Follow Slashdot blog updates by subscribing to our blog RSS feed


Forgot your password?
Slashdot Deals: Cyber Monday Sale Extended! Courses ranging from coding to project management - all eLearning deals 20% off with coupon code "CYBERMONDAY20". ×

Comment Oh, Com'on Robin (Score 3, Interesting) 82

The very best thing you could have done with that particular posting of Eric's would have been to ignore it, and run the story about that nice woman without mentioning it. She can stand on her own and nobody but Eric should be held to account for what he said.

Comment Re:Good idea (Score 4, Insightful) 260

focusing on making a really good browser is a great idea.

Except all the effort they've put into Firefox over the past couple years has been making things worse, breaking things, using more resources, copying the worst parts of Chrome, pushing away users and lowering market-share. I'd rather their focus goes elsewhere, until somebody realizes their mistake.

Comment Re:...would smell as shitty as any browser (Score 1) 115

NoScript is a HUGE improvement, in my experience. However, it's also a big pain in the ass to use, so I wouldn't foist it on my wife's computer for instance.

I think you're just doing it wrong... NoScript can easily be configured to allow everything by default. Then you just tell your wife it's a big "toggle" button she can click on when a site does things she doesn't like... (e.g. pop-ups/pop-unders, autoplay, and other annoyances)

Comment Re:Discussed before (Score 1) 291

Some libraries like libx264 are built for multiple cores especially when used with ffmpeg.

I know them quite well... I've done some minor dev work on them. But with multi-threading, you get diminishing returns and can't fully max-out all the cores, no matter how much you want to, you'll have some idle CPU time on some cores... increasing amounts as you increase the core count.

This is why Firewire and eSATA and Thunderbolt are preferred. All of them have independent chips which handle the work so the CPU doesn't have to do it.

A completely specious argument... The overhead of a USB transfer doesn't drag-down a modern CPU. Take just a few dollars of the money you'd pay for Firewire versions of your gear, and use it to get a slightly higher-end CPU. Now the overhead of USB is covered, you've saved some money, and you're better-off all-around.

Firewire is around because of legacy DV cameras. If Firewire was any good, eSATA wouldn't have ever appeared. But it's still around mostly because it had several years' head-start on USB3.
Thunderbolt is only a thing because Apple has a dog in the fight, so they insist on forcing it on their users like Firewire before it, or PPC processors before it, or now with H.264 and blocking Flash and WebM, or a million other examples.

I've said it is a poor choice for realtime work

Except I already said USB isn't good for real-time... But with most use cases USB can easily do faster-than-real-time, which eliminates the issue.

Your example would be to argue that Toyota Camry is fine for all home improvement because they are everywhere.

No, my argument would be that you've so far been unable to name even ONE single case, where your Toyota Camry has been insufficient for the task...

Comment Re: uh? (Score 1) 142

I found people without degrees or took extreamly specialized classes tend to be good in a small area, then be grossly inadequate in others.

I previously had an incompetent Jr employee under me, who happened to have a doctorate...

Meanwhile, the billionaire founders of Google, Microsoft, Oracle, Facebook and Dell are college drop-outs.

Comment Re:Discussed before (Score 1) 291

The problem with USB is that it is CPU bound meaning that contention isn't as easy to remove as you think. For most video production work, the CPU is busy with processing the video

On a tech site, I'd think you'd know about multi-core CPUs and such... They're pretty common these days.

the practical rate is around 400MB/s (less than 60% actual efficieny).

That's not a question of efficiency, and theoretical gripes don't matter to anyone. In the real world, it's plenty fast, and you'd be hard pressed to point to an actual scenario where someone tried and found they were unable to use USB.

Comment Re:Proprietary charging cables are devil's work (Score 1) 291

Why would you need 12V for it to work for laptops?

Because at 5V you need to pull 3X as many amps over the wire to meet the power needs... Probably MORE, with conversion losses.

So, for a laptop that draws 90W (pretty common), at USB's 5V, that's over 20amps. Do you know what a USB cable that can supply 20amps will look like? Here's a hint: Those big long heavy orange $20+ extension cords are only rated for 15amps.

Comment Re:Unitasking cables are dumb design (Score 1) 291

If you are one of these deluded people who thinks we can do everything through wireless then you couldn't be more wrong.

I never said that, and your feigned outrage as you distort what I said isn't helping your argument.

When it comes to phones, yes, we could eliminate data cables. We certainly don't need them often, so it's ridiculous to pretend we need data (rarely used) and power (frequently needed) combined.

Perhaps you aren't old enough to remember every frickin' cell phone vendor shipping their own unique power cable.

Yes I am... and they were ALL combined power + data. If they didn't tack data onto it, they could have used simple barrel connectors. That's how you end up with proprietary connectors... tacking-on new mis-features.

Who gives a shit?

You do, genius. Don't you remember what you JUST SAID?

"USB just needs to settle on a single un-keyed connector that can carry enough power to run a laptop" -sjbe

Single function cables are idiotic, wasteful and unnecessary in the vast majority of cases.

No, in fact we've got decades of history to prove they're the best options, all-around, every time.

Demonstrably not true and completely missing the point.

Nobody has ever demonstrated that, because you're spouting nonsense. Tiny pins can't take a fraction the force of big solid metal barrels. And you're missing the point that broken USB cables, and worse, jacks, are a HUGE and ridiculously common failure mode, resulting in tons of money being wasted, and a great many phones being thrown away for no other reason.

Having to carry a special quasi-unique power cord around everywhere is idiotic design.

Laptop connectors don't have many variations, and they could easily settle on a single one, if just a fraction the government interest that was put on phones to get micro-USB MANDATED, was directed towards other devices like laptops.

Barrel connectors have their uses but powering a laptop, tablet, cellphone or other mobile device should not be one of them. It is wasteful, unnecessary, and provides no meaningful performance benefit.

You're actually describing USB... It has its uses, but it's a clumsy and fragile piece of shit as a power supply. Barrel connectors are nearly the perfect design for just such a task. Hell, the headphone jacks one cell phones tablets and laptos, which nobody ever has trouble with, is proof enough of the durability of such a simple connector design.

Comment Re:Discussed before (Score 1) 291

For years, firewire buried usb in terms of hard drive transfer speeds. (solid 800mbps vs laggy 480mbps) But it was during this time that PC motherboard manufacturers stubbornly refused to put firewire ports on their boards. Also during this time, it was difficult to impossible to boot a windows computer from an external drive. These factors led to almost a decade of time where the macs were the only computers that commonly used high performing external storage.

Uhh, not really... Even Apple balked at the expense of Firewire-800, and mostly stuck with 400, except in their most expensive products.

When Firewire-800 was introduced, hard drives couldn't even keep-up with Firewire-400 or USB2 speeds. Even the 15,000 RPM Seagate Cheetah drives introduced around 2002 had maximum throughput of 60MB/sec, while their average was lower.

Firewire-800 was 2002... Around 2004, higher-end PCs started getting eSATA connectors that were at worst TWICE AS FAST as Firewire-800. And yes, as above, eSATA was much faster than necessary, too. Most hard drives, particularly external 2.5" ones, weren't seriously bottle-necked by USB2 speeds, which is why USB3 took so long to materialize.

Comment Re:Discussed before (Score 1) 291

But I thought that USB 2 was inadequate, or rather, that High Speed USB was inadequate, and that they needed Super Speed USB (3) in order to get video transfers enabled over USB

USB2 is just fast enough to about max-out the transfer speed of a Class-6 SD card (i.e. real-time), which is on the high-end and more than fast enough for 1080 video recordings (can usually get away with Class-4).

You'll only need speeds above that for all I-frames/stills in 1080 (e.g. for studio use) or 4K recording... And even then, you could still transfer the data over USB2 if needed, just at less-than real-time.

Comment Re:Proprietary charging cables are devil's work (Score 1) 291

Power and data can and should go over the same cables.

We no longer need ANY CABLE for data. Your saddling yourself with restrictions to satisfy a long-lost need. And...

Proprietary charging cables are wasteful, annoying, redundant, and unnecessary

It's only where POWER and DATA go over the same cable that we end up with horrible proprietary crap! If phones had power-only cables, they'd have been simple, standard barrel connectors.

Now USB just needs to settle on a single un-keyed connector that can carry enough power to run a laptop

Laptops have had that forever... Their simple barrel connectors can pull 200W+, no trouble at all. And no USB connector will ever be 1/100th as durable as a tough, simple, basic barrel connector.

There's only ONE THING I give USB credit for... It sucked the air out of all the subtle variation of voltages around its range. Instead of devices that needed 3V, some that needed 6V, and others that needed 4.5V, and even with some that wanted 7.5V, now all those devices will have mini-USB plugs (usually not micro-usb), and run on 5V.

Of course if every smartphone charger on the planet switched to 12V, it would have the same effect on laptops, and all other devices that use any voltages anywhere around that range.

Comment Re:Not replaced: serial and parallel ports. (Score 1) 291

It is also unnecessary 99.9% of the time. Nearly all RS232 devices and hosts will work just fine with TTL voltages (+5V/GND).

I wish that were true, but it sure as hell isn't!

Sure, MANY devices accept TTL voltage RS232, but a big number DO NOT. So if you depend on that USB-RS232 adapter, you'll be left standing around, looking like a moron.

Just had to configure a brand new smart PDU a while ago... Absolutely no activity to my company laptop via USB. Bad cable? Incorrect wiring? Defective PDU? Nope... Plugged that cable to a server with an actual RS232 port, and it works flawlessly.

I wish I could find USB-RS232 adapters that actually output the proper voltages...

A rock store eventually closed down; they were taking too much for granite.