On 3/1/19, Dirk Munk <[email protected]> wrote:
> Lee wrote:
>> On 2/28/19, Dirk Munk <[email protected]> wrote:
>>> I've set the following cache parameters with about:config :
>>>
>>> 1. browser.cache.use_new_backend = 1 (true)
>>> This activates a 'new' cache mechanism, that seems to be faster and more
>>> stable than the old one. It is unclear why this isn't the default
>>> setting.
>> It's a programmers' law: There is no fix as permanent as a 'temporary'
>> fix:
>>    https://bugzilla.mozilla.org/show_bug.cgi?id=913806#c8
>>
>> The pref that is enabled by default is
>> "browser.cache.use_new_backend_temp"
>>
>> I still have the defaults for
>>    browser.cache.use_new_backend  set to 0
>>    browser.cache.use_new_backend_temp  set to true
>> And all my cache files are under the cache2 dir which agrees with
>>    https://www.janbambas.cz/mozilla-firefox-new-http-cache-is-live/
>
> Did you read this in that article:
>
>      Enabling the new HTTP cache by default is planned for Q4/2013.

That and the date in the bug report - hence the snark about the
'temporary' fix that's still there.

> I assume that Firefox is now using the 'new' cache as the only cache
> mechanism, so why shouldn't Seamonkey do that as well?

What makes you think it isn't?

I just tried exiting both SM & FF, deleted all the cache directories
except for safebrowsing & started the browsers.  I have a cache2
directory for both.

I don't get a cache directory created until I put 'about:cache' in the url bar

>>> 2. browser.cache.memory.capacity = 4194304 (4 GB)
>>> This sets the *maximum* memory capacity of the cache to 4 GB. It does
>>> *not* mean that Seamonkey will always use 4 GB of cache memory, it
>>> merely means that the cache memory is allowed to grow up to 4 GB *if*
>>> Seamonkey needs it. For that to happen there must be many, many tabs
>>> open.
>> http://kb.mozillazine.org/Browser.cache.memory.capacity
>
> Did you read the table on that page?
> It says that using the -1 setting will give you a memory cache of 32 MB
> if your system has 8 GB or more RAM.
> The default setting for Seamonkey is 200 MB at the moment, I'm using 4 GB.
> That page was written in the dark ages.

^shrug^  maybe so, but I'm not having a problem with seamonkey
stalling or maxing out a cpu.  The defaults are working fine for me.


>>    For e-mail and newsgroups (i.e., Thunderbird and SeaMonkey),
>> messages for IMAP accounts are cached as well in either disk or memory
>> cache, unless synchronized locally already. This reduces the amount of
>> network activity to reload previously viewed messages. This preference
>> controls the maximum amount of memory to use for caching decoded
>> images, messages, and chrome items (application user interface
>> elements).
>>
>> Maybe if you haven't compacted your mail in a while & all the deleted
>> msgs are still in the file?  Or you're looking at newsgroups with a
>> long retention period?  Because it seems like the only web pages that
>> might need >10 MB of cache are if videos are cached.
>
> On the one hand it may be interesting to know why Seamonkey is using so
> much memory cache. On the other hand, I don't care. I want to use
> Seamonkey the way I'm using it. So I make the settings fit for my use.

Sounds good to me :)


>>> 3. browser.cache.disk.enable = 0 (false)
>>> This setting *disables* the disk cache. After I made this setting,
>>> Seamonkey became extremely fast compared with an active disk cache.
>>> However, keep in mind that you should only use this setting after
>>> increasing the memory capacity of the cache.
>> I still think it's a bad idea, but I don't have a gigabit speed
>> internet link or <10 millisecond response time to the web sites I
>> frequent like I recall somebody claiming they had.
>
> I think I have 300 Mb/sec download at the moment.
>
>>
>> https://lifehacker.com/speed-up-firefox-by-moving-your-cache-to-ram-no-ram-di-5687850
>>    Update: One of the folks over at Mozilla laid out a few downsides to
>> using this method. It's not a bad idea, per se, but it's good to be
>> informed about what this does vs. the default settings (and how future
>> plans for Firefox will work with this tweak).
>> ---- links to
>> https://groups.google.com/forum/?_escaped_fragment_=msg/mozilla.dev.apps.firefox/nqYLKTsOAbs/Fh7XO2PVUn0J
>>
>> Lee
>
> Again, that is an article from 2010!!
>
> But let's see if that article is still useful:
>
> 1. It will slow down plug-ins like Adobe reader. I don't notice that.
>
> 2. The size of the memory cache is capped at a much lower number.
> Perhaps with a 32 bit browser, but the standard size of the disk cache
> is 350 MB, I'm using 4 GB in memory!!
>
> 3. The disk cache persists across restarts. That is a horrible argument.

Not so horrible if you have a metered connection.  The only internet
connection my brother in law had for a long time was wireless that
cost $$$ if he went over his traffic quota.  For people like him, a
persistent cache helps save money.

> If there is anything I hate, then it is taking junk from a previous
> session to a new session. When I was still using Windows 98, I often had
> the Blue Screen of Death. The stability of the system was greatly
> enhanced after I made a registry setting that cleaned the page file
> during the shutdown procedure.

$WORK kept doing that to my laptop.  It was most annoying having to
wait 10-20 minutes for the damn thing to shut down.

>It's the same thing with Seamonkey. After
> Seamonkey crashed, I often deleted the profiles folder in appdata >
> local > Mozilla > Seamonkey. It made Seamonkey much more stable, since
> this folder also contains the disk cache.

Try it again.  See what cache directories get re-created.

> 4. I see no reason to use a disk cache if you have a proper memory
> cache. It's very simple, never do on disk what you can do in memory.

Whatever works for you.

Regards,
Lee
_______________________________________________
support-seamonkey mailing list
[email protected]
https://lists.mozilla.org/listinfo/support-seamonkey

Reply via email to