Re: faster rendering of eps pictures in YAP?

cguttman wrote:
Alan Ristow wrote:

Indeed, this is what I would really need. That a DVI viewer can store the pictures and only if they change would they be rendered again. To everyone: isn't there a tweak out there that would allow such a solution? Maybe there is already an option in GScript that allows this?

Chris


Maybe it would be possible to just show the preview from this EPS file? This should be much shorter? Is this possible?


As far as I know, no existing DVI viewers can do that. And of course, the EPS file would have to contain a preview in the first place for that to work (and the file located and loaded and the preview read and displayed somehow, which might take time). Personally, none of my EPS files ever contain previews because I've found that previews often interfere with running LaTeX successfully, but that's just me....

Alan

The basic thing to focus upon is that this is a physical process which takes time. Images on the hard drive take more time to load than images
which are stored in the much faster ram memory. Some images could be
stored on a server and pre-processed. That avoids processing time but
then there is transmission time of the signal from the server to client.

http://www.linuxjournal.com/article/3513
"Using an OPI server allows a client station to offload the processor and I/O-intensive work of creating a high-resolution PostScript file. The OPI server creates and then forwards the PostScript file to the RIP. RIPs are powerful workstations that convert huge PostScript files into ultra-high-resolution laser recorder instruction sets. These files are imaged to film or plate and ultimately used on offset printing presses."

Remember again, that physical time-consuming instructions are processed
by the cpu to implement algorithms, rules to accomplish some request.
So using a more efficient algorithm (maybe preview-latex) saves time.
Buying a system which has a faster hard drive seek time, cpu and very
fast ram memory will make the process go faster.

Drawing pixels on the screen is a physical task. Lower resolution and smaller images will take less time, but then quality will be sacrificed.
There is nothing to do from the command line to change the efficieny
of a given algorithm. But you can give that algorithm less work to do
so that it takes less time. For instance when some program is compiled,
the ./configure takes switches or options. If you choose no debugging,
the build proceeds faster. But you can't make a silk purse from a pig's
ear. It would take time invested in another area to compensate for any
time save later. Your most practical solution is to try preview-latex
or buy a computer twice as fast. It is like two different printers that
I'll call methods; a new printer usually prints much faster than an old
printer given the same document and desired quality of output. I mean
suppose you had some trick to speed up your old printer. That same trick
can be used on the new printer so it boils down a fixed production rate
limitation that comes with your fixed internal method capability.

..eps files are fairly large. On web pages, .png images are often used
because they load faster because there is less information to process. They usually render a poorer quality display. There is a basic physical
size and resource limitation imposed that can be measured by time taken.
One might compromise by using .jpg graphics files which maybe are twice
as large as the same .png file format, so take longer, but give better
quality; .eps files are often converted for viewing web pages.

The reason you keep getting No answers is that you are hoping there is an intrinsic software solution, like a magical -optimization switch
that overcomes what is a basic *physical* limitation. It seemed like
you didn't understand the implication when the term "cache" was used.
I have a 1Gb cpu, only 256mb of memory and it takes like 4/5 seconds
for Yap to load and display. With a 4gigmz cpu, a gig of memory and a
very fast hard drive with big secondary^cache, it would take a second,
a little more or less. [Follup to tex.de removed]

Regards,
Stephen


.

Relevant Pages

  • Re: FastcodeScaleDown B&V 0.2
    ... The quality isn't that good IMHO, I posted examples in the attachments NG, ... where the algorithm in the B&V tool is matched against ... This means the quality of the algorithm may appear suitable on images with ... The issue then becomes how much quality loss is acceptable... ...
    (borland.public.delphi.language.basm)
  • Re: best image enlargement process/ algorithm?
    ... These folks have an algorithm wrapped up in Photozoom Pro 2 that ... *apparent* quality of your images for, for example, printing. ... Like all the other enlargement algorithms methods though, ...
    (sci.astro.amateur)
  • Re: Image Quality
    ... adjustment for download speed/picture quality. ... > Zone Alarm Pro can block Images in IE ... > Scroll down to the Multimedia section and make sure the box>Show> Pictures> ...
    (microsoft.public.windows.inetexplorer.ie6.browser)
  • Re: Image Quality
    ... adjustment for download speed/picture quality. ... >> Zone Alarm Pro can block Images in IE ... >> Pictures Are Not Displayed on Web Sites in Internet Explorer ...
    (microsoft.public.windows.inetexplorer.ie6.browser)
  • Re: Image Quality
    ... I am using IE 6 and Windows XP home edition. ... > 'suboptimal quality' is. ... > Zone Alarm Pro can block Images in IE ... > or contains entries that are blocking access to the desired site. ...
    (microsoft.public.windows.inetexplorer.ie6.browser)