longer an option.
heart. For those who wish to try, this is relevant:
I will try to use batch mode as much as possible, and it buys some time.
yes -- I have been doing a lot of that.
appears to buy some time.
>
> On Mon, Jul 11, 2016 at 2:03 AM, Aryeh Weiss <
[hidden email]
> <mailto:
[hidden email]>> wrote:
>
> Hi Curtis
>
> Thank you for your reply. It is good to know that there is hope...
>
> Is there anything that I can do in either my Fiji configuration or
> my Java configuration to improve performance?
>
> Best regards
> --aryeh
>
>
>
> On 11/07/2016 1:25 AM, Curtis Rueden wrote:
>
> Hi Brandon & Aryeh,
>
> I am fairly confident that the cause is attaching AWT menus
> bars, as I
> outline here:
>
http://forum.imagej.net/t/mac-java-speed-issues/304/4?u=ctrueden>
> Someone with a technical background needs to investigate the
> potential
> solutions to the issue that I mention in that post.
>
> In particular, using the Apple-specific
> "Application.setDefaultMenuBar"
> method is highly likely to fix the issue. However, as I
> outline here:
>
http://forum.imagej.net/t/mac-java-speed-issues/304/8?u=ctrueden>
> It would be a bit of work to add code to ImageJ2 to
> hot-translate the
> ImageJ 1.x AWT menu bar to a Swing JMenuBar. The good news is
> that doing
> this would have multiple advantages:
>
> * The "grinding to a halt" performance problem would likely
> disappear.
> * We would have a richer menu bar with icons for ImageJ2
> commands that
> provide them. ;-)
> * The menu bar flashing when you switch windows would no
> longer happen.
>
> The downside is that it is more complex internally, since we
> would need to
> keep the JMenuBar internally synchronized with the menu bar
> that IJ1
> believes is its menu bar (but in actuality is now never
> actually attached
> to any windows).
>
> There might also be unforeseen technical obstacles with my
> proposed
> solution here. All in all, it would probably cost me several
> days to
> explore and implement this fix, and I currently have other
> more urgent
> priorities. But I will try to make time to do this work early
> this fall.
>
> Regards,
> Curtis
>
>
> --
> Curtis Rueden
> LOCI software architect -
http://loci.wisc.edu/software> ImageJ2 lead, Fiji maintainer -
http://imagej.net/User:Rueden> Did you know ImageJ has a forum?
http://forum.imagej.net/>
>
> On Sun, Jul 10, 2016 at 5:15 PM, Brandon Hurr
> <
[hidden email] <mailto:
[hidden email]>>
> wrote:
>
> We've been going back and forth on it a bit on the forums too:
>
http://forum.imagej.net/t/mac-java-speed-issues/304>
> I think we really need to help pin this down for Chris and
> Wayne so they
> can make a change. I've simply not had the time recently
> to dig further,
> but I intend to when the field season here settles down.
>
> It is absolutely related to the changing of windows, but
> what exactly is
> causing it is unclear to me. It does not happen when in
> batch mode. Only
> when in interactive mode.
>
> On Sun, Jul 10, 2016 at 4:23 AM, Aryeh Weiss
> <
[hidden email] <mailto:
[hidden email]>> wrote:
>
> About a year ago I wrote about Fiji slowing down
> during use with Java 8
> and OSX.
>
http://imagej.1557.x6.nabble.com/java-8-and-OSX-td5013642.html>
> As this is still happening, I want to find out if
> there is any additional
> information about this.
>
> I am now using OSX 10.11.5 with Java 1.8.0_92 (64-bit).
> I have 16GB RAM.
>
> Wayne had noted that menu updates are problematic, and
> added a parameter
> SetMenuBarCount: to follow that.
> I note problems when this is over 50, but I cannot say
> when it really
> becomes unusable and I have to quit and start again.
>
> One thing that I noticed -- if I allocate 12GB to
> Fiji, then the problems
> will start much sooner than if I allocate 10GB to Fiji.
> So I suspect that some resource that Fiji uses, which
> is outside of Fiji,
> is using up memory (but I dont know how to check this).
>
> Do people who are running on high-powered Linux
> machines have this
>
> problem?
>
> --aryeh
>
> --
>
>