Life as an Artificial Lifeform

Adobe needs some fire under its pants and while those designers in Adobes arms are rarely looking for alternatives its still refreshing to see that there is a thriving market for Adobe program alternatives. I know that there are quite a few photoshop contenders out there that are shaping up to become real contenders (and I am not talking about Gimp which some argue is already there as PS replacement - I don´t), what I didn´t know is the huge market for lightweight vector programs of all sorts and kinds and most are programmable - something Adobe will not include into Illustrator in a million years (except for AfterEffects and maybe Flash (out of necessity) Adobe has avoided direct scriptability that they can´t control). The list in the linked article is long and intriguing and covers not only Mac apps but also a LOT of Linux apps and some Win apps - yes I am eying Linux if people haven´t noticed yet - Apples DRM crap and pure consumer orientation is coming to a tipping point for me very slowly (I have been using Apple since 1994 btw). So back to topic - the list of programs is stunning and includes everything from chart drawing to 2d cad oriented programs to experimental vector programs like NodeBox which lets you Python program your art (something very dear to my heart) and then lets you save it as Quicktime animation or PDF still. If you are a graphic designer and have ever needed a vector program you should have a look at this list and see if you might want to replace the worst program in Adobes lineup or finally put your copy of Freehand 10 to the old peoples home.

20 Vector Graphic Editors Reviewed

Posted by fALk at 11:15 | Permalink | Comments (0) | TrackBacks (0)

Apple has opened up the DRMed, stripped down propitiatory miniDisplay port that is to be found on all MacBooks for free licensing.
I beg the gods of hardware heaven to have Apple fail with this endeavor. For humans not intelligent enough to learn from past mistakes there is no place on earth.
This "idea" has failure written all over it. First its a display port standard from apple. Something they have tried at least two times already without any kind of success - just leaving a lot of customers in the dust with exotic connectors that you can not buy any adaptors for anymore (I still have a fully functioning G5 with three adaptors dangling on its back just to get to VGA because a ADC to VGA adaptor is hugely expensive and hard to come by (so its ADC to DVI to VGA) . Secondly - I would be all over this spec if they actually managed to improve upon a spec that is far more versatile and robustly designed to begin with - HDMI. They could be in heaven if they would have made display port not broken by design and integrating degrading DRM in it. You know make it the antithesis of HDMI. But but what they do is try to release a new spec just after they have been blasted with criticism all over the netosphere about their inclusion of DRM so this is a now a hostile market and they think anyone is picking in up? HDMI - while having the same DRM problems at least does not have a marketing problem at the same time. Oh and what about freaking audio? At least the novelty of the failed ADC adaptor was to reduce cable clutter. Now apples "mini" version of the "standard display port" is ommiting the audio spec so you still have to add an audio cable to your clinical clean apple desk? (nothing I mind personally I would like to add - cable clutter has its charme ;)
It will be an epic failure again and I am totally unsure why Apple is putting itself in the same hot waters again as they must have learned from ADC at least a tiny bit - and ADC had a LOT more going for it then the miniDisplayPort - no bad press, not broken by design, less cables.

Posted by fALk at 19:06 | Permalink | Comments (2) | TrackBacks (0)

The more I read and think about it I think there is a big issue with the new display port that Apple is trying to put under the rug. They want to plug the analog hole once and for all. I think its not only to do with copyright shit - so I am reasonably believe that is one of the bigger underlying issues - its also the floppy drive issue. Uh? Yes I am actually of the generation where apple removed the floppy drive without asking anybody. A huge outcry and then a year later people applauded them for the move - you know there was nothing really nice about floppy drives and cd burners where just better and looking back it was one of the smarter moves for apple - pushing the whole industry forward - for the few that needed a floppy drive they could just attach one through USB - sure for them it was the more expensive road but it must have been a tiny minority - I know for myself that I never looked back to the floppy era again.
Fast forward eight years later and Apple brings out a Laptop that for the first time is missing any way to connect analog video to it. Mind you I am a VJ and the only way to do our thing is through analog video cables - but also I am vfx professional and go into any professional video studio these days and have a hard look around and I promise you - you won´t find any kind analog video anywhere anymore. Now also the projectors that do not have a digital video input option of some sort also seem to have died out. So in quite general I would say that yes its a move that from a very very far perspective makes sense on Apples part - ditch analog video and the world sees innovation an crisper pictures and maybe soon wireless digital pictures. Now there is only one big big problem with that thinking and I can see that Apple is not seeing that angle - because they rarely come into contact with it - the whole event bussiness needs to bridge ultra long distances with video cables and as of right now the only digitial standard that can bridge 50+ meters is digital SDI - a very professional option that almost no beamer under $5000 on the market supports nor does apple have a video out solution for that either - especially not on their portable laptop line. I am sure with budgets above 10.000 Euro per event you can cook up a all digitial solution already but this is not the market where 95% of events are in - also the whole backend has not switched to an digital format either - unlike the Floppy that was already replaced by the CD burner and shortly later by the USB flash drives - there is no standard in the video world for digital video. It starts with the cables, goes to compression and codecs through framerates and resolutions - there are about a trillian combinations and no market standard has prevailed so far. So forcing a switch over from analog to digital video at the moment is not a clever idea - no matter what your motivations are - and when half your motivations are bad already (DRM evil evil DRM) then this switch is going to alianate a LOT of people - even if the event market might only make up 0.5% of Apples laptop buying population I think people needing analog video still at this moment approach more 5% (old beamers, old TV sets, bars wanting to show a video etc. pp) and that is a huge chunk. I am quite sure Apple has to come out with a native adapter at one point. The current solution when you ask an Apple employee? Get an old DVI to video adaptor and get minidisplayport to DVI adaptor and chain them together - very elegant Apple - very professional - very reliable such an adaptor chain. BTW - just 3 years ago you could just plug a SVHS cable right into your powerbook without any kind of adaptor - apple used to be about simplicity - these days seem to be waning.

Posted by fALk at 12:30 | Permalink | Comments (0) | TrackBacks (0)

Word is out that the new Apple MacBooks and MacBookPros with their new display port technology are lovely looking environmently "friendly" trojan horses of the MPAA (that is the movie mafia of america). When you buy a movie through the iTunes (not only music anymore) store and play it on these new machines - you have to have a copyright safe monitor attached to actually view the movie. That is a mafia approved monitor. If you do not you get a nice little warning sign saying that you are not allowed to play the movie.

Now this raises an interesting question: Is the lack of a miniDisplayPort to analog video adapter something that will never be rectified because it would mean that this would breach HDCP copyright protection? Anyone ponder what this would mean to the Apple using VJs? Not to mention that the movie mafia does the exact same mistake as the music mafia had been doing - by actually forcing people to crack movies just to watch them on their devices - while at the same time teaching millions of people how to actually crack a movie who then find it so easy and enjoying the freedom afterwards that they don´t ever buy one again.

This decision of Apple and the movie mafia is so wrong on so many levels its mind boggling that in the current environment such mistakes are made. What is Apple thinking? They are not above the godline just yet - they must know that these things are creating a huge backlash on the internet - just after their movie store was heralded as something great.
Watch the blog storm about this in the coming days with the citizens demanding DRM free movies and apple apologizing and putting a software patch out in a three month timeframe and the movie industry realising their mistake in less then one year (when they need a bailout).

To the people thinking this stuff up I can only say one thing: Stupid fucking idiots - you must have won your harvard degree in the lottery.

To quote a commenter on the Engadget thread:

I'm not really down with stealing, but when they decide they are going to control how I watch my movies and listen to music they can go die in a fire.

Posted by fALk at 10:49 | Permalink | Comments (0) | TrackBacks (1)

I have had enough of doing secretly behind closed door dirty little things with code and such so since the state of affair is at the point where I wanted it originally over a year ago but also much further and somehow not as far I decided to make the future progress on the live site. Now this is all cryptic and such because there is no content - no meat so to say and I will not make more eyes look on it then needed. Generally its maybe a 0.1b then anything else but as right wing newstalker Bill O'Reilly so lovely proclaimed - when technology fails: fuck it we will do it live... So if you know what I am talking about have a look - if not - forget about it there will be a proper announcement when we reached 1.0...

Posted by fALk at 17:17 | Permalink | Comments (0) | TrackBacks (0)

After getting some frustrations with the old movable type installation and lots of broken links and an outdated spam system I have taken steps to upgrade the private ProtoBlogs to MovableType 4.2.
The installation went smoother then I thought with the exeption that I actually had to restructure the site a bit and that means a devastating loss of links from external sites. If you are an external site deep linking to an article on prototypen.com/blog/ this link will very likely not work anymore, maybe you have the time to find it again and update your link (one can dream right? ;).
The new structure is much better but I would have loved to avoid this problem and it will cost us probably a lot of traffic. In return all articles are now under their category and google is most likely able to rank our content better in the future. Furthermore Categories have their own rss feeds and all the other niceties that I have not dug into that can make it to our blogs now.

Update: old deeplinks working again..... its double content now (old and new links) - I know that google aint liking that much but better then loosing the links I guess...

Posted by fALk at 13:38 | Permalink | Comments (8) | TrackBacks (0)

The Blinkenlights iPhone application is released and available on the appstore so you can watch the stream of pixels as it appears (or should appear ones its working) on the house on you iPhone with different viewpoints.

iPhone BlinkenSim on the Appstore

Posted by fALk at 15:49 | Permalink | Comments (0) | TrackBacks (0)

In my ongoing series about absolutely stunning developments from Siggraph this year comes a techdemo /white paper that is so simple yet so powerful. Every serious 3D artist knows that it is hard to capture real life textures and replicate them with lots of details and try to make them appear haptic. Tricks like making fake bump/displacement maps by extracting a range of values in a certain color channel only get you so far and touch up is time expensive.

Now Mashhuda Glencross and Gregory Ward from the University of Manchester in the UK and Dolby Canada in Vancouver developed a dead simple - everyone can do it way to extract the normal map of any photo that is shot twice - one time with normal exposure and one time with the flash turned on. From that the get the depth data of the surface. Look at the video of how simple this is - no commercial product yet but believe me that this takes about two weeks to surface (if it isn´t patent locked in some closet).

Watch the video to see how it works and be amazed:

Original Article over at the NewScientist

And I found this tutorial in the comments over there on how to fake this with 4 photos and a lightsource other then camera flash.

Posted by fALk at 13:23 | Permalink | Comments (0) | TrackBacks (0)

Besides Apple trying to keep anything and everything under wrap there are things I wish they would just publish loud and open. I had to look about an hour today to find the maximum texture resolution and the general graphic specs for the iPhone implementation of OpenGL ES - since I am not developing but producing content for a developer I have no access to the specs but needed them badly. I eventually found some light info on geeks3d.com which refers to an pretty extensive article called "OpenGL and Mobile Devices: Round 2 - OpenGL ES for the iPhone and iPod Touch" by Richard S. Wright Jr.
that is available from Dr. DobbsPortal:

There are a few limitations you should know from the start:
* There is no stencil or accumulation buffer.
* There are only two texture units.
* The maximum texture size is 1024×1024 (use power of two only).
* The maximum space for textures and surfaces is 24MB.
* Only 2D textures are supported.
* There is no software rendering fallback.

Posted by fALk at 12:44 | Permalink | Comments (0) | TrackBacks (0)

A website lets users speak out what frustrates them with Adobe apps and the new open Adobe Company is responding in depth and length that I have not seen any other company doing ever - I think thats quite great.

Read the following part about a Linux Port of Photoshop in the discussion:

Linux - there are a lot of people there wanting Linux versions of your leading apps. And yet that's been glossed over time and again. And while it wasn't me that added that particular gripe, Photoshop and Lightroom really ARE the one and only reasons why I can't ditch this POS Windows operating system for Ubuntu.

[I can't speak for other products, nor do I want to give you false hope. Having said that, the architectural investments we're making will make the Photoshop codebase more flexible and portable over time. The fundamental problems with moving to Linux are A) sales to Linux users don't represent growth, they represent replacements of Windows units, and B) Linux use is heavily based in antipathy towards non-open-source commercial software. --J.]

I mean this is some reasonable thought on the issue that Linux users might even understand.

There is much more juice with questions and answers about prices of the Creative Suite, User Interface consistency between apps. While some answers are a bit inward looking instead of outward looking there are so many details in there its hard to recount them all here so head over first to the dearadobe.com website and add your gripe then go to the dear adobe top 25 problems (its a must read and if you ever come into contact with adobe apps I am sure you agree with 26 of the 25 points being made.) and finally go to the official Adobe Photoshop insider blog of John Nack to see a huge company open up to the world in a way I have not seen before. Especially follow the discussion after the blog entry where there is a healthy back and forth between users and developers - I hope there are some other big time developers out there taking a cue.

Posted by fALk at 18:10 | Permalink | Comments (0) | TrackBacks (0)

I have had a problem that bugged me since weeks. The finder on my MBP (dual 2.4GHz) was taking up about 100% processor speed - always. That not only reduced my battery life to about 30 minutes it also helped fry the graphic card (which I am still not sure if it is actually fried or just acting up occasionally).
You boot up and have nothing on and still your computer feels like it has a single 100Mhz processor in it. I was about to replace the finder with some other finder like tool today when I accidently stumbled upon the fix - on the mac rumors forum. (Somebody say those rumors sites are not good for nothing).
Anyway the fix is simple - the problem it seems is that something is calculating all along - the theory is its calculating the volume size of your main volume all the time. It only happens so (and then repeatably) when you have "Show Item Info" on your desktop turned on and showing your volumes on your desktop (don´t know if that can actually still be turned off - you used to be able to).
So the fix:

1.) click on your desktop
2.) go to "Show View Options" under your "View" menu or hit cmd+j
3.) unselect "Show Item Info"

voilá cpu usage drops from 100% down to 0.2%

Now dear Apple FTFF.

Posted by fALk at 18:22 | Permalink | Comments (0) | TrackBacks (0)

This OS is going to be my darling I can say that already. Something that I have made a feature request in every major OSX release since 10.0 is coming finally true.

from roughlydrafted:
Sources also indicate Snow Leopard will expand upon Font Book to provide full Auto Activation of any fonts requested by any application, using Spotlight to track them down.

and its even better then one thought. Imagine the computer to be intelligent enough to find fonts on your harddrives or servers for you instead you firstly trying to put fonts in an orderly fashion then upgrading systems then a year later opening a document then figuring out there are fonts missing then searching for the fonts on all your available resources then putting them in the font book app then activating them. (yes I am also using linotype font explorer - but the autoactivation feature is so buggy its almost unusable and I lost the organizational database twice after reinstalls already and remaking that takes a lot of time). Now open a document and its activates all the fonts if they are somewhere on your mounted disks - hope that this will be fast and stable.

Posted by fALk at 10:17 | Permalink | Comments (0) | TrackBacks (0)

After awaking from the enormously boring keynote this morning I actually headed over to the Snow Leopard page that apple put up and what do my weary eyes see? Quicktime X. It really does seem that Snow Leopard is the OSX version that OSX should have been from the beginning - 8 years after its introduction. Yes we got a ton cool interface features in the past years, some great functionality and stuff - but OSX lacked speed right from the beginning and never quite recovered (I still think my Titanium Powerbook running OS9 was way way faster (in everyday use that is - of course not rendering) then the current top of the crop Powerbook running Leopard). Then also it seemed that some very important core components have been left out of the development cycle all together just fixing them up to run on OSX and the modern computers but still the behind seemed to carry so much baggage and if you went a bit deeper then a normal user would you felt this baggage dragging you down into the muddy bottom of the ugly code sea.
Two of these components - for me at least - are so essential for my work and everyday normal use that sometimes I was questioning the sanity of the OSX developers for overseeing these components so long. Its the much hated Finder and even more importantly Quicktime.
Now last year they announced an "all new Finder" and everyone was filled with joy - yet what we got was a great new interface for the finder - that still lacks the spacial OS9 Finders quality - but definitely got visually/interface wise on the right track - yet the core of it seemed to be the same old and the Finder is still sometimes idling with 50% processor use now and then (that is me doing nothing and the finder using 50% processor for doing nothing or so it seems - that is happening daily and I witness that even on an 8 core mac not only on the Powerbook -> MacBookPro). So I really really do hope the Finder get new underlying code for SnowLeopard but thats not the topic of this post - the topic is Quicktime.
Now Quicktime is a GREAT tool in itself. Lets split out what Quicktime actually is.
What a normal user normally sees as "Quicktime" is the the Quicktime Player - that is just a fancy front end to the underlying core technology called Quicktime - this technology is capable of producing and playing Quicktime containers - and a dozen and some other Media enclosing container formats - such as MPEG1,2 or 4 (video or audio) but also picture files (go ahead and see for yourself how you can open a JPG with Quicktime Player). Hence quicktime is pretty much responsible to display something or create something whenever there is an audio, video or picture involved on OSX. In this day and age where "MultiMedia" is something people need as much as oxygen (ok I am exaggerating for dramatic reasons here) Quicktime is the Oxygen mask to make you survive in the toxic environment that is riddled with format wars and thousands of different formats.
Quicktime - through extensibility - is one of the best technologies out there because it simply can play such a HUGE range of formats right out of the box - going back to play videos created in 1990 (yes I have one and it still plays). The codecs support (codecs are algorythms that compress video or audio to smaller sizes) is tremendous and it includes every codec ever delivered on Quicktime itself. Yet as we know with windows - backward compatibility comes with a price. Making upgrades without breaking stuff becomes harder and Quicktime has felt this. I would venture to say that Quicktime has not seen a significant update since 1999. I am not sure as I am not a developer but I would say that raw multiprocessor support was built in back then already (for two threads that is). The only update since then worth mentioning is Javascript support slapped on so you could actually interact with Quicktime films through webpages - if you have used this feature you know why Apple is not trumpeting it more on its front page.
Now what is missing? Why am I so happy to see that Quicktime gets the full overhaul?
1. True multiprocessor support.
Video encoding is the single most CPU intensive task beeing done by more then one community (only trumpeted by the Scientific Community with its hardcore calculations and the 3d community by its rendering tasks). As it stands it seems some codecs are capable of more then two threads but most Quicktime tasks are tied to two threads and that is unbelievable.
2. Bugs, crashes, bugs, slowness. Demandingly working with quicktime (meaning opening and rendering 100 videos or 2 hours videos or 2 hour enhanced podcasts etc) you can just wait for spinning beachballs outright crashes, files that stop at two gig when rendered over network and outright causing kernelpanics (try opening 600 movies in QT player (yes that is something I do have to do at times)).
3. The web...
Apples neglect has most cetainly cost them a huge opportunity to overtake the online video deliver market - now completly dominated by Flash - which is about the worst container for video performance wise. You know right after the browser wars in the mid to end 1990s you had the media wars which where basically fought between Apple with Quicktime, Microsoft with Windows Media and Real with Real media. There was no clear winner but then everything changed. Real - the clear front runner completely fall back with having not enough resources to throw out for free services without overloading them with ads - which turned off a lot of people, Windows Media also somehow lost contact with Microsoft concentrating on Windows Vista and Apple clearly had an opening but blew it because Quicktime on the web sucked bad, the quicktime plugin just used soo much resources - still out of this race Apple came out on top (supported by the success of the iPod and the improved installed user base of Quicktime as a result) but somewhere in between Flash added video capabilities (based on the old sorenson codec apple used to push 4 years earlier). The thing it what gave flash video tracktion was the installed user base which has always been much higher then any other plug in. Then came YouTube and embraced FlashVideo and the battle seemed over. Then Adobe bought Macromedia and included MPG4 (baseline) and shortly later MPG4 (h.264) into Flash - making Flash a pure player application rather then a container format. Apple in the meantime has grown the installed user base through iPod and more importantly the iPhone - which does NOT include flash (for a reason I will tell you) yet.
Thing that was problematic for apple with quicktime on the web was not even quicktime itself so much but rather the lack of information of how to embed quicktime and how to make pages without Javascript that don´t load the whole movie itself before the user is clicking on the play button (that still has not been ironed out yet btw). And a customizable player interface (that is rather halfheartedly ironed out through Javascript which sometimes even works).
These are the most important things Apple needs to adress to make me a satisfied customer when it comes to Quicktime. There are probably more things - mostly relating to

Now Apple has one absolutely tremendous asset to actually turn around things and that is the iPhone. The phone is set to become the defacto standard to browse the web on the go (and if you have used one ever you know what I am talking about). Apple has not included flash in the first gen iPhones and while I believe that speed was a concern (the one they put out officially as the reason they didn´t support flash) as this can be witnessed by much faster Laptop computers spinning their fans with two youtube videos open - I still think the bigger picture is that they see the iPhone as the Quicktime on the web comeback device. Lets be clear - I actually have a lot of love for Quicktime as a container format - its so extensible it has great support for interactive media even outside its own container (I would argue that Quicktime is the only system that can play the MPG4 interactive spec as it is on paper) and the number and quality of codec is unrivaled - the encoding quality is maximum of what a codec can deliver and the usage is easy as pie even for more complex stuff. So Quicktime on the web will have another shot and I truly believe I will see a Quicktime version of Youtube in my lifetime.
So Quicktime X has to clean up the underlying codebase (a rewrite comes close and has already started with Quicktime becoming a core component). It has to make this Javascript stuff work 100%. The latest iterations have been better all around but still when you are going outside safari it lags a bit. Especially if you want to make your own controller - to get a feedback from the plugin you have to sprinkle some serious voodoo and then it still fails at points (try scrolling through a window on apples site with apples controller - I mean stress test that scrolling - it will fail at one point or get sluggish or crash your browser or whatnot and that is the best implementation there is on the web - everything else I have seen is even worse).
And QTX has to be fast fast and as bug free as possible and utilize all threads and the graphic cards (maybe even with CUDA pretty please).
And QTX has to come with more authoring application. A QTXCode would make me dance. What for? Interactive content! Quicktime has great capabilities with QT VR, Tracks, Sprite support and whatnot to actually replace websites (yes you heard me right) the potential is tremendous - I tried that once and I loved the general idea but making it work with like walking barefoot on ice-spikes while your head is worked on with a flamethrower. Now the Javascript is there to make an interaktive quicktime (or MPG4!) interact with an HTML page. And while Quarz composer is great to make neat animations and interfaces it does not replace an actual timelined authoring tool.

So in general for me Quicktime X is the announcement of the day and I hold out big hope for it to become THE media plugin surpassing Adobe Flash in a heartbeat (and I believe they will have true vector support in Quicktime X too . it would make sense since OSX itself already has vector support throughout). And there is no problem with installed user-base as mentioned above - iTunes need Quicktime - iPod iPhone needs iTunes.

Posted by fALk at 8:02 | Permalink | Comments (0) | TrackBacks (0)

I just read an article on Ars Technica making the case that the next MacOSX version will be called Snow Leopard and I didn´t need to read any further to see what it is about (i did read it in the end of course). Apperantly (and no I am not in the know as of yet) OSX gets a much much much needed codecleaning. I can not say how happy I would be if thats indeed the case. No more featuritis just a clean lean fast stable system. And while I think there is going to be an outcry of die hard old time Mac users for the dropping PPC support I think its a good move to consolidate the code base (less code to watch over mean less bugs creeping in). And what I would cheer as loud as I could is the dropping of Carbon - about fucking time. And to you three carbon developers - get a grip - you earn money with programming you can surely learn something new and make your codebase meaner, leaner, cleaner, faster in the end and everyone will be happy (yes that means work - no free money is there?).
I really pray to the finder gods (who are utilizing ridiculous 40% of the processor pretty constantly at the moment) to make this rumor true and bring us a true new finder (not just a new interface), a quicktime that doesn´t suck, graphic card drivers that actually use the graphic cards fully, stopping ram leaks, stopping hanging system calls etc etc etc.. If thats true its one heck of a smart move on apples part - but then again I believe it when I see it - we were promised a new finder since 10.0 (that was 8 years ago)…

Posted by fALk at 11:36 | Permalink | Comments (0) | TrackBacks (0)

Small software developers are great. You communicate with them tell them your wishes your frustrations and they listen, they implement they remain calm they even consider outlandish feature requests and tell you reasons why they might not work - you can engage with them in a chatter and see a program grow to your liking in front of your eyes - and in the end everyone benefits.
Ubercaster qualifies definitely for that. Eberhard - the guy writing the app - has now answered my third feature request filled mail in a row going into detail on every piece - it feels great to be a customer that is not a number and in return be able to do your thing more efficient.

So if you are doing a podcast - especially if you are doing an enhanced podcast I can suggest Ubercaster as a great solution - its not quite 100% there yet but with such an aware developer I am sure this is going to be THE podcast program on the mac hands down when it hits 3.0.

Find more information at ubercaster.com.

Posted by fALk at 15:29 | Permalink | Comments (0) | TrackBacks (0)

Well everyone ever comlaining about the lack of development on the 3D tracking market can rejoice. Autodesk just bought out Realviz - known mostly for their Matchmover 3D tracking software but recently also for their Stitcher VR panoramic stitching application (you know shoot photos all around and get a QTVR stitch - the new hype in 3d animaton at the moment) and their pretty new Movimento Motion Capturing Solution that works with real camera input (instead of the expensive other mocap solution called VICOM that uses highspeed infrared cameras to try and capture body movement).
Seeing that matchmover will not coma as a standalone product anymore under the autodesk brand it gives hope that we see this technology integrated into the two flagship autodesk 3d programs - Max and Maya. As for Maya finally replacing the "Maya Live!" tool that has not been updated for over 5 years (and is about the worst solution if you want to do any serious 3d tracking that you can find).
I really think that is one of the first aquisitions that I read on in a long time that actually makes a lot of sense and will actually help moving things forward - especially since matchmovers interface and core program also lacked development even so the underlying algorythm are said to be some of the best out there. I really look forward to have a 3d tracking solution that works inside Maya - the confusion with scale, different interpretation of how 3d cameras work, 3d data exchange that still has no real standard that actually works etc have been driving me nuts. Working integrated uncomplicated 3d tracking together with a simple and cheap motion capturing solution is one of the last core things missing before the whole 3d world can move to realtime rendering ;)

Read the press releasehere.

Posted by fALk at 11:01 | Permalink | Comments (0) | TrackBacks (0)

This is something the mac community might have been looking for for a long time without realising the need - a quicktime component that solves most codec woes and put your personal Quicktime Installation into the "I can play ALL codecs" category without VLC or other 3rd party apps that are fine but never good enough for some reason...

Download it here.

Its piss easy to install, no restart required, you get a very nice pref pane where you can update the component when you feel like its necessary or have it done automatically for you and it just generally works great and feels very integrated and doesn´t lag the quicktime app with difficult files.

From the about page:

Perian aims to provide a single package for all your playback needs. It is a collection of QuickTime components incorporating several libraries:
• libavcodec, from the ffmpeg project, along with code from the old FFusion component:
• MS-MPEG4 v1 & v2
• DivX
• 3ivx
• H.264
• Flash Video
• Flash Screen Video
• VP6
• H263I
• VP3
• HuffYUV
• ffvhuff
• MPEG-1 & 2 Video
• Fraps (up to v4)
• Windows Media Audio v1 & v2
• Flash ADPCM
• Xiph Vorbis (in Matroska)
• MPEG Layer I and II audio
• DTS Coherent Acoustics audio
• Snow wavelet video
• libavformat, from the ffmpeg project. along with AVIImporter.component:
• AVI file format
• FLV file format
• libmatroska, along with matroska-qt.component:
• MKV file format
• Subtitles:
• SSA file format
• SRT file format
• liba52, via A52Codec:
• AC3 audio

Posted by fALk at 11:52 | Permalink | Comments (0) | TrackBacks (0)

Cinebench

old desktop:
OGL: 2268
CPU-MP: 2760

new desktop:
OGL: 6125
CPU-MP: 18579

:)

Posted by fALk at 14:24 | Permalink | Comments (0) | TrackBacks (0)

My love for the WiiMote goes way beyond Raymans Raving Rabbits and other fun games as I embrace the gyroscope filled controller for next generation interfaces and 3d tracking in various forms. 3d tracking with the WiiMote has but one big problem. While it is quite great to get the bank and roll data out of the white brick it is only possible to know where it is using infrared lights. Infrared lights tend to behave like real lights as they can get obscured or not in the right place when you need them. Sure you can try to get the location of the WiiMote by using the accelerometers and gyroscopes but the result - and I have tried that - is quite bad. You get what people call a drift very very badly especially when you move around the room with the thing and attach a camera to it.
Now the Boston USA based company Motus Corporation claim to have developed a WiiMote of their own that can be used without the infrared sensor bar and still get the location right. They developed it in the form of a Samurai sword handle and made it primary for the target group of golfers (their are known for golf training gear) but if any of their claims hold true this is the revolution of the sensor revolution. The gyro package send new data every 30 Milliseconds is called Darwin (probably because the software to read out WiiMote data on the computer is Darwin Remote?) seems semi compatible with the WiiMote (the protocol is the same but there is different data available). The firmware has error correction build in and my uneducated guess is that they slapped another 5 or so accelerometers in there to get a more accurate average reading. Surprisingly the price is "in the range" of not beeing overly gold encrusted (its going to be somewhere between $79 and $99) and if the claims hold up its a bargain because then you can build a real realtime camera tracking solution and with a couple of these also a motion capturing solution that rivals about everything out there in the price range 10k+. Lets hope they find a distributor soon (or figure out that in these days you can self distribute this and still be successful)

Posted by fALk at 6:52 | Permalink | Comments (0) | TrackBacks (0)

I am not the fiddler when it comes to computers as I am already pissed off to no end with crappy buggy software all day - so I like it clean and as stable as possible. Every time I put in stuff in my mac that was not 100% spec approved it made the machine more crashy in the years to come (and I have tried to mod my macs as much as possible in the past).
But for those who just want a computer to play around with and want to try out MacOS X or those who just don´t want to fork out the Apple premium to get OSX and have no problem with some weekly kernel panics and general hacklery along the line (especially if they want to stay up to date on their OS) there is an article on wildwobby with a detailed description and not so legal links to all parts you need to build a beigebox mac for ultra cheap. Also the article points to a forum thread for more help if you get stuck building your ultra cheap mac over at the InsanelyMac Forum.

Posted by fALk at 8:43 | Permalink | Comments (0) | TrackBacks (0)

Keynote over and I think I was pretty spot on with the MBNano - äh Air. 1,6 Ghz, No Tablet, 5 hours battery life, no FunkStrom - sadly - otherwise "thinnest" notebook on the market - great - pricey - not great. Its geared toward executives, those that actually show off with their macs, those who need a mac to work on will choose one of the other two options.
Only "surprise" was that the trackpad is now "multitouch" which it of course already is - its just a software hack as the trackpad on all MacBooks already recognize a second finger. So I see this coming - at least via a hack - also to the other Books in the line - and I think its sweet but ultraconservative of Apple to go down that route - but this is how they are nowadays. I still hope that the developers catch on and put multitouch gestures in their apps.

AppleTV I was right - no mouse, no new iApp.
What I found great about todays keynote:

A company admitting they made a mistake (Apple TV 1.0) and saying exactly how they combat that mistake - it was very well presented and made sense.

Movie Rentals - ja we want to rent movies and ja you are in the ballpark pricewise - in the upper ballpark but you know you wouldn´t be apple if you wouldn´t. Signing all Major Hollywood studios must have cost some sweat. I say its a success on their hand - finally no forgetting of bringing back a DVD (poor DVD rental places). Of course there needs to be IndyMovies somewhere for a bit cheaper as SOON as possible because your early adopters are more the people renting Indy Movies then the Hollycrap.

what i didn´t like:

A overpriced basestation with a harddrive inside - not even a freaking raid 1.

Conservative Apple goes the save route - don´t want another cube do we?
I am not sure if the MacBookNano will sell its one of those products in between that noone really cares for - only Higherups with lots of dough would love this - but seriously 1.6Ghz in 2008? I don´t like it and I also don´t like the form somehow...

Generally -> Yawn.. but lots of dough coming towards Apple in 2008 anyways.

Posted by fALk at 19:11 | Permalink | Comments (0) | TrackBacks (0)

Well I can not help myself but to give out my predictions for the Steve Keynote this year. Its a very difficult matter to judge whats behind the black cloth covering the ads in the expo buildings in San Francisco today. What is almost certain is that there will be a new line of portables coming out of Cupertino. The question is just how portable it will be. The wild predictions on the net make it this über slim multitouch capable tablet book with flash harddrive, BlueRay, Wireless USB and Wireless Power and an antigravitation device (hence the name AirBook or MacBookAir), I think judging from apples past 3 keynote that everyone might be hugely dissapointed when the ultra slim MacBook with 1,6Ghz Wireless Power and no DVD drive, no multitouch and non tablet with alumnium foam casing (hence air?) will be released. It will probably be ultra sexy, ultra expensive (for its speed), have xxx amount of battery hours and can play HD film on its "backlit on demand" screen - but no multitouch no tablet. Wireless Power via an induction pad might come but I highly doubt it (yet I could see this coming as a marketing gimmik "look this comp doesn´t need any cables" - kinda thing - you know you gonna read in all those nontech newspapers "Apple has done it again and released a cable less computer as a worlds first" along those lines). But why the f*$k don´t I think multitouch is coming? Its developers stupid. To make the multitouch truly usefull and therefore a computer with it truly usefull the whole User Interface Experience must be though over. Coverflow in the Finder is something that is definitely going in this direction but not much else. If Apple brings out just one computer in its offerings with multitouch there is almost no incentive for software developers to make use of such a novel idea - hence Apple would have to supply all apps that take advantage of this themself for the time beeing - that would not incite people to buy the machine (the anti theory to this is that this computer is for developers to make such apps but I doubt it) - multitouch comes across the line or not at all. That would mean if the MacBookAir or whatever it will be called has multitouch then we also see an iMac with Multitouch and an Apple Cinema Display with multitouch, because the pros would love multitouch more then anyone else (see my earlier posting about the autodesk lab). I highly doubt that the technology is at that point yet - but hey would I love it :)

So if the new Book is a tablet then across the line updates (iMac with Mutlitouch and CinemaDisplays with multitouch).

Will the Book be Dockable? Its a very Apple like intriguing idea, BUT with the docking concept that was floating around you "through away" 13 inch of viewable display area and you would have to pay for much thicker screen that does nothing without the book - while nice idea I don´t see that coming, except if the screen in itself is an imac and the adding of the book gives the iMac more power and you can somehow flip open and use the screen of the Book in addition to the one in iMac - even then its a small small market and would be hugely costly in R&D - seeing the ultra conservative road apple takes after the cube massacre I don´t see it coming.

What else will we eventualysee? Maybe a sneak peek at the New OS 10.6? That would be lovely but I think Apples OS developers are on a break after the horrendous busy 2007 - I think we will see that at WWDC in Juli. AppleTV update? Very likely or they will forget about this idea and slowly phase it out - I vote for an update here with specs similar to the current offering but a build in blueray player (or maybe even recorder) and some extension capabilities (like playing other video codecs (unlikely) and generic browsing (with a innovative interface). Generally I think Apple is seeing the TV as the centerpiece of entertainment as NOT the strategy to move forward (but what do I know).

A new mouse similar dissapointing to all Apple mouse over the past 9 years but with wireless power.

A new iApp that manages all your media needs and adds automatic metadata in an iFriendly manner.
:
A 3G iPhone - given Apples traditional product update cycle one is in the cards - but who knows. Maybe with wireless power but its unlikely as this would piss off people in Europe and with the piss off lecture they already had with the iPhone price drop they rather not go down that route again and just delay the announcement until spring time.

iPhone Nano (Air or whatever they call small scale devices these days) is a likely possibility if they manage to cram the interface on half the screen and make it still usable - half the price half the storage - but I think the keyboard functionality would go out the window and that might be the show stopper for such a device.

So quite generally I see not much coming other then an ultra thin but otherwise pretty conservative MacBookNano (and if they called it Air I will still call it nano because its unlikely as light as air ;) everything else is a complete wildcard today - but be prepared to be underwhelmed with the new conservative Apple.

Posted by fALk at 7:39 | Permalink | Comments (0) | TrackBacks (0)

I just had a very inspiring email exchange with Scott Sheppard from the deep underground laboratories of Autodesk Inc. - makers of two major 3D applications (Maya - Max), all major CAD Applications (AutoCad, Studio) and some much more professional application in the design and visual fx market. I stumbled across a post on the Autodesk Labs_ Blog concerning the use of the WiiMote to be used as an input device for their applications. I asked him about the possibility of making the headtracking I posted yesterday available inside Maya (and on OSX). Never expected a reply but I eventually got one. Basically he is saying that they see the WiiMote as a big step forward for a cheap revolution of interfaces and that he actively discussed this with the person responsible for MultiTouch in the lab. They come to conclusion to integrate something like this it would need an interface inside Maya that is not there yet. I kind of disagree as it probably possible to integrate this with Python - but sadly my Python skills lack "a bit" to make a proof of concept.

Anyway what was important is that a major software developer (I would say they rank in the top 10 if not even the top 5 software developing companies in the world) is understanding that the current method of interface input devices need a big overhaul, that they embrace going the "cheap" route instead of developing stupid proprietary stuff and that its not only the WiiMote that is of interest but also multitouch (they even had a mutlitouch summit). Now if the headtracking I posted yesterday together with multitouch input doesn´t make for a great input future of computing devices I don´t know what would. Just imagine how "clay like sculpturing" would work here. You move around and closer just by moving your head and use your ten fingers to sculpt the model...

I am greatly exited.

Video of WiiMote controlling Autocad
corresponding blog article
Multitouch and AutoCad Video on Youtube(A must see if you are a designer and can´t envision what wultitouch brings)
We Have the Touch introduction of Multitouch at the Autodesk Lab

Posted by fALk at 9:27 | Permalink | Comments (0) | TrackBacks (0)

Holy Crap. If you thought that your flat screen TV is just flat and that to get into the 3rd dimension you need a new screen that is about 10 years off this video will prove you wrong.
Made with the unbelievable useful WiiMote and the sensor bar or sensor bar "like" safety goggle which have their white "headlight" LEDs replaced by Infrared ones you can make a headtracker. Then you can make your normal screen look like its an actual 3d window by analysis the motion of the head and translate it to you "virtual camera".

How much its only perception that makes you "think" 3d is demonstrated enormously well. Besides using it for games I SO WANT THIS to work in Maya. Moving around an object or getting a closer view by moving your head has to be the second coolest interface since äh - since - multitouch of course ;)

Posted by fALk at 8:17 | Permalink | Comments (0) | TrackBacks (0)

Ugh... Berlin company Mental Images - mostly known for its Mental Ray Rendering Technology which is included in the three biggest 3D applications out there (max, maya, softimage) and used in a lot of major film productions throughout the world has been bought out by NVIDIA the one of two graphic cards company (not only but you get the point).

The combination of mental images and NVIDIA united some of the greatest talents in the visual computing industry. This strategic combination enables the development of tools and technologies to advance the state of visualization. These solutions are optimized for next generation computing architectures and create new product categories for both hardware and software solutions.

This means only one thing: Realtime Raytracing - or less technical Realtime super duper photorealism - is around the corner very soon now - otherwise this takeover would not make sense. I love Mental Ray for its realism - the rendertimes are unbelievable high at the moment - even for small scenes. Now that NVidia has snatched the code I expect that they put the turbo in the rendering software via their graphic cards that have become more general purpose processing devices then graphic cards lately.

This is good. :) Hail the day when there are no rendering times anymore.

NVidia Press release

Posted by fALk at 15:06 | Permalink | Comments (0) | TrackBacks (0)

Everyone knowing me knows I am a Flashhater since day one. I saw flash as a dying species when it first came out as the arcane nonstandard scripting language (why not use javascript or python?) never appealed to me, neither did the "plug in to the web" approach instead of an "open standard" open code open everything approach. Also I did not like the interface mentality of macromedia (and a recent look at firework cs3 has gotten me those flashbacks) so I never felt good working in Director or Flash or any other macromedia app - except for freehand (which I used before it was a macromedia product - Aldus anyone?).
But I do like vectors and especially do I like fonts - especially crazy nonstandard fraktur fonts. So the ability to include fonts into Flash files and make a layout that playes exactly as the author intended on all computers is an appealing one. My hatred for Flash has trumped the need for alternate fonts and true vectors to this day and will likely into the long long future.
Then about a couple of years ago (1999) Adobe System proposed a standard to the W3C - so did Microsoft to break the Flash supriority on the web with an open standard - both standards fused in t one and became SVG. SVG sounded good back then and still does - true open documented XML document format, truly open source of the generated files, and backed by the W3C - easy to generate with just a text editor if you are so inclined. I was dancing on my chair back then when I remember correctly.

There was two major problems - no easy way to generate content (other then a text editor) and no browser compatibility.

Adobe themselves tried to eliminate the first problem by putting out the ill fated program "flame" or whatever it was called. It sucked - it sucked hard - and it wasn´t for the mac. Then there was years and years of silence until there was an obscure "export to svg" menue point in illustrator in CS1 (I think)

On the browser side it didn´t look better - Adobe did offer an SVG plugin - but this eliminated one of the biggest superiorities over flash. Neither first incarnation of Firefox could play SVGs neither did Safari do a good job using SVGs (so it did at least recognized it in some sort or form without putting out horrendous errors all over the place) - Internetz Exploder I have no idea but my guess that it does not even know what an SVG is to this date without the adobe plugin.

Then Adobe bought Macromedia and one of the biggest reason for the takeover was Flash. Adobe must have seen that not being able to deliver platform/client independent interactive vector graphics with embedded fonts would sooner or later give them a big disadvantage and that making such a technology from scratch is not as easy as it sounds - especially with the browser developers trying to get the basic W3C standards to work first before implementing something flashy as SVG.

As my heart jumped at the introduction of SVG my heart plummeted when I heard about the Adobe takeover. I thought that thats it and flash is taking on the world without anyone stopping it. Flickr and PooTube just have aggravated these fears - I was waiting for adobe trying to make Flash a W3C standard or something along the line - yet something else and surprising was happening all of the sudden. Firefox and Safari started to add build in SVG support - unusable at first but nevertheless progress. Also Illustrators SVG Export became more and more sophisticated and - CLEAN.

Then we entered the post AppleIntel Adobe CS3 Microsoft Silverlight (why they did not choose SVG I can not figure out) world. A world where Safari has full SVG 1.1 support Firefox claims it has 1.0 support and Adobe Illustrator CS3 outputs clean sufficient SVGs.

Thats the day I jumped into it to see the true power of it unfolding before my eyes. I stumbled into it accidently. All I wanted to do is making a form that I use very often in Indesign more "interactive". The form in question is a template to fold you DVD cover out of a single piece of paper without glue and be able to print on the front and back side. Its one of these great origami secrets :)
Well that form has one big problem - all text and graphics are printed out at an roughly 35.3 degree angle.
First obvious choice was to make it a PDF form. After trying it out with Acrobat 8 and whatnot of small utilities I gave up because - you can can only rotate form fields in a PDF 90, 180, 270 degrees - no 35.3.
Ok I was about to give it up but I really want this form that I use so often with an easy interface without opening Indesign everytime I burn a DVD.
Next I was thinking about a database solution - Filemaker in special. Obviously I tried to angle a field there right in the beginning just to find out that there as well you can not do so either.
I did not have a solution in my head anymore. CSS? hmmmm no. CSS3- hmmm is supposed to have a rotation value but the support is rather sketchy even if you are developing for just one browser (which would be sufficient for this form as only I wanted to use it and I use Safari all day all night).
I tried to get the idea out of my head - unsuccesfully. Then I remembered SVG and briefly looked into the documentation of Illustrator CS3 and found lots of good things about - I thought that is worth a try.
I layouted the piece and exported through the "Save for Web and devices" dialog. The first thing I saw made me smile "Include Font - for used Glyphs only - or for all Glyphs or none or commenly used Glyphs". YEY fonts!!!!
Then I looked at the code - XHTML :) I was feeling all home right from the beginning.
But how do I make it interactive? I decided to go a PHP, Javascript route.
Then is when it became apparent that even all looks rosy it isn´t. There is virtually no documentional howtos on SVG on the net. The W3C documentation is so geeky that its over my head and the few howtos available where either extremely old and didn´t work or just didn´t work. I delved into it.
Four long days later I figured it out - scrapped the Javascript route and went a pure PHP route - after I found out how to integrate PHP into SVGs and after I figured out how to manipulate the DOM tree in an SVG with Javascript that sits outside the embedded SVG. In short: it was a pain!
The coordinate system in SVG is not easely understood and I had to print out 40 pages to make the resulting SVG print out correctly (only on Safari3final I might add - it does not work on any other browser) because safari scales stuff so it fits on a page - which was counterintuitive with this document. The code is intermangled and inside to outside javascript integration is not robust in Safari. Dom updates are not as straight forward as in plain vanilla XHTML. Deleting parts of the dom tree and reintroducing it also did work only after extensive fiddling (the main reason I went with pure PHP).
But the biggest biggest problem was the handling of text - how can any standard that displays text in any sort or form NOT support a line break! Yes you heard right - for having multiline text you can NOT use something of the < br > sort - you have to calculate the linespacing yourself and make new lines through < tspans > with a "y" value that specifies where you next line of text is! HOW INSANE!
In the end though I am impressed at the output because I could include a font in the document and now have a nicely working form to print out my DVD labels. And changing Layouted text on the fly through a webinterface is something novel and great.

I surely hope SVG is making progress now that its at last almost usable - but I do not expect it to gain widespread adoption until the linebreak issue is resolved - its a HUGE showstopper.

One big question that came to my mind was a copyright issue. Fonts are to design what RIAAsongs are to the ear - its a copyright regime fighting hard against the users since day one.
So if I use my owned copy of Solex and embedd it into an SVG do I break the licensing?

Why? - you might ask - Flash has done this for years - Yes, - I reply - but in Flash you do not have the easy option to extract the font from the files - in SVG every vector of the font is being readable in plain text - you can even change vectores around if you want to - its just a matter of time until someone writes an easy to use "font translator" plugin that can read the SVG and generate a TrueType font out of it. Fontshops galore will have want to have a say in this I am sure!

Overall the future for open vectors with embedded fonts on the web does look brighter after this year and death to flash and all :P

Next project regarding SVG for me will be to try to embed video in it.

Posted by fALk at 16:19 | Permalink | Comments (0) | TrackBacks (0)

Or the nitpicky Leopard longtime user review

As some of you know I am actively involved/enrolled with the Appleseed which is apples closed beta testing group for a lot of their software. I have been doing just that since about seven years now and its always a lot of fun mixed with a lot of frustration and a lot of pleasure if you see the bug that you reported fixed or that enhancement request you did make it into the final version. Recently I have been actively involved in Leopard testing - for the last 14 month that is.

As we Appleseeder are halfly lifted from our non disclosure agreement with the release of the final version I would like to talk about a user perspective in the OS for a short paragraph.

Its been a slow coming and a lot of builds I could not test because my equipment was not compatible but in the end just a month before the release they send us a build where magically a lot of problems went away - yet the first time I installed Leopard I got this urge not to go back - even so I did not see any major improvement that would justify this reaction. But its the small things and now after the 14 month I can safely say that I have never ever seen such a productivity improvement when upgrading to an OS.

My absolute favorite - and once you get used to it you gonna see why - is: SPACES.
Designate spaces for different work conditions (one for 3D, one for video editing, one for still picture editing, one for office stuff, one for system stuff f.e.) assign programs to these spaces and remember what space number is what and all of the sudden all messy window overlapping is past – forever. I can not stress how great this works and urge everyone to take the time to set up individual spaces (drag and drop apps you use into the space preference panel and assign a space number). Spaces is whicked fast even on my old (now dead) Titanium Powerbook 1Ghz.

Time Machine is a mixed blessing – while I like the idea the whole thing is a typically 1.0 version. I personally would not use it and I know that some testers had serious problems with it. The lack of fine graned controllability is a real bummer – an "advanced setting" button is missing – I hate it.
Safari had been a surprise – I though it would suck first but the new search is amazing. It is the first search that real gives you a visual feedback – I love it totally.
Also the new DEBUG menu goes long ways – not Firebug yet but already very close with full inspection of boxmodels, load times, page code etc – great (you have to enable the debug menue to see this in action).

Bluetooth and syncing is vastly improved in speed and stability.

The new dock is a step up I guess but nothing revolutionary. I do like to use the stacks as launch managers for programs – for anything else they are useless as they only display a certain amount of files and I find myself to just use the open in Finder button inside the stacks all the time for folders that change content a lot – spaciality is not Apple core competence these days and as soon as you do serious stuff that is still a big problem.

BUT they are making headway's here and there. The "new" finder had been a huge disappointment for me at first as they put up a new face without fundamentally change the underbelly – this seems to have changed a tiny bit so in the latest releases after a big backlash inside the seeder community. I can care less if we have coverflow in the finder, but I see the reason why it in there. For me the only possible reason to put coverflow inside the finder is to use it with a multitouch input device because coverflow with a mouse is just painfully slow but I can imagine it to be great to flick you fingers through your files this way. Generally the finder window is MUCH better then before. It seems more ordered and once you get over yourself and accept new things bring change it is actually quite usable. And, hey, individual windows REMEMBER there setting now. So when you set folder to list view next time you open it it actually opens in list view. It took them 7 years to get this functionality in since the seeders posted a million and one Enhancement requests about it but hey better late then never. Sadly so it still does not remember where the window was placed last time you opened it – this is still the single biggest problem for me in the finder interface. Oh and still no tabs in finder windows :( The underbelly of the finder did improve magically in the last two - three builds. Still no "live" finder but you can at least reliably force reload folder contents by reopening the folder (no that did not work before). And generally it seems a tid faster when copying and it does not lock up when there is a server missing (at least not more then 20 seconds – much better then the 10 minutes it used to under tiger).
Screensharing rocks - its solid mature and fun. Not much more to say about it. Its one of the things you ask "why not earlier" but the clean integration and the speed and the quality of the screen cast is just what you would expect nothing more nothing less and all without installing anything – I can see Apple going a "virtual desktop" route with this sooner or later, I strongly expect more on this front (true virtual desktop as they had in NeXt). I have a space with my server and my two g5 on my Powerbook and can control the whole office with just clicking "ctrl 5" which gets you to space 5 and to full c to your comps – again GREAT.
The rest is all small things scattered throughout the system. Especially the System Preferences have seen some great improvements, like better Networking features. An advanced tab for user management to set you home folder to somewhere different (YAY! no more netinfo mangling). Filesharing with access lists also works incredibly well and you can set sharepoints which is an appealing concept (so it does have its rough edges).

There is a LOT of things for developer, since I am not one I can not comment to detailed on them. But one thing that is great for me is that you can now use RSS in Quarz Composer files that are not Screensavers (something that was disabled in Tiger because of the fear of intrusion through that door). So you can make a Quarz Composer Composition with an RSS feed and have that play through Quicktime on a webpage.

And while I am at it: Quicktime. Its the same old f*$k s%&t under the belly – even worse I have reason to believe it has seen a serious performance decrease. If you open lots of movies (above 200) it crashes it drags down the system and generally I would hope one day Apple takes the gagantuan task to replace it with modern code - they might do just that with the QTKit but its not beeing used by QTPlayer yet it seems or they just copy and pasted code from old Quicktime to QTKit. They should just use the codebase of VLC and make Quicktime opensource and have the coder/decoders closed or something - the current state of affair is not acceptable.
BUT I do like the new overlay interface acrros the board. Very great interface desging not getting in the way and absolutely minimal but usable. Great. Same goes for the new DVD player overlay interface.

Oh and I a