Jaguar: Portrait of Xena
Valid HTML 4.01 Transitional

Acer Aspire E5-573G
Subsystem Checkout

Jim Carter, 2016-02-07

When the new machine was received, I checked out these features. In summary, most of the features that the machine was supposed to have, performed with no hassle out of the box. But there were a few annoying exceptions, and some required significant effort to get working.

In particular, the Wi-fi and touchpad would only work with kernel 4.4.0, not with 4.1.12 which is the current kernel for OpenSuSE 42.1. (See workaround for the touchpad.) These are tagged with a yellow background. But I was aware ahead of time that kernel 4.4.0 would be needed, and they worked out of the box once the more recent kernel was installed. Other devices worked OK with kernel 4.1.x, including the SuSE installer's kernel (4.1.5).

Legend

Box Worked out of the box.
Conf It needed editing a configuration file or running a setup GUI.
Add-On Needed additional or patched software: download it; follow instructions to compile and/or install it; it works.
Fiddle As with Add-On, but it required some workarounds in the procedure, which might seem minor to an experienced user but which might prevent a newbie from making it work.
Hack So you call yourself a guru? This one will challenge your skills. But this guru eventually got it working.
Fail I wasn't able to get this one to work, at least so far.
N.T. Not tested.
* A prerequisite did not qualify for the Box or Conf rating, but once that was set up, the listed component worked with no further hassle.

Checklist of Features

Does it boot at all? Update
  As delivered it had BIOS version 1.25, which has a bug that prevents reading a partition table with Linux partitions. You must upgrade to 1.31 or later (I have 1.35). Once that's done it legacy boots the installer and the installed kernel with no problems. This is OpenSuSE-42.1, with either kernel 4.1.15 or 4.4.0.
UEFI Boot Fail
  It can UEFI boot Windows, and the SuSE installation media. However, so far, all attempts to UEFI boot the installed Linux system have failed; the BIOS finds the EFI partition but does not recognize the Linux booter as existing. You have to use legacy booting. See here for gory details.
Processor: Intel Core i5 5200U Box
  Cores (each hyperthreaded) 2
  Technology 14nm
  Nominal clock speed 2.2GHz
  Turbo clock speed (1 core) 2.7GHz
  Minimum clock speed 0.5GHz
  Bogomips 4390
  Jimc's benchmark (Mb/sec) 81.3
  MPEG-4 playback (ffmpeg), CPU% 5.4%
Memory Box
  Memory parameters reported by memtest86+ and /proc/meminfo, with 2x 8Gb memory sticks. (The machine comes with 1x 8Gb.)
  L1 Cache size 32Kb
  L1 Cache speed (byte/sec) 1.46e11
  L2 Cache size (per core?) 256Kb
  L2 Cache speed (byte/sec) 3.33e10
  L3 Cache size 3Mb
  L3 Cache speed (byte/sec) 2.48e10
  RAM on this machine 2 x 8Gb
  Main RAM speed (byte/sec) 1.43e10
  Video RAM reservation for Intel GPU 1Gb est'd
  Sockets for RAM 2
  CPU's addressing limit 16Gb
Graphics: Intel HD Graphics 5000 Oops
  With the Intel GPU and a 1920x1080px display (1080p), the machine can perform MPEG-4 encoded media with no problems: way more than enough CPU power and no dropped frames. Gaming performance is also adequate. But this machine is equipped with the nVidia GeForce 940M (GM108M) GPU which could not be used. The nouveau X-Windows driver only goes up to the GM107 GPU. The current nVidia proprietary driver for OpenSuSE (v361.28) is locked to kernel 4.1.12. I finally got it to install, but it turns out to not support the GM108M GPU either. There are also political and security objections to closed source drivers. More details here.
Display Box
  Display size is 1920x1080 pixels (1080p). The display functions out of the box. However, it does not use IPS technology. For viewing angles: Vertical: there is not really a range where the display colors are constant. Decent rendition +30deg to -20deg. Legible: +45deg to -35deg. (+ is above.) Horizontal: Colors are reasonably constant in ±20deg. Legible in ±60deg.
Disc
  As delivered: Western Digital 1 TB 5400RPM SATA 3Gb/s 8 MB Cache 2.5-Inch Internal Notebook Hard Drive, model WD10JPVX-22JC3T0 Box
  Upgraded: Samsung 850 EVO 250GB 2.5-Inch SATA III Internal SSD Box
LAN Box
  Realtek RTL8111/8168/8411 gigabit ethernet (IEEE 802.3). r8169 driver. PCI ID 10ec:8168. My net uses MOCA and so is limited to 100Mbit/sec.
Wi-fi Kernel
  The device works out of the box given a driver and firmware that supports it. It is a Qualcomm-Atheros QCA9377 (802.11-ac and backward compatible with b/g/n). PCI ID 168c:0042. ath10k_pci driver, which supports this chip starting in kernel 4.4.0. There is also firmware to be uploaded. I tried to backport the ath10k_pci driver and friends, but the result was only partially successful.
Bluetooth Box
  USB ID 0489:e09c by Foxconn / Hon Hai. Using Blueman I paired a keyboard and a headphone; the keyboard generated keystrokes and the phone played music. Nobody says whether Bluetooth and Wi-fi are coordinated to avoid transmitting at the same time.
Keyboard Box
  It has 103 keys (104 counting the power button). The feel of the keys is fine, with normal negative resistance. Key spacing is standard size, except F1-F12 etc. on the top row are shrunk about 15%. There is a numeric keypad on the right side, normally laid out. You don't get the block of Home, End, etc. with arrows below; the arrow keys are below the right shift key, and Up and Down are split into one key cell, which some people may complain about. Home, PgUp, PgDn, End, Del are in the top row with the function keys. Ins is Fn-Del.
Touchpad Kernel
  Name: ELAN0501:00 . The touchpad is civilized in its feel; out of the box a quick swipe gets you across the screen with room to spare, while a slow swipe moves halfway across: speed and acceleration are set nicely out of the box. For buttons (two) you press the pad corners; see the Setup page for how to get three buttons in Linux. A two finger gesture works for scrolling, vertical and horizontal if the application can do that.

The only nasty issue is, there is a pair (at least) of bugs in the pad's firmware, which surface from time to time and prevent multitouch mode. Reported fixed in kernel 4.0.8, not working for me in 4.1.5 and 4.1.12, reported failing in 4.4-rc4, fixed (and working for me) in 4.4.0, still working in 4.4.3. In BIOS set the pad for Basic mode and it will work in 4.1.x (and probably all the rest), including acceleration and scroll gestures.

Audio Box
 
  • Intel Broadwell-U Audio Controller, and…
  • Intel Wildcat Point-LP High Definition Audio Controller
    Only one is accessible; the loaded drivers are for HD audio.
  • Analog stereo (speakers): tested and works, but far from audiophile quality.
  • Analog stereo (wired headphone): tested and works.
  • S/PDIF on HDMI: avaiable but not tested.
  • Microphone: records speech.
  • It has a four-wire jack for a headset including a microphone, but I didn't test the wired mic.
Webcam Hosed
  By Chicony, USB ID 04f2:b51f. First the good news: Skype, using limited 32bit GStreamer, is able to open the webcam and show video. The bad news is, I can't get 64bit GStreamer to do the same thing for other apps. I suspect it's my fault involving version skew, but I put a lot of work into the problem and failed to solve it, so the rating on the webcam remains magenta.
USB Ports Box
  Bus 2 has controllers for USB-2.0 (EHCI, 480Mbit/sec) and USB-3.0 (XHCI, 5Gbit/sec). On it there are two type A (full size) jacks on the left side in blue (USB-3.0), and one on the right side in black (USB-2.0). Bus 2 also serves the Bluetooth HCI, the webcam, and the card slot. Bus 1 has a USB-2.0 (EHCI) controller. It's not clear what, if anything, is ever connected to this bus.
Special Features
  Random number generator (use rng-tools package) Box
  Virtualization (using qemu) Box
  Wake on LAN N.T.
  Watchdog timer from iTCO N.T.
  PXE (network) boot N.T.
Suspend and Wake Oops
  The machine wakes on the power button or lid opening. In the BIOS it can be configured to wake from S3 (not S4) on LAN. It suspends to RAM on lid closing. Times to suspend and to wake, in seconds, with the SSD:
  S3 (suspend to RAM): You need to leave it asleep or awake for 5 mins; with incubation of only 15 sec it fails to wake on about 10% of trials. 2 / 1 sec
  S4 (hibernate to disc, speed varies with RAM used); 100% reliable so far. 7 / 21 sec
  S5 (power off and reboot) 8 / 42 sec
  Greeter to XFCE desktop 3 sec
  Onboard devices including graphics can suspend and wake successfully.
  Beware of a BIOS bug that fails to shut down the non-boot processors if you're on AC power. Windows compensates; Linux also needs a hack script.
  The reliability issues and hack script are discussed in detail here.
Power consumption in Linux, and estimated battery life (* = measured). Windows does better. More power measurements here. OK
  S3 (suspended to RAM, 2x 8Gb) 0.65W, 55hr
  Idle, screen off (DPMS) 12W, 3.0hr
  Idle, screen on 15W, 2.4hr
  Playing MPEG-4 1080p video 18W, 2.4hr*
  Maxed out, both cores 24W, 1.5hr
  Playing game 33W, 1.3hr*
  Charging an empty battery (and time to charge) 22W, 3hr*
  Power brick provided 65W
  Battery: 2.47 amp-hr x 14.4V = 35.6 watt-hr.
Jaguar: Portrait of Xena