Monday 31 January 2011

3D Eye Depth Map and Gradient Maps

Now we have a laser cutter I was intrigued to see what sort of 3D objects it could make. The way in which is works is when it is in 'Raster' mode, it works like a dot matrix printer, creating more and more sparse spots the lighter the area is. So the blacker an area, the more dense the dots and the further into the material the laser should go. I think they'll be a slight rough texture to it though.


My idea was to just have sections of the face imprinted into acrylic or some such transparent material. My main focus was on the eye section of a face. I thought of taking the already made 3D me but after many trials it wasn't good enough. Here's some pictures


Sooo, I got a picture of Anaïs' eye and made in mudbox (it's good practive anyway) and here's the result:


I was quite pleased with this, but perhaps I should say how I got the depth map. So, Mudmox doesn't do Depth Maps. I exported from Mudbox as an OBJ and imported into Maya. I had to create another sphere for the eye. I got a new camera and pointed it at the eye. It was quite hard working out how to do a depth pass, but I found it here. Particularly Emil3d - 03-26-2007, 03:31 PM.

So, go to the render properties, check the "Z-Depth" option, and also make sure it saves it as IFF format. Do a batch render and somewhere you'll find an .IFF file. Open this in FCheck and it'll list the channels that were rendered. Check the z-depth one (or something similar) and you'll see the Depth Map. Save this as a TIFF.

I then put this into Photoshop and re-adjusted to get the whites - white and the blacks - black. But I also thought that the lasercutter could only cut 8mm at a time, so I'd need to separate the picture into segments for each cut. I did this by having another curves layer and creating a straight line from bottom to top, for just 1/4 of the graph, 4 times. Images look like this:
 
 


So 8mm X 4 = 32mm = 3.2cm. I think this is about right for the depth of Anaïs' eye?

Another problem would be the 'dithering' that it need to do. To achieve different depths it doesn't vary the power of the laser, just the amount of dots. So this might make it look pretty weird. But it's something we'll need to trial and error.

Whilst looking into the gradients, I found 'Gradient Mask'. It looks at the gradients in the image, and applies whatever gradient over the top of it. Works nicely with depth maps. An example is a multiple black and white gradient based on this depth map. It could be used to make stuff look like terrain maps:

I also overlaid it to keep the depth.

I also thought about adding texture for the eye, but thought it'd look good with skin as well. So I took the original photo of Anaïs and overlaid it.



Sunday 30 January 2011

Creatures - the game - fixes and docking station

Well, Jakeh has remembered the creatures game he played many years ago, and how much he enjoyed it. So I decided (he made me promise) to install the game on Windows 7. Doh! That was a mistake, doesn't seem to like Windows 7 as it was originally built for Windows 95. Even tried in compatibility mode!! Found this website which is the games development website involved in Creatures. They've got all the downloads. I tried one of the patches but something goes wrong with unzipping it.

I did find a free game download called "Creatures Docking Station" which looks like it'll work in newer versions. I installed it and it comes up with the game in the background, but in the foreground a username and password is required. The link for registering doesn't work. Only found out  a day later the registration website has moved to this. But after filling in the form it doesn't let me log in! So I've emailed them to sort it out. I did find some fixes for this common problem here. Although someone does say in the forum that it does work!!! I've not tried the fixes yet. I've also found there is a patch (crack) for the docking station.

I shall add to this blog when I find out more

Monday 24 January 2011

Scanning with the UMAX PowerLook



Scanning with the UMAX PowerLook 1000

 

 

 

 

In ‘Photoshop’ go to START > Programs > MagicScan > MagicScan 4.7

 

This will load up the scanning application as seen below.


  1. Click ‘Preview’ to preview the image 
  2. Select where you want to save your file (your home drive)
  3. Select the area you wish to scan
  4. Click ‘Scan’ to scan image
  5. Click ‘Exit’ when finished



 

 Next you can open the file you have scanned in Photoshop.


Scanning with the Microtek Scanmaker



Scanning with the Microtek ScanWizard Pro

 

 

 

 

In ‘Finder’ go to Applications > ScanWizard Pro > ScanWizard Pro 7.12

  1. Click ‘Overview’ to preview the image
  2. Select the area you wish to scan
  3. Click ‘Scan’ to scan the area
  4. Give it a filename and click ‘Save’


Scanning with the HP Scanjet 4370



Scanning with the HP Scanjet 4370

 

 

 

 

In ‘Finder’ go to Applications > Hewlett-Packard > HP Scan Pro

 

  1. Click ‘New Scan’ to preview the image
  2. Select the area you wish to scan
  3. Click ‘Accept’ to scan image
  4. Give it a filename and click ‘Save’


Canon CanoScan LiDE 60 for XP

Canon CanoScan LiDE 60

 

To use this scanner you need to load the Application called “ScanGear Starter” found in the Start Menu > Programs > Canon > ScanGear Starter



1.    Select the area to scan

            2.    Check the settings (300DPI is good)

            3.    Click Scan

 



You can simply scan as many as you want, and when you’re finished it, you can close the application and it will ask you where you would like to save the images:


4.    Select your Networked home drive and click save.


John Kings Website

John Kings Website


Purpose

Primarily to advertise himself as an artist and give information about his art. This information would also include workshops being held, gallery information, and any other events.

Potential expansion:
Selling artwork
Booking workshops through website
News

Audience

General types of people whom this website might be aimed at, taking into consideration:
  • Audience Characteristics
  • Information Preferences
  • Computer Specifications
  • Web Experience

Gallery's interested in artwork

Information wanted:
  • Looking to find out about the artist
  • Looking for examples of work
  • Looking for previous exhibitions
  • Interested in the workshops
  • Contact info
  • Possibly the latest news

People interested in the workshops

Information wanted:
    • Looking to find out about the artist
    • Looking for workshop information, dates and times
    • Contact info
    • Examples of work

    People interested in buying work

    Information wanted:
      • Looking to find out about the artist
      • Examples of work
      • Where to buy from
      • Contact info

      Fellow artists

      Information wanted:
        • Latest news
        • Examples of work
        • Newest pieces

        All audiences are presumed to have at least a 17" screen with computer up to 4 years old. Could be running Internet Explorer 6, although might have bought new machines more recently and so have at least Internet Explorer 7. Not particularly computer 'savvy' so won't be running other browsers (FireFox or Opera). Might also have Apple Macs running Safari.

        Content

        Content would be primarily aimed at these audiences, so the content they would like is as follows:
        • Background information about Johns work
        • Possible work history (previous exhibitions)
        • Contact information (email and possible phone?)
        • Workshop general information.
        • Possibly specific dates and times when they become available
        • Examples of work, which could be expanded to encompass all work. Possibly categorised

        Information Architecture

        Site map:


        Home
        About me
        Contact
        Events/News (workshops)
        Work (categorised)

        Workshops could have its own page, but it does come under the "events" type banner.

        Thoughts on the general architecture of the site:

        First page:
        Should have a nice introduction to the website, maybe an explanation of why there is a website, and perhaps a little about the artist; and show lots of examples of work.

        Work Page
        After trying to design the layout for artwork on this page, I have concluded it would mean more to the audience and John if information about the pieces was available. So this would contain images of work and synopsis' of work as well.

        Design ideas

        Websites of inspiration:

        LIKE

        http://www.voodoochilli.net/ - NOT SURE WHAT THIS IS
        http://www.artistportfolio.net - BIT IMPERSONAL?
        http://www.thibaud.be/ - CLEVER, LIKE IT
        http://www.joshuaellingson.com - BIT BUSY ON FIRST SIGHT

        Things to avoid:

        DISLIKE - A BIT TWEE FOR ME



        Notes


        John creates a broad range of art and so would prefer it kept simple, in order that no one art takes precedence or might look odd or out of place.

        He prefers clean lines, simple text and an uncluttered look.

        The top three of his list are his favourite.

        For navigation a horizontal list is preferred as it is quite common and takes up less room.

        A neutral colour scheme so that his artwork does not conflict

        Name might be called "www.the-art-works.co.uk"

        Good webdesign inspiration:
        http://www.monkeydevil.co.uk/website/portfolio
        http://www.csszengarden.com/

        keywords

        Art studio
            Tools
                Paint
                Brushes
                Palette
                Scalpel
                Chisel
                Projector
                Table
                Easel
                Donkey Easel
                Palette Knife
                Cloth
                Water
                Hands
                Knife
                Graphite
                Charcoal
                Fingers
                Oil Pastels
                Chalk
                Rubber
                Anything
                Ink
                Saw
                Art book
            media
                Canvas
                Clay
                Wood
            Light
            Work Space
        Art works
        Gallery
            Space
            Neutral
            Light
            Quite
            Clean
          

        Thinking about the art studio. Like the thought of using the tools which made the art in some way. Maybe have them in the background, some focal blur, and the main page in the middle, close to the user, giving a sense of depth.



        John wants the website to focus on the art works, so why not have the whole website consist of art. You can have access to the art where-ever you are in the website. But a stumbling block was always - how to lay this out. I tend to a more interactive layout, what would be better then fully interactive, moving each picture as if it were a canvas. In the middle of the page would be the text and "The Art Works" logo would be at the top (possibility for moving). But I don't want this to be a novelty, it is only there so you can view all the art works where-ever you are.

        I would like a way to filter through these and create some sort of order though, may be difficult as a there would be some programming involved. Not sure how friendly it would be to screenreaders though (although would hard of seeing people want to visit the sight?). And the w3c validation?


        Done some research and found these little websites:
        http://www.walterzorn.com/dragdrop/dragdrop_e.htm#flexlyr
        http://dunnbypaul.net/js_mouse/


        I was quite pleased with the results of this. For the third example I though of a combination of these two designs. More formal and organised, but with constant access to all the art work. Working with grid based design. The content for each page would be in a designated area in the middle. This area is also used for looking in more detail at the work.




        Examples

        Standard




        A scrollable website, with navigation at the top. You will find example work on the "Work" page, where you can scroll up and down and filter the work using the filters on the left. Each image has a description next to it. Clicking on an image will show a larger preview.


        Free form




        This website was designed around the thought of always being able to see/access the example work. Each image can be moved around using the mouse, and re-ordered. Clicking on any image will bring up a larger version of it, with description. You can also filter these images down to view images with certain attributes. Clicking on a link at the top (navigation) will display different information in the middle, keeping the arranged images in view.

        After much looking over it I decided to move things about a little. My main worry was the title was too far to the left and looked a little sad. I also didn't like how the tabs on the upper right were a little out of the way and looked slightly odd with that thin white section connecting them. So I've centred the title and moved the content and tabs lower and in to the middle.



        Grid Design




        Based around a neater version of the Free form design. Images can be seen at all times. Navigation and content is placed in the middle of the images,  in a nice little box. Images are filterable (top right) and clicking on one will load it up in the content box (where the text usually is) with a nice little bit of information about it.

        I've been thinking about making the image sizes changable, therefore being able to see more or less of them on screen at any one time. This could be implemented in all of these designs.

        After much consideration and he chose the free form design. I mentioned if he wanted it adapting at all but he seemed pretty certain he wanted it kept like that, although as part of the implementation I shall design it so it can be changed easily.

        Technical design and Implementation


        Well, no specific time frame so I shall dedicated about 5 hours a week, which should work out at 2.5 hours a night for 2 nights. The hardest part will be in the technical aspects of interactivity.

        Possible technologies to use in the website:

        HTML - goes without saying,
        Flash - might give this a miss as it's not "iPhone" friendly!
        Java - A good alternative to flash, even found some that will do moving boxes/pictures!
        HTML 5 - a little ahead of itself so I probably won't
        CSS - definitely for styling
        XML - a lot to learn but very good for information architecture. I could store image name, information which it can load up from a xml file. Needs investigating.

        So HTML to hold the page format and everything else. Using java to control interactive pictures. CSS for styling and XML for information handling. Easy peasey!

        A quick note; maybe the links could go round in an arced corner? Maybe give it a go. Might look interesting.



        Anyway, this is the basic structure of the html page with DIV tags. So first we have one for everything. In this will be first the title; next the navigation, then the Content.

        In Content will be the text and the filter. The images will be mixed in with this, with the images contending with the space of the text and filter.

        So hierarchically speaking:

        container - height and width 100%
            Title - Width = 100%, Height = ?
            Content - height = expandable, with = container
                filter - height = expandable with filters, width = ?
                NestedContent - height and width = ?
                        Nav - Width, height = nested content
                        gumph - Width, height = nested content
                pictures - I guess they should be in the content layer somehow moving about?

        Note - How do they maintain a randomness if the pictures are ordered by the web browser? Surely they will need to be laid out by hand with absolute (or nearly absolute) position.

        Note - Don't forget that these images are going to be interactive and movable, Controlled by JAVA (i think)

        Page fitting


        How will the page fit to different window sizes. This is a good question. I have reduced the size to 1024 X 768. Perhaps I should make it a little smaller as someone with that resolution would need space for the bar at the top and start menu at bottom. Now keeping it to width of 950px.

        So, basically, when someone expands the screen I think the page should be pretty static and stay in the middle. It can expand downwards with content. If it widened with the browser then I'm not sure how to move the images about. I don't think their position will be able to change with more space available. If it can though it would look good.

        Page Loading


        There are several different ways loading pages could work:
        • clicking a link on the nav panel would load up another html file (page). A couple of problems. The filtering might be reset and the position of pictures. Also might mean a slight delay in loading the new page.
        • clicking a link on the nav panel could hide the current nested content and display another nested content. Not sure if this would be good for screen readers or HTML validation?
        • Clicking a link on the nav panel could look up a XML document and display some information in it. A down side is that at present I know nothing of how XML works, or what its' capabilities are and its weaknesses.


        It appears the best solution is for the link to hide and reveal layers of information. Easy to do and does not not upset page setup with filters and position and doesn't take any time to load. Sorted.


        Drag and Drop Images


        Well, this must be the most difficult bit. To see how it'll work and then make it work. General plan is to research different techniques and evaluate which is best for this purpose.

        Research

        Drag and Drop


        This looks like the proper way to drag and drop items in a list. Also the pictures could be in a list, although note sure how to position. Anyway, it looks good but probably wouldn't work because it's not in a list. But it does have solid code which might be worth looking at.

        http://tool-man.org/examples/dragging.html It has constraints and can drag these anywhere. Definitely worth looking at.

        drag and drop things - http://www.javascriptkit.com/howto/drag.shtml

        Looks good, not quite as nice a website though. Javascript which is easy to install and you just give an element a class name and it becomes movable. Doesn't have constraints like the above

        Nice short and sweet script easy to download. Works like the previous

        Note - A problem we might come across is the z order of things. I would like the last thing that was clicked on to be the front most image.

        Re-order page with drag and drop java - http://www.dhtmlgoodies.com/scripts/arrange-nodes-2/arrange-nodes-2.html

        Very, very complex script which can work server or client side. It's nice because they've seen the need to re-organise web pages.
        Also http://www.dhtmlgoodies.com/scripts/floating_gallery/floating_gallery.html for dragging images in a list to re-order them


        Fancy Image loading

        Well, I remember James saying that there was a facility somewhere to make it so when you click on an image it'll load up full page and shade the background out. Heres some research.

        http://visuallightbox.com/index.html#example

        Looks good and does a lot of stuff, but might need to buy it for John Kings website. Looks like it links in with Flickr which might be good for john to upload to.

        http://www.dhtmlgoodies.com/index.html?whichScript=picture-zoom

        looks good for use with johns website. Nice and easy, a little less stylish though.

        http://www.huddletogether.com/projects/lightbox2/#how

        Looks just as good as the previous, but is totally free.

        Think I'll be using lightbox2 over picture-zoom
        §

        Shadows

        So, I've just thought about shadows. How will we accomplish this shadow effect with CSS2? Well it'll have to be some old fashioned (not that old though) bodging.

        So we could use transpared PNG's. Not compatible with IE6 but really nobody should be using this anymore. Other then this I see no problem. So the main shadows are on the nested content with the tabs. If I save an image of what the nested content looks like this could be the background. The tabs could have two looks, infront and behind. Might be able to use the same picture for all tabs.

        No real transparency needed if I'm to keep the background the same colour.

        Implementation


        So, here we go.

        First the basics of HTML, header, title and all that. Then the body. Put everything in nested DIV's. Next work out which scripts to put in. First is to test Lightbox2. Not sure which of the drag and drop scripts to use, think I'll try the second one first.

        Doh! most important is to work out how many pixels some of the things are!! Need to know:
        • Content width - doesn't matter, just centre the content
        • nested content width - 236px
        • filter width - 214px

        Tabs!!!


        Well, it's the first time I've actually added pixel styling to elements not in tables. It's very interesting. So, I originally though I'd put a background borders on and see how it goes but it's a little more complicated. So I finally had a look on the internet and found this gem:

        http://24ways.org/2005/centered-tabs-with-css

        It's all relevent and tells you how to define the height (width not important) and finally how to put a background on the tabs. Here it shows good use of <span> which I've never used before.

        So because the <li> are inline, you have to make adjustments to the <a href> elements. The span is used so you can put a background left and a background right on the link, which is a little complex to think about at 22:00 on Saturday night. Doh! I'm giving up at the moment.



        so, after many hours, here it is:

        This is the HTML for the tabbed list
         
                        
                          
        • Home
        •                     
        • About Me
        •                     
        • Events
        •                     
        • Contact


        • As you can see there is a lot more here, but most of it if for the change of properties for viewing what is in each tab. The main points to note are teh class="nav" and the .

          Here is the CSS that goes with it:

          .nav ul li, .nav ul {
          display: inline;
          list-style-type: none;
          margin: 0px;
          padding: 0px
          }
          .nav ul li a {
          background: url("Assets/Tab-right.png") no-repeat 100% 0;
          text-decoration: none;
          padding-top: 20px;
          padding-right: 5px;
          padding-bottom: 5px;
          padding-left: 0px;
          margin: 0px;
          }

          .nav ul li a span {
               background: url("Assets/Tab-left.png") no-repeat;
          padding-top: 20px;
          padding-right: 2em;
          padding-bottom: 5px;
          padding-left: 2em;
          }

          I found the the "background" for ".nav ul li a" had to be that format for some reason? Oh, and here are the files:

          Tab left ->
          Tab right ->

          I also tried to do the same for the body of the text but it was too hard, so I just made it the background.


          Tab Navigation


          To navigate through the tabs I used "Behaviours" in Dreamweaver to change the properties of the said element when clicked. So, when the Home link was pressed, all other elements were set to "display=none" while home was set to "display=block". It took me a while because throughtout my trials I had another script always showing home!! Doh!


          Fancy image browsing

          This was pretty easy. I found "thickbox" and "lightbox" the best and free. You download their scripts in folders, and follow the instructions. Pretty simple. I put the lightbox scripts in js for reference. I'm using thickbox for now, but I might change that.

          Had a small problem for thickbox, but found a solution. Apparently just change thickbox.js line 79:

          TB_TempArray = $("a[@rel="+imageGroup+"]").get();
          TB_TempArray = $("a[rel="+imageGroup+"]").get();

          Drag and Drop

          This caused no end of headaches, and has taken the longest to work out (or give up on). So, with tool-man I tried and tried but in the end could only ever get it to move the first picture. Could not for the life of me work out how to add more (although I'm not sure if the script allowed it). It looks like you can move one thing for each class.

          So next I tried "dragtest1", which I worked out only worked if you didn't put the DocType thing at the top, which basically means it breaks convention which I definitely didn't want.

          Finally I tried Dragtest6 http://luke.breuer.com/tutorial/javascript-drag-and-drop-tutorial.aspx, which works but conflicts with thickbox! doh! But I shall try some other time when I am less fed up.

          Filters

          So now on to filters, which might just be the hardest!!! Oh, and I don't even think I've done any research into it either! So heres some research:

          Wow, just found this: http://www.askthecssguy.com/2009/03/checkbox_filters_with_jquery_1.html  Exactly what I wanted. Trying it as we speak. Someone has tried it in their website > http://www.gjunkie.com/labs/

          I'm having a little trouble displaying the images inline after the tickbox has been unchecked and checked again. The gjunkie website looks like they've solved it. Must look into this.

          Also, the filters is generated and placed within the pictures div, which is what I didn't want. Must look at the later example and find out how he did it.

          Copied gjunkie script which works much better, with a better fade and better for inline lists.

          Just re-arranging it now, which is one of the hardest things.

          So, after quite some rejigging I found that by putting the "Gumph" inside the "filterThis", but not in the
        • element  and telling it to float right, that the list does carry on around this. The "filters" are told to float right. Done.



        • So far: 

          Things to do
          • Change colour of non-highlighted tabs to a lighter gray
          • Make tabs bigger
          • delete unwanted tabs
          • add email address
          • make the links list 'normal'
          • make background image
          • make footer image
          • put on as background
          • play about with "visual arts" and put next to "Wall Space"
          • add all filters
          • add all captions
          • export thumbnails slightly bigger 100px
          • make thumbnails slightly bigger
          • make slider to control thumbnail size
          • make actual footer on webpage
          • put heading on the filters
          • remove unwanted filters (and duplicates)
          • re-order images


          So it's been a while and i'm slightly regretting using the filters script from that guy as now I want to customise it some more and I haven't got a clue where to start.

          After some browsing I found this great little forum describing how to use checkboxes to show and hide things! Brilliant. Might just do it myself from scratch!

          Wednesday 19 January 2011

          Adobe Acrobat OS X Active Directory Crash

          So after fixing Adobe Acrobat admin password problem acrobat 8 loads up, but crashes after 20 seconds. I remember fixing this exact problem with Marketings' computer, Louise Webster I think.

          Doh! I didn't write the solution in the blog!!! Lesson learned over and over again. So here's the brief explination I told Paul Jenkins by email:


          Hi Paul,
          I went over at 12 today to have a look. It appears to have problems on any active directory accounts which aren’t administrators. Its a permissions issue because acrobat (and lots of other programs) write temp files to the users home drive with the wrong permissions. The NHR program was installed but only redirects caches and fonts folders to the local drive, but still didn’t fix the acrobat problems. I found another folder which needed re-directing - “~/Library/Application Support/Adobe/Acrobat/”
          I made a folder in “/Users/Shared/Websterl/” and created an alias in her “/Library/Application Support/Adobe/Acrobat/” so it gets re-directed to the local drive. It only works on this machine for only her at the moment. If any more people have the same problem it would be worth adding an entry in the NHR program for acrobat.
          I also noticed the programs and operating system weren’t up-to-date. I’ve set it updating adobe products. Louise said she’d get back in contact if she had any more problems. I didn’t get around to updating the operating system though.
          Cheers
          Stephen
          I had a look at this folder and came up straight away with error! I remember not it was a folder within this folder:  /Library/Application\ Support/Adobe/Acrobat/9.0_x86. So after redirecting this one it works. So to fix this problem for everyone, NHR program will need to do this as well! I'll find out how to do that tomorrow!



          Unfortunately, NHR the full program redirects too many folders, i.e. Microsoft User Data, so I'll have to adapt the loginhook, logouthook, clean script files etc. But then I can copy them to all machines and it should be fine.

          Adobe Acrobat administrator passwords on OS X

          I've been having this problem for a very long time. I install adobe Acrobat professional 8 on the master image and it asks for the admin password the first time it is run by anybody! This is OK if you know the admin password but of course students don't. So, here's the main part of my investigations and one solution (there are more).


          I found a few forums discussing the same issues and it apparently occurs on version 8 on OS X. It looks like the application needs to run a fixing program if it can't see a few files where it expects them, and this fixing program requires admin password. Forum here.


          The forum gets a little confusing and I really don't want to read it again, a lot of crossed wires and such. So, the files involved is: ~user/Library/Preferences/com.adobe.acrobat.80.sh.plist

          So I logged in as an admin user, ran Acrobat, put in password and if fixed itself (for that user). Then I copied this file and put it in an active directory users library and loaded up Acrobat. It worked! So, the easiest way to stop it is to make sure users have this file.

          My plan is to put it in Workgroup Managers > Managed Preferences. Then it should work for all users.

          ... the next day ...
          Yay, I tried it with the managed preferences and it worked. Looks like there is another file for the newer version: ~user/Library/Preferences/com.adobe.acrobat.90.sh.plist

          Monday 17 January 2011

          OS X Bootpicker randomly not working

          I have a computer suite with Dual Boot apple macs, when you switch them on they boot into OS X and give you the BootPicker menu. Rather annoyingly, only in this room, after rebooting bootpicker doesn't load on some, and after another reboot bootpicker works on these and fails on others. It's been happening for years and caused a lot of annoyance (only with me though). I must admit I do hate random problems.


          So I thought I'd take the time to properly look into this problem. I noticed that bootpicker needs the "BSD e.g. Name:disk0s2" of the Windows partition to work. Now on these Apple Macs (Mac Pro's) there are two hard drives, one scratch disk and one operating systems disk (partitioned into two). Now for the Mac Pro's or OS X, it just randomly orders the disks, probably classed on how fast they respond on bootup, to either "disk1" or "disk0"! And so bootpicker fails. The command I found which lists the windows partition disk is:
          diskutil list | grep 'Microsoft'
          So, solutions! Well, there are two apparent solutions. Either get bootpicker to work not from bsd names but something unchanging like mount points (which are always the same) or UUID's, or get OS X not to randomly name the disks.

          To get OS X or mac to keep disk names the same is to do with the /etc/FStab file. Now in OS X it doesn't need this file but if you make one it'll use it (I don't like it though). So using this file you can tell OS X which disk with particular UUID should be mounted where. Here are some posts I've found about it:

          http://stackoverflow.com/questions/2278714/look-up-a-volume-by-uuid
          http://discussions.apple.com/thread.jspa?messageID=5820852&tstart=0

          To get the UUID you just type in terminal:
          diskutil info /Volumes/Untitled
          Untitled is the mount point for the windows partition .

          To get bootpicker to use something which doesn't change you've got the choice between the mountpoint and the UUID. At the end of the day, the way bootpicker works is when you click on "Windows" it runs the command:
          /usr/sbin/bless --device /dev/ --setBoot --legacy --nextonly
          You'd need to edit the /Library/Preferences/bootpicker.plist file. There is a "windowspartition" section where it normally says "/dev/disk0s3". I'd like to change this to the UUID or mountpoint. I'm not sure if the mountpoint is relevant at bootup, I think that's more of a OS X thing, so the UUID might be more relevant.

          Thursday 13 January 2011

          OS X Active Directory login issues - delete parallels

          I keep coming across Apple Macs which have been working in Active Directory and yet not letting users log in (or even use kerberos in Firefox). The general symptoms are:

          • "Some network accounts available" at login screen (i.e. active directory not working, open directory working)
          • not letting active directory users log in
          • showing active directory working in "Directory Utility"
          • letting you re-bind to active directory with no problems
          • trying to use firefox with kerberos authentication comes up with "unable to ......." (can't remember the exact error)
          I tried the simple fixes such as reset the PR Ram and fsck -yf. I was going to do a disk permissions fix but thought it'd take too long. I looked in the console and found a few kerberos errors.

          I found that there were a few errors with the network settings. I found that because the image was taken from a computer with Airport, it hadn't updated this and thought the ethernet connection was in fact an airport connection. I deleted the "Chesterfield College" location and made it "automatic" but deleted the airport which fixed that. I also noticed two Parallels shared connection network interfaces, which were set up when I was trialing Parallels. On deleting these and logging out, active directory accounts worked. But on reboot the devices were remade. So there was some files somewhere which remade the connections. I thought it be quicker looking for help rather then looking for files, and on a quick search I found some helpful stuff. Before I come to that I did notice that Parallels was actually still installed. So I recommend first you delete Parallels from the Applications folder.

          Next would be deleting parallels library files. I found this forum, and down the bottom they mention:

          cd /Library/
          sudo rm -rf Parallels
          cd /Library/Preferences/
          sudo rm -rf Parallels
          cd /Library/StartupItems/
          sudo rm -rf ParallelsTransporter
          cd /System/Library/Extensions/
          sudo rm -rf prl_hid_hook.kext
          sudo rm -rf prl_hypervisor.kext
          sudo rm -rf prl_vnic.kext
          sudo rm -rf prl_usb_connect.kext
          sudo rm -rf prl_netbridge.kext
          cd /Users//
          sudo rm -rf .parallels/
          sudo rm -rf .parallels_settings

          I of course did all of this, but I think just deleting the Parallels application should do the job.




          .... well, had a little play around and it turns out it's best doing all of it and heres ARD send command thing (under root)



          rm -R /Applications/Parallels
          rm -R /Applications/Parallels\ Desktop.app
          rm -rf /Library/Parallels
          rm -rf /Library/Preferences/Parallels
          rm -rf /Library/StartupItems/ParallelsTransporter
          rm -rf /System/Library/Extensions/prl_hid_hook.kext
          rm -rf /System/Library/Extensions/prl_hypervisor.kext
          rm -rf /System/Library/Extensions/prl_vnic.kext
          rm -rf /System/Library/Extensions/prl_usb_connect.kext
          rm -rf /System/Library/Extensions/prl_netbridge.kext
          rm -rf /Users//.parallels/
          rm -rf /Users//.parallels_settings

          It's impossible to delete network devices via command line, but doing the above disconnects them, which is the next best thing.

          Mac OS X Power Saving Schedule via Command Line

          I now have, or want to have, all computers booting up on weekdays at 8:30 which will give me a good idea of which ones are working and which aren't via Remote Desktop. Also I have them shutting down at 16:50 as they will not be needed after this time (apart from times when they might be doing updates/scripts but I'll just have to remember this problem)

          I also plan on changing the schedule for Half Terms and holidays so no extra power is wasted.

          To control it all the xserve is set up to manage machine preferences and so through Workgroup Manager I change the preferences. A problem I'm having is that not all machines boot up at 8:30 even though they all appear in good working order. To check/manage preferences from Remote Desktop (to troubleshoot also) you can use the command line:

          pmset -g sched

          This works well and gives you what the schedules are. So far they are telling the right schedules:

          N251-04899 (192.168.32.22)
          Repeating power events:
            wakepoweron at 8:30AM weekdays only
            shutdown at 4:50PM every day
          Scheduled power events:
           [0]  wakeorpoweron at 01/14/11 08:30:00 by Repeating
           [1]  shutdown at 01/13/11 16:50:00 by Repeating

          I'll update this more when I find out more.

          Tuesday 11 January 2011

          Final cut pro "Preview Disabled"

          So yesturday a student was capturing film from a DV deck. Logged out for break and logged back in and low and behold when in "Log and Capture" it said "Preview Disabled". So I tried different tapes, different computer, different user (cove), different PAL settings and refresh rates to no avail. Only thing I didn't try was different tape deck (all were in use).


          A quick search on google came up with a few results:



          1. Switch from PAL to other and then switch back (should refresh it)
          2. View drop-down menu in FCP and clicking on 'Refresh AV devices'
          3. Trying leaving the camera on and plugged into your Mac and then reboot your computer
          after this guess I'll do some more searching (maybe even in console)