This sure is impressing, but nobody said Flash doesn't have some good features. The question is about whether anybody who wants to can implement a player for this page.
Neat. This example is nowhere near as pretty as the Flash examples, but it's interesting that my CPU is barely doing anything even when walking around. In the Flash version my CPU is pegged just sitting there doing nothing.
however, suppose you are about to design a video site for the common public back in 2005, what technology would you choose? quicktime? realplayer? windows media player?
i dont think flash video is unneeded any soon, even if html5 has a 70% adoption rate, people would still need flash for h.264 fallback
Lack of universal acceptance was one of the cons you either accepted, or naively overlooked, back when you decided dvorak was the way to go. It'd be nice to see it work in more places, sure, but it just ain't... you're conflating 'should' and 'could'.
The issue with ignoring dvorak is that any web browser (heck, nearly any application) will handle it just fine. Flash apparently doesn't, or makes doing simple things (like capturing +/- keystrokes) hard enough that people don't do them in a generic way.
Given that Flash is primarily used in the web browser, and most of the world doesn't use QWERTY, it's rather shameful.
A quick test on my end shows that the Flash example indeed does not respect dvorak, while a Javascript key event captures the same code regardless of layout. Including Japanese. How's Flash handle that?
Alternative keyboard layouts work for pretty much everything. Almost all applications rely on the OS to translate key mappings, so there's virtually no incompatibility to worry about. I don't think I've ever come across an application that didn't work just as well in Dvorak as Qwerty.
It's just plain displacement mapping. It's a very simple (and old), but versatile technique allowing a large amount of effects to be done. The first time I saw this was in 93 or 94, rendered in real time in 256*256 in 25 fps, on an Amiga running a 14mhz 68020 processor. THAT was impressive to me, but seeing this 15 years later in 15-20 fps in Flash on a 2ghz machine just isn't.
Of course the real difference is a lot higher as it's not just about the clock rate - an awful lot of improvements were made in areas such as the pipeline, caching, floating-point support etc. so I wouldn't be very surprised if the real performance ratio was higher than 1000.
Yeah, 16 Mhz ==> 160 KFlops, if you believe wikipedia. Supposedly the i7 has a peak of 79 GFlops. Sure, you can't get all of that performance out (either then, or now), but 79,000,000 / 160 is a bit larger than 143.
You used to measure the number of cycles an integer multiply costed, in the dozens or hundreds. Now it takes 1/4 of a cycle. A factor of 10,000 might be a fairer number
I remember something exactly like this except it was an Asian lady and it looked like a real photo. Also instead of looking away from the cursor her eyes were following the cursor.
It's not slow on modern computers. It's slow on these environments that weren't made to execute code quickly, much less do any pixel manipulation. How many orders of magnitude more instructions is it to set a pixel on an Amiga, versus setting it via JavaScript?
Because it's running in canvas? In other words, it's running within abstraction layer upon abstraction layer, with various other abstraction layers involved to help make sense of the fact that there are too many abstraction layers.
The amiga code was probably pretty close to talking to the graphics driver directly. Canvas is cool, but in terms of performance it's never really going to compete with native code, or even with flash for that matter since flash can remove some complexity by only building for a single, predictable, optimized environment.
> it's running within abstraction layer upon abstraction layer, with various other abstraction layers involved to help make sense of the fact that there are too many abstraction layers.
"Every problem can be solved by adding another layer of abstraction--Except for the problem of having too many layers of abstraction"--Alan Perlis
"Graphics driver" still sounds too complex. The code was probably poking bytes to the memory segment mapped directly to the display - just like in DOS days, where each byte of the 0xA000 segment represented a pixel when using a VGA 320x200 resolution, 256 colors palette.
A bit of context: this is a demo showing parallax mapping, which was a soon-to-be-released feature of the Flash-based Alternativia engine at the time. It's was never supposed to be a state-of-the-art demonstration of Flash.
45 comments
[ 0.26 ms ] story [ 97.2 ms ] threadand
http://blog.alternativaplatform.com/ru/files/2009/02/hero.sw... (press space to rotate)
http://www.benjoffe.com/code/demos/canvascape/textures
http://www.ponticstar.com/projects/burning-words/
Try set the font to 100, and see if it's still as smooth as you thought.
http://vimeo.com/10553088
But hey, Flash is dying, etc
However:
No matter the codec or operating system, we should not NEED Flash to view video on the Web.
however, suppose you are about to design a video site for the common public back in 2005, what technology would you choose? quicktime? realplayer? windows media player?
i dont think flash video is unneeded any soon, even if html5 has a 70% adoption rate, people would still need flash for h.264 fallback
Wolfenstein, meet Doom.
I'd say the bunker example above is pushing Quake 2 levels.
I agree that Adobe have dropped the ball with Flash, but "HTML 5" is no where near where it should be...
Given that Flash is primarily used in the web browser, and most of the world doesn't use QWERTY, it's rather shameful.
A quick test on my end shows that the Flash example indeed does not respect dvorak, while a Javascript key event captures the same code regardless of layout. Including Japanese. How's Flash handle that?
http://en.wikipedia.org/wiki/Motorola_68881
Yeah, 16 Mhz ==> 160 KFlops, if you believe wikipedia. Supposedly the i7 has a peak of 79 GFlops. Sure, you can't get all of that performance out (either then, or now), but 79,000,000 / 160 is a bit larger than 143.
You used to measure the number of cycles an integer multiply costed, in the dozens or hundreds. Now it takes 1/4 of a cycle. A factor of 10,000 might be a fairer number
http://29a.ch/2010/3/24/normal-mapping-with-javascript-and-c...
The amiga code was probably pretty close to talking to the graphics driver directly. Canvas is cool, but in terms of performance it's never really going to compete with native code, or even with flash for that matter since flash can remove some complexity by only building for a single, predictable, optimized environment.
"Every problem can be solved by adding another layer of abstraction--Except for the problem of having too many layers of abstraction"--Alan Perlis
http://blog.alternativaplatform.com/en/2008/01/04/character-...