nanogui: Thread: Re: Mozilla runs on Nano-X

Some statistics on the memory front... Peter and crew have
almost completed a full-blown Mozilla port on Nano-X.
(see the screenshots on the web page).  Anyway, here's
what he as to say about the relative sizes of Mozilla running
on X vs Nano-X:

: Just a quick note about memory. We had just done some (rough) comparisons
: with Mozilla(xlib version) + X, and Mozilla(nanoX version) + NanoX.
: 
: To put is simply, xlib version + X uses (approx) 30M more REAL memory than
: the NanoX version + NanoX.
: 
: 
> 
> Basically, the 18M version of Mozilla on Nano-X takes 48M
> on X??  Is that right?

:Actually mozilla+X compared to mozilla+NanoX. The big benefit is due to
: not needing X. Although the NanoX build is 4Meg smaller than the X version

This is great news.  I'm extremely pleased to hear that Nano-X has
come to the point that it can run such a sophisticated application...

Regards,

Greg

Please note, sources and patches will become available within a week from
now. The address will be posted when they the site is up

On Wed, 4 Oct 2000, Greg Haerr wrote:

> Some statistics on the memory front... Peter and crew have
> almost completed a full-blown Mozilla port on Nano-X.
> (see the screenshots on the web page).  Anyway, here's
> what he as to say about the relative sizes of Mozilla running
> on X vs Nano-X:
> 
> : Just a quick note about memory. We had just done some (rough) comparisons
> : with Mozilla(xlib version) + X, and Mozilla(nanoX version) + NanoX.
> : 
> : To put is simply, xlib version + X uses (approx) 30M more REAL memory than
> : the NanoX version + NanoX.
> : 
> : 
> > 
> > Basically, the 18M version of Mozilla on Nano-X takes 48M
> > on X??  Is that right?
> 
> :Actually mozilla+X compared to mozilla+NanoX. The big benefit is due to
> : not needing X. Although the NanoX build is 4Meg smaller than the X version
> 
> This is great news.  I'm extremely pleased to hear that Nano-X has
> come to the point that it can run such a sophisticated application...
> 
> Regards,
> 
> Greg
> 
> 
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: ####@####.####
> For additional commands, e-mail: ####@####.####
> 

> Some statistics on the memory front... Peter and crew have
> almost completed a full-blown Mozilla port on Nano-X.
> (see the screenshots on the web page).  Anyway, here's
> what he as to say about the relative sizes of Mozilla running
> on X vs Nano-X:

Where? http://igelaus.com.au/ is empty and http://www.igelaus.com.au/ is
mostly "About IGEL Australia Sorry, nothing here yet."


//Björnen.

On Thu, 5 Oct 2000, Bjorn Eriksson wrote:
> > (see the screenshots on the web page).  Anyway, here's
> Where? http://igelaus.com.au/ is empty and http://www.igelaus.com.au/ is

The Microwindows web page. To be more specific,
ftp://microwindows.censoft.com/pub/microwindows/ScreenShots/Nano-X-Mozilla.jpg

-- 
--------------- Linux- the choice of a GNU generation. --------------
: Alex Holden (M1CJD)- Caver, Programmer, Land Rover nut, Radio Ham :
-------------------- http://www.linuxhacker.org/ --------------------

Anyone got statistics on the speed front? 

> Some statistics on the memory front... 
....
> > Basically, the 18M version of Mozilla on Nano-X takes 48M
> > on X??  Is that right?
> 
> :Actually mozilla+X compared to mozilla+NanoX. The big benefit is due to
> : not needing X. Although the NanoX build is 4Meg smaller than the X version
> 
> This is great news.  I'm extremely pleased to hear that Nano-X has
> come to the point that it can run such a sophisticated application...
> 


Rob Taylor
Flying Pig System

On Fri, 6 Oct 2000, Rob Taylor wrote:

> Anyone got statistics on the speed front? 
> 

The speed is not too bad, but with True Type fonts, it is VERY at the
initial rendering of a page. I am looking into it.

From Peter

: The speed is not too bad, but with True Type fonts, it is VERY at the
: initial rendering of a page. I am looking into it.

Currently, Nano-X doesn't cache any information about
Truetype font sizes, etc.  This means that Nano-X will [re]load
the font for every call.  Some browsers (ViewML used to do
this) call GetTextExtent and such many times per line, and
that's one reason for the slowness.  Either a caching patch
to Nano-X or a cache in Mozilla's Nano-X driver will
most likely fix this problem.

Regards,

Greg

Greg Haerr wrote:

> Currently, Nano-X doesn't cache any information about
> Truetype font sizes, etc.  This means that Nano-X will [re]load
> the font for every call.  Some browsers (ViewML used to do
> this) call GetTextExtent and such many times per line, and
> that's one reason for the slowness.  Either a caching patch
> to Nano-X or a cache in Mozilla's Nano-X driver will
> most likely fix this problem.

Hello all, we have not been too active lately on this list, we
are getting closer to release dates etc...  But that doesn't mean
that we are not working on microwin/Nano-X!

We are in the process of developing a client side cache for some
operations that causes client/server traffic to eat cpu for our
application mix.  My guess is that these would help the Mozilla
project as well.

From our initial tests, its looks very promising, but it is not
done yet.  Give us a few days and we can get back to you on this
issue when we surface around here.

I will discuss with Greg how best to merge our code with the
new versions.

Best regards,
Morten Rolland, Screen Media, Norway.