Stuff that fries your potatoes with olive oil of HAPPY.

[Spectere]

Holy fucking shit, the IE team has been pulling some overtime.  IE9 is officially far more interesting and progressive than both Firefox 3 and 4 combined.

Only thing I wish they'd do is move the address bar up and the tabs down, or vice versa.  Or at least give us a means to do that.

But major credit is due; between the sheer speed of that motherfucker (started up Chrome-fast, no joke), the silky smooth scrolling that it manages thanks to its new hardware acceleration, and its amazing progress that the IE team made in terms of standards compliance, I think Trident just about has enough fuel to dethrone Mozilla in the Windows world in terms of rendering engine quality and performance.  IE8 was a good step forward; IE9 is just a miracle.

Now I need to give it a whirl on a sub-standard platform.  Say, my laptop.

Edit: FYI, here's the error listing that I get in IE9:

Code: [Select]

---------------------------
Message from webpage
---------------------------
Failed 5 tests.
Test 26 passed, but took 577ms (less than 30fps)
Test 69 passed, but took 2 attempts (less than perfect).
Test 75 failed: Object doesn't support this property or method
Test 76 failed: expected '0' but got '100' - Incorrect animVal value after svg animation.
Test 77 failed: Not implemented
Test 78 failed: expected '90' but got '1.9237771034240722' - getRotationOfChar(0) failed.
Test 79 failed: Not implemented
Total elapsed time: 2.47s
---------------------------
OK   
---------------------------
The only time that the test really stalls for a perceivable length of time on my system is test 26, and the length of the stall varies.  Total elapsed time is anywhere from around 1.4s to 2.5s.  It's not quite as smooth as Chrome (which gets a perfect score on my system in both performance and, of course, rendering) but performs about as smoothly as Firefox 3.6.9 stable.

The only rendering error is that the shadow under the "Acid3" text is not displayed and, of course, the blue mark isn't completely blue due to the failed tests.

Edit: Here's a SunSpider benchmark: http://www.favbrowser.com/ie9-platform-preview-4-beats-safari-5-acid3-95100/.  IE9 platform preview 4 (which is what the beta is based on) kills Firefox 3 and 4 in JavaScript performance, edges out Safari, and closes in on Chrome.

The browser wars just got a little bit more exciting, I think.

Of course, real world use is far different than synthetic benchmarks.  So far, with my limited interaction with IE9, it definitely feels far less clunky than IE8.  I'm definitely going to have to test it on lower-end hardware to get a good idea of how well it's going to run.

Edit: Just loaded IE9 on my laptop and tried out a few demos.  My laptop's GPU doesn't support IE9's GPU acceleration, so we're stuck with software rendering here.

Overall performance feels closer to Firefox than IE8.  Firefox was fairly smooth with a bit of choppiness on certain sites with this system and IE8 just stuttered every few lines.  Definitely a marked improvement over IE8.  Both browsers still fall short of Chrome, which somehow manages to allow this system to scroll quickly and smoothly despite my dumb ass turning Aero on with only a RADEON X300.

I went over a few demos from Microsoft's IE test site to compare the speed of Chrome, IE, and Firefox.

In the pinball demo available on Microsoft's site, Chrome ran consistently fast, even with tons of balls on the screen.  IE9 ran at a consistently slow speed, pulling maybe 10-15fps and staying roughly around there...hard to say whether it's the JavaScript engine or the rendering engine at fault there, though I'm guessing that it's the former struggling with the physics engine.  Firefox 3.6 was definitely faster than IE9, but it behaved far worse from a user perspective, going back and forth between sluggish to zippy as the rendering engine failed to keep up with the JavaScript engine.  If I had to guess (there's no framerate counter on the site) Firefox was going back and forth between 15fps and 45fps rapidly, which made me want to shoot myself.  Firefox 4 was slower than Firefox 3.6 (!!) but held a steady framerate.  It was slightly faster than IE, but it certainly wasn't a speed demon.

Another demo, the "speed read" one, showed a very different set of results.  IE9 was able to maintain 2-3 fps by drawing one frame every ~400ms.  Chrome stayed at a whopping 0fps because each frame too 1150ms to draw.  Firefox 3.6 wound up in the middle at 750ms a frame, but I disqualified it because it didn't actually draw anything, so that doesn't really count.  Firefox 4 not only actually rendered the animations (WOW) but its speed matched IE9 on my system.

Next demo I tried was the "flying images" demo, using the "all browser" setting with 36 images and the mouse cursor kept to the far left side at the vertical center of the page.  IE9 stayed within a range of 8-11fps, averaging 10fps.  Chrome pulled off 13-14fps, averaging 14fps.  Firefox 3.6 managed 5-7fps, averaging at 6fps.  Firefox 4 rendered a tad slower than 3.6, managing 4-6fps and averaging 5fps.

Next up is the EMCAScript5 Breakout demo.  There was no framerate counter on it so I'm pretty much going based on my observations.  IE maintained a fairly consistant framerate, with smooth movement from all three balls.  Chrome started out extremely choppy, smoothing out as more blocks disappeared, then exceeded IE9's performance at the very end of the test.  Firefox 3.6 does not have EMCAScript5 support (which is understandable, seeing as it was released a month after it was published).  Firefox 4 performed identically to IE, smoothly animating the balls and maintaining a steady framerate.

And my final test for the night is the IE Beatz test.  It's basically an HTML5-powered drum sequencer with some basic fullscreen graphical effects.  IE managed a "score" (the beats per second that the JavaScript tracker is able to maintain multiplied by the framerate that the render can maintain -- the higher the better, obviously) of 280-300, Chrome managed a score of 20-30, and neither version of Firefox could run it because it can't natively play MP3s (perfectly understandable due to the patent issues).

I'm sure that the Microsoft demos are probably more intended for IE9 than other browsers, but given how far performance drifts back and forth, particularly between IE and Chrome, it seems like they are probably fairly reasonable tests in general.

In conclusion, I think it's safe to say that, despite IE beating Chrome in a few instances, Chrome is quite a bit faster overall.  Firefox 3.6 and 4.0 were an odd pair.  Firefox 4.0 was very close to IE9 in terms of speed and performance, though 3.6 did edge its younger brother out in a few tests.

It's also very telling that Chrome 6 -- a stable release -- overall managed to outperform both of those cutting edge, beta browsers.  It's even more telling that the Internet Explorer team was not only able to catch up with, but in some cases even beat the Mozilla team in just a year and a half (IE8 was certainly no threat for Firefox 3.0 when it was released; IE9 will most definitely be a threat to Firefox 4.0).  I hope for Microsoft's sake that they keep the IE team going in full swing (I remember hearing that the reason IE suffered so badly through versions 6 and 7 was because the IE team was essentially turned into a skeleton crew) and continue to crank out quality releases.  If they manage that, IE10 could very very well give Chrome a run for its money.  If development continues at its current feverish pace, by the time IE9 is released it could very well end up being one of the best browsers in the Windows world.

[Bobbias]

Which chrome release were you using, out of curiosity? I wonder if there've been any changes between that one and the dev release. (aside from the obvious design ones, and such).

[Spectere]

I was using the latest stable release of Chrome 6.  I'm not at home so I can't get the exact version string.

Edit: WAHOO!



Apologies for the annoying lack of focus; my iPhone 3G has a fixed lens.  And yeah, I took a picture of an iPod touch with an iPhone.  Too many i's.

[Bobbias]

You've given apple quite a bit of your money, lol.

[Spectere]

Only $2400 total, haha.  PowerMac G4 in 2002 for $2000 and now an iPod touch.   The phone isn't technically mine; it's my work phone*.

About damn time, too.  I've been wanting an iPod since they actually started offering a decent amount of storage. XD

*Yes, a work iPhone.  It works with Exchange like a dream, surprisingly.  Way better than the Blackberry I used at my last job, too.

Edit: Wow, the Retina display is beautiful.  960x640 on a 3.5" screen?  Smooooooooth!  Looks like I do have a dead pixel, though.  Oh well.

I can still make out individual pixels (apparently my vision hasn't been completely destroyed by a couple of decades of computer use), but it's still the sharpest display I've ever seen.  Apparently an Android phone will also be getting that LCD panel soon, so the, uh, Retina is spreading?

[『フエハデ』]

That I'm finally understanding electrical theory. I am so happy I took on that EE major. :3c

[Bobbias]

That I'm finally understanding electrical theory. I am so happy I took on that EE major. :3c

Which part? (I say that because there's so damn much to electrical theory... And I'm in Electrical Engineering myself )

[『フエハデ』]

How you get watts from volts and amps, Kirchhoff's laws, how resistors work (fucking Ohm's Law how does it work), voltage division, current division, etc.

I've had a very basic working knowledge of electricity before, but the theory gives all of it an overarching sense of logic. I'm four weeks into this class and already quite happy with it, so we'll see where we get to by the end of the semester.

[Spectere]

Looks like I do have a dead pixel, though.  Oh well.

Nope, it just looks a little odd when that part of the screen is black.  Looks fine with every other color.

How you get watts from volts and amps, Kirchhoff's laws, how resistors work (fucking Ohm's Law how does it work), voltage division, current division, etc.

How about the seemingly misunderstood notion that milliamps kill, not volts? :p

[Bobbias]

How you get watts from volts and amps, Kirchhoff's laws, how resistors work (fucking Ohm's Law how does it work), voltage division, current division, etc.

I've had a very basic working knowledge of electricity before, but the theory gives all of it an overarching sense of logic. I'm four weeks into this class and already quite happy with it, so we'll see where we get to by the end of the semester.

Ahh, ok. Yeah, there's tons of stuff out there. All the stuff dealing with capacitance, the laws dealing with magnetism, etc.

How about the seemingly misunderstood notion that milliamps kill, not volts? :p

Well, both are important. Ultimately, it IS the current that kills, but because of Ohm's Law, the voltage is part of the equation. You can't have amps without enough volts. Skin can have a resistance anywhere between 100 kohms to as little as 500 ohms (when high voltage breaks the skin down), so it's not too hard to end up drawing dangerous currents. But lower voltage tends to stay on the skin (due, in part, to the "skin effect", among other things), which means that you could very easily end up drawing what would be a lethal current over your skin harmlessly. Of course, if you have a cut or something, that can change matters. Open wounds allow the current to travel through your circulatory system, meaning that you are REALLY at risk.

The other aspect of electrical shock is that you can be burned by high voltages. Basically, the higher the voltage, the deeper the burns will be.

This is part of the reason I don't want to work on hydro stuff... I'd like to keep as far away from lines that have voltages measured in the kV range, especially ones in the hundreds of kVs. I'd much rather deal with what you see in industrial settings: 600 volts usually the highest you'll see, unless your dealing with stuff like mining equipment, and such.

[Sqthreer!]

That reminds me of something I saw about a guy who's face melted off from an electrical burn and survived. Apparently it's so intense that it actually dissolves the bone and tissue?

Anyway on a happy note, I am feeling more secure about my finances since I got my job back

[Spectere]

post

Yes, I'm well aware of that.

I'm just saying that laymen tend to overreact when they hear about anything -- low-current or not -- throwing out multiple kVs.  If you could die simply from getting zapped with a 50kV source every single person that gets tased would die.

I've gotten shocked by a high-voltage, low-current source (roughly 1-2kV before my hand helped it discharge).  It surprises the hell out of you, but as long as it's a low-current source and it doesn't bridge your heart there's really nothing to worry about.

Anyway on a happy note, I am feeling more secure about my finances since I got my job back

Woo, grats!

[Bobbias]

That reminds me of something I saw about a guy who's face melted off from an electrical burn and survived. Apparently it's so intense that it actually dissolves the bone and tissue?

Well, more or less, yeah, depending on the voltage, current, and whether or not your in the middle of an arc flash.

In case you don't know what an arc flash is, it's essentially an explosion. Only thing is, they are an explosion of superheated copper (or whatever else is currently conducting the rather large current) plasma. They are the REALLY dangerous things, and they can be VERY hard to avoid, since sometimes it can be nearly impossible to check if something is under fault conditions.

That is a particularly good example of an arc flash.

Yes, I'm well aware of that.

I'm just saying that laymen tend to overreact when they hear about anything -- low-current or not -- throwing out multiple kVs.  If you could die simply from getting zapped with a 50kV source every single person that gets tased would die.

I've gotten shocked by a high-voltage, low-current source (roughly 1-2kV before my hand helped it discharge).  It surprises the hell out of you, but as long as it's a low-current source and it doesn't bridge your heart there's really nothing to worry about.

Well yes, the internal resistance of the current source is one of the things that factors into things (which is generally what limits the current, unless the circuit was designed with a current limiter). I've been shocked by a 100kv hand held tesla coil before. It was interesting. I've also been shocked with 120v house power (twice :/). A lot less fun. Pretty damned scary, actually. Now, low voltage current limited sources are interesting. I've likely had 5 or 10 volts running through me a number of times, and I likely never even noticed.

The problem I have with high voltage stuff isn't so much the voltage involved (though it makes it a hell of a lot easier to get zapped) but the fact that 44kv power lines for instance, may be designed to conduct 100 amps, or more. By the time any sort of relaying trips, you can say goodbye to whatever part of you just got hit.

[Sqthreer!]

Wow shit. That guy caught in the explosion looked like he just vaporized o_o

[Bobbias]

If you follow the discussion at the bottom, you'll see people saying that from what they can gather, he's still alive, so that's good. But yeah. What happened is that he was putting a circuit breaker for some high voltage lines in. Those have 3 big posts at the top that fir into a thing inside the cabinet. If there's a fault on the hot side of the circuit, and you rack a breaker in, if the 3 posts don't all begin conducting at the same time, there's the possibility of what happened there happening.

And there's no way to detect if the hot side is under fault conditions.

There are remote breaker racking devices, but they are REALLY expensive, and still quite new, so most places, people still have to do that by hand.

And that's one more reason I don't want to work for hydro companies.