Change kodi cache settings movie#

Uses about 330MB ram 10 minutes buffered. Uses about 230MB ram 5/6 minutes buffered

Uses about 180MB ram - 2/3 minutes buffered

Change kodi cache settings movie#

Uses about 110MB ram - 30 minute bufferedДўnd test: 7G movie DTS - Network maxed out around 10,000KB/s filling the cache. Uses about 10MB ram, very little buffering - about 1 minute. No file - Defaults = 20MB cache readfactor 4 However, I do wonder if Kodi has changed the algorithms for Kodi 17 as I can see no reason why the 3 x memory rule applies. (The RPi3 only has 736MB available due to GPU memory allocation, 250 MB it used for the system, leaving only 480MB to play with) I've just done some tests on a RPi3 7.95.3 using Ethernet and everything is working as expected. I know it's the advancedsettings that's not working. I'll continue messing around with readfactor but is there anything else you suggest for me to do? I know it's not the server or the Pi because it works perfectly on Jarvis with the advancedsettings but, without it it's the same on Jarvis. Just to check, I ran iperf3 and go pretty good speeds and I have 300 Mbps link so it's not the network. If you must use it, adjust it incrementally and check your data. Personally, I think the default LE settings i,e NO advancedsettings.xml should work for most users. Kodi buffering will NOT help a bad network/server, nor will it help a slow processor (Wait times loading the data). I have a feeling that if P1 can supply more to U1 than R2 can accommodate, then problems will occur (Dam burst), which might explain my problems with readfactor of 60. Adjust the Video Cache Size, CURL Timeout, Read Buffer Factor, and Buffer Mode to the recommended settings and press Apply These Settings. R2 will fill as quickly as possible but if P1 or R1 cannot supply enough, then lagging and buffering will occur. So as long as enough water is coming through P1 to satisfy U1 then you should have no problems. The amount of water fed into the reservoir is determined by the amount being used by U1 multiplied by the readfactor. R2 is a reservoir (Kodi buffering) - The size of the 'reservoir' is set by memorysize. P1 is a pipe (The bigger the pipe the more water can be supplied per second) (Your network). R1 is a reservoir (your NAS/Internet streaming provider) The way I think of it, is as a water supply. Use something like nmon/iperf to monitor the amount of data coming in and determine the maximum size received.

Use the tools available in the LE Repository, Network and System Tools. Remember a RPi only has a 10/100base ethernet and max speeds are only in the region of ~40Mbs. You need to check the usage on all sides. I've never had any issues with buffering, even with very large files. I use a Cubox (orig) running Archlinux as a NFS server with a Gigabit network, going to wifi on a 300Mbit extender back to Gigabit into a Cubox-i4 running LE. (unsure where the 3 times memorysize comes into it). The amount of memory used, is going to be based on the size of data required for playback multiplied by the readfactor. The memory is being used, just not ALL at once.