq:28 ioport:4000(size=256) memory:fa200000-fa200fff memory:c0000000-c001ffff(prefetchable)

Notice the driver=ndiswrapper+netmw14x after configuration:. If you get that you should be in business! ๐Ÿ™‚

Backing up / migrating your hard disk data with a Ubuntu live CD

Recently two hard disks died in my household, both of which were Windows boot disks, just as if I needed a reminder how quickly one can lose important data. This and some troubles while upgrading my Gateway M-6750 notebook from Ubuntu 9.04 to Ubuntu 9.10 (“Karmic Koala”) prompted me to get some spare hard disks and do a complete backup / upgrade.

I upgraded my notebook from 250 GB to 500 GB, and I can now keep my old drive as a safe snapshot to go back to if anything should ever go wrong with the new drive.

The most important tools needed for this were an Ubuntu 9.10 live CD from an ISO image (downloaded with uTorrent) and the NewerTech Universal USB 2.0 Adapter, which lets you hook up just about any IDE or SATA drive to a USB-equipped computer.

My notebook has Windows Vista on it, for which Gateway did not ship an install DVD – theres’s only a recovery partition. For installing Ubuntu I had originally shrunk the Vista NTFS partition on the 250 GB drive to make space for a 60 GB Linux partition, which provides a dual-boot feature via the GRUB loader.

I ordered some WD Scorpio Blue 500 GB notebook hard disks (WD5000BEVT) from Amazon and they arrived the next day. These are good work horse drives. I hooked one up to the NewerTech Universal USB 2.0 Adapter and booted into the Ubuntu Live CD.

From there I started a terminal window and used the “dd” command to make an image copy of the drive, see below. Be very careful with the source and destination specification. Use fdisk -l to verify which drive is which and in some cases the USB drive will appear as /dev/sdc instead or /dev/sdb. For IDE (PATA) drives the names could be /dev/hda, /dev/hdb, etc. If you specify the wrong drive as the destination you could wipe out all your data!

sudo dd if=/dev/sda of=/dev/sdb

This will copy every single byte from the first SATA hard disk to the second SATA (or USB 2.0) hard disk on the machine. You won’t have to worry about what partitions there are and which one is bootable, because it will simply copy all of them.

I started it last thing before I went to bed. It took about 2 hours to copy the entire 250 GB of the existing to the existing WD2500BEVS to the new drive. In the morning I shut down Ubuntu, opened the bottom of the case and transplanted the new drive inside, which took a small screw driver and about 10 minutes. Then I fired up the machine again and it successfully booted both Windows Vista and Ubuntu 9.10 off the new drive.

Since the new drive had more space, I wanted to resize the partitions so that each could take 250 GB instead them having to share that space. GNU utility gparted lets you do that. It can shrink, grow and move partitions pretty any way you like. You can tell it the new size of a partition and how much space to leave before and after it.

I first did not know that gparted can move partitions as well as resizing them, so I decided to remove the Ubuntu partitions, grow the NTFS (Windows) partitions and the reinstall Ubuntu into the remaining space. Be careful when removing an Ubuntu partition in a dual-boot system, because the active loaded and its menu file will be in the Ubuntu partition. Thus if you shoot it without first making the Windows partition the active boot partition you won’t be able to boot off that hard disk any more! Either boot of your operating system install CD/DVD or off an operating system recovery partition (if available) and select the command prompt. From there use fixldr /mbr (2000/XP/2003) or bootrec /FixMbr (Vista / 7) to rewrite the Master Boot Record. This will disable the Ubunto bootstrap loader. Make sure Windows will boot without the Ubuntu boot menu coming up. No you can remove the Ubuntu
partitions.

With sudo gparted on the Ubuntu Live CD you can resize the NTFS partition for Windows to any sensible value. In my case I got an error because some NTFS data structures were in an inconsistent state. To fix that I had to boot Windows and run chkdsk /f from a command prompt and then restart. Windows fixed the problem and one restart later I was back in Ubuntu live and gparted was able to resize the partition

After that I installed Ubuntu into the free space that I’d left for it on the 500 GB drive.

Given the low cost of SATA and USB 2.0 drives versus the time and data lost when something goes wrong with your hard disk, I do recommend a full image backup like mine. The dd command also lets you copy drives or partitions to files, so you can back up multiple machines to one large drive.

Computer power usage: AMD, Intel and VIA

The Kill A Watt EZ Electricity Usage Monitor P4460 by P3 International is a popular gadget in North America for measuring power usage by electronics and electrical appliances. You simply plug it into the wall socket and plug the appliance into the device and it will give you instant read-outs of power usage in Watt, electricity consumption over time in kWh as well as electricity costs (if you enter the price the utility charges per kWh). Knowing exactly how much electricity each device consumes encourages smart choices about when and how you use them.


Watt Checker Plus (2022-04)

While Kill A Watt models have been available for 115V and 230V markets in North America and Europe, I could not find any mention of a 100V model for Japan. As it turned it out, the device is made by Prodigit in Taiwan, who do make a model for Japan (2022-04) which is sold here under the name ใƒฏใƒƒใƒˆใƒใ‚งใƒƒใ‚ซใƒผPlus (“Watt Checker Plus”) by Keisoku Giken, Co. Ltd. I bought mine through Amazon.co.jp for JPY 5,670 (about US$63, $28 more than in the US).

Here are some preliminary results:

  • Acer Aspire M5201 (desktop: AMD Athlon X2 5000+, 2.60 GHz, integrated Radeon HD 3200, 4 GB DDR2 RAM, 320 GB + 1000 GB 3.5″ SATA HD):
    – 68 W at idle
    – 120 W at 100% CPU (both cores loaded)
  • eMachines T6212 (desktop: AMD Athlon 64 3200+, 2.00 GHz, discrete Asus EAH3450 256 MB, 3 GB DDR RAM, 160 GB 3.5″ HD):
    – 69 W at idle + “performance on demand”
    – 75 W at idle + “maximum performance”
    – 90 W at 100% CPU (only core loaded)
  • Dell Dimension 3100C (desktop: Intel Celeron D 331, 2.66 GHz, 1 GB RAM, 160 GB 3.5″ SATA HD):
    – 78 W at idle
  • VIA MM3500 (desktop: VIA C7, 1.5 GHz, 2 GB DDR2 RAM, 2 x 1 TB SATA WD10EADS):
    – 41 W at (almost) idle
    – 69 W loaded
  • Gateway M-6750 (notebook: Core 2 Duo T5450, 1.67 GHz, 3 GB DDR2 RAM, 250 GB 2.5″ SATA HD):
    – 24 W at idle
    – 48 W at 100% CPU (both cores loaded)
  • Mac mini (desktop: Core 2 Duo T5600, 1.83 GHz, 2 GB DDR2 RAM, 80 GB 2.5″ SATA HD):
    – 21 W at idle
  • Lenovo S10e (Notebook: Atom N270, 2 GB DDR2 RAM, 160 GB 2.5″ SATA):
    – 20 W at 90% CPU load
  • Dell Latitude CPx J650GT (notebook: Intel Pentium III Mobile, 650 MHz, 512 MB PC100 RAM, 60 GB 2.5″ IDE HD):
    – 13 W at idle, screen off
  • Dell 2408WFP (monitor: 24″, 1920 x 1200):
    – 48 W at brightness 0%
    – 60 W at brightness 12%
    – 120 W at brightness 100%
  • Dell 1905FP (monitor: 19″, 1280 x 1024):
    – 27 W at brightness 0%
    – 32 W at brightness 50%
    – 36 W at brightness 100%
  • Epson PM-A950 (inkjet printer / scanner / copier):
    – 3 W at standby (display off) or soft “off”
    – 15 W at idle (display on)
    – 19 W while scanning
    – 25 W while printing
  • Mitsubishi 25T-SY3 (colour TV: 25″, CRT, 110 W):
    – 80 to 100 W (depending on brightness of scene)

The inefficiency of the Celeron D was to be expected. It’s based on the infamous Pentium4 architecture that was later abandoned by Intel in favour of the more efficient Pentium III / Pentium M derived Core architecture.

Also expected was the frugality of the Atom N270 netbook, but I was positively surprised by how little power the Pentium III Mobile machine (Dell Latitude 650) consumed. Neither of these machines is a scorcher, but given its age the 8 year old Dell is doing surprisingly well as a temporary web browser and e-mail machine for my wife after her Dimension 3100C’s hard disk failed last week.

My biggest surprise was that the Mac mini uses only half the electricity of the VIA server I built (21 W versus 42 W). Admittedly, it only has one notebook drive (80 GB 2.5″) instead of two high capacity 3.5″ drives, but that should only account for about one quarter of the advantage of the faster Intel Mac over the slower VIA. The 80 GB single platter Hitachi TravelStar 2K250 in the Mac only draws a tiny 0.55 W at idle, but the Western Digital WD10EADS in the VIA server are doing quite well for a drive with 1 TB and 3 disks at only 2.8 W each at idle.

Note also that my Mac mini is last year’s model (2008). A newer model that came out in early 2009 (2.0 GHz Core2 Duo, 2GB DDR3 SDRAM, 320GB HD, GeForce 9400M video) is rated even lower, at less than 14 W at idle. Apple calls the Mac mini the “most energy-efficient desktop computer” and not without cause.

I also expected my old CRT TV to use considerably more power than my flat screen monitors. That was true for the 19″ monitors, but not necessarily for the 24″, depending on screen brightness settings.

Another surprise was that it makes no difference whatsoever whether I switch off my inkjet printer when I’m not expecting to print anything soon. After powering it on or after any printing, copying or scanning, the small LCD display will stay lit for a few minutes, with the printer drawing about 15 W (idle mode). After that interval, the display goes dark and the printer goes into a sleep mode from which it will awake on any button press or print job. In this state it is drawing the same 3 W as after switching it off using the On/Off button but leaving it plugged in. All the Off-switch seems to do is to stop it from responding to print output or to buttons other than the On-button. To get rid of the last 3 W you would need to unplug it or switch it off at a switchable power strip. Note that inkjet printers should only be fully disconnected from the mains this way while already switched off using the On/Off button.

I wish that power draw figures at idle and load were readily available for every computer on the market, so that consumers could make informed decisions.


Watt Checker Plus (2022-04)

Acer Revo R3600 and other dual core Atom 330 NVidia Ion nettops

The new Acer AspireRevo R3600 (Acer AspireRevo R3610-U9012) that combines a dual core Atom 330 processor with the Nvidia Ion platform was introduced at the IFA consumer electronics fair in Berlin in September. It is just one of several interesting new nettops coming out now that will offer significantly more processing power, especially for video decoding, while still using little electricity.

Low cost, low power Atom CPUs in small desktop cases (nettops) such as the Asus EeeBox 202 first became popular about a year ago, following in the footsteps of their mobile cousins, netbooks such as the Asus Eee PC range. The latest generation of machines are adding new features and more performance, which will expand the market for low end machines.

Other machines with similar specs to the AspireRevo (dual core Atom 330, NVIDIA Ion chip set and 2 to 4 GB of RAM) include the Asus EeeBox EB1012, the ASUS EeeBox EB1501, the ASRock Ion 330 / Valore ION 330 and the Zotac MAG HD-ND01. In addition, people are building their own low power Atom 330 NVIDIA ION desktops based on mini ITX motherboards such as the ZOTAC ION ITX A Series or the ASUS AT3N7A Atom 330 motherboard and small cases such as the M350.

So what makes this latest bunch of machines so interesting?

First, they use the dual core version of the Atom, the 330 which will speed up multi-tasking as well as single applications that are multi-threaded (click here for a table comparing performance of the N330 to other CPUs).

Second, they abandon the rather pedestrian Intel 945GC chip set (which is basically a four year old design by now) and replaced it with the NVidia GeForce 9400 chip set (Intel Atom + NVidia 9400 = NVidia Ion). The new chip set not only supports DirectX 10 for Windows Vista and Windows 7 but also hardware decoding of digital video. This dramatically reduces the CPU load in software video players that take advantage of it, so that even a humble Atom CPU can keep up with high definition 1080p video streams.

Video performance may have been less of an issue on small netbooks with tiny 1024×600 pixel screens and lacking optical drives, but nettops and other desktops are more likely to use 20 inch screens and bigger that lend themselves well to watching video clips and movies.

Still, NVidia Ion is not a perfect solution for video yet. Amongst others, high definition Adobe Flash videos currently may still have problems because Adobe does not yet take advantage of decoding hardware even if present. For now, only Core 2 and other faster CPUs can cope with all video formats with all software, but Adobe has announced support for hardware decoding for Flash video before the end of the year, at least for the Windows version of Adobe.

NVidia Ion is also much more energy efficient than the Intel 945 GC Express chipset that was used in some earlier Atom nettops. While the Mobile 945 chipset used with single core Atom N270 netbooks (945GSE) is reasonably efficient, the desktop version of the 945 chip set used with the single core Atom N230 and dual core Atom 330 burns more than 20 Watt, over 5 times as much as the modest 4 Watt of the Atom 230 itself and 2 1/2 times as much as the 8 W of the 330. In fact the 945GCE is so inefficient that the cooling fan on the first Intel Atom desktop board had to be mounted on the 945 chip, not the N230 CPU which could be cooled with a passive heat sink alone. Less power than a conventional desktop means not only a lower electricity bill and a smaller carbon footprint, it also means less fan noise and heat.

A dual core Atom with the Ion chip set will actually consume less power than a single core N230 with the 945GC chip set (see GeForce 9400M Versus 945GC – Review Tom’s Hardware : Nvidia’s Ion: Lending Atom Some Wings for a full comparison of the two chip sets).

Another major benefit of the NVidia chipset is that it supports up to 4 GB of RAM while the 945GC and 945GSE are limited to 2 GB even though the Atom was capable of more. This limitation didn’t get much attention before because most Atom machines were shipped with Windows XP, which Microsoft did not allow to be bundled with machines that had more than a single GB of RAM, even though most of these machines could be upgraded to 2 GB by the user. However, if you add a 2 GB DIMM to a 945GSE board that already has 512 MB installed and one free slot, you will still only have 2 GB available, not 2.5 GB. The Ion removes this artificial barrier. Also, Ion boards typically have two DIMM slots while many 945 boards have only one. More memory is welcome because it often means less disk swapping, with a direct boost to performance. Ion offers better memory bandwidth too, which does help when both the CPU and the video chip have to share access to the main memory.

Most Ion boards have 3 internal SATA connectors and also one eSATA connector, while 945GC boards tend to have only two SATA and no eSATA ports. Having an eSATA port is great for using an external drive such as a Blue-Ray disk player or a an external hard disk subsystem such as the Guardian MAXimus external RAID-1 solution to provide robust Network Attached Storage (NAS) via a network-connected Ion machine. Ion nettops also tend to offer HDMI (a digital video link to digital TVs and monitors) and S/PDIF (digital multi-channel audio). Most have 6 or 8 USB ports and draft-N WiFi (802.11b/g/n). On top of that both 945GC and Ion support Gigabit Ethernet.

I’ve been checking online retailers for actual availability of the dual core Ion machines, but things have been moving slowly. I wonder if manufacturers have been holding back until after the Windows 7 release date on Oct 22, to avoid upgrade hassles. Who knows? For example, newegg.com stocks the ASRock ION 330 NVIDIA ION (which comes without any operating system) and the single core AspireRevo AR1600-U910H (which comes with XP), but any of the 2 GB or 4 GB dual core machines that for now are supposed to ship with Windows Vista, are not available yet. Likewise, Amazon sells a 2 GB single core version of the AspireRevo with Linux or a 1 GB single core version with XP, but no dual core version of it at all yet. Whatever the reason, for now you still have to be patient.

Hopefully more machines will gradually start hitting the stores by November and I certainly expect them in volume before Christmas. With a dual core CPU and the superior NVidia chip set, these carbon-saving small desktop machines are becoming viable for many new purposes, whether running Windows XP, Windows 7 or Linux.

Memory upgrades for Lenovo S10e, Aspire M5201, eMachines T6212

Last month my Gateway GT4014j suddenly died and I replaced it with an Acer Aspire M5201, which came with 2 GB of memory installed. That was less than the 3 GB I had in the Gateway and it affected some very memory hungry apps I use. So three weeks later I did a memory upgrade – a series of upgrades to be more precise. I bought two sets of memory upgrades and upgraded three machines.

In my experience, most computers don’t become obsolete because of CPU clock speeds that are too slow but because either there is too little main memory (RAM) for newer, more demanding (i.e. more bloated) applications or because the machine runs out of disk space. The latter used to be more of a problem than it is now, because of the availability of external USB drives, but lack of RAM is still a problem.

Giving your computer enough memory usually is the best way to give it a performance boost because insufficient memory will result in disk swapping. Without enough memory your applications will spend more time waiting for data to get swapped to and from disk, which slows even the fastest CPU down to a crawl such that even a medium or low end CPU with sufficient RAM could run circles around it.

When Vista came out, many manufacturers still sold 512 MB configurations with Vista Home Basic, while advertising mainly the processor speed. Buyers would have been far better off picking a somewhat slower CPU but equipped with a whole GB of RAM. The same is basically still true today for 1 GB machines (usually with Vista Home Premium) vs. 2 GB machines: More main memory almost always beats more Gigahertz!

Back to my Aspire M5201: It has four DIMM sockets (2 banks) for up to 8 GB of memory (2GBx4), but unless you run a 64-bit operating system such as a 64-bit Linux, Vista 64-bit or (from 22 Oct 2009) Windows 7 64-bit, you can’t actually make use of more than 4 GB. Therefore I opted to go as far as 4 GB only for two 1 GB DDR2 PC2-6400 DIMMs from Crucial Technology (CT12864AA800, from Tsukumo.co.jp). They were 1,980 yen each (about US$22) including sales tax.

I also ordered one 2 GB DDR2 PC2-5300 SODIMM (CT25664AC667) for 3,800 yen (about US$42) for a Lenovo IdeaPad S10e. The S10e is a typical Atom N270 netbook with 1 GB of RAM. One 512 MB SODIMM base memory is soldered in while the other can be replaced with either a 1 GB or 2 GB SODIMM. You end up with 1.5 GB or with 2 GB (not 2.5 GB because the 945 GSE chip set is not capable of addressing more than 2 GB of total memory). The Intel Atom itself supports up to 4 GB, which is why nettops with the NVIDIA Ion chip set such as the Acer Revo 3600 can handle up to 4 GB of RAM. The memory upgrade went very smoothly, especially the netbook. The Acer Aspire M5201 (AMD 780G chip set) now shows 3.75 GB of RAM in Windows while the Lenovo shows 1.99 GB.

Encouraged by that I removed the 1.5 GB of DDR DIMMs from my 2005-vintage eMachines T6212 and transplanted the 3 GB of DDR DIMMs (1 GB x 2 and 512 MB x 2) from the dead Gateway GT4014j to it. DDR memory was superseded by DDR2 3 years ago, but now that old machine that originally came with 512 MB shows 2.75 GB of memory in Windows.

If you’re not sure what memory is right for your machine or how much memory you can fit, Crucial have a convenient Memory Finder application that will figure it out for you.