- This topic has 16 replies, 3 voices, and was last updated Dec 21-3:39 am by SamK.
-
Posts
-
andfree
You have 2 swapspaces, on both the P4 and Celeron.
Yes, I verified this for the celeron system, too. I also noticed that the size of the zram0 partition changes slightly every time I reboot.
A couple of links that might be helpful to you:
Tips for Improving Performance on Ancient Kit
ZRAM Misconceptions and DoubtsI have already reduced the shared RAM to the lowest value, switched off CUPS, WICD & Bluetooth, I use the min-icewm desktop, I use Palemoon for common browsing, I have set conky off. After all, I have seen a significant improvement in speed & performance. Now, I’m testing with the older kernel & swap issues, but I can’t say yet if these two have improved the performance of the system any more.
Many thanks for your help.
SamK
…also noticed that the size of the zram0 partition changes slightly every time I reboot.
That is expected and normal.
The size of the zram swapspace depends on the amount of available RAM at the time the swapspace is created. Because zram swapspace exits only in volatile RAM i.e. while the system is powered up, the zram swapspace is recreated at every boot-up. The zram swapspace creation occurs after other software has begun, or is in the process of loading. During boot-up small variations in the working of the hardware combined with all the software that is loaded in the boot process, means the amount of RAM available for zram may slightly differ from boot-to-boot. This can be seen as a small difference in the size of the zram swapspace.
I have already reduced the shared RAM to the lowest value, switched off CUPS, WICD & Bluetooth, I use the min-icewm desktop, I use Palemoon for common browsing, I have set conky off. After all, I have seen a significant improvement in speed & performance. Now, I’m testing with the older kernel & swap issues, but I can’t say yet if these two have improved the performance of the system any more.
It might be the previous tuning you have done has hit the sweet spot for you. Assessing zram in isolation is not the most helpful or revealing way to tweak system performance. It is only one tool is a range of measures you have already tried and seen the benefit of using. As you are discovering, it is the right combination of tweak measures, specific to you and your systems, that produce the worthwhile results.
Many thanks for your help.
That is most welcome. You also should share the credit. The quality of your research, testing, and reporting, is a good model for others to emulate.
Off Topic
Some thoughts to close this discussion. You have already begun to discover that breaking free of conventional, restrictive ideas can pay large dividends.Consider using more lightweight apps and slightly modifying the way you perform your daily tasks. Together they can produce further worthwhile improvements in the performance of your systems. You have already mentioned using the Palemoon web browser. If you use it to watch Youtube videos, try using Streamlight instead. You will find it uses a tiny amount of CPU and RAM compared to any heavyweight web browser.
Streamlight is just one example of a different way of achieving a desired goal (e.g. viewing a streamed video or downloading it). There are plenty of other examples in antiX that perform well while using very few CPU and RAM resources.
The written user guide is available in the antiX FAQ:
Online
http://download.tuxfamily.org/antix/docs-antiX-17/FAQ/streamlight.html
Locally installed
See the Help section of the antiX main menuThe video is available on Youtube
https://www.youtube.com/watch?v=2M5B5aXBDwk
- You must be logged in to reply to this topic.
