Profiling JavaFX rendering

I have written a small tool for profiling the rendering of JavaFX Script application
Read about it in my blog:
[http://www.nobel-joergensen.com/roller/java/entry/profiling_javafx_rendering]
Please let me know what you think :-)
- Morten Nobel
Edited by: Mortennobel on Aug 3, 2008 2:09 AM

I don't use NetBeans, but I had some success in profiling memory of a (standalone) JavaFX program with VisualVM, if it can help.

Similar Messages

Profiling JavaFX in Netbeans

Hi,
I'm blowing memory in my JavaFX app - I've never 'profiled' with Netbeans before (am using 6.7.1) but I'm getting no joy.
I'm trying to analyse memory usage but it just sits on the "Connecting to the target VM"
I read that you have to calibrate first - but the platforms available are only JDK 1.6(default) and JDK 1.6
Nowhere does it say JavaFX SDK on Java1.6(default)
and I read on others problems regarding my problem (general Java, couldn't find JavaFX one) that its because the wrong platform has been chosen ?!
Could someone help me out with my noobness and JavaFX profiling and tell me step by step what I need to do ?
Cheers
ScottyB

I don't use NetBeans, but I had some success in profiling memory of a (standalone) JavaFX program with VisualVM, if it can help.

Profiling JavaFX project with Netbeans

Hello,
I'm trying to profile my application with the built-in netbeans profiler.
But when I click the 'Run' button, the message 'Connecting to the target VM...' does not disappear anymore...
What could be the problem?
thanks

I've installed a 64 bit JDK and additionally a JRE , where I'm not sure if this is 32 or 64 bit.. I thought having downloaded a 64 bit version, but the system wanted to locate it in the 'Program Files (x86)' folder (Win 7).
The jvisualvm does not really support my javafx applications.
They're listed as <Unknown application> and eg. at the 'System properties' tab only the message 'Not supported for this JVM' is shown.
Tell me if you need some more informations viewed in jvisualvm!
thanks
P.S. i had massive problems with installing the java staff under Win 7. After first installation java did not work in Firefox. So I had to experiment with uninstalling / installing again and the env variables. Now it works with Firefox too and all my apps can be run in netbeans. But the Java symbol in the Control Panel does not work, and the 'java -version' command results in " Error: could not open `C:\Java\jre6_x86\lib\amd64\jvm.cfg' "
amd64 in the path ?? I have Intel. I'm really scared .....

How to set up Print Specs and Profiles in CS5 Suite

I am no color management expert but I do want to get my printer to print the same colors as my monitor displays in PSD, AI and InD applications.
My Intel iMac and Cinema Display running OS10.5.8 are both color calibrated with Spyder 3 Elite to achieve correct color reproduction.
After days of searching and testing, this is my solution which I offer to help those new to CS5.
1. Start: PHOTOSHOP:
     Set colour settings (Edit>Color Settings) as follows:
          RGB - Adobe RGB 1998
          CMYK - US Web Coated (SWOP) v2 (This setting does not effect printer).
          Leave Grey and Spot as default 20%
     Color Management Policys:
          RGB, CMYK, Gray; all set to Preserve embedded Profiles
     Conversions Options:
          Engine - Adobe ACE
          Intent - Relative colourimetric ( my choice for AI artwork. Choose Perceptual for photos in PSD).
     Go to Save, give profile a name, and note the Conversion intent so you can create another profile for Perceptual and PSD.
     (I called mine Epson R2400 and noted the Relative Colorimetric intent).
     Finally, Go to Settings at the top of the Color Settings dialogue and select the color profile you have just named and saved. Click OK
2. Go to Bridge>Edit>Creative Suite Color Settings. Find your saved profile and select it and click Apply.
     Now AI and ID have the same colour preferences.
     Users without Bridge will have to set these items in Color Settings for each Adobe application.
3.   Photoshop>File>Print: Color Management
          Color Handling - Photoshop Manages Color
          Printer Profile - Choose the ICC Profile to suite your printer and paper stock. I used Epson canned ICC SPR2400 Premium Glossy with great results.
          Rendering Intent - Relative Colorimetric for artwork or Perceptual for photos.
4. At top of Print dialogue box, select Printer (Epson Stylus Photo 2400 in my case).
     Below, Click and open Print Settings;
     Layout - select Paper Size and number of pages (copies). Layout is one of several items in a drop down menu.
     Open the next drop down item, Color Matching. Check that he two options, Color Sync and Epson should be grayed out.
     Move to Print Settings.
          Media Type - depends on your paper. (mine is Photo Quality InkJet).
          Color - select Color or Black
          Color Settings - OFF, This turns off color management in Epson Driver leaving Photoshop in charge!
          Mode - I chose Advanced to get fine quality print. Automatic OK for general use.
      Click Save. Click Print and you should get a print matching your Monitor.
ILLUSTRATOR CS5
Bridge will ensure that Color Settings match Photoshop as 1. above. If you don’t have Bridge Go to Edit>Color Settings and specify the same as      Photoshop. Note Settings, at top of Color settings dialogue box requires the profile you created in Photoshop.
     Go to File >Print. There are 3 steps to check.
     In the window at left are headings General, Marks and Bleed, Output, Graphics, Color Management, Advanced and Summary.
     These are self explanatory. If you do not use Bridge, check Color Management as below:
1. Printer: your printer. Color Handling: Let Illustrator determine colors. Printer Profile: select your monitor RGB icc profile.
   Unlike Photoshop, do NOT use printer/paper ICC. Rendering intent: to suit your image.
2. Bottom left of Print dialogue box is Page Setup. Set your size etc.
3. Bottom Left is Printer. MOST IMPORTANT. Here are the similar setting options as PSD 4 above which require different handling.
     Select Layout, go down to Color Matching>select COLORSYNC not Epson. This ensures Apple ColorSync does color conversion.
     Profile - Automatic should appear with name of your preferred Printer/paper ICC below. ( be careful here as I have found AI sometimes finds the wrong      paper icc. Close Print and start again).
     Move down list to Print Settings - Basic:
          Page Set up - Standard; Media type - Your choice, (Photo Quality InkJet for me).
          Color - Color
          Color Settings OFF to ensure Epson printer driver does not try to manage color.
          Mode - Advanced if you want fine quality print. Automatic is OK for general use.
4. Click Print to close Printer dialogue box. Click Print to close the Print dialogue box and you should get a print to match your on screen image.
INDESIGN and General Comments.
Always select your Printer first as settings vary according to printer type.
InDesign Print set up is very similar to AI. Select Color Management at left and the Printer, Printer Profile and Rendering Intent require identical settings.
Go to bottom left, Page Setup. Specify Printer and paper size. Select Printer and the settings required are the same as AI.
If you switch between these apps there is considerable danger that the different Profile choices will trip you up. Photoshop color management requires the Printer/paper ICC, AI and InDesign require Monitor ICC. My primary error was to select my professionally scanned printer/paper ICC profile which resulted in weird colours on AI and ID. The clue that led me to choose my Monitor ICC profile instead was the ColorSync option in Color matching in AI. I hope you will find printing runs a lot smoother for you and I heartily invite better informed users than I, can add to, or qualify this post.
May the Force be with you!

Regarding print brightness at output. The Settings of monitor calibration determine white point and black point. At the time you run Spyder 3 or whatever your monitor calibrator is, the software sets a white point of your choosing. current iMacs are far too bright and prints resulting from unmodified Mac monitors will appear too dark because of the excessive screen brightness. My iMac at lowest reading is 139 candelas. To get prints which match the screen I specify 80 candelas. The Spyder software also reads ambient light in the room and makes a calculation which effects the white and black points registered for both viewing and printing. Since paper reflects a light level according to lighting conditions, we each view prints in different conditions according to illumination source. This effects also effects screen viewing so conditions determine what we see.
In simple terms, ambient light reading also modifies the software calculations for colour output. If it is too high, prints are pale, too low and prints are too dark. If the Monitor calibration has been done properly there should be no problem. The point here is to make certain your calibration process is accurate. Personally I have calibrated in a dark room at night-time and I have calibrated In a shaded room at midday. Hugely different conditions. Under both circumstances I have found the calibration has resulted in good prints. So I suspect it is user white point settings or calibration error which results in prints being too dark or too light. I understand the data used by the calibration software applies to the ambient light reading to the monitor for reading conditions and uses other color gamut data for the printing process. So printing error must surely originate with the user.
I welcome comment from more experienced people in this matter.
Cheers,

Possible solution for problems printing with ICC profiles - esp. R2400

(N.B. This is long because I've decided to go in to details about the background of the problem etc.. Also note that whilst my experience is with the Epson R2400, anyone with problems printing using ICC profiles in Aperture may find this post helpful, as will be explained further down the post.)
Ok, here's the situation. I've been an Aperture user for over a year, and an R2400 owner for half a year. In that time I have done a huge amount of experimenting, but I've never managed to get Aperture to work perfectly with Epson's 'premium' R2400 ICC profiles - the ones you can download from their site which are better than the ones provided 'in the box'. This hasn't been too big a deal because, in fact, the R2400 does a rather good job just set to 'System Managed' in Aperture and 'Epson Vivid' with a gamma of 1.8 in the printer driver. Nevertheless, it really annoyed me that something that should work wasn't, which is why I've spent a lot of time trying to figure out what's going on. Having said that, I have come across a method which will give you pretty good prints out of your Epson R2400 using the premium profiles in Aperture - it's not perfect, but it's the best you're going to get if you want to use those profiles in Aperture. I understand the words 'it's not perfect' aren't what photography experts would probably want to hear, however, I have seen a few anguished posts from R2400 owners in here before, so I think some people may find it useful.
The whole reason why Aperture is hopeless at using the R2400's premium profiles is because - unusually - their default rendering intent is set to 'relative colorimetric' rather than 'perceptual'. You might say 'but that's good - it means you get more accurate colours!', and if you do, you're right... however, there's a snag. To get an image to reproduce well using Epson's premium profiles and relative colorimetric rendering, you really need to use black point compensation. This is where the trouble lies: Aperture's black point compensation is diabolical to the point of being unusable when used with relative colorimetric rendering - I feel I need to be awarded compensation every time I've ever tempted to use the setting. So because BPC in Aperture is unusable, that effectively makes the premium profiles unusable too, because Aperture always uses the default rendering intent specified in the profile.
The solution? Use perceptual rendering instead. Ok, so you can't change the rendering intent in Aperture, which makes that sound a tad difficult. However, as I said in the above paragraph, Aperture always obeys the default rendering intent specified in the profile... so you can see where we're going with this: we need to change the ICC profiles' default rendering intent from 'relative colorimetric' to 'perceptual'. I did some digging around and found one or two expensive pieces of software that could do that... but then I found that, lo and behold, the Mac OS has a command-line utility which can do the job for us, for precisely £0.00. It's called SIPS or 'Scriptable Image Processing System', and you can find out some information about it here: http://developer.apple.com/technotes/tn/tn2035.html#TNTAG58 For those who don't like reading technical jargon however, here's what you need to do to convert a profile's rendering intent. First go to terminal, then type in the following command:
sips -s renderingIntent perceptual
Do not press 'enter' yet. Instead, add a space after 'perceptual', find the ICC profile you want to modify, and click and drag it into the terminal window. You should then find that your command looks something like this:
sips -s renderingIntent perceptual /Users/yourname/folder/RandomProfile.ICC
At which point you can then press 'enter', and the command will execute, giving you an ICC profile which will now make Aperture use perceptual rendering.
There is just one further thing to be aware of after doing this: for some crazy reason, you then need to turn on BPC in Aperture for the prints to come out as good as possible. Black point compensation shouldn't make any difference when using perceptual rendering as the idea of perceptual is that it takes account of things like that anyway, however, in Aperture BPC does make a difference, so remember to turn it on to get a half decent print. In general, I find that prints made using this setup come out pretty well; they almost perfectly match prints made using the profiles with a perceptual intent in Photoshop Elements, except for the fact that Aperture blocks up the shadows a bit more than Photoshop. However, if you can live with that, you might find this is quite a workable solution.
Now, I said near the beginning of this post that all the above can apply to other printers too. Most printer profiles have 'perceptual' set as their default rendering intent, in which case everything I've just said won't be of much help. However, If you are reading this because you're having problems with ICC profiles in Aperture, but you don't use an Epson R2400, find your problematic ICC profile, double-click on it, and take a look at the window that opens: specifically, at the 'Rendering Intent' the window mentions. If it doesn't say 'Perceptual' then it may well be worth trying the steps I've outlined in this post to set it to perceptual, to see if doing so produces an improvement when using the profile in Aperture.
Finally, just one note of caution: if you decide to try out the steps I've detailed above on a paid-for custom-made profile, please back your profile up before messing with it. I haven't experienced any problems when using SIPS to change a profile's rendering intent, but I obviously can't guarantee that it won't do something weird and corrupt your expensive custom-made profile.
If you have any questions, feel free to ask, although (contrary to any impression I may give) I am not a colour-management expert; I'm just someone who doesn't give up when they have a problem that should be solvable.
Thomas
Mac Pro 2.0GHz with 30" ACD; 15" MacBook Pro 2.0GHz Mac OS X (10.4.10)

Thomas
Wow - thanks for such a comprehensive post.
I have Aperture and a 2400 so this information is exceptionally useful to me.
Again - thanks for caring and sharing
Brian

On differences between JavaFX CSS and the w3c standard css

This is a discussion topic which came to mind after reading comments in this thread Re: FXML, css and -fx-font-family
Feel free to comment or share your thoughts on the topic.
Note that these comments relate to the naming and semantic meanings of css tags, and not the syntax and parsing of css (as both w3c css and javaFX css seem to be equivalent in the later regard).
You know, sometimes I find JavaFX's lack of certain css features refreshing. It is nice to have all, well, almost all, of the JavaFX CSS documented on one page. I think if you were to do the same thing for w3c css then it would end up with a much larger, more difficult to understand document and even what the contents of that document should be would probably be pretty controversial, in the same kind of way that the HTML5 specification ended up being. Microsoft contributed over 7000 tests to w3c just to cover only a subset of the css rules available in browsers today. The webkit project lists almost a thousand open CSS bugs: https://bugs.webkit.org/buglist.cgi?product=WebKit&component=CSS&resolution=---. So implementing even partial w3c CSS support is a complex project.
When I first started using JavaFX CSS I found the differences to w3c CSS quite jarring, and it was hard to account for them. Now that I am used to JavaFX CSS, I don't have as much of an issue. The difference in names and semantics is going to be an impediment to designers working on JavaFX who are familar with w3c CSS - but the designers will quickly also discover that FXML is not HTML and Java is not JavaScript and JavaFX deployment is not HTML deployment. So, perhaps, in the larger picture, it is not as big a deal as it would seem. Still, one can't help think that any impediment to people easily picking up and adopting JavaFX is doing the technology a disservice.
There are other advantages in having the JavaFX CSS in it's own namespace to w3c css, in that it can evolve seperately, it is not expected to be exactly the same because it has a different name, it is not expected to fully implement w3c css as browsers do because it's clearly a different thing by name, etc. There is even precedent for it in the use by mozilla of moz- CSS prefixes and webkit css properties http://css-infos.net/properties/webkit - the web as not as standard as a lot of people propose. JavaFX CSS is backing a JavaFX rendering engine and not an HTML rendering engine. It is quite remarkable that Oracle was able to build and make available a CSS model for JavaFX that feels as familar as it does to w3c css, as w3c css is targeted very strongly toward styling a completely different technology set (HTML markup and document object model).
It would be useful to have a tool which translates w3c css to approximations of JavaFX css and vice versa, or the ability for JavaFX to have a mode (perhaps a boolean value when a stylesheet is loaded) to perform an automatic alias or mapping of w3c css to javafx (at least for the subset of w3c css which would make sense to automatically translate to JavaFX css). I haven't checked the JavaFX jira in detail, so perhaps a request for such support already exists - perhaps it is under the private http://javafx-jira.kenai.com/browse/RT-9272 jira.
The JavaFX css model is really powerful and I have found really useful some of the additions that it adds above the baseline 2.1 css that is found consistently in browsers. With the upcoming Java CSS object model http://javafx-jira.kenai.com/browse/RT-17293, you will also get better programmatic access from Java.
So, nice work on building and implementing this complex feature . . .

Thanks for the feedback. Closing the gap between JavaFX CSS and W3C CSS will be an evolutionary process. It would be nice to use a standard's based stylesheet but there is not always a 1-1 mapping to JavaFX. But for those properties that can be mapped, it is something that should be supported. For example, we should be able to handle "font" or "-fx-font".
I have made http://javafx-jira.kenai.com/browse/RT-9272 public.

Printing and Proofing with multiple profiles

I have a book project where two kinds of papers is used.
One more yellow than the other.
For both of these papers I have a different profile.
Is there any way that I could define different output/proofing profiles for different parts of the document?
If not then what about combining 2 documents into a book.
Can I print/export the book so that it ends up having two different profiles in the resulting pdf?

Solution for the OP, as tested by CS2:
Convert by Photoshop:
Image1 RGB to CMYK1 (for instance ISOCoated for white paper)
and embed profile CMYK1
Image2 RGB to CMYK2 (for instance for yellowish paper)
and embed profile CMYK2
Define Document CMYK space CMYK0 in InDesign.
CMYK0 is CMYK1 or CMYK2 or something different.
Place CMYK-image1 and assign explicitly profile CMYK1
Place CMYK-image2 and assign explicitly profile CMYK2
Assignment by
>Object >Image Color Settings >Profile and Rendering Intent
InDesign does not honour the embedded profiles automatically.
Without explicit assignment the document color space is
implicitly assigned (CMYK0).
These are the settings for Export to PDF:
No Color Conversion
Include All Profiles
No Output Intent
No compression or ZIP/LZW compression
Downsampling as required (here 288ppi if above 288ppi)
The PDF contains the two images, and the CMYK profiles
are embedded. Style like PDF/X-3. Test by Preflight.
For offset printing the profiles are not required.
So far it's assumed that both profiles CMYK1 and CMYK2
are valid for the printing process with the same ink
but for different paper. CMYK-numbers are print-ready.
For proof printing one needs a RIP with this feature:
For tagged images use the embedded profile as input
profile (as defined by PDF/X-3).
Output profile is that of the proof printer.
RIP Colorgate ProductionServer5 works as described,
additionally per image for RGB and Gray images.
How to test this workflow:
Source image is the same AdobeRGB(98) color pattern,
first converted to CMYK1=ISOCoated, then to CMYK2=Toner-
Printer-Profile.
The AdobeRGB image and the two CMYK images were printed
by inkjet as described. For patches which are in-gamut
for all profiles the appearance is the same (we have the
same physical colors but different encoding).
Best regards --Gernot Hoffmann

Hidden color profiles

A profile which appears in Windows/System32/Spool/drivers/color does not appear in the Printer Profile Window of the Print Window. A second profile does appear there but not in Windows/System32/Spool/drivers/color. This is after twice rebooting my WIndows XP system, running Photoshop CS3. Any suggestions, please?

Catpearl,
On my WinXP system, my default color profile directory is WINNT/System32/Spool/drivers/color. First check that. I was on Win XP Home, and it varies from one version to another. It's probably right the way you have it, but double check to be sure.
I assume you are printing from Photoshop using "Print with Preview"? If so, you can instruct Photoshop to select a profile and rendering intent in the PS Print with Preview dialog box and it should find the profiles if they are in the right location. If you install profiles when PS is open, you may have to shut down PS and restart so it sees them. If you set the profile in PS's PwP, then turn off color management in your printer driver so you don't do double color management and conversion (butt ugly).
If you are selecting profiles in your driver, it may or may not see custom profiles. Not sure, since I always use PS print with preview when printing from Photoshop and select profiles there, or I use the PS print plugin that comes with my Canon iPF6100 printer (which bypasses the O/S completely, uses a 16 bit print stream, and talks directly to the printer...very cool and well thought out).
It could be you have profiles in the wrong location, your printer driver doesn't see them, you have a corrupted file, or there is a glitch in the driver installation. Just guesses. Also, it could be your driver settings....if you select certain settings, it may just look for the "canned" profiles that were installed with the driver. Grasping at straws, since I don't know you printer model, settings you are using, what you are using to print, etc.
Hope this triggers some ideas and investigation that pays off.
Lou

Pro-100 Custom Printer Profiles

Has anyone had success in using a custom printer profile with this printer. Glossy generic Canon paper profiles work ok but are not completely on the money. I get a little too much red. I had a custom profiles created by reputable 3rd party provider for a variety of Canon papers. When I use them to print the test sheets the anomalies are dramatic. Instead of smooth transition of colors steps and anomalies are are printed throughout the test sheet. When I print the Macbeth color chart an example of how far the colors are off one could not find the red square on the printed output. Thinking the provider had a bad day I asked him to do it again. There was no change in the results. I then went out and bought an Xrite Color Munkie. I have the same results as with the 3rd party provider's color profile.
Previously I used custom profiles with the i9900 for years. Still using the same computer and software; CS5 and Vista 64 bit OS. I let Photoshop manage the colors, I select the custom profile, check Manual on the print driver screens, and select None for Color Correction on the Matching tab. Am I missing something?

I tried the Color Correction to None and when i started from scratch and redid my profiles it worked! Thank you
What I find troubling is where is this explained in the manuals or help? If this is a "professional printer" then custom profiles are going to be used.
Where is the expalnation for the other options on this screen particularly ICM. When it is chosen I only see Paper company profiles. Why do I not see my custom profiles (see link on previous comment)? When I press Help I get the following:
"ICM (ICC Profile Matching) Adjusts the colors by using an ICC profile when printing. Specify the input profile, printer profile, and rendering intent to be used." If your major software products from Adobe and Apple do the color management why is this here? Somewhere in the internet, not at Canon site, I see its purpose is for applications that do not control the color management. Which begs the question how many users who use that kind of software are going to look, find, and understand this screen and color profiles.
Sorry after losing a lot of time and money I have to vent at Canon. I do appreciate the help. Thanks again.

Which perfomance improvements will JavaFX 8 deliver to us?

Hi there,
my whole company consists already of fans of JavaFX 2 - it's fun to work with it and slowly but steady, the team seems solves on the most annoying bugs. But one thing in the 2.x versions seems to be a problem for us. Even on recommended combinations of hard- and software, performance isn't really predicable yet. Sometimes, our applications perform well on 2-year old hardware while we have issues using high-end desktop components. Generally, i think that there is some potential to improve the performance of JavaFX.
For example, i read in Jira that the developers where thinking of pre-rendering nodes as bitmaps and cache those to gain some performance wins. I didn't find this discussion again, but i think that this hasn't been implemented yet.
So what i really would like to know is if there are upcoming performance improvements in JavaFX 8 and if the answer is yes, what will they look like?
Best regards,
David
Edited by: TheJeed on 04.02.2013 02:49

There are some preliminary performance numbers for JavaFX 8:
http://fxexperience.com/2012/09/preliminary-performance-numbers-in-javafx-8/
The fxexperience post also includes descriptions of some of the work to improve performance in JavaFX8.
Also there is a note on memory consumption:
I would have to check our performance numbers to be certain, but I believe actually we have also reduced
the memory usage over the same period, primarily due to making Control a Region and removing one extra
node for most Controls (for example, a Button is now a Button -> Text, instead of Button -> StackPane -> Text).I think some of the major work involves allowing the JavaFX rendering thread and JavaFX application thread to execute concurrently.
i read in Jira that the developers where thinking of pre-rendering nodes as bitmaps and cache those
to gain some performance wins.
I didn't find this discussion again, but i think that this hasn't been implemented yet. Perhaps it has. Have you tried out the Node setCache and setCacheHint APIs?
http://docs.oracle.com/javafx/2/api/javafx/scene/Node.html#setCache%28boolean%29
http://docs.oracle.com/javafx/2/api/javafx/scene/Node.html#setCacheHint%28javafx.scene.CacheHint%29
Additionally, you can manually do this by using node.snapshot and then rendering the resulting image in an ImageView rather than the original node.
http://docs.oracle.com/javafx/2/api/javafx/scene/Node.html#snapshot%28javafx.scene.SnapshotParameters,%20javafx.scene.image.WritableImage%29
Depending on the usage situation, the caching APIs can make a huge difference to application performance.
Running a grep for the word performance on the JavaFX 8 issue tracker gave =>
RT-5130     Mechanism to report FPS of an arbitrary app (app doesn't call PerformanceTracker) in Prism
RT-5475     RT-5474 Optimize picking performance
RT-6475     Need new hints to control how Text node is rendered
RT-8037     Need to complete a temporarily performance fix on filter effects (DropShadow, InnerShadow and SepiaTone)
RT-8243     Memory Leak (sun.awt.geom.CurveLink)
RT-10753     prism-sw: ColorfulCircle bm performance drop in case circle radius is equal 2 or 4 for Inner & Outer Stroke positions
RT-13324     Text needs to support letter spacing
RT-13365     poor performance running "colorful circles"
RT-14065     DirtyRegions: different performance results for equivalent dirty regions
RT-14122     Camera Style Transform
RT-14867     b35: 20% (-2.47 fps) performance regression in TextBox-text100k run with hw pipeline comparing to b34
RT-15566     too big area repainted if the test updates more than 12 checkboxes
RT-15687     TreeItem expand/collapse performance in hardware pipeline is almost 20 fps worse than in j2d pipeline
RT-16725     API to manipulate scrollbar on TableView
RT-16853     TextArea: performance issue
RT-16924     WebNode reload test can't work in sandbox mode
RT-17510     Improve performance of rendering a TRANSPARENT stage on Windows 7
RT-17551     MacOS: Optimize using of lockFocusIfCanDraw
RT-17960     Transparent stage resize flickers on Mac
RT-18067     fx2.1-15 (b01): up to 90% regression in Controls.TextArea-select_all-255x1000 in j2d pipeline
RT-18354     Animation perform down observably when i add the dropdown effect to root
RT-18612     Glass Mac: ClipboardAssistance.actionPerformed()
RT-19106     LCD text blending isn't accounting for over-lapping glyphs.
RT-19271     J2D pipeline does not implement proper handling of the subimage "isolate mask" control
RT-19305     Animation performance may be very unstable on Mac
RT-19605     The fix for RT-19532 results in 65% (317 fps) performance regression in Controls.Buttons-adhoc-mouse-nodes1008-skipAll in fx2.1-controls-scrum-b345 build
RT-20356     PresentingPainter and UploadingPainter disregarding dirty clip rect
RT-20405     Improve Path rendering performance
RT-20475     Allow to select videocard (or choose the best one automatically)
RT-20616     TreeView creates new cells every time root is changed
RT-20978     fx2.2-graphics-scrum-h175: performance regression on Mac (initially 50% but now unknown)
RT-21672     Implement Occlusion Culling
RT-22076     CSS: Treat style class as a bit mask in Selector for better performance on selector matching.
RT-22244     Pisces Renderer shows huge performance win when coded in C
RT-22422     Cannot sign JAR files
RT-22567     Minor tweaks to FX/Swing painting
RT-22913     Implement and enable accelerated compositing to improve WebView rendering performance
RT-23413     FXMLLoader should be able to cache imports and properties between to load() calls
RT-23450     Improve performance of Prism rendering and clipping
RT-23458     Improve ellipses performance
RT-23459     Reduce node count in Controls by allowing a Skin to manipulate the Control's children and layout directly
RT-23463     Implement Region image based caching
RT-23467     Evaluate Native Text Engines
RT-23469     Analyze Layout Performance
RT-23562     Optimize Region by using immutable State, caching on CSS level
RT-23574     Add support for tiled rendering of textures (both for performance and functional reasons)
RT-23582     Update Caspian CSS to use better performing selectors
RT-23588     Investigate using less precise but faster min, pref, max size computations for controls
RT-23616     Avoid processing CSS for VirtualFlow cells that are CSS clean
RT-23621     Check use of HIGHP with es shaders.
RT-23704     Use DirectX 9Ex mechanism to avoid calling cv.save()
RT-23725     Beagleboard: Execute fragment shader on the GPU causes significant drop in performance
RT-23738     fx3.0-graphics-scrum-h45: Webkit synchronization for 3.0 caused -55% performance regression in WebNode.ColorfulShapes-Blur2Circle
RT-23743     Evaluate the performance of computing and uploading individual glyphs one at a time versus in a batch
RT-23746     Consider optimizing libjepg for performance rather than quality (using already existing build flags)
RT-23773     Improve caspian.css to use child selectors where possible, rather than require css engine to perform needless busy-work
RT-23839     Lists,Trees and Tables do not work with wrapped text
RT-23873     Investigate (and improve) ListView / TreeView / TableView performance
RT-23879     Mac: Clicking in a Popup/ContextMenu/ComboBox steals the focus from the primary stage in applet
RT-23905     SQE: Provide JavaFX CSS3 support
RT-23913     SQE: Optional FX Packages
RT-24010     WebNode.GUIMark2-Vector fails with NullPointerException in multithreaded mode
RT-24012     Text performance of the hardware pipeline must be equal or better than the software pipeline
RT-24013     Multi-Core scalability
RT-24017     3.0-controls-scrum-58: 100% regression in NodeMemory.TreeItem and 96% in Controls.TreeView-Expand
RT-24018     3.0-controls-scrum-60: Controls.ListView-Keyboard and Controls.TableView-Keyboard benchmarks are broken
RT-24143     Ant task fx:deploy expectes all JARs to have a MANIFEST.MF and fails with NPE if at least one doesn't
RT-24148     Private controls Weak* classes should clean up in more situations
RT-24298     3.0-graphics-scrum-h142: memory usage of ColorfulShapes.Blur2Circle-Transparent has been increased by 625% (246,890.67 kb)
RT-24320     WebView draws entire back buffer on screen upon every repaint
RT-24437     X11Pixels and LensPixels are ill-implemented wrt managing the native object
RT-24480     Use FilteredList for Parent.unmodifiableManagedChildren
RT-24557     ImagePattern is slow on embedded systems
RT-24582     High frequency refresh and Heavy but low priority updates in the same app (multithreading render, multiinstance...)
RT-24587     Changing a single child of FlowLayout is slower than changing all children
RT-24624     prism-sw pipeline is up to 90% worse than j2d pipeline
RT-24741     WebView is not using accelerated compositing for http://javaweb.sfbay.sun.com/~epavlova/Performance/JavaFX2/mexsantos opacity transitions
RT-24965     8.0-graphics-scrum-h82: Controls.TableView-XmasTree and DirtyArea performance regressions caused by RT-23725 fix.
RT-24966     8.0-graphics-scrum-h99: up to 30% regression in Guimark2.Bitmap
RT-24992     8.0-controls-scrum-h51: Controls.CheckBox and Controls.RadioButton benchmarks are broken
RT-24997     8.0-controls-scrum-h44: up to 35% regression in Controls.TableView and Controls.Buttons benchmarks
RT-25036     8.0-h26-b56: Controls.TableView-XmasTree benchmark crashes with SIGSEGV on MacOS in sw pipeline
RT-25078     Mac: dirChooser.showDialog() returns immediately "null" but somehow remains open and appears to work (...when being executed from a Button onAction Eventhandler in new Stage)
RT-25118     8.0-controls-scrum-h51: -23% performance regression in Controls.TreeView-Expand.adhoc-items1000-wide benchmark
RT-25120     8.0-graphics-scrum-h99: up to -24% performance regression in Controls.ListView and Controls.TableView
RT-25166     Path updates in a ScrollPane where content has a Scale transform are 100 times slower
RT-25233     Javafx applet with preloader crashes under Windows / Java7u7 / Chrome
RT-25350     poor animation performance on PathTransition when part of the Stage sits outside of desktop screen
RT-25382     8.0-h38-b58: up to 30% performance regression in Image rendering in sw pipeline
RT-25403     8.0-h38-b58: 67% (-21.7 fps) performance regression in Charts.Bubble in hw pipeline on MacOS-Low end machines
RT-25486     Debian policy requirement for 'Installed-Size' not set -- users get warning when installing: NetBeans built JavaFX .deb package for Ubuntu
RT-25684     GridPane.layoutChildren() hangs when wrapText is true
RT-25685     GridPane.layoutChildren() hangs when wrapText is setup in CSS style class
RT-25801     8.0-controls-scrum-h81: 25% performance regression in Controls.RadioButton on mac-low end machine
RT-25864     New "shared textures" do not share pixel update flags as well as they should
RT-25919     8.0-graphics-scrum-h242: 48% (16.70 fps) performance regression in GUIMark2.Bitmap on MacOS
RT-26024     Scene Builder hangs when adding column
RT-26113     8.0-graphics-scrum-h278: disabled accessibility feature triggers dependency on uiautomationcore.dll
RT-26140     Node Orientation - API issue: Should effective orientation be a property?
RT-26221     8.0-graphics-scrum-h276: 33% regression in NodeMemory.TableColumn-rendering-on benchmark
RT-26438     8.0-graphics-scrum-h342: up to 30% of memory usage increase in bunch of performance benchmarks
RT-26537     8.0-graphics-scrum-349: deadloack in WebNode.GUIMark2-Bitmap benchmark
RT-26702     Poor DisplacementMap effect performance on Mac
RT-26716     Performance of scrolling TreeView tail is much more slowly when scrolling TreeView head
RT-26894     String rendering is less performant than java2D one
RT-26909     Regions are being cached even when extremely tall
RT-26999     8.0-controls-scrum-h122: up to 20% regression in some Controls.TableView benchmarks
RT-27073     8.0-graphics-scrum-h458: more than 50% regression in Controls.TextField benchmark
RT-27079     8.0-graphics-scrum-h449: RT-25055 caused up to 80% regression in Charts.Bubble benchmark
RT-27111     8.0-controls-scrum-h138: up to 30% regression in Controls.TreeViewExpand benchmarks
RT-27143     8.0-graphics-scrum-h471: TableView-Keyboard fails with NullPointerException
RT-27508     8.0-h96-b68: 68% regression in SpiralText in hw pipeline caused by enabling DirectX 9Ex
RT-27667     [TreeTableView] Multiple cell selection issue.
RT-27982     Mac: 8.0-graphics-scrum-h595: JavaFX crashes on MacOS
RT-27985     8.0-graphics-scrum-h595: 42% performance regression in Guimark2.BitmapCanvas on Windows-Mid-Range machine
RT-28064     A new Scene accumulates all PointLight's from previous Scene's

Choosing profiles for forced RGB to CMYK color conversion

When a mixed RGB/CMYK PDF is opened in Illustrator CS6, Illustrator forces a conversion to one color space or the other. See this screenshot: http://imgur.com/sK8iEdn
I assume this is a limitation of Illustrator and there's no way to keep both color spaces. Under that assumption, Is it possible to choose the profiles used for the conversion from RGB to CMYK? Can Illustrator be made to use the RGB and CMYK profiles defined in its Color Settings to make this conversion?

I did some experiments with Illustrator CS6 and the MacBeth RGB test chart and verified my results with Photoshop CS6. I discovered Illustrator is (mostly) doing what it should be doing, within a 1% error (probably rounding) on the output CMYK values. Here's a summary, in case anyone else needs this info:
Assuming the source file's elements are all untagged, when a mixed RGB/CMYK PDF is opened in Illustrator and CMYK mode is
chosen, Illustrator will use the profiles and rendering intent defined in
Color Settings to make the color conversions from RGB to CMYK. Thus, we
have control over the profiles used for this conversion.
If the RGB elements in the PDF file have embedded ICC profiles,
Illustrator will use the embedded ICC profile instead of the RGB profile
defined in Color Settings. This ONLY happens, however, if the PDF file
also includes the correct CMYK output intent profile.
If the PDF doesn't contain a CMYK output intent, Illustrator will fall
back on the Color Settings RGB profile for RGB->CMYK conversion. I believe it
would be more correct for Illustrator to use the embedded RGB profile and
the CMYK profile defined in Color Settings, but that's not how it seems to work.
If the PDF contains the incorrect CMYK output intent, Illustrator will
ignore the Color Settings and respect the embedded RGB and CMYK profiles
for the conversion, as might be expected.

Is there a way to import Raw files into LR catalog ..including..any edits performed in an external raw file editor such as Capture One?

Is there a way to import Raw files into LR catalog ..including..any edits performed in an external raw file editor such as Capture One?
I can import the Raw file successfully but cannot see any edits that were applied in Capture One. Im assuming no but just checking.

Your assumption is correct, Capture One, Lightroom and other third party raw processors have their own proprietary processes and profiles for rendering the raw data.
They do not make permanent changes to the raw data and store the changes to an .xmp file or to a catalog file like Lightroom.

Lightroom 1.3.1 Printing Problems with Apple OSX 10.5 (aka Leopard)

Whilst the majority the majority of Leopard related print issues were addressed when Adobe updated Lightroom to 1.3.1 and printer vendors released updated drivers some issues remained and it has been left to users to establish workarounds. One such proven workaround is presented below:
Information and text by Jason Hicking
I've been having exactly the same issues running a nearly identical system but on a MacBook Pro with an external Eizo 24" monitor for critical work. My results also mirror yours ie slightly greener and darker prints from LR than PS.
I've spent literally hours troubleshooting this as I own a digital imaging company that offers printing and colour management to our customers, many of whom have the same setups and same issues. Well I'm glad to say that I seem to have found a (temporary) fix. Read on and let me know your thoughts...
If you go into the Apple ColorSync Utility (in the Utilities folder on your hard drive) then follow these steps I'm pretty sure you'll get accurate prints that match Photoshops output.
Step 1 Open ColorSync Utility and click on the Devices button at the top of the window.
Step 2 Click on the grey arrow to reveal the Printers list
Step 3 Click on the grey arrow next to "EPSON Stylus Photo R2400"
Step 4 Locate the profile that relates to the media setting that was used when you created your custom profile. For example I create most of my custom profiles for gloss or satin media using the Premium Semi Gloss media setting so in Colorsync Utility I choose "SPR2400 PremiumSemigloss"
Step 5 Click on the arrow next the text "Current Profile:" and choose Other...
Step 6 From the window that opens, choose the profile that you are using to print through in Lightroom, as an example I choose "R24 IlfSmGls PSG NCA" as this is the profile I made to print on Ilford Smooth Gloss and the one I'm selecting in Lightroom.
Step 7 Click on OK to accept this change and ensure that it's actually changed in the ColorSync Utility window. If everythings OK then close Colorsync Utility and return to Lightroom.
Step 8 Print from Lightroom, choose the exact same profile and rendering intent that you would normally, click on the Print... button, make sure the preset is set to Standard then select the matching media setting and quality settings that were used to create the profile (as you would when printing from Photoshop), remember to check that No Color Adjustment is set in this part of the driver.
Step 9 Click on the Print button and in a minute or so you should get a print that matches your Photoshop output and more importantly matches your 'calibrated' screen.
Step 10 Presuming this works save this setup as a template in Lightroom for future use.
This may seem long winded in text form but it's a one time operation and I wanted to make sure that every step was included for clarity.
What I feel is happening is that there is a double profiling issue with print stream from Lightroom to the the Epson driver (v6.12) and by selecting the identical profile in both places we are ensuring that no conversion takes place in the driver, something that simply setting "No Color Adjustment" usually achieves in Photoshop and other older applications that use a different print stream to Lightroom. Also note that even when your profiles are set this way in the Colorsync Utility there is no difference in accuracy or colour when you leave it set like this and print as usual from Photoshop.
Attached screenshot of ColorSync Utility provided by Ian Lyons
A Lightroom print tutorial is also provided at: http://www.computer-darkroom.com/lr_13_print/lightroom_print.htm

Well, maybe, CS3 is the only one fully compatible, but Photoshop CS and CS2 will work...mostly. There are a couple of niggling little problems, but I have been able to do what I want, one way or another, with CS1. Elements 2 is, like PS7, a dead letter in Leopard, but Elements 3 and 4 work. Of course, these older programs run in Rosetta, being non-Intel native, and so my MacPro runs them at the same speed as my old G4 dual 800. I'm looking forward to a speed boost when my Xmas upgrade to CS3 is installed.
Francine
Francine
Schwieder

RGB to CMYK to commercial printer

We are using CS2 InDesign and Photoshop 7 on a PC. I am putting together two
large books. One insert section of the first book has color photos. I have
to prepare these color photos for a printer located in China. They want all
images to be CMYK .tiff.
I have an assortment of about 100 images from various sources around the
world, there are .tif, .jpg. .bmp, and .psd. All of them appear to be RGB.
There are images with: 1) no color profile, 2) sRGB IE60966-2.1, and Adobe
RGB 1998. Do all images need to have the same RGB color profile before
converting to CMYK?
I have been converting all of these to .tif (and using a single dpi/pixel
resolution for
all images).
Do I just select/save as color mode CMYK? Or are there special
settings/profiles I need to know about. Is there anything special for China?
The printer's English is not that clear . I was hoping that someone in this
group would be familiar with RGB to CMYK conversion.
Also, it appears that the final PDF pages (with embedded Tiff images) will
be very large. The first complete book for this project (there are two
books) looks like it will be about 20GB. The second book looks like it could
be 12GB. Both books together equal about 1,000 pages and have about 1,000
images. 90% of the images are grayscale. What are the current technologies
for sending large projects like this to a printer?
Thanks
ps, the Photoshop group suggested that I write to this group.

Tech....
In the conversions, are you referring to conversions of B&W or color images? I'll assume color for the moment.
Assuming your monitor is properly calibrated, and assuming you are viewing your color images on your monitor, your conversions should keep "in gamut" colors the same. You will, however, probably see some loss of dynamic range, color, and saturation as you enter CMYK. If the color and dynamic range of the original is all within the destination CMYK gamut, you should see very little, if any, shift. The sad fact is that CMYK on press has limited dynamic range and color, especially compared to a bright monitor. This is generally true, though CMYK can print some colors that lie outside the typical monitor's gamut.
If you have a brightly colored original in RGB mode, then you should experiment with both perceptual and relative colorimetric rendering to see which one looks best. This is an image by image thing. Relative colorimetric will reproduce all "in gamut colors" as accurately as possible, then will take out of gamut colors and bring them to the closest printable color. This can cause loss of detail and 'piling up of colors' as you near the boundaries of the image's color gamut in the destination CMYK space. Perceptual scales ALL colors and preserves the relationships between colors, which sacrifices color accuracy and saturation, but sometimes looks more natural, especially if there are a LOT of out of gamut colors. Pick whichever ones looks best overall.
Then, once in CMYK, make minor curves adjustments to tweak if for CMYK. Just be careful not to push it too far, especially in the deep shadows, since you may push beyond the ink limit of the press. For example, if you convert to US Web Coated SWOP v2, the ink limits are set at 300 total. If you drastically darken the shadow areas, you could end up with an ink limit of 320, 330, 350, etc. Knowing the ink limit of the press will help you stay within those limits. Final tweaking in CMYK is usually a good thing.
How did you determine that the images were dull with too much magenta? Did you print a proof on a calibrated printer, or is this just your screen display? Or is this from a printed image off the press? I have a well calibrated monitor and accurate custom profiles for my inkjet. If I am sending a job to a press that supposedly prints to US Web Coated SWOP, I can proof that file on my inkjet and get a very good match. We need more information to know exactly what you are doing and how you are assessing your images.
Normally, I do the conversion in Photoshop, using the rendering intent that looks best. Then I tweak the image in CMYK to get it looking its best. I usually have the press profile, but sometimes work with a standard profile if no custom profile is available. I leave the image tagged with my conversion profile. Then, I usually place these images into InDesign, with the profile intact (color management turned on in InDesign). I right click on the placed image to be sure that the profile and rendering intent are set to what I want. Then, I can either supply the InDesign file to the printer, or convert to PDF (leave color unchanged) and make sure that all profiles are included. InDesign should export each image to the PDF properly, along with profile and intent. I have never sent to job to press using Photoshop...only InDesign or Illustrator.
BTW, rendering intent is only used when getting an image from one color space to another, such as a conversion from RGB to CMYK. You do that, choosing the one that looks the best. Once it is in the destination CMYK space, rendering intent is no longer needed, since all the colors and tones have already been remapped into the final space. Rendering intent is just used to help us handle those out of gamut colors.
Hope this helps.
Lou

Colors in print preview not matching colors in soft proofing

Hi There,
Just wanted to print a new photo and realized that the colors in print preview do not match the colors in soft proofing. In both cases I selected the same icc profile and rendering method. The print colors matched the colors in print preview. I never had a problem so far. All new prints will be checked with soft proofing and adjusted when necessary. I never paid attention to the color rendition in print preview and all prints perfectly matched the colors from the soft proofing. I was surprised when my print came out of the printer and the colors weren't matching the soft proofing colors, but that of the print preview.
I don't understand why Photoshop renders the colors differently in the first place. Please see attached screenshot for the difference in the blue/cyan colors.
I would appreciate if anybody could point me in the right direction in what is causing this difference. I don't care if the print view colors will match the print, but I do care when soft proofing is not working.
Thank you.
Best regards,
D.

Here are some addtional details:
PS 13.1.2
Mac OS X 10.8.4
12 GB Ram
60 GB free disk space
I printed the same photo from two other computers (MacBook and iMac) with different PS versions (CS4 and CS5). The prints turned out identical to the first one which matches the print preview color rendition on my main computer (MacPro) running CS6. Strangely the colors in print preview of CS5 on the iMac renders the colors identical to the soft proofing colors.

Profiling JavaFX rendering

Similar Messages

Maybe you are looking for