ECO versus EJB versus 8I

Hi,
I want to compare ECO upon OAS, EJB upon OAS and EJB in Oracle 8i (8.1.5).
For that, I wrote a little component which has 3 methods:
- send a string of 50 000 caracters
- open an Oracle Session via JDBC
- read 2 000 lines in the table travel (Hotel example)
The client is a Java application connected to the server via 16Mb.
Oas and O8i are on the same server (NT 4)
The results:
ECO sendString 206 ms openSession 224 ms readTravels 1154 ms
EJBOAS sendString 306 ms openSession 329 ms readTravels 3567 ms
EJB8i sendString 918 ms openSession 157 ms readTravels 3560 ms
The same but in Client/server 2 levels
Local sendString 78 ms openSession 562 ms readTravels 1485 ms
Conclusions?
ECO better than EJB (RMI is the reason perhaps?)
EJB8i not optimal even in reading databases?
Question for Oracle Team:
Now,I start a project built upon OAS and Oracle 8i and Java. EJB in OAS and 8i doesn't seem very mature, isn't? Have you benchmarks for comparaison beetwen ECO and EJB ? Can I expect better performances in next release? When ?
Thanks.

The double quotation marks strictly unnecessary ... unless you are dealing with objects that don't adhere to the standard naming convention.
So, if your table is called MYEMP then these three statements are the same...
SELECT * FROM myemp
SELECT * FROM MYEMP
SELECT * FROM "MYEMP"
/ But if someone has been ill-mannered enough to name the table using lower or CamelCase (which inself neccessitates the use of " in the CREATE statement) we must also reference it using double quotes...
SELECT * FROM "MyEmp"
/ cheers, APC

Similar Messages

  • EJB versus servlet

    Hello everyone,
    I'm hoping you can point me in the right direction. I'm looking for
    articles or books that cover the pros and cons of implementing solutions
    using EJBs and/or servlets. More specifically, when and where it makes the
    most sense to use EJBs or servlets or both, and how the two compare. Also,
    how well regarded are EJBs in the marketplace (seems to be a love/hate
    relationship).
    I've read some of Jason Hunter and Richard Monson-Haefel's older material
    but I'm hoping you might know of (or have) something more recent.
    Thanks in advance!
    Joe

              Of course the best article is the one that Xiang suggested, J2EE blueprints.
              If you want something quickly to read, I enjoyed this one very much:
              "How to Build a Really Big Distributed J2EE System Using Tools You Have Around
              the Office"
              http://www.javasuccess.com/design_wp.html
              And of course this one:
              "J2EE Design Considerations for Weblogic Server"
              http://www.bea.com/products/weblogic/server/papers.shtml
              Regards,
              David
              "Joe Martins" <[email protected]> wrote:
              >Hello everyone,
              >
              >I'm hoping you can point me in the right direction. I'm looking for
              >articles or books that cover the pros and cons of implementing solutions
              >using EJBs and/or servlets. More specifically, when and where it makes
              >the
              >most sense to use EJBs or servlets or both, and how the two compare.
              > Also,
              >how well regarded are EJBs in the marketplace (seems to be a love/hate
              >relationship).
              >
              >I've read some of Jason Hunter and Richard Monson-Haefel's older material
              >but I'm hoping you might know of (or have) something more recent.
              >
              >Thanks in advance!
              >
              >Joe
              >
              >
              

  • EJB versus Hibernate

    Hi Everyone,
    I heard that ejb is not the final solution for all your database opertions in all cases. Sometimes, you need to use JDo or Hibernate.
    Can anyone compare theses three technologies and the predict future market for them please.
    Thanks and regards,
    Girtish Varde

    Thanks for your prompt reply and sorry for delay in me replying.
    I really appreciate your prompt reply.
    I tried to Google for Hiobernate, but not fortunate enough to get hard core extensive comparision and actual strengths and weakness areas of the three. I was confused because of the opposite views.
    Actually, i am in a hurry and i have to choose among ejb and hibernate.
    One company which is offering me is large and using ejb, while the other is relatively young but uses hiberbate and strongly canvasses it, We are having discussion on chat / mail, but i can not take that as final word, it might be motivated.
    Do you really feel hibernate has bright future and i should choose it against ejb?
    Does it provide all other features such as messege driven bean, session beans etc. in ejb or just for persistance?
    I need guidance from experienced people like you.
    Can me send me some links or referrences please.
    I will continue my digging in the meantime.
    Still, thanks and regards
    Girish Varde

  • SCN versus EJB 3 @Version for optimistic locking

    Does anyone have an opinion on using the standard EJB3 @Version annotation and column compared to using the Oracle SCN?
    Could there advantages in using the SCN over an explicit version column?
    Is there anyway to use the SCN as the version column with EJB3?
    Are they mutually exclusive solutions to the problem?
    Any ideas?
    Thanks
    Michael

    Does anyone have an opinion on using the standard
    EJB3 @Version annotation and column compared to using
    the Oracle SCN?
    Could there advantages in using the SCN over an
    explicit version column?
    Is there anyway to use the SCN as the version column
    with EJB3?
    Are they mutually exclusive solutions to the
    problem?
    Any ideas?
    Thanks
    MichaelI have a similar question.
    1) Has anyone answered the prior question?
    2) I had a Java EJB3 Developer tried to use ora_rowscn in Hibernate, but he said it failed, and his reason was this ora_rowscn column must be write/update-able and that is not possible. I am not sure if it is Hibernate limitation or its configuration.
    3) Can Oracle's TopLink product (the licensed and/or free version) utilize ora_rowscn column for version, thereby avoiding the need for an extra column for version ?
    Any other way to avoid using an additional version column for supporting EJB3 / JPA applications?
    Thanks.
    Mitesh

  • 120 versus 240hz versus HD and other technical questions

    We do not and will not TV cable. But we do watch movies via DVD. We do want to upgrade to Blue Ray. LED / 3D are out of price range so it's going to be an LCD.We already know we want 1080.
    Can anyone tell me what I should be looking for in a new TV? There a lots of technical data.
    What are the benefits for 120 vs 240 vs 600Hz? Do we need speed for movies?Plasmas have faster refresh rate but they are a dying breed it appears. Shouldwe consider a plasma?
    We have a box for the Wii connection. Do we need more than 2 HDMI connections?
    Thanks for any help you can provide.
    Dan
    Solved!
    Go to Solution.

    Basically - Most DVD/Blu-Ray content is 24 frames per second.  The only benefit of 120 Hz over 60 for non-3D is that 24 divides evenly into 120, but no 60, so can cause slight motion judder artifacts when displaying 24 fps content on a 60 Hz display.  Most people won't notice this even in a side-by-side test.  3D basically doubles framerate requirements - so slight benefit of 240 over 120 for 3D (48 doesn't divide evenly into 120), but no benefit of 240 over 120 for non-3D.
    There is no content in existence with framerates over 60 Hz.
    *disclaimer* I am not now, nor have I ever been, an employee of Best Buy, Geek Squad, nor of any of their affiliate, parent, or subsidiary companies.

  • IPhone versus iTouch versus G1

    I really love my G1 but am going back and forth as to whether to cancel my contract with T-Mobile, take the early termination hit but can make it back easily by selling the G1 online. Then I can pickup an iPhone and have an all-in-one product.
    I currently pay about $75 month for the G1, this includes all services and data fees (I never text or IM). As I said, I don't make many calls at all. I use it mainly to surf, listen to music and screw around with the apps. Any idea what a plan may costs with the iPhone? And, more importantly, how fast is their G1 network?I have wifi at my house and also live in NY so my service provider, Optimum, provides free wifi whenever and wherever it is available, so using the wifi service with the iTouch is a no brainer.
    Besides the lack of the phone capability, how many apps are available (not games), how fast does it surf the Net? I bought the G1 for a few reasons - my wife already was a t Mobile subscriber and for the G3 speed when surfng the Net.
    The major problem with the G1 is that is does not synch with the USB port in my new car. I have a 2009 Civic EX and any of the newer ipods can easily sync with the USB port and I have tons of tunes, books, podcasts that I can listen to.
    Sorry for the long post - thanks in advance,
    slider

    this is the iPod touch forum, you may get a better response re: tariff prices in the iphone forum.
    at the moment you can have 9 pages of apps (16 apps per page, plus 4 'main' ones which are at the bottom all the time). probably worth bearing in mind that about 15 of these are the pre-installed apps, and as of yet you can't delete these.
    if you meant how many apps are available, i have no idea, thousands and growing all the time!
    i'm not sure about the speed, i think it depends on your network.

  • Pci versus pxi versus compactRIO

    Is it true the only advantage to move from a pc platform to PXI or CompactRIO is the deterministic of an acquistion?  Currently am using PCI7332 and read the encoder position and analog input all of which takes about 4ms for each iteration.  If I move to a PXI or compact RIO I don't get any advantages in speed?  I would if only the processor is faster than what I'm currently using?
    Thanks for your time......
    dg

    DG, 
    It will really depend on what you are trying to accomplish with your application. If you are just performing simple contour moves or anything general, then you would probably want to stick with your current set up. Yes, the main advantage is a deterministic acquisition along with other benefits that come with having an embedded system. As far as speed goes, it is generally true that it will come down to the processing speed of your computer for comparison. There won't be a large difference between PCI or PXI, but the cRIO should be the "fastest" of the three. But that term is very relative, and it will mostly come down to what your application is performing. For rugged outdoor environments, specialized I/O configurations, and want to have FPGA programming then the cRIO and Soft Motion would be the better choice. 
    You might find this Developer Zone article to be of some help if you are looking for more detailed information:
    Building an NI Motion Control System
    http://www.ni.com/white-paper/12127/en#toc7
    Huntington W
    National Instruments
    Applications Engineer
    ***Don't forget to give Kudos and Accepted as Solution where it is deserved***

  • EJB deployement: jar versus ear

    WLS 6.1 SP2, NT4
    Hi,
    if I deploy an ejb-jar-1 to be used by a Java client for instance and
    have requirements for a new web client using ejb-jar-1 and a new ejb-jar-2
    six months later,
    what should I do ?
    1/ Deploy a newapp.ear including my war file, my ejb-jar-2 and again
    ejb-jar-1.
    This ejb-jar-1 will be referenced as newapp.ear\ejb-jar-1.jar compared
    to the one previously deployed (ejb-jar-1)
    In this case, how does JNDI work if I lookup "MyBean" especially if both
    ejb-jar-1
    and newapp.ear\ejb-jar-1 have one common JNDI entry. Do have I to
    include a new ejb-jar-1-bis instead and change the
    jndi entry in its descriptor ?
    2/ Deploy a newapp.ear including my war file and only my new ejb-jar-2.
    3/ Deploy a newapp.ear including my war file and only my new ejb-jar-2 but
    referencing/attaching the ejb-jar-1.
    4/ Anything else ?
    Generally speaking, I am a bit confused when comes the deployment of several
    applications
    potentially using common components. Can we have duplicate deployed
    components ? Is there any JNDI issues
    associated with such practice ?
    Could somebody clarify how does it work when comes to deploy single EJBs
    versus applications including EJB ?
    What is the best practice etc ... ?
    Thanks
    Arnaud

    STARTMODE is only available in WLS 6.1 and not WLS 6.0.
    Deepak Vohra wrote:
    Tom
    Edit the startWeblogic Command Script.
    set STARTMODE=false
    Deepak
    Tom Gerber wrote:
    Is STARTMODE available in 6.0??? I thought that was new to 6.1.
    I'm running on NT so do I set this on the Java command line as a
    -DSTARTMODE=false or is it a regular environment variable?
    Tom
    Deepak Vohra wrote:
    Tom
    In startWebLogic script set STATRTMODE=false
    STARTMODE=false starts server in development mode.
    Deepak
    Tom Gerber wrote:
    We are using wl 6.0 sp2 rp2 - with 2 WL instances on the same box.
    We have always seemed to have had this headache with WL - getting EJB's
    to reload consistently.
    Does anyone have a sequence of steps they follow to get good consistent
    reloading results from both an EJB in its own JAR or an EAR file?
    If we just copy the file over sometimes it seems to reload and sometimes
    not.
    If we try to use the console to "install an application" again sometimes
    it deploys the EJB's to the wrong server instance!
    Also is there anyway in the Weblogic deployment descriptor to specify
    which instance the EJB should be deployed against?
    Thanks
    Tom
    Rajesh Mirchandani
    Developer Relations Engineer
    BEA Support

  • SONET versus DWDM

    A little confusion surrounding metro technologies. When should you deploy next-generation SONET versus DWDM versus Ethernet, etc.?
    TIA
    Pam

    The use of one doesn't necessarily rule out the concurrent use of the other (or any of the others).
    DWDM is neutral as to the type of signalling that it can carry; it's just a way of increasing the apparent bandwidth over a given pair of SM fiber. In a high-density metro area, it's likely to be far cheaper to deploy WDM / DWDM than to pull additional media. For an existing plant, the bandwidth multiplication frequently translates to more efficiency and lower per-unit transmission costs for a given span.
    In a large campus environment, using DWDM to "multiply the bandwidth" is usually a faster solution to implement than trenching-in new media or pulling more fiber into the existing raceways.
    WDM / DWDM is just a way to "fatten up" the fiber.
    FWIW
    Scott

  • Why migrate from toplink 10.1.2 to EJB 3 JPA instead of to toplink 10.1.3?

    We're in the situation that we're using Toplink 10.1.2 and are making plans for our next release. I'm wondering wether we should 'just' migrate to the 10.1.3 version of the Toplink API or wether we should migrate to EJB 3.0 JPA API (with Toplink as implementation).
    Advantages and drawbacks I see are the following:
    Toplink API 10.1.3 versus EJB 3 Persistency on Toplink
    + maybe the only way to access Toplink specific functionality: i.e. historic queries. (Is that a fact by the way?)
    - In the long run it will be easier to hire people with knowledge of EJB 3.0 Persistency
    + Maybe smaller effort to migrate. (Is that true? I guess we'd have to add a lot of annotations)
    Are there more drawbacks/advantages?
    Regards,
    Joost de Vries
    Netherlands

    Hasam,
    We do have a uix to JSF migration utility that you can get through the check for update center. The missing part there right now is that it doesn't support uix template migration.
    If you are using jheadstart you might want to post your question on the jheadstart forum as well they might be able to do the migration through their repository.
    And remember that if your application is working now and your customers are happy then you don't have to migrate it.

  • Q: General : When will you use Web Service as opposed to just Servlet and/or EJB?

    Hi,
    I don't know if this is an appropriate question to ask for this class as it's not
    workshop
    specific. I was wondering how will I decide when it is appropriate to use a web
    service as opposed to say servlet or an EJB for my application? What is a good
    rationale to use?
    Thanks,
    Sadhana

    Hi Adam,
    Thanks a lot for giving such a lucid explanation of my vague question!
    I understand now the difference between when to use EJBs/Servlets/Web Services.
    Thanks again!
    Sadhana
    "Adam FitzGerald" <[email protected]> wrote:
    >
    Hi Sadhana,
    Web service versus Servlet versus EJB. Which one to use? That is a very
    good question.
    EJBs should be the core desription and processing of your business rules.
    They generally
    have good performance, are easy to integrate into your security set up but
    they require
    access using the RMI-IIOP protocol. This means that generally EJBs are not
    accessible
    outside of your firewall and their clients have to be written in Java (other
    enterprise
    objects or applets or GUI applications).
    Servlets often form a web-based front end to EJBs. They also have good performance,
    integrate into security and only require HTTP access. This means they are
    accessible
    outside your firewall and any client with a HTTP or web API can call on
    them, however,
    there is no standard contract or rules for this type of interaction so you
    will need
    provide direct input into development of the client.
    Web services are also accessible via HTTP and so can be accessed from outside
    your
    firewall. In fact, Web services internally within Workshop and WebLogic
    actually
    operate by using servlets and EJBs but the main advantage is that the mechanism/contract
    for interacting with the web service is a well defined standard (WSDL) and
    can actually
    be registered or retreived dynamically (UDDI). Since web service standards
    are open,
    most popular programming languages already provide client APIs to interact
    with them
    thus freeing the web service developer from direct interaction with the
    client developer
    and letting the client be developed on any platform. The penalty for this
    openness
    is a small performance hit (XML translation) and a slowly maturing security
    model.
    So in summary, use EJBs for business operations that have Java clients within
    your
    enterprise, use servlets to provide web style access to your business operations
    (like web pages) and use web services to enable automated clients that can
    be written
    on any platform either inside or outside your enterprise (firewall). In
    fact, it
    is common to write the EJB first and then to expose some parts of it outside
    of your
    enterprise by writing a web service wrapper (maybe using the EJB controls
    that we
    saw in class).
    I hope this helps.
    Adam
    "Sadhana Jain" <[email protected]> wrote:
    Hi,
    I don't know if this is an appropriate question to ask for this class as
    it's not
    workshop
    specific. I was wondering how will I decide when it is appropriate to use
    a web
    service as opposed to say servlet or an EJB for my application? What is
    a good
    rationale to use?
    Thanks,
    Sadhana

  • Basic JNDI questions, mostly

    Below are some questions I have in regards to EJB development and deployment. Mostly
    these are JNDI questions, with some WebLogic ones mixed in. I have a web app that
    I created, and which works, but I’m not exactly sure why it works. I’d like answers
    to these questions, but I’d also like to know if this set of questions indicates
    a particular set of training that I’m missing, and, if so, where I could get it.
    Q: In an EJB method that uses another EJB, when do I use “java:comp/env/ejb/PersonBean”
    versus “ejb/PersonBean” versus just “PersonBean”? The web app that I wrote uses
    “PersonBean” and it works.
    Q: Why do we have to use PortableRemoteObject? Why can’t we just directly cast
    it to “PersonHome”? How does it know to narrow the name “PersonBean” to the home
    interface, and not to the remote interface instead?
    Q: For the DataSource, I’m looking up “java:comp/env/jdbc/bahiawebDB” in the code
    instead of just “bahiawebDB” like I do for the EJB. Why the inconsistency?
    Q: In “ejb-jar.xml” the “ejb-name” is just “PersonBean” instead of “ejb/PersonBean”,
    but my “res-ref-name” is “jdbc/bahiawebDB”. Why?
    Q: On the WebLogic server admin tool, under “mydomain > JDBC Data Sources”, should
    my data source Name or JNDI Name be “jdbc/bahiawebDB” or just “bahiawebDB”? Why?
    Q: On the WebLogic server admin tool, under “mydomain > JDBC Data Sources”, what’s
    the difference between “Name” and “JNDI Name”? Who uses “Name”? Is it just for
    humans? Is the JNDI name supposed to exactly match the “res-ref-name” in the ejb-jar.xml
    file?
    Q: On WebLogic, why are Data Sources and Connection Pools separate? Why would
    you want to map multiple Data Sources to the same Connection Pool?
    Gary Kephart
    New Century Mortgage
    [email protected]

    Below are some questions I have in regards to EJB development and deployment. Mostly
    these are JNDI questions, with some WebLogic ones mixed in. I have a web app that
    I created, and which works, but I’m not exactly sure why it works. I’d like answers
    to these questions, but I’d also like to know if this set of questions indicates
    a particular set of training that I’m missing, and, if so, where I could get it.
    Q: In an EJB method that uses another EJB, when do I use “java:comp/env/ejb/PersonBean”
    versus “ejb/PersonBean” versus just “PersonBean”? The web app that I wrote uses
    “PersonBean” and it works.
    Q: Why do we have to use PortableRemoteObject? Why can’t we just directly cast
    it to “PersonHome”? How does it know to narrow the name “PersonBean” to the home
    interface, and not to the remote interface instead?
    Q: For the DataSource, I’m looking up “java:comp/env/jdbc/bahiawebDB” in the code
    instead of just “bahiawebDB” like I do for the EJB. Why the inconsistency?
    Q: In “ejb-jar.xml” the “ejb-name” is just “PersonBean” instead of “ejb/PersonBean”,
    but my “res-ref-name” is “jdbc/bahiawebDB”. Why?
    Q: On the WebLogic server admin tool, under “mydomain > JDBC Data Sources”, should
    my data source Name or JNDI Name be “jdbc/bahiawebDB” or just “bahiawebDB”? Why?
    Q: On the WebLogic server admin tool, under “mydomain > JDBC Data Sources”, what’s
    the difference between “Name” and “JNDI Name”? Who uses “Name”? Is it just for
    humans? Is the JNDI name supposed to exactly match the “res-ref-name” in the ejb-jar.xml
    file?
    Q: On WebLogic, why are Data Sources and Connection Pools separate? Why would
    you want to map multiple Data Sources to the same Connection Pool?
    Gary Kephart
    New Century Mortgage
    [email protected]

  • Unit Testing and Code Coverage

    Is there any way to see graphs and charts of how much code was covered during Unit Tests in OBPM 10GR3 from the CUnit and PUnit tests?
    We use Clover Reports in Java Projects.
    Any such tool for OBPM 10GR3 projects?

    Here are some more
    Triggers and DB links are not available in Oracle Explorer - it would be great to have them in there - I found triggers under tables - but I would much prefer them to be broken out under their own node rather than nested under table
    I think others have mentioned this but when you query a table (Retrieve Data) - it would be great to be able to specify a default number of records to retrieve - the 30 second timeout is great - but more control would be nice - also a way to control the timeout would be nice as well
    I noticed that I get different behavior on the date format when I retrieve data (by selecting the option from the table menu when I right click on a table) versus getting the same data through the query window - why isn't it consistent?
    Also - with Intellisense - can you make the icons different for the type of objects that the things represent (like tables versus views versus functions)
    I noticed that I couldn't get dbms_output to show up with Intellisense - I had filtered out of Oracle Explorer all the System objects - does that somehow affect Intellisense as well? I know that the account I am using has access to those packages.
    Also - more control over collapsible regions would be nice - I love that feature of VS - but for ODT it seems to only work at the procedure level (not configurable with some kind of directive etc...)

  • CS5.5 Gripes/CS6 Wish List (from the perspective of an FCP switcher)

    I have been a Final Cut Pro user for more than 10 years (starting with version 1.2 on a 500 MHz G4).  Final Cut Pro X is a disaster, but that’s a well-covered topic for a different forum.  After experimenting with Premiere Pro CS5.5 for a few months, I am ready to make the switch for all of my future projects.
    I make a living editing video.  I don’t, however, work for a big company, and I don’t have a lot of money to spend on third-party software or hardware.  Much of my work is shot on DSLRs and delivered online.  That may put me in the category of “pro-sumer” to some, but I wouldn’t be making the switch if I didn’t require a more professional alternative to FCPX to do my job.
    I also don’t think “pro” has to be synonymous with clunky and ugly.  (I’m looking at you, Avid.)  When Final Cut was young, it was fun, intuitive, and sturdy.  It became a robust, professional NLE over time.  I think Premiere Pro is on that path now.
    I like the look, the feel, and the functionality of CS5.5.  It’s not quite Apple-slick, but it’s very much at home on a Mac.  Overall, it’s an upgrade to FCP7 and a very strong alternative to FCPX.  I have high hopes for CS6.
    But this is not a comprehensive review.  It is just a list of negatives:  my gripes, wishes, and personal preferences.  My list of positives would be much longer, but my concern right now is with CS6 and the improvements that I hope it delivers.  As I am new to Premiere, it's quite possible that I am mistaken about certain functionalities or lack thereof, but I thoroughly researched each point in the help docs and forums before posting this.
    My primary system is a 3.2 GHz iMac 21” (2010) with 8 GB RAM, 512MB VRAM (ATI Radeon HD), running Mac OS X 10.7 with external Firewire 800 drives and a DisplayPort-to-HDMI external display.
    MAJOR ISSUES
    Hardware acceleration support for ATI GPUs
    There are rumors that Apple will be switching back to nVidia, but all recent iMacs have ATI and only ATI cards.
    Background rendering (or at least improved rendering options)
    Background rendering is probably FCPX’s most impressive feature and Adobe needs to catch up.  However, even if true background rending can’t be achieved, there should be an auto-render option (after a set idle time), partial rendering (if you cancel a render, keep everything that has been rendered up until that point), and more render options (e.g. render all and render selected). Regarding that last point, I think the whole work area concept should be dropped.  It makes sense for other apps, particularly for animation, but it just gets in the way of more important timeline functions while offering little functionality beyond being a clumsy way to control the area to be rendered.  At the very least, have a way to hide it.
    Full-screen preview
    Maximizing the program frame just isn’t the same (although the grave accent key function may be reason enough to switch to Premiere).  Full-screen preview isn’t just a nifty function for demo-ing sequences.  It’s a big part of the way I work.
    DisplayPort/Thunderbolt out to HDMI
    It may be that a third-party card is required for proper color correction on an external display, but there’s no reason this feature shouldn’t exist.
    Thumbnail images/show frames bug
    This one drives me crazy. I’ve tested this on 3 different Mac systems with various hardware configurations. Thumbnail images in the bins and frame images in the timeline seem to be recreated every time a project is opened, even though the thumbnail image files in the media cache folder don’t actually appear to get rewritten. It’s as if there is no cache at all (even for the most recently viewed bins and timelines segments).  It can’t be an intentional functionality for saving hard drive space because the cache files continue to take up more and more space.
    Open multiple projects simultaneously
    I know that you can cut and paste between projects but being able to open multiple projects at the same time is a very useful feature of FCP7.
    Magic Mouse/Magic Trackpad scrolling
    Premiere Pro is a cross-platform system and impressively so, but there need to be a few Mac-specific interface adjustments, the most important of which is support for the Magic Mouse and Magic Trackpad scrolling.  I use a Magic Mouse.  (It’s an irritating device sometimes, but once you get used to it, it’s hard to live without it.)  When I’m in a window, I expect a flick up or down to scroll up or down -- in every situation, every time, including the timeline.  Unless I stop using all other Mac apps, I will never get accustomed to the timeline suddenly flying left or right when I want to scroll up or down. Also, here’s a chance for Adobe to fix a problem that FCP7 shared:  When you scroll up or down in the effect control window using the mouse, the drop-down effect controls sometimes twirl all over the place if the mouse happens to float over them. As it is, it very easy to throw settings into random disarray without even realizing it just by scrolling through the pane.
    The timeline:  selection indication, icons, and general improvements
    It is very difficult to discern at a glance what, if anything, is selected in the timeline. Transitions always look selected!  Audio tracks should be a different color or otherwise more distinct. There should be an option to show frames only without any text. The icons for track options are small, crowded, and ugly.  Some of the editing icons are too similar, namely edit and ripple edit, although the excellent status bar at the bottom of the application goes a long way towards making up for this.  Simply put, the timeline could use some polish.  Don’t be afraid to steal from Apple on this one.  Keep the tracks; just make it easier on the eyes!
    Conforming audio
    The ability of CS5.5 to work natively with DSLR footage is awesome, but it’s not fully DSLR native if it has to pre-render a major component of the footage before it can play, even if it’s just the audio.
    Media loading
    When launching a project, it can take a long time for all the media in the project to “load” as tracked in the status bar.  I’m sure there’s a reason for this, but since the program is able to determine which source media files are missing before this step (and gives you the option to reconnect them), what is it doing and why must it load every clip in the project, even those not in use by any sequences?
    Page up/page down and arrow key commands
    For starters, the current page up/page down key functions should take the CTI to the next cut in the sequence, not the next cut in the track that happens to be targeted, which can be way down timeline.  In FCP7, I frequently navigated cut to cut with the up and down arrow keys.  I’ve avoided customizing the keyboard commands to match FCP7 --  I would rather endure some hardship and learn the proper Adobe commands -- but this one is flat-out backwards.  The page up/down keys should page through the timeline (left to right and right to left), while the up and down arrow keys should take over the clip to clip function (assuming the targeted track issue is fixed).  Currently, the down arrow takes you to the very end of the timeline, a function already duplicated by and better suited for the “end” key.
    Bin management and clip relationships
    Deleting an item from a bin should not delete it from the timeline, but this is part of a bigger problem, which is the whole master file versus instance versus subclip versus dup clip thing. There aren’t any good indicators regarding the relationships of these clips, and there are few ways to adjust them.  If clips are going to be connected, then they should be fully connected.  For instance, changing the name of a clip in the bin should change it in the timeline.  Adding an effect or trimming should affect the corresponding clips between bin and timeline.  Otherwise, the clips should just be completely separate instances.  I haven’t been able to find an option to turn a clip into an independent instance, although I have to think it exists.  Also, editing an instance of a title does change every other instance of that title throughout the sequence.  I would prefer to be able to edit them separately without having to duplicate them in the title editor.
    Bin effects
    Related to the above, it should be possible to add effects to clips in bins or keep effects on clips added to bin from timeline.  It would also be nice to be able to group effects together in bins for a specific project.  The effects window is better suited for global collections.
    Clearer visual indicators that a clip has been adjusted in the effect controls panel
    Even with the effects panel visible, you have to take a close look to see if there have been any adjustments to the standard settings. An indicator on the clip in the timeline itself would be useful.
    Snapping should include the CTI
    It should also include the blade tool, although the Cmd-K option to cut at the CTI position does make this less of an issue.
    Smoother scrubbing
    Scrubbing is pretty awful.  While this is understandable with native footage, FCPX somehow manages to make this silky smooth at full resolution.
    Color correction shape mask and better color correction in general
    Simplify the primary functions of the 3-way color correction effect.  Keep all the rarely used adjustments out of the way.  I use Magic Bullet for some purpose, but I’ve always done most of my color correction within FCP.  Put the most common adjustments front and center (or up top, as it may be).  The highlights/mid-tone/shadows drop-down is inconsistent in what controls it pertains to.
    FCPX’s color correction is actually one of its more underrated features.  The way it allows you to layer corrections is dead-simple yet as powerful as anything in FCP7 or Premiere.  Stick with the 3 wheels, but rely less on hard to control tonal ranges and add simple keyframe-able shape masks for secondary color correction.
    Better support for shared media access
    AVID is king in this arena and Adobe needs to improve.  It should be possible for multiple editors to safely work on the same project files and share media over a server.
    MINOR ISSUES
    More vertically compact playback/edit control area in source and program panes
    Make room for either a taller timeline or bigger previews.  The virtual scrubber and shuttle controls are the nifty things you drag with the mouse the first time you ever use an NLE and never touch again.
    Larger icons in icon view
    Remember size and position of bin windows
    Playback resolution setting indicator
    There should be an always-visible indicator of the playback res setting in the preview pane that’s easy to adjust without right-clicking.
    Better markers for sequences and clips
    Include colors and more keyboard shortcut control.
    Clip sliding with keyboard commands
    “Opt-,” (that’s Option-comma) should cause a clip collision, not overwrite.  Either that, or “,” and “.” should slide, while “Opt-,” should overwrite.  Opt-arrow should only extend an adjoining clip if in the middle of two clips.  Basically, this whole arrangement should be re-thought.
    Project browser should auto refresh/sort.
    Option-drag on a Mac should always be copy, not move.
    Copy is currently command-drag in the project browser.
    Stop auto-save from interrupting adjustments in timeline.
    Ideally, the auto-save should just be a background function, but it should at least wait until you’re not in the middle of dragging something.
    Add ability to select a cut directly and add default transition.
    CTI control
    If you move the CTI while it’s in play mode, it should continue to play from that spot after you release the mouse, not stop there.
    Show number of frames being adjusted during keyframe adjustments.
    Enable/disable specific effect parameters/keyframes.
    Clearing an effect should clear keyframes as well.
    Or have an option to clear both.
    Position controls
    There should be a preference to make the default position 0.0 x 0.0 (as opposed to 50% of whatever the resolution happens to be), and include a reset button.
    Through-edit indicator in timeline
    And a quick way to join clips, such as a right-click menu option.
    Border controls for images
    Auto save location preference and functionality
    In FCP7 I set the number of auto saves to keep to the maximum of 100 and Premiere can go even higher.  I have always used this feature as an additional backup and archive system, which has come in very handy.  Those files add up, however.  So, I prefer to auto-save to an external drive.  Also, auto saves in Premiere continue to occur whenever there are unsaved changes.  This means that if you make a small adjustment and then leave Premiere for a while, it will keep auto saving the same iteration, which is not only inefficient but leads to my next point….
    Tame the bouncing dock icon on auto save
    While using other apps, there’s no need for the dock icon to bounce every time Premiere performs an auto save.
    More/clearer control over cache locations
    There should also be an option to reset the cache/render file locations to their defaults.
    Export source range
    When exporting, remember the last-used setting of the export source range (or just dump the whole work area thing as I previously suggested).  It’s too easy to cut off a portion of the video when the export defaults back to the work area every time.
    Larger timecode display
    After spending hours and hours editing, the current timecode for clips and sequences is something you want to be able to track without squinting.  Use the letters h,m,s and f instead of colons.
    Simple slug
    Creating a black matte works okay, but it involves a few extra steps.
    Snapping toggle
    FCP allows you to quickly toggle snapping while dragging a clip.  Upon releasing the clip, the snap toggle returns to its original state.  Premiere could use this fucntion, but I suggest something simpler.  Holding down the “S” key should always turn snapping on regardless of toggle state, while releasing it returns it to however it was set before.
    Ability to dock the audio meters along top
    I like to give my timeline as much horizontal space as possible.  I’ve noticed a lot of users keep the toolbar up there, myself included.  I think it would be a good place horizontally aligned audio meters as well.
    More detailed tool tips or hover explanations in preferences
    For example:  The options regarding XMP data should make it clear that the original files will be modified by Premiere.  This caused me problems in other applications.  (After researching these settings, I understand that various Adobe applications use this information to share resources, but it’s still very unclear what the specific benefits are or what functions are lost without this option.)
    More detailed support documents
    The online support documents on the Adobe site very good.  They are nicely arranged and easy to search, but they could be more detailed and offer fuller explanations.  (Peruse the support forums and you will discover all kinds of debates that could have been easily cleared up with one concise line in a help document.)
    Icons and button design
    Icons and buttons are often too small, too similar, and too crowded throughout the interface.  Take some style tips from Apple on this one.
    NEW FEATURE WISH LIST
    Footage auto-analysis:  shot recognition, color correction, and color match
    If only Apple had just added these features and others to the FCP7 framework.  Adobe can do it better though, by using smart folders in addition to the traditional bins that we know and trust (anything besides those dreadful iMovie-style “events”).
    Effect previews
    In the 64-bit era, there has to be a quicker way to preview effects.
    Ability to render in alternate formats
    Namely ProRes or DNxHD.  This would save me considerable time on exports by allowing me to select the “use preview files” option.
    PluralEyes functionality
    Okay, I suppose I should just purchase PluralEyes...again.
    Many, many more effects!
    How about starting with the missing vignette effect?  Just because an app is “pro” doesn’t mean you should have to custom build every effect.
    Many, many more looks!
    Will most of them be cheesy looks I would never touch?  Sure...but give me some templates to play with before I tweak the look down to my exact specifications.
    More speed!
    FCPX, for all its flaws, blazes on any modern Mac.  This probably comes back to hardware acceleration, but while Premiere Pro on a Mac isn’t slow, it doesn’t blow you away.

    You hit the nail on the head, Peter.
    I recently started giving Premiere Pro CS5 a try, and was shocked by the amount of bugs and usability issues. I'm not even going as deep as you in the features, but the most glaring evidence of the lousy interface is that timecode offset bubble that appears everytime you move a clip in the timeline... the bugger appears just under your mouse pointer and won't go away, so if you try to just move a clip down a track quickly, and let go of the button ONTO the bubble, it acts as an obstacle and your drag/drop fails ! seriously, Adobe... don't tell me nobody has stumbled upon this one during beta-testing ?
    Ditto on the autosave that basicly interrupts anything you're doing. No background saving in 2011 ? At first it bothered me to no end so I disabled the autosave. And then a few hours later, Premiere crashed, and didn't even try to recover my work. I lost 2 hours of intricate work. I put autosave back on, and learned to endure the constant, annoying save dialog. And it's not like the save process is instant... even the simplest of projects takes 5 to 10 seconds to save... on an SSD... come on... let me work already...
    Of course, a headache-inducing implementation wouldn't be complete if the autosave didn't trigger even 30 seconds after a manual save. If I choose an interval of 5 minutes, just autosave 5 minutes after a manual save !
    In thumbnail view, you can reorder the clips manually. It's all fine, but you can't reorder them by name or date or length, unless you switch to list view and of course, lose the thumbnails.
    Dragging and dropping a clip in thumbnail view is a chore, because the palette doesn't scroll when the mouse reaches the edge. You know, when you want to move something at the top all the way to the bottom... so you have to drag, let go, scroll one screen, drag, let go, scroll one screen... it's a joke. Even the timeline can do it. Heck, it's a software standard, nowadays.
    Renaming a clip and pressing ENTER takes you back to... the top of the folder !?! How many times I've renamed files that were at the end of the thumbnail view, and for each clip, I had to scroll back down. Time waster.
    Want to locate a clip from the timeline, in the project ? the locate function takes you to the folder. But not to the file, you have to navigate towards it manually.
    Ditto on the up/down arrows. I haven't found a way to jump from cut to cut like in FCP7. How often do I need to jump to the starting/ending points ? much less often than jumping to a nearby cut.
    How come I can't reorder filters in the effects pane ? sorry, I can actually reorder the filters... provided I take them from the bottom and move them up. Moving effects down doesn't work. It's driving me crazy.
    Conforming happens more or less anytime. Without reason. And even if the Media Cache is already full of conformed files.
    When moving the boundaries of the work area, no timecode/offset appears. You need to let go of the button and THEN hover, then you'll know the exact time/duration.
    Try scrolling up in the timeline, it scrolls to the left (which is stupid, it should scroll up). Now scroll up on the tabs of a palette : they scroll to the right !?!
    Can't select an active track. Direct consequence : copy-pasting a clip overlaps anything that's on track 1. Say I want to duplicate clips on the same track... I can't do it.
    Copy-pasting between projects doesn't keep the transitions, only the order of the clips.
    Have you tried doing a frame freeze on a reversed clip and setting proper in/out points for the freeze ? Good luck.
    "Duplicate" command, when you right click on a thumbnail, is way too far from the cut/copy/paste commands at the top, even though they're quite related.
    The other day I transcoded footage with different audio parameters. Just the audio had changed. Upon opening the project, Premiere wanted me to locate the files, which I did. And it failed miserably. Instead, I opened the project ignoring the missing files, and then re-linked them. I pointed Premiere at the first missing file, and had to confirm the replacement. Premiere saw all the other missing files in that folder, and began prompting me FOR EACH SINGLE FILE !!! two hundred and forty three of them, precisely. Can't I have a "yes to all" button in 2012 ? and why does the relinking work inside the project and not when you open a project ?
    That was just from 3 days with Premiere Pro CS5. I still can't believe it. Just to think people bashed Final Cut Pro X, praising Premiere in comparison... Sorry, but they both have serious issues. And Premiere doesn't even have the excuse of novelty.

  • Should I open sent PDF's through PS (flatten then save as .eps) & then save as a PDF thru distiller

    Hi, I'm coming from a print background with newspapers where it was necessary to do the above with sent PDF's for print so that the fonts wouldn't bomb at print. I'm now working with a high gloss magazine and am wondering if it is necessary to do the same now. I have been doing it as routine with sent PDF's for print, once I ge the email I open in PS, make sure they are CMYK then flatten the image and save as .eps to open in Distiller and save as a PDF again. Again this was to make sure no fonts would bomb at print as per I was tought. (Used Quark when working with newspaper and thankfully am using InDesign for magazine). Is there any negative to doing this still? Even if the sent PDF's are fine (all True type font)? Could doing this, flatten and resave cause the iamges or fonts on the PDF's to lose quality at print? I'm pretty sure it wouldn't but I could be wrong with this higher quality print.
    Appreciate any advice here, as I'm slightly worried it may cause some loss of quality at print (fuzzy border around text - light but there). Thanks.
    Programs used are PS CS5, InDesign 5, Adobe Distiller 9. Final proofs are PDF's from InDesign to printer.

    The “workflow” you cite is absolutely not recommended for any purpose whatsoever. The only thing that can be said for it is that it produces consistently low quality results with extra steps and overhead. Forget about transparency and color management! The newspapers in question must have been either (1) working with antiquated, buggy RIP software from the previous millennia or (2) had personnel with mindsets and education from the previous millennia!
    Adobe strongly recommends use of direct PDF export from InDesign using the PDF/X-4 preset – never create PDF by distillation of PostScript!
    Although we strongly encourage use of OpenType fonts for their typographical features and cross-platform compatibility, in terms of output quality and reliability, use of Type 1 versus TrueType versus OpenType is irrelevant.
             - Dov

Maybe you are looking for

  • With the latest version of itunes, are you able to search your library by purchase date?

    With the latest version of itunes, are you able to search your library by purchase date?  My prior itunes software had Name, Time, Artist, Album, Genre, Rating, Plays and Purchased.  With Version 11.0.3 of itunes, 'Purchased' is not a tabbed option t

  • Will not start - get a white screen and  gray folder with ?

    I tried to erase the drive and reinstall snow leopard - which got interupted.  Now when I try to start - I just get a white screen - sometimes with a folder and question mark in the middle of the screen. I tried rebooting with original OS disc - whil

  • XI question on two different sender

    Hello all,   I want XI to pick up two files and send them to two different receivers based on some conditions. This condition will be done via message mapping. Is it possible to do it without BPM? If we can do it with BPM how? I don't want to use tra

  • Applying css rule to column in a html table?

    Im trying to get variable images to be resized with in a table, in one column, to have a standard size on display. Ive googe'd just about every thing i can think of to try find an answer. Being newb to dreamweaver i might not know the correct wording

  • How do you delete duplicates

    ??? help i need to delete my dupilcates and i dont know how?