Lee wrote:
On 3/3/19, David H. Durgee <[email protected]> wrote:
Dirk Munk wrote:
With all the changes I made so far, Seamonkey is unrecognisable fast now.
Have you done any testing to optimize these settings?
I installed the apache2 package on a debian system at home & put a
link to a 23MB jpg on the index.html page
Create a brand new profile for seamonkey, click on Tools / Web
Development / Toggle Tools & select the network tab so I can see how
long it takes for SM to load the file
tl;dr loading a big file from my local network isn't a good test.
I'm open to suggestions on how to make it a better test (ie.
demonstrate how changing settings changes the experience)
Earthlights.jpg ( 23,174,231 bytes)
browser.cache.memory.capacity 200,000 (default)
2047 ms
1953
1953
1953
1969
browser.cache.memory.capacity 1,000,000
browser.cache.memory.max_entry_size -1
1968 ms
1968
1968
1953
1968
click back, DO NOT click tools / clear private data
no time, Transferr... cached Size 0 B
browser.cache.disk.enable false
1953
1952
1968
1968
1968
Lee
Web pages usually consist out of many small objects, all those objects
must be stored in the cache(s). In real life the caches(s) may contain
thousands of objects, so you have to emulate such a situation.
Pipelining for instance is there to lessen the influence of latency, so
your test set-up must include latency.
Setting up a proper test environment is not at all easy, it seems even
Mozilla hasn't been able to do that. Trying to understand what is going
on, making educated guesses etc. is a better method at the moment. With
that approach I came to these improvements, that can be noticed by any
one who implements them. I think the results will be the same on all
platforms.
_______________________________________________
support-seamonkey mailing list
[email protected]
https://lists.mozilla.org/listinfo/support-seamonkey
