Cut down execution time

So I need to take the execution time of the attached VI down Greatly.  it sits at about 10-15 seconds and i need it to be more like 500ms.  the issue seems to be with extracting the registers from the modbus return and then formatting them to look correct on the VI.  I Have some ideas on cutting this down but they have proved nominal at best.  I could really use some suggestions on this.  it me last major piece on this project.
thanks,
Mark R. 
Attachments:
Yaskawa data collection.vi ‏83 KB

You haven't supplied all the SubVIs... But I have some suggestions:
I notice that the "MB Ethernet Master Query.vi" SubVI is re-used several times for each of your Drives.  If the VI isn't re-entrant then each call will block between drives.
It can take time to open a connection. It looks like this VI is part of a larger application - you are better off keeping the connection open up-front and only performing the reads when needed.
The register conversion logic won't be slowing you down; this will execute very quickly. It is more lilkely that the modbus reads are where time is being spent. But again, without the rest of the SubVIs it is hard to give any further guidance.
It is hard to tell without the SubVI but it looks like you are using a time-out of 10000 seconds for each modbus read. Are you sure one of your requests isn't timing out and adding 10s to your execution time?

Similar Messages

  • Cut down on Rendering time

    Hello,
    I am looking for any tips on how to cut down on rendering time. I usually do weddings and it seems, for the ceremony portion, I have to render each cut with each transition. These are just cross dissolve transitions. I am using 3 camera's, so the footage from each one is on V1, V2 & V3. If I do a transition involving the footage on V1...that particular cut always has to be rendered...taking up time...Is there a way to get around this? Thanks in advance.

    That's because there isn't one. v1 did not have much RT capabilities. We'd need to upgrade to a newer version for more productivity. The current version with LiveType is great for event producers.

  • Aperture 3 shuts down every time I try to use it after only using it for one minute. Aperture closes then a pop up screen asks if I want to reopen Aperture. If somrone at Aperture didn't keep pushing a stop button for it to close,it wouldn't need to ask t

    Aperture 3 shuts down every time I try to use it after only using it for one minute. Aperture closes then a pop up screen asks if I want to reopen Aperture. If somrone at Aperture didn't keep pushing a stop button for it to close,Apple wouldn't need to ask the stupid question"Do you want to reopen Aperture?" to begin with. Anyone out there having the same annoying problem?? I like Aperture or used to anyway,but if this nonsense continues, I will be forced to buy some other editor besides "No hanging fruit - Adobe photoshop CS5" or "I don't like you so I'm shutting down - Aperture".

    Hello,
    sorry to hear about your troubles, but we will need to have more technical information about your problem to be able to help you:
    There may be several reasons that can cause Aperture to crash on start up:
    You may have imported an image (or video) in an encoding that Aperture does not support, or that is larger than 2 GB, or that is corrupted and Aperture cannot process it
    Your Aperture Library or Preferences may have been corrupted
    Or you are having a system wide problem with your machine.
    Or you may simple be running out of disk space.
    It would help, if you could post the crash log from your last crash. Also, more details on your Aperture Library: Is it managed or references? how large is it? Is it on your system drive or on an external drive?
    What have you done, shortly before the trouble started? Have you imported new images or videos? Did you install any plug-ins or any new software, lile Final Cut?
    Try, if you can get Aperture to launch, if you defer the generation of Previews: Hold down the SHIFT-key immediately after you start Aperture to prevent preview generation for that session. If a damaged image in your library is preventing normal opening, this may allow you to start Aperture. If this succeeds, try to identify the broken image or video among the images you recently imported and remove it from the library.
    Regards
    Léonie
    Don't you think your user name is a little bit tactless in an Apple provided forum?

  • I need to edit out parts of a song to cut it down to two minutes for my daughter's talent show. I went to info and it will cut down one part but only one part. I need to put three parts together. I tried adding duplicates of the song to the list but if I

    I need to edit out parts of a song to cut it down to two minutes for my daughter's talent show. I went to info/start time and it will cut down one part but only one part. I need to put three parts together. I tried adding duplicates of the song to the list but if If I change the start time on one it will change it on all of them.

    In your library, right-click the song you want to edit.
    select "get info".
    go to "options" and select the start and stop times for your first section of the song.
    click "OK".
    Find the shortened version of the song in your library. It may take a minute for it to show up. right click it again and select "create (????) version". (the ???? is different for different formats).
    The library will spit out a new, second version of the song which you can rename. I suggest you use the original title and add a 1 to the end.
    Now you can go back to the other version and repeat the process with a different  start/stop time.
    Once you are done editing, you can burn all of your versions to a disk, just make sure your interval time is zero so there are no gaps between edits.
    To keep the orginal song on your library, just go back to the original and put the start/stop times back to the original settings.

  • In preview all is good...burning dvd the music cuts out 2 times in 4 songs...i have turned off all other programs, sleep and energy mode for battery...multiple tries... same result ...help

    IDVD - in preview all is good...burning dvd the music cuts out 2 times in 4 songs...i have turned off all other programs, sleep and energy mode for battery...multiple tries... same result ...help

    Hi
    No audio on DVD disc.
    Most common origin.
    1. Imported audio from iTunes.
    • In iTunes copy out as an audio-CD .aiff
    • Use this in Your movie project
    2. Low - Free Space on Start-up/Boot/Internal/Mac OS Hard disk makes it hard for iMovie to work as intended.
    Down to 1.3 GB and it doesn’t work - especially audio don’t migrate over to Media Browser
    (iM’08 & 09)
    3. Material in iMovie’08 & 09 - Shared to Media Browser and here selected as Large
    large.m4v. It silenced out this project in iDVD. By making a slight alteration - provocing it to ask for a new Share/Publish to Media Browser and here selecting 640x480 - and audio was back when a new iDVD project was created including this movie.
    176x144
    480x360
    640x480
    720x540    960x540
    4. Strange audio formats like .mp3, .avi etc.
    • Change them to .aiff. Use an audio editor like Audacity (free)
    5. Main audio is set to off in System Preferences - Does this by it self - Don’t know why
    Cheque Audio-Out resp. Audio-In
    6. Ed Hanna
    Had the same problem; some Googling around gave me a kludgy, but effective fix
    Downgrade Perian from recent versions back to version 1.0.
    That worked for me, and so far I haven't encountered any deficiencies—except it takes more advanced versions of Perian to enable QuickTime to handle subtitles in .srt format—that I have noticed.
    7. GarageBand fix.
    In this set audio to 44.1 kHz (Klaus1 suggestion)
    (if this don’t work try 48 kHz - me guessing)
    Before burning the DVD.
    • Do a DiskImage (File menu and down)
    • Double click on the .img file
    • Test it with Apple DVD-player
    If it’s OK then make Your DVD.
    Burn at x1 speed.... (or x4)
    • In iDVD 08 - or - 09
    • Burn from DiskImage with Apple’s Disk Utilities application
    • OR burn with Roxio Toast™ if You got it
    Yours Bengt W

  • Elements 9 ACR 6.5 is a cut down program

    I've used ACR with CS4 before to process my RAW files but recently bought Lightroom 3.5 and Elements 9. I've upgraded both programs to ACR 6.5 to support my camera and noticed for the first time that the Elements plugin is a cut down version of ACR 6.5 with many of its features missing. If ACR 6.5 is a free download and available from thrid party websites why isn't the full program available when opening a RAW file withing Elements 9? I've tried downloading it from Adobe and using it as a standalone probram but when opening it up I'm told that it isn't appropriate.
    George

    From what I’ve seen, while many adjustment tabs and sliders are not in PSE-hosted ACR, the underlying RAW converter is still the same, so if you have made adjustments in LR and save into an XMP then open in PSE, or just use Edit In Photoshop Elements from LR, the ACR that opens the file on the way in to PSE will apply all the adjustments that were made in LR.  And so since you can make all the adjustments in LR there’s no need to make anymore adjustments in the ACR on the way into PSE is there?
    Is your experience different than this, where if you do something in LR that doesn’t have a corresponding slider in the PSE-hosted ACR, it doesn’t get applied.  For example, if you make a manual lens-correction for perspective that is easy to tell whether it happened or not, in LR and then do a Edit in Photoshop Elements from LR, does the perspective-correction not happen?
    The only thing that should be different is that you have to be in LR to make all the adjustments, not use PSE-hosted ACR to make them.

  • Execution time of a flat-sequence

    Hello there -
    Is there any way to get a measurement of how long each part of
    the flat sequence takes to execute?  Anything like matlab's "tic" and "toc"
    commands in labview?  I have been playing with it for a while now and
    have yet to discover if Labview has this functionality.  Anyone know of
    anything like this?
    I currently have a VI that controls the realtime acquisition of a CCD camera via Firewire and a USB spectrometer.  The VI collects data from each of these devices (triggered by an external source at 10Hz), and dumps them into a Matlab script which does analysis on the CCD image and spectrum.  The bulk of the VI sits inside a while loop, which continues to run until the user presses the stop button.  Inside this main loop is a flat-sequence.  The sequence goes:    ACQUIRE DATA --->  PROCESSING DATA ---->  MATLAB SCRIPT ----> PLOTTING GRAPHS -----> OUTPUT DATA TO FILE.   
    The problem here is that the VI runs at 5Hz, while we are triggering it at 10Hz.  Originally, it was my thought thought that the matlab algorithm was to blame, but I used the matlab commands "tic" and "toc" to determine that the matlab algorithm runs in 15-20ms.  I did this by putting a "tic" command at the top of the matlab algorithm and a "toc" command at the bottom.  The problem, as I have now discovered is that the rest of the labview code takes ~180ms to execute.  (This was discovered by putting the "tic" at the bottom of the program, and the "toc" at the top of the program, thereby measuring the execution time of everything except the matlab algorithm).  Each time a trigger signal from the external source comes in, it starts the flat-sequence structure (which takes ~190ms), and then waits for another trigger signal, always missing every second signal.  My eventual goal is to reduce the bloat, and get the algorithm down to less than 100ms, so that I can run the VI and acquire data at 10Hz rather than 5Hz.  If anyone can offer some help with this, it would be much appreciated!
    Eric
    P.S. - I have attached a copy of the VI that I am working on, but unfortunately, it most likely will not run on your computer....the VI will not run unless it is connected to a triggered spectrometer and CCD camera....but I have attached it anyways incase anyone who can help might want to take a look.
    Attachments:
    RTSpider.vi ‏376 KB

    can we divide the program into 2 parts and use background process for acquisition and front end process for analysis?
    I mean, create 2 VIs from the present VI and then launch the acquisition program dynamically as a background process and fire events in Main VI from acquisition VI and process it.  not sure how much it is going to reduce. lets give a try....
    Anil Punnam
    CLD
    LV 2012, TestStand 4.2..........

  • Execution time issues with SU01 demo script

    Having worked with Scripting in a Box for a while now I wanted to try out the examples there. I read FM: SO_USER_LIST_READ or another one explaining why my attempt to narrow the returned users failed (Craig, did you find out why the functionality was removed?) and Re: Issue with "Scripting in a Box" seeing that  Harry had the same problems with only ~200 users in his system. However, Craig's original post states he successfully managed with over 400 users. I'm a bit confused...
    I included some simple timing stuff and found out that processing of one user in the loop takes about 1.7 seconds - little surprise then that every script times out. This seems to be due to the additional calls to GetStatus() and GetValid() - by commenting them out I get the whole list rather quickly.
    Unfortunately commenting them out also means no nice icons for 'Status' and 'Valid', which is not desired. I probably could create a Z FM to deliver me the userlist with these two fields already added (which would save on rfc-calls, assuming the operation is much quicker on the server directly), but I hoped to get a solution based purely on PHP, not own ABAP coding (being aware that Craig also used a Z FM anyway, but still...)
    I'm a bit unsure now how easy it is to actually create useful frontends in PHP, with such long execution times. I assume this will happen in other occasions as well, not only for user lists. Is there an alternative? Or a general way to do those things quicker?
    :Frederic:

    Craig: you say it's easy to go from 1.7 seconds per user lookup down to a small fraction of it? Then apparently I'm lacking these skills. Could you please give me a hint what should be done there?
    I though about creating a Z function, but having to write custom wrappers - possibly for about any transaction in this style - I wanted to avoid.
    Bala: the two functions only take one user as input, not a list. So w/o modifying the ABAP side I can't feed the whole list in there. I wonder how much it would improve the result time anyway, so perhaps I'm trying it. It's just not a solution I'd prefer.
    Paging is a good idea, the actual call to get the whole userlist is quite quick. Having like 20 users displayed at a time is manageable - still slow, but the script won't timeout anymore. I think I'll implement this today.
    About AJAX: yes, I want to play around a bit with AJAX, however having the two columns valid and status removed and the information only displayed on mouseover etc is a bit like cheating. And 1.7+ seconds waiting for a hoover info is too long. So I'd like to optimize on the rfc-calling side primarily.
    Craig: surely it was just a demo, and I'm just trying to get it to work for understanding it
    :Frederic:

  • To reduce execution time of a Business Objects Dataservices job.

    The  issue that we are facing-
    Our goal-  To compare a record from a file  with 422928 records from another table on basis of  name & country, if a match is found then you take some specific columns of that matched record (from the table ) as our ouput.
    What we are doing-  We are at 1st  removing duplicates by doing matching on the address components (i.e.- addr_line1, city, state, postal code & country), here                    the break key for match transform is country & postal_code, its taking 1823.98 secs. Now the record count is 193317
                      Then we are merging the file record along with the 193317 records to put them in the same path and send them for matching.
                      The match criteria is the firm name & iso_country_cd,
                       the break key for match transform is the  iso_country_cd & the 1st  letter  of the name.
                       It took 1155.156 secs.
    We have used the "Run match as seperate process' option for the match to reduce the time.
    The whole job took  3038.805 secs.
    Please suggest how to reduce the execution time.
    Edited by: Susmit Das on Mar 29, 2010 7:41 AM

    This really is impossible to help with without seeing your code.
    Replacing while loops with Timed Loops will not help. Timed Loops are used for slowing while loops down in a controlled manner. You would use them to synchronise with a clock rate, and can set other parameters for priority and CPU allocation etc. These will not likely help to reduce your execution time.
    If you are seeing code execution of 1 second then you will need to optimise your code to reduce the execution time. There are LabVIEW guides on how to improve your code execution time, just search around for them.
    Also try using the Profiling tools to learn which VIs (I presume your code is componentised and each while loop contains subVIs?) are hogging the most CPU time.
    If you cannot share your VI then there it is very hard to know where your code execution bottlenecks are.
    Thoric (CLA, CLED, CTD and LabVIEW Champion)

  • Execution time is too high.

    Hi,
    I got several xsql-files that uses aprox. 1 minute to execute by them selves. I want to run one xsql that includes them all, but it times out, and I get an server-error. Is there any way I can get the execution time down, or change the timeout-setting?
    I am running the xsql that comes with 8.1.7
    Terje K.

    If Oracle8i JServer is included in the Oracle 8i package, then yes. The database itself is not large (aprox. 50Mb with data), but the results of the queries can get somewhat large. Here is an example:
    1. first I made a view:
    create view view_section2_issue as SELECT SPVSPC.OPN, OPE, PCA, PCS, PCSR, OSC, PWTT, OVSM, PLWD, PCSOD, PDC, TO_CHAR(ISOD,'dd.MM.YY') AS ISOD, TO_CHAR(PCSOD,'dd.MM.YY') AS OD, PSCN, PMDP1, PMDP2, PMDP3, PMDP4, PMDP5, PMDP6, PMDP7, PMDP8, PMDP9, PMDP10, PMDP11, PMDP12,PDT1, PDT2, PDT3, PDT4, PDT5, PDT6, PDT7, PDT8, PDT9, PDT10, PDT11, PDT12, PMDC, PMCA, PMMDP, PMDT, PMLWD, PMWTT, PMSCN, PMNS, PMWTH, PMSCH, PMOD
    from SPVSISSU, SPVSISS2, SPVSPCS2, SPVSPC
    where SPVSISSU.OPN = SPVSPC.OPN
    and SPVSISSU.ISS is not null
    and SPVSISS2.OPN = SPVSISSU.OPN
    and SPVSISS2.ISSUE = SPVSISSU.ISR2
    and SPVSPCS2.OPCS = SPVSISS2.IOPCS
    and SPVSPCS2.PCSR = SPVSISS2.IPCS_REV
    then I made the query (with some cursors):
    SELECT OPE, PCA, PCS, PCSR, OSC, PWTT, OVSM, PLWD, PCSOD, PDC, OD, PSCN, PMDP1, PMDP2, PMDP3, PMDP4, PMDP5, PMDP6, PMDP7, PMDP8, PMDP9, PMDP10, PMDP11, PMDP12, PDT1, PDT2, PDT3, PDT4, PDT5, PDT6, PDT7, PDT8, PDT9, PDT10, PDT11, PDT12, PMDC, PMCA, PMMDP, PMDT, PMLWD, PMWTT, PMSCN, PMNS, PMWTH, PMSCH, PMOD,
    CURSOR( SELECT PNS, POD, PWTH, PSCH
    FROM spvspcs4
    WHERE spvspcs4.opn = view_section2_issue.opn
    and spvspcs4.pcs = view_section2_issue.pcs
    and spvspcs4.pcsr = view_section2_issue.pcsr ) as wallThickness,
    CURSOR( SELECT PELM, SST, PDSTD, PFS, PTS, PTY, PMN, MDS, ESK, PRM, PAGEBREAK, PMELL, page, start_remark(opn,pcs,pcsr,pel,pell) starten,
    end_remark(opn,pcs,pcsr,pel,pell,start_remark(opn,pcs,pcsr,pel,pell)) as slutt
    FROM spvspcs6
    WHERE spvspcs6.opn = view_section2_issue.opn
    and spvspcs6.pcs = view_section2_issue.pcs
    and spvspcs6.pcsr = view_section2_issue.pcsr ) as elements,
    CURSOR( SELECT PVELM, VDS, PVFS, PVTS, PVRM, PMVELL
    FROM spvspcs7
    WHERE spvspcs7.opn = view_section2_issue.opn
    and spvspcs7.pcs = view_section2_issue.pcs
    and spvspcs7.pcsr = view_section2_issue.pcsr ) as vType,
    CURSOR( SELECT PBLP, PAGEBREAK, LTXT
    FROM spvspcs5
    WHERE spvspcs5.opn = view_section2_issue.opn
    and spvspcs5.pcs = view_section2_issue.pcs
    and spvspcs5.pcsr = view_section2_issue.pcsr ) as kommentar,
    CURSOR( SELECT count(*) as tot
    FROM spvspcs5
    WHERE pagebreak = 'P'
    and spvspcs5.opn = view_section2_issue.opn
    and spvspcs5.pcs = view_section2_issue.pcs
    and spvspcs5.pcsr = view_section2_issue.pcsr ) as kpages,
    CURSOR( SELECT count(*) as tot
    FROM spvspcs6
    WHERE pagebreak = 'P'
    and spvspcs6.opn = view_section2_issue.opn
    and spvspcs6.pcs = view_section2_issue.pcs
    and spvspcs6.pcsr = view_section2_issue.pcsr ) as tpages
    from view_section2_issue
    where OPN = {@opn}

  • Oracle NoSQL YCSB - continuously increasing execution time

    Greetings,
    currently I am testing several nosql databases using YCSB. I am new to that type of databases, but I have already tested few of them. I am using VM with 2GB RAM and hosted on Win 7. Even though it is not recommended, since I am working in now capacity environment, I am using KVlite. But my problem is confusing and I can not find the reason. So, I have successfully loaded data and tested Oracle NoSQL using different workloads. However, with each execution, I get higher execution time. For example, if at 1st execution I get 20 seconds, if I shut database down and next day execute same workload again, I get 35 second execution time and so on.
    Do you have any idea of what may be causing that? Like I said, I have been researching some nosql databases but I have never had that strange results.
    Regards.

    To add to Robert's comment, the NoSQL DB documentation on KVLite states the following:
         KVLite is a simplified version of Oracle NoSQL Database. It provides a single-node store
         that is not replicated. It runs in a single process without requiring any administrative interface.
         You configure, start, and stop KVLite using a command line interface.
         KVLite is intended for use by application developers who need to unit test their Oracle NoSQL
         Database application. It is not intended for production deployment, or for performance measurements.
    Per the documentation, you can use KVLite to test out the API calls in your performance benchmarking application, but you should not use to it perform the actual performance testing. For performance testing, please install and use the Oracle NoSQL Database server.

  • Execution time increase

    hai  guys,
    Im currently developing a code that run 24 hours 7 days a week. My problem is the execution time increases as days pass by. As im tracing down the execution time ,the 1st day i start the to run the program execution time is 0.04 second but after 3 day it increases to 4.5 seconds .
    then i indually tracked down the subvi that causing problem and find out that the problem is on writing to ini file.
    The subvi function is to REPLACE the data (20 data )in a file every 10 second*not accumulating data just replace the old data  . I open file replace data and close file every itteration and as day pass the execution time for this subvi increases.
    Can any one help me to get a solution for the problem as time is a very important factor in my programming. the execution time is limited for less that 10 second.
    thanks
    regards
    kokula 

    As David said, post the code.
    Is there any reason why your are loggin to an .ini file?  Does that mean you are using the LabVIEW config file functions?
    I don't know if deallocating memory would help, or is even possible, without seeing your code.  You either have a problem with resources continuously being created, but not being closed out.  Or you have an ever growing array.  This takes more memory over time, and slows down the code because LabVIEW has to move memory around to account for the ever larger array.
    I bet if you let your program run long enough, it would eventually crash due to running out of memory.
    By the way 47MB to 350 MB of memory consumption means the same thing as RAM usage increasing.

  • LabVIEW2012 has long execution time to write string than LabVIEW2011 on XP SP3

    in for loop, each execution will output string and display it on front panel by value property.
    But after upgrade to LabVIEW 2012, the execution time doubled on Chinese XP OS, I really want to know this difference between LabVIEW 2011 & LabVIEW 2012, pls advice.
    "I think therefore I am"

    This is no way to measure the execution time. Your parallel while loop spins as fast as the computer allows, consuming all CPU it can get in the process, and starving everything else. Your code is highly flawed!
    Eli: actually I add 10milliseconds delay inside both for and while loop. 
    In addition, get date/time in seconds is also a relatively expensive function, not to mention all these local variables! Your benchmark is completely meaningless! All it does is slow down your regular code, nothing else. How do you know in what order things start? Where is the "start time" initilized (where is the terminal?!)
    Eli: I am using the states machine, and set the "start time" during the program initialization, then calculate "test time"in next case (running block). 
    A proper benchmark uses a three-frame flat sequence with to high resolution relative seconds timers, on in each outer frame and the code to be benchmarked in the inner frame. (here is an example). The difference between the two timers is the exection time of the inner code in seconds. Make sure to only have wires in the benchmarking code. All controls and indicators belong outside the sequence. Also make sure that no other code can run in parallel to the sequence.
    Use this and you'll be surprise how fast your code actually runs.
    Then you should also disabele debugging.
    Please attach your actual code (including the subVI) so we can see what else you are doing wrong.
    Eli: if possible, give me your mail address, then I can sent more actual code to you separately. 
    "I think therefore I am"

  • CVI XML Functions Execution Times Increase When Looped

    I have written multiple functions using CVI that read XML files. I have confirmed in the Resource Tracking utility that i have cleaned up all of my lists, elements, documents, etc. I have found that when I loop any of the functions I have created, the execution times increase. The increase is small but it is noticable and does effect my execution.
    Are there any other sources of memory that I need to deallocate? It seems that there is a memory leak somewhere but I am unable to see where this increase is located.
    I am currently running LabWIndows/CVI 2009 on Windows 2008 Server. I have looped my functions using TestStand 4.2.1. Any help would be appreciated!
    Thanks in advance,
    Kyle
    Solved!
    Go to Solution.

    HI Daniel,
    Thanks for the quick response.
    It is indeed slow down in execution speed when we loop. When looped, the XML reader is overwriting variables, not adding to an array. Our application is structured differently than my test case. We run a CVI function from TestStand that contains a series of commands, which contains the XML reading. The XML looping is really done in CVI. I used TestStand in my test case just to get execution times. Our psuedocode for the CVI function is as followed:
    For loop (looping over values, like amplitude or frequency)
    Reading the XML
    Applying the data from the XML to set up some instrument(s)
    Do something...
    End loop
    I can confirm that the instrument set up is not the cause of the slow down. We have written the same XML reading in C# and applied the values to the instrument setup and do not experience the slow down.
    I tested with On-The-Fly Reporting enabled and the execution time continued to slow down.
    I hope that answers all of your questions!
    Thanks,
    Kyle

  • Java needs a cut down "media" VM

    What would it take for java to take on flash in the internet media space?
    My thoughts on why flash dominates java:
    The java download is far too big, java is ~7MB, flash is ~1.8MB
    The vm takes far too long to startup
    You get an ugly grey box when the vm is starting up in place of the applet
    What is needed is a cut down java "media" vm. It should:
    Be no larger than the flash download of ~1.8MB
    Have greatly improved startup time. On a modern pc it should be almost instantaneous.
    Be fully upgradable to the "full" vm. For example if the full vm is ~7MB then if you have the media vm installed the full vm download should be ~1.8MB smaller
    As the name implies it should be focused solely on rich media (steaming movies etc)
    It should not cause the java icon to popup on the taskbar by default. It should be transparent to the user that java is running.
    Should not display a grey box in place of the applet when starting up, the box should be black or non-existent.
    The aim would be for sites like youtube to be able to realistically replace flash with java. What do you think?

    >
    Not true, you should probably read a few tutorials. You can already replace that with an image or an animated image.Well I have not written any applets for a long time it is just what I am seeing out there. I have been developing java for over 10 years so I don't need to read any tutorials.
    This is a large development space and java should be taking it over. JVM modularity is the first step in the same direction.

Maybe you are looking for