Slow web browsers

[linuxheid]

Hi all,

I have been given an old Gatewat ZA8 laptop by my good lady's dad to try to do something useful with it. I have installed Linux Mint xcfe 19.1. So far that seems to work reasonable well for the very basic testing I have done. The stand out issue though is how slow web browsing runs. I was initially using Firefox but this was so slow that captcha anti spam wouldn't load. I have since tried installing a couple of other browsers that I read online are lightweight to try to isolate the root cause of the issue to Firefox. However both these browsers, (Falkon & Midori) show similar symptoms to Firefox. Could this be seom sort of network setting on the laptop I need to optimise?

When I switch on Task Manager, CPU seems normal when I don't have a browser running. Even with a browser running, CPU seems not too bad if I am not actually using the browser. When I send data to the internet though until the screen has fully refreshed a page with the data that comes back, CPU shoots up and flatlines at 100%. Memory seems to be consistently OK.

Very brief use of other apps such as word processor seem OK. The wifi I am connected to is home wifi which runs great for a variety of other connected devices.

Is anyone able to offer any suggestions please on how to tackle this? Any help is appreciated.

[karlchen]

Hello, linuxheid.

Not everyone of us will be aware what the technical details of a Gateway ZA8 laptop are. I admit I am not.
Therefore an inxi -Fxz report, created on your Gateway ZA8 laptop, would come in very handy, because it will give us a concise overview of your technical details and of the system configuration.

• Please, open a terminal window. Make the window fullscreen.
• Execute the commandline

  Code: Select all

  inxi -Fxz

  precisely as it has been typed here.
• Mark the complete text output, which inxi will display with your mouse.
• Press the keyboard shortcut <ctrl><shift>C to copy the marked text into the clipboard.
• Paste the clipboard content into your next reply, please.

Best regards,
Karl

[linuxheid]

Hi Karl,

Thanks for your obviously constructive reply. My apols for the delay in replying and that I didn't find a means initially for posting the info you suggest. I've done whatI believe is what you recommend. Here is a copy of the reply text.

Thanks again for your help, it's much appreciated.

Code: Select all

System:
  Host: csb Kernel: 4.15.0-46-generic x86_64 bits: 64 compiler: gcc v: 7.3.0 
  Desktop: Xfce 4.12.3 Distro: Linux Mint 19.1 Tessa 
  base: Ubuntu 18.04 bionic 
Machine:
  Type: Laptop System: Gateway product: LT31 v: N/A serial: <filter> 
  Mobo: Gateway model: SJM11-YK serial: <filter> BIOS: Phoenix v: 1.3103 
  date: 05/27/2009 
Battery:
  ID-1: BAT1 charge: 21.3 Wh condition: 27.2/48.8 Wh (56%) 
  model: SIMPLO UM09B73 status: Discharging 
CPU:
  Topology: Single Core model: AMD Athlon L110 bits: 64 type: UP 
  arch: K8 rev.F+ rev: 2 L2 cache: 512 KiB 
  flags: lm nx pae sse sse2 sse3 svm bogomips: 2393 
  Speed: 1197 MHz min/max: N/A Core speed (MHz): 1: 1197 
Graphics:
  Device-1: AMD RS690M [Radeon Xpress 1200/1250/1270] 
  vendor: Acer Incorporated ALI driver: radeon v: kernel bus ID: 01:05.0 
  Display: x11 server: X.Org 1.19.6 driver: ati,radeon 
  unloaded: fbdev,modesetting,vesa resolution: 1366x768~60Hz 
  OpenGL: renderer: ATI RS690 v: 2.1 Mesa 18.0.5 direct render: Yes 
Audio:
  Device-1: AMD SBx00 Azalia vendor: Acer Incorporated ALI 
  driver: snd_hda_intel v: kernel bus ID: 00:14.2 
  Sound Server: ALSA v: k4.15.0-46-generic 
Network:
  Device-1: Realtek RTL810xE PCI Express Fast Ethernet 
  vendor: Acer Incorporated ALI driver: r8169 v: 2.3LK-NAPI port: a000 
  bus ID: 03:00.0 
  IF: enp3s0 state: down mac: <filter> 
  Device-2: Qualcomm Atheros AR9285 Wireless Network Adapter vendor: Foxconn 
  driver: ath9k v: kernel port: a000 bus ID: 04:00.0 
  IF: wlp4s0 state: up mac: <filter> 
Drives:
  Local Storage: total: 232.89 GiB used: 9.90 GiB (4.2%) 
  ID-1: /dev/sda vendor: Seagate model: ST9250315AS size: 232.89 GiB 
Partition:
  ID-1: / size: 225.80 GiB used: 9.75 GiB (4.3%) fs: ext4 dev: /dev/dm-1 
  ID-2: /boot size: 703.5 MiB used: 149.6 MiB (21.3%) fs: ext4 
  dev: /dev/sda1 
  ID-3: swap-1 size: 1.75 GiB used: 0 KiB (0.0%) fs: swap dev: /dev/dm-2 
Sensors:
  System Temperatures: cpu: 46.0 C mobo: N/A 
  Fan Speeds (RPM): N/A 
Info:
  Processes: 169 Uptime: 12m Memory: 1.70 GiB used: 644.9 MiB (37.0%) 
  Init: system

[Pjotr]

Apply these safe speed tweaks:
https://easylinuxtipsproject.blogspot.c ... -mint.html

[karlchen]

Hello, Linuxheid.

Your CPU is a single core only, even though 64-bit.

Code: Select all

CPU:
  Topology: Single Core model: AMD Athlon L110 bits: 64 type: UP 
  arch: K8 rev.F+ rev: 2 L2 cache: 512 KiB 
  flags: lm nx pae sse sse2 sse3 svm bogomips: 2393 
  Speed: 1197 MHz min/max: N/A Core speed (MHz): 1: 1197

Browsers like Firefox and Google Chrome e.g. increase their speed by spawning 4 to 8 parallel processes.
4 to 8 parallel browser processes will not only increase the amount of memory which they occupy substantially. (memory hogs)
4 to 8 parallel browser processes will also increase their speed, provided the CPU's can handle the parallel processes quickly enough.
For a single core CPU this approach of course means overkill.
How should a single core CPU service 4 to 8 web browser processes in a performant way in addition to servicing the workload which the operating system and other applications put on it?

I do not know if and how it can be done in Google Chrome. But on the most recent Firefox installations, I have limited the number of parallel browser processes to the number of available CPU cores.
This tendency of assigning all available memory and CPU resources to the browser only is ridiculous.
Instead of optimizing their code the browser developers simply hog all available resources.

Owners of older and slower hardware will be the losers, in particular if all the work has to be done by a single core, which is not even capable of hyper-threading.

Even following Pjotr's tuning advice, which he had linked to, is unlikely to give you any acceptable performance experience on this specific machine. At least I would be very much surprised to learn otherwise. Yet, you may be more patient than I am.

Best regards,
Karl

[kevin987]

Try out Slimjet, Midori and Pale Moon browsers. They normally use less resources.

[karlchen]

Try out Slimjet, Midori and Pale Moon browsers. They normally use less resources.

I cannot comment on Slimjet and Midori, because I have never tried them. Yet, with respect to PaleMoon, my experience is this: It may use less memory than Firefox and Google Chrome. Yes, it does not create several browser processes, but uses threads inside its own process. As a consequence, the total CPU load, which PaleMoon creates, is not necessary less than the load created by Firefox e.g.
Bloated webpages lead to high resource usage by web browsers.
(This forum and the Linux Mint webpages are no good examples for bloat by the way. )

[linuxheid]

Apply these safe speed tweaks:
https://easylinuxtipsproject.blogspot.c ... -mint.html

Hi Piotr and thanks for this. I've had a try at these. So farno noticeable different. There is though one change suggested dependent on the amount of memory.

---
Putting the Firefox network cache into the RAM
7.1. By moving the Firefox network cache from your hard disk to the RAM, you diminish the amount of disk writes. This'll probably make your Firefox noticeably faster. The price you pay is small: it'll only "cost" you 200 MB of your RAM.

Note: don't do this when your computer has less than 2 GB of RAM! Because with very little RAM, even 200 MB can't be missed.
---

The PC has 2 GB RAM in total but 0.3 GB of that is video ram. Should I or should I not apply this change?

[linuxheid]

Try out Slimjet, Midori and Pale Moon browsers. They normally use less resources.

Thanks Kevin for replying,

1) I've tried Midori already and also Falkon without any major improvement. Does Pale Moon run on mint? I ask since Mint doesn't seem to be an option at this link

https://software.opensuse.org/download. ... e=palemoon

2) I've downloaded this

https://www.slimjet.com/en/postdl.php?v ... 4&type=zip

Are there install instructions anywhere? So far i a unable to find any.

Thanks again for helping out.

[linuxheid]

Your CPU is a single core only, even though 64-bit.
[...]
Even following Pjotr's tuning advice, which he had linked to, is unlikely to give you any acceptable performance experience on this specific machine. At least I would be very much surprised to learn otherwise. Yet, you may be more patient than I am.

Hi Karl,

As per replies to Piotr & Kevin so far you are correct. If I get a breakthrough from the outstanding actions I'll post an update.

[karlchen]

1.) [...] Does Pale Moon run on mint? I ask since Mint doesn't seem to be an option at this link
https://software.opensuse.org/download. ... e=palemoon

Your Linux Mint 19.1 xfce 64-bit is based on Ubuntu 18.04 64-bit. Therefore in Steven Pusser's notation you need Palemoon for xUbuntu 18.04.

[kevin987]

1) I've tried Midori already and also Falkon without any major improvement. Does Pale Moon run on mint? I ask since Mint doesn't seem to be an option at this link

https://software.opensuse.org/download. ... e=palemoon

Are there install instructions anywhere? So far i a unable to find any.

From the link you have here, choose Ubuntu. You can grab the binary package by downloading the deb package from the xUbuntu 18.04 link, choosing the amd64 option.

If you like the browser, you can also add it to your repository.

[linuxheid]

Hi Karl & Kevin etc,

I've done the pale moon install as was suggested. I received an error message in doing so that seemed to me the install had failed. However Pale Moon is installed and browses the net. I'll check out how it goes performancewise.

Thanks everyone who has helped.

[Hoser Rob]

Hello, Linuxheid.

Your CPU is a single core only, even though 64-bit.

Code: Select all

CPU:
  Topology: Single Core model: AMD Athlon L110 bits: 64 type: UP 
  arch: K8 rev.F+ rev: 2 L2 cache: 512 KiB 
  flags: lm nx pae sse sse2 sse3 svm bogomips: 2393 
  Speed: 1197 MHz min/max: N/A Core speed (MHz): 1: 1197

Browsers like Firefox and Google Chrome e.g. increase their speed by spawning 4 to 8 parallel processes.
4 to 8 parallel browser processes will not only increase the amount of memory which they occupy substantially. (memory hogs)
4 to 8 parallel browser processes will also increase their speed, provided the CPU's can handle the parallel processes quickly enough.
For a single core CPU this approach of course means overkill.
How should a single core CPU service 4 to 8 web browser processes in a performant way in addition to servicing the workload which the operating system and other applications put on it?

I do not know if and how it can be done in Google Chrome. But on the most recent Firefox installations, I have limited the number of parallel browser processes to the number of available CPU cores.
This tendency of assigning all available memory and CPU resources to the browser only is ridiculous.
Instead of optimizing their code the browser developers simply hog all available resources.

Owners of older and slower hardware will be the losers, in particular if all the work has to be done by a single core, which is not even capable of hyper-threading.

Even following Pjotr's tuning advice, which he had linked to, is unlikely to give you any acceptable performance experience on this specific machine. At least I would be very much surprised to learn otherwise. Yet, you may be more patient than I am.

Best regards,
Karl

Unfortunately I agree with all of this, you may wantr to try Lubuntu or another distro aimed at old or weak hardware.