David: it IS in master now, no need to patch manually to test it  ;)<br>So waiting for your advice (but if you didn't even noticed the change yet, it's good sign… )<br><br><br><div class="gmail_quote">2011/8/3 David REVOY <span dir="ltr"><<a href="mailto:davidrevoy@gmail.com">davidrevoy@gmail.com</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><u></u>

  
    
    
  
  <div bgcolor="#ffffff" text="#000000">
    I tried to apply the patch to test it , but the terminal answer me :
    <br>
    <br>
    "<br>
    patching file krita/plugins/paintops/libbrush/kis_brush.cpp<br>
    Hunk #1 FAILED at 625.<br>
    1 out of 1 hunk FAILED -- saving rejects to file
    krita/plugins/paintops/libbrush/kis_brush.cpp.rej<br>
    "<br>
    <br>
    Maybe the modification can be applied to master branch ; it will be
    easier to test.<div><div></div><div class="h5"><br>
    <br>
    <br>
    <br>
    On 03/08/11 01:27, Timothée Giet wrote:
    <blockquote type="cite">So for me that's ok, test looks good (I tried painting
      with my presets that use rough predefined brushes and several
      sensors, and with simple small round dots, result looks fine; and
      of course it's much faster).<br>
      Waiting for Ramon and Deevad opinion to confirm.<br>
      <br>
      <br>
      <div class="gmail_quote">2011/8/3 Sven Langkamp <span dir="ltr"><<a href="mailto:sven.langkamp@gmail.com" target="_blank">sven.langkamp@gmail.com</a>></span><br>
        <blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204, 204, 204);padding-left:1ex">
          <div>
            <div>
              <div class="gmail_quote">On Tue, Aug 2, 2011 at 5:09 PM,
                Sven Langkamp <span dir="ltr"><<a href="mailto:sven.langkamp@gmail.com" target="_blank">sven.langkamp@gmail.com</a>></span>
                wrote:<br>
                <blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204, 204, 204);padding-left:1ex">
                  <div>
                    <div class="gmail_quote">On Mon, Aug 1, 2011 at
                      11:42 PM, Sven Langkamp <span dir="ltr"><<a href="mailto:sven.langkamp@gmail.com" target="_blank">sven.langkamp@gmail.com</a>></span>
                      wrote:<br>
                      <blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204, 204, 204);padding-left:1ex">
                        <div class="gmail_quote">
                          <div>On Sat, Jul 30, 2011 at 11:41 AM, Silvio
                            Heinrich <span dir="ltr"><<a href="mailto:plassy@web.de" target="_blank">plassy@web.de</a>></span>
                            wrote:<br>
                          </div>
                          <div>
                            <blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204, 204, 204);padding-left:1ex">
                              Am 29.07.2011 23:30, schrieb JL VT:<br>
                              <div>> I don't know how fast Krita is
                                in the other branches (I haven't tested)<br>
                                > however I'd like to ask if there's
                                a good date to start feeling really<br>
                                > worried about speed?, I don't want
                                2.4 to be a single bit slower than<br>
                                > 2.3, so I'd like to start hacking
                                to try to find where our current<br>
                                > bottlenecks are; but I don't know
                                which parts deep in Krita to touch<br>
                                > without stepping on other
                                developer's toes, moreover, my fears may
                                be<br>
                                > unfounded, maybe we're just a
                                couple weeks shy of a branch merge
                                solving<br>
                                > all these problems, but, again, I
                                haven't kept up enough with our IRC<br>
                                > backlogs and Krita branches to be
                                sure.<br>
                                ><br>
                                > Should I be worried?.<br>
                                ><br>
                                > Is there an upcoming branch I
                                should be testing to help with the speed<br>
                                > bottlenecks instead?.<br>
                                ><br>
                                <br>
                              </div>
                              As far as I know we have two big bottle
                              necks.<br>
                              The first is the creation and
                              transformation (rotation, scaling) of the<br>
                              brush masks. The last time I profiled
                              krita it spent 20-30% (I can't<br>
                              remember exactly anymore) of the whole
                              processing time while painting<br>
                              with this operations.<br>
                            </blockquote>
                          </div>
                          <div><br>
                            <br>
                            Depends on which brush is used. Autobrush
                            shows about 30% of the time in processing
                            the mask, but it's not using 100% cpu. I
                            think there is another bottleneck that
                            callgrind doesn't show. Painting with
                            predefined brushes shows a big performance
                            bottleneck in scaling the brush (callgrind
                            file: <br>
                            <a href="http://depot.tu-dortmund.de/get/syvg6" target="_blank">http://depot.tu-dortmund.de/get/syvg6</a>
                            ) in the stroke benchmark. There is a single
                            method that takes most of the time, so that
                            should be the first target when trying to
                            speed up things.<br>
                          </div>
                        </div>
                      </blockquote>
                    </div>
                    <br>
                  </div>
                  I have attached a patch that turns off interpolation.
                  It should give about double the performance when
                  painting with predefined brushes. Can the artists here
                  check how big the impact on quality is?<br>
                </blockquote>
              </div>
              <br>
            </div>
          </div>
          As first tested showed that it my patch also reduces some
          artifacts too, I have commited it. Please check out if it's
          ok. If not I will revert it.<br>
          <br>
          _______________________________________________<br>
          kimageshop mailing list<br>
          <a href="mailto:kimageshop@kde.org" target="_blank">kimageshop@kde.org</a><br>
          <a href="https://mail.kde.org/mailman/listinfo/kimageshop" target="_blank">https://mail.kde.org/mailman/listinfo/kimageshop</a><br>
          <br>
        </blockquote>
      </div>
      <br>
      <pre><fieldset></fieldset>
_______________________________________________
kimageshop mailing list
<a href="mailto:kimageshop@kde.org" target="_blank">kimageshop@kde.org</a>
<a href="https://mail.kde.org/mailman/listinfo/kimageshop" target="_blank">https://mail.kde.org/mailman/listinfo/kimageshop</a>
</pre>
    </blockquote>
  </div></div></div>

<br>_______________________________________________<br>
kimageshop mailing list<br>
<a href="mailto:kimageshop@kde.org">kimageshop@kde.org</a><br>
<a href="https://mail.kde.org/mailman/listinfo/kimageshop" target="_blank">https://mail.kde.org/mailman/listinfo/kimageshop</a><br>
<br></blockquote></div><br>