Tuesday, October 28, 2008

Surf The Crowds: Character Contest

Alright, we now have a clear forerunner in the contest for funniest character head (but I think we can still do better). If you don't have a clue what I am talking about, check out yesterday's post.

Click here to view the current leader.

Some people didn't quite get the joke... here is a hint:

Monday, October 27, 2008

Surf The Crowds: New Game from the Basement

I'm very excited to announce that the Basement's first game went live today. Please check out SurfTheCrowds and let me know what you think (good, bad, and even the ugly). You can also find the link at KFCRocks.

I had a little fun with it this afternoon and took the liberty of placing some familiar faces on the character. I tried to keep a Flash theme but I might have strayed a bit:
Joshua Hirsch
Grant Skinner
Keith Peters
Chris Allen
John Grden (\m/)
Seb Lee-Delisle (you shouldn't put those types of pictures online, it may come back to bite you ;) )
Sarah Palin (couldn't resist)
Jesse Warden
Lee Brimelow
Aral Balkan
Dr. WooHoo (hair and all)
Steve Fulton (thx to both Steve and Jeff for giving their feedback/advice early on)

And then I can't post all those heads without doing it to myself... so here you go.

If you come up with some good characters, please post a comment with the link (you can obtain the link by e-mailing it to yourself from the win/loss screen) or send them my way: mark AT thebasement.tv. I am looking forward to seeing what you guys come up with, and of course I'll share what I think are the top ones.

Also, if you think you got a high score, please feel free to brag. Would like to see how high people can get it up to. Hints: You score more for being more accurate and hitting the arrow in the center. You get docked if you hit the keys extra or hold them down.

Hope you enjoy.

Sunday, October 26, 2008

Head Conference Wrap Up: Day 3

Previous wrap-ups:
Day One
Day Two

I didn't get off to a good start on day three. Caught only the last 10 minutes of Mario Klingemann's presentation... there is a big DOH! But will definitely be catching his presentation later (what I did catch was very interesting)... who starts these presentations before noon anyways????

The first full session I caught was Papervision Simplified by Seb Lee Dilisle.... sorry Seb, I had to. It is really spelled Seb Lee-Delisle, and he is a good sport about it). Seb is the Technical Director at Plug-in Media and manages Brighton's Flash user group.

He started off by showing us his Plug-in Media site, which has some very cool AS2 3D effects going on. There is a great 3D soccer game buried within the site... all done in AS2, which is pretty amazing.

From there he got into some of his papervision projects. He showed a Baseball simulation for MLB (someone brilliantly asked why the hell are Brits doing a baseball simulation??? which led to some fun heckling). Anyways, the simulation was very cool. He pointed out a few tricks along the way, such as baking lighting into textures to save processor as well as using a skybox to avoid huge ambient scenery. He then went on to show us a Big/Small character game. Its a very slick interactive game targeted towards 4-6 year olds (but you can bet that I will be playing it too :) ). It looked incredible. One example he showed us he was doing some finger paintings and then just tossed the paper on the floor. For the rest of the game, his painting just lay there in the center of the room. Is the little features like this that really make projects shine. Looking forward to when it goes live.

There was not much time for QA, but he and Mr. Doob (who was attending) did briefly touch on the Flash Player 10 branch of papervision. The bad news is that the new branch does not see any performance increases from using Flash's new native 3D capabilities. The good news is though, that it does improve how it looks.

Apparently the speech was similar to his FOTB's presentation and if you would like to view it in full, please check out Peter Elst's blog post here.


Next up on the agenda was Carlos Ulloa on "The best way to predict the future is to invent it". He mainly focused on project timelines and pointed out the importance of adding time for research and prototyping in the timeline. He described the current project timeline as starting with some art concept work done in photoshop to be proof of concept. The developer may have some say in what works and what does not, but at this point in the project he may not truly know. From there the developer is expected to produce the project, but in many cases unforeseen problems arise and the final product is different than the photoshop file that was originally shown to the client.
His solution is to not worry about the art until you have a working prototype. Spend a significant amount of time at the beginning of the project on having the developer do some research and then produce a working prototype of what the final product will be. The benefits here include:
- the client should have better expectations of the final product.
- The Art can be made for the prototype. This is key. Instead of finding out late in a project that the art needs to be scaled back or compressed, everything should be made to fit what is already known to work. The Designer and Developer should be working very closely together at this stage in the game.
- By leaving time to research, you may find ways to make the final product even better than expected.

During the Q/A session he mentioned that there is no minimal size for a project to fit within this scheme. Even if its only a 1 month project, you could use the first week on prototyping.


Unfortunately Ralph's session did not happen. Not sure why, but both his presentation and Peter Elst's presentation will be recorded and shown later (similar to Keith's). At least that gives us some more content to look forward to :).


Next I attended Dr. WooHoo's Generating Artwork. Holy Crap! All I can say was that it was mind blowing. He was using Flash swf files (via ExtendScript and Adobe PatchPanel) to create tools and panels to extend the capabilities of photoshop, illustrator, and the rest of the Creative Suite. I had no clue you could even do this... it was ridiculous. He even cracked open Maya to create some 3D coordinates, copied some of their xyz data, and then used them in one of his custom tools to create this wicked Generated Art Video. Its hard to really describe what all was shown or said. For most of the session I just sat there with my jaw dropped, thinking wait, you can't do that. Oh man, do I have some research to do.


I ended my conference experience by watching Simon Wardley redo his talk on Why "open" Matters from innovation to commoditisation (which he did on day 1). And by redo, I mean just show the video from Day 1 while he heckled himself and took questions from the audience. He even made paper slides to show on his webcam during the presentation and was able to "RickRoll" Aral... priceless.

I'm glad we ended with this presentation. It really displayed one of my favorite aspects of the Head Conference. It wasn't what he said, but just how everyone interacted. Everyone joked around. You could type questions or jokes and would get a reaction right away from the presenter (this was especially seen in Seb's presentation... he was easily distracted... which made it fun). I had assumed that an online conference would lose some of the interactive feel, but I think we actually gained some via the chat and webcams. Overall it was a great experience and one that I plan on doing again next year. I highly recommend it.

Saturday, October 25, 2008

Head Conference Wrap Up: Day 2

You can find Day 1's wrap-up here.

Day 2 was very interesting to say the least. How many people can say they viewed a talk in which the speaker was sitting in a car in Manhattan while stealing internet connection on her laptop? I missed why Aleks Krotoski was stuck in this position, but it was very entertaining to watch to say the least.


I started my day by watching a great presentation from Joshua Hirsch, the Minister of Technology at Big Spaceship. He warned that people may think he is just promoting his company, which of course he is doing in part, but he is also giving some great insight into how a very prestigious creative firm works. His presentation struck close to home since The Basement is going through some similar situations that Big Spaceship faced roughly six years ago. He hammered home the importance of staying small, working close together, and focusing on producing quality work instead of making money (money will follow).


From there I took a break and then attended Keith Peter's presentation. Out of all the presentations, I have been looking forward to Keith's the most. His Bit-101 lab was one of my main inspirations when I first learned how to code. Unfortunately, the "curse of Room 3" got the best of Keith and he was not able to have audio or video for more than 10 seconds at a time. Soooo, it quickly become an QA chat session and he promised to post his presentation online at a later date. Everyone seemed to take it well and were good sports about it.
We still were able to get a few nuggets from the QA session:
- His next book is coming out early 09.
- In CS4 IK Bones are mainly used on the design side and are barely references in the AS API. (of course you can make your own IK system).
- Infrared5 is wrapping up the following site.


The final presentation I attended was Chris Allen's talk on Red5. Luckily for Chris he was not in the dreaded Room 3 so his audio and video did work. Unluckily for Chris, Adobe Connect used some of the same ports that his Red5 server did and therefore he ran into some troubles. Taken out of context some of his quotes where quite hilarious... "Red5 is impossible to install." and "Red5 application does nothing". This was of course being said while drinking a Hoegarden on being heckled in the background by Keith and the rest of the Red5 crew. Needless to say it was a very entertaining speech... unfortunately I did not take away much from it due to the distractions. I did like the fact that they are making it so you can program server side code in AS3 (everything gets translated into Java). I also enjoyed hearing that Red5 will be a plugin for Eclipse (if it isn't already). And I found it interesting that in contrast to Big Spaceship (where all 45 employees work in house), infrared has 7 in house and 7 working remotely.


My favorite part of the HEAD Conference so far is how easy it is to interact throughout the session and during QA. Everyone acted very polite and mature and any heckling was does in good nature and only added humor instead of distracting from the conference. It was very easy to throw out a question and presenters were able to answer them almost immediately. I was very impressed at how "interactive" and "engaging" an online conference could be.

It may not have the networking that other conferences offer. But I was happy to run into a few familiar names. Hopefully I will "see" more of you tomorrow.

The Basement Design + Motion: Graffiti

Brian and some of his friends did an excellent job spray painting the entrance to our new office...

Its even more impressive in person... hard to capture a hallway with a camera.

Here are a few other randoms photos from us working hard at the Basement.

Friday, October 24, 2008

HEAD Conference: Day 1

Unfortunately I was not able to attend to much of the conference today since I was at work. I did have it on in the background, and was able to jump over and catch a few interesting tid bits.

So far I have been really impressed with the ease of logging in and ability to view and interact with the speakers as well as other attendees. My main complaint is the audio quality. At times its great, but it had a tendency to cut out, get feedback, or just get garbled. I was also thrown off a bit by the mysterious London Hub Schedule hidden away in Room 1. It had a great lineup, but was not displayed on the main schedule page. Luckily I caught it fairly early on.

As far as content goes, I got the most out of Simon Wardley and Tim O'Reilly. I only got to catch part of Simon's presentation (hope to catch the rest later), but what I did watch was insightful and entertaining. He made some interesting points about how innovation is built on the commodization of innovation.... say that three times fast. He also had a couple brilliant slides that described cloud computing and used an analogy of our economy to show how it could break down.

Tim O'Reilly closed the London Hub's schedule by discussing a wide variety of topics. My favorite nugget of information from this discussion was his answer to how Web 2.0 occurred and when will we reach web 3.0. He stated that Web 2.0 is not actually version number and should not be treated as such. It was first used after the dot com bust to state that the web is back... thus Web 2.0. It wasn't a milestone or a version, but rather a statement. To me, that makes complete sense. I never liked the whole web 2.0 garbage. The web will continuously evolve and it is almost impossible to set version numbers on something so dynamic/organic.

That covers day one. Hopefully I will get time to go back and watch some of the presentations with my full attention.

Now I am looking forward to two more days worth of information. Here are the main presentations I want to watch live... some of the rest I hope to catch later:

1pm: Big Spaceship Digital Creative Agency

3pm: Advanced Action Script Animation (Keith Peters)
4pm: Building Red5 Applications (Chris Allen)

10am: Here be Pixels (Mario Klingemann)
11am: Papervision3D, Simplified (Seb Lee Delisle)
12pm: The Best Way to Predict the future is invent it (Carlos Ulloa)
1pm: Flash 2D & 3D Effects (Ralph Hauwert)
2pm: Generating Artwork (Dr. WooHoo)

I'd also like to watch the Game+Play topic, but I can't pass up an opportunity to listen to Keith Peters, especially when he is going to cover his book which he is almost finished with. And I would watch the SWFObject 2 Master Class... buts its at 9am. Lets be realistic here :) Hopefully the recordings will be good enough to watch later.

If you are attending this weekend and see me in one of the rooms. Feel free to drop me a message.

Saturday, October 18, 2008

Flash Game Performance: Browser Comparison

Its getting crunch time for The Basement's first Flash game. We are under an non-disclosure, so my hands are tied as far as releasing details about the who, what, when, where, why, and how. (Although you can check out some of the character art at Amy Kingman's blog).

With the deadline quickly approaching, I am spending a majority of my time squashing bugs and trying to squeeze out a few more frames per second out of the game. One thing I noticed while testing was the vast differences in performance between browsers.

I am running XP and am comparing Chrome, IE 6, and Firefox 2. I am using two different builds of the game: an older one from Friday vs the latest build that has gone through a few optimizations. I made sure that I had no other programs running and only opened 1 tab in 1 browser at a time to ensure there was no noise or discrepancies.

Firefox 2:
Friday Build - 22 FPS
Latest Build - 25 FPS

IE 6:
Friday Build - 29 FPS
Latest Build - 32 FPS!

Friday Build - 24 FPS
Latest Build - 29 FPS

I got to admit I'm all about Firefox... but its hard to argue against these numbers. I haven't taken the plunge into Firefox 3 yet, mainly due to the lack of Flash Tracer support, but once I do I will update this with my findings (also will have to jump into IE 7 as well to make it fair... altho IE seems to be doing fine without updates).

It is also interesting to note that both IE and Firefox increased by 3 Frames Per Second while Chrome jumped 5. Not sure why... which isn't helpful. I can tell you that most of the improvements involved creating smaller bitmap areas to be drawn, using object pools, cutting back the items in the pool, and removing clips from the stage when they were not being used. Not sure why Chrome would handle these improvements differently, I guess I need a better understanding of the bottlenecks of each browser.

If anyone has any insights on the matter, I would love to hear them.

After being reassured by the comments below that Flash Tracer did actually exist for FF3, I went ahead and downloaded it as well as IE 7. Here is what I found:

IE 7:
Friday Build - 29 FPS
Latest Build - 32 FPS

Firefox 3:
Friday Build - 28 FPS
Latest Build - 34 FPS

IE 7 didn't seem to add any performance boost to the game, but IE 6 was already so fast that it did not matter. Firefox 3, on the other hand, gained six fps on the old build and a whopping nine on the Latest build! The swf was published at 35 fps, so the 34fps from Firefox 3 is incredible.

Sunday, October 12, 2008

Papervision 3d Project: What went wrong, What could be improved

I spent some time this summer working on a new homepage for the basement using Papervision. It was mainly a learning experience and the project never saw the light of day... partly due to the fact that it hogged cpu and partly due to the face it didn't turn out quite like we had hoped. You can see for yourself here.

The only objects in the screen are 3D planes and custom 3D geometry. The oscillating billboard accepts animated movieclips as materials. All movement is either with the Camera or Tweener.

After taking Grden's class I think I could have improved performance in a few different ways:

1. Instead of using animated movieclips as the material, I should have used bitmap materials. I could have maintained the animation by using blitting techniques and storing each frame as a bitmap.

2. I should have made all bitmap dimensions to be a power of 2 to take advantage of Flash's native mipmap capabilities. For example: 16x32, 64x64, 128x256... etc.

3. I should have set the Stage's quality to Low. With papervision, Low will look the same as Medium and High and will only require one render pass.

4. I should have used separate viewports for the foreground and the background. In the sites current stage, the entire page is constantly being rendered even when the camera is not moving. The foreground needs to be constantly updated since the "billboard" continuously rotates. The background, however, only needs to be rendered on camera movement. By separating the two viewports I could have saved some cpu when the camera is not moving.

I don't have the time to implement these techniques on the project mentioned above. But I will definitely take them into consideration for future papervision projects.

Sunday, October 5, 2008

Uses of Flash

One of my pet peeves lately has been reading articles, posts, and comments about the limitations of Flash and how it should only be used as icing on websites and not the cake. In the Flash Community, most developers know better, but venture outside of this bubble and you will discover a scary world, a world similar to idiocracy (but HTML pages have electrolytes), in which everyone views Flash as evil and should be used as little as possible.

One of my first tastes of this outside world came when I joined Triiibes, an online community focusing on marketing. After a few heated debates within Triiibes, I wrote this post on why Flash sites should not be overlooked by the marketing community.

Yet still, the majority of the online community still views Flash as a redheaded step child only to be let out in small doses. Just search "uses of flash" and you will see what I mean. The top result is an interesting article from GoogleWebmasterCentral that states how Flash is only meant for Video, icing, and I hate to even say it... Splash Screens. ugh. Seriously? We realize splash screens were a mistake, do we still have to go there?

This negative attitude seems to be spilling over into the Flash community as well. I recently read Lee Brimelow's summary of FOTB. The last paragraph mentions how Jonathan Harris ranted about the current state of the Flash community being too focused on experimentation. Lee does a nice job of pointing out that experimentation "is only a small fraction of what Flash is used for."

Sure we have our fair share of Artsy Experimentation (at lot of which is bad ass by the way... just check out Natzke and you will see what I mean.) And I will admit, Flash does a damn good job w/ video... it is the backbone of youTube and even Hulu. But to say its not cut out for the real world is just plain naive.

All it takes is a few minutes looking at past conference winners to see the variety of excellent work out there. You will see everything from animated shorts, to games, to e-commerce, to educational, to 3D, and beyond. And these types of sites are popping up daily, just make a trip over to theFWA and you will see some premier sites hot off the presses. Whats more is that there is an entire underground group that uses Flash off-line, which is rarely recognized by most people. These underground projects include installations, presentations, and even complex visualizations used by the Department of Defense. I am guessing not many people realize just how far reaching the Flash community exists.

Still, there are going to be haters out there who say, "but HTML has electrolytes (SEO) and Flash should never be used for a full website or application." Really? Can your electrolytes pull off Nike's site?. A site riddled with true interactive experiences backed with deep linking and global language support.

It is about time the outside world realizes there is more to Flash than Splash Screens and that SEO is not the end all be all to what makes a website successful.