Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).


Comment: Re:LOL! (Score 5, Informative) 446

by gilgoomesh (#38951857) Attached to: Tapeheads and the Quiet Return of VHS

As a video software engineer, I know your pain from a slightly different angle.

Your "waterfall effect" is over quantization of DCT blocks (in rare cases it could also be misuse of the deblocking filters). It's pretty easy to avoid and most encoders can actually give feedback about quantization rates and whether artifacting will be visible in output frames.

The problem is that people don't know how to use their encoders correctly, use them with completely the wrong settings and then don't inspect the output to see the result.

The MPEG4 High Profile 4.1 used in BluRay discs is capable of practically flawless encoding at any motion rate if operated with a little care. MPEG4 allows custom and dynamic quantization and a two pass encoder can use the second pass to fix any mistakes by adapting the local bitrate and quantization method.

I actually suspect though that you're seeing MPEG2 video getting pumped at an MPEG4 bitrate which is causing massive over quantization. This generally happens when studios have MPEG2 encoding hardware but no MPEG4 encoding hardware but they are told "keep your video at X bitrate" – even though this leaves half the disc empty and the video looking like a stream of 8x8 shiny cubes.

Of course, some decoders don't implement deblocking algorithms correctly and actually *increase* blockiness in some cases. This would be the fault of your BluRay player – you'd need to play on a good software player and compare.

And don't get me started on interlacing in digital video. It's a "feature" that has only ever made digital video worse and is somehow part of most broadcast standards. Aaarrgh!

Comment: Re:I was at the announcement (Score 1) 416

by gilgoomesh (#38758502) Attached to: Apple Unveils Software To Reinvent the Textbook

The iBook app lets you take notes and highlight sections right on the page. It's one of its key features as a "student reading app". You can also export the highlighted sections and notes to create separate study notes.

I don't know how well it would work, of course, but they're trying to capture the same workflow.

Comment: Re:They are talking desktop screens (Score 1) 556

by gilgoomesh (#37728394) Attached to: Is Apple Pushing Away Professionals?

Glossy is also better brightness since every extra bit of coating (and matte is a coating) reduces the light permittivity. Granted though – brightness changes are not matte's primary effect.

However with truly bright sources behind you, particularly direct sunlight, matte can reduce contrast so much that you literally can't see anything.

Comment: Re:Not boron nitrate. . . (Score 2) 50

by gilgoomesh (#37659686) Attached to: Graphene 'Big Mac' — One Step Closer To Microchips

Slashdot correctly quoted the University of Manchester article. It used "boron nitrate" repeatedly and only used "boron nitride" in a quote. I don't know what MagusSlurpy is talking about when he mentions the abstract though it doesn't mention either.

I'm far more disappointed that nothing mentions why we would expect this to replace silicon as a semiconductor.

Comment: This is not the "Australian Censorship Scheme". (Score 1) 212

by gilgoomesh (#36639972) Attached to: Telstra Starts Implementing Australian Censorship Scheme

The title and summary are a little misleading. They imply that this is related to the Australian government's proposed mandatory censorship scheme. It is not the same scheme and it is not being done in the same way. If there is any relation, it is that this scheme is intended to pre-empt any effort by the government to pursue mandatory censorship.

This scheme being implemented by Telstra is the exact same scheme already implemented by UK ISPs BT, O2 and Virgin.

Unlike the Australian government's mandatory scheme, this is *not* a hidden secret blacklist with no opportunity for objection. Multiple law enforcement bodies must agree before anything is censored. There is an appeal process in place. They are only censoring illegal child pornography and only where the victims are clearly underage (guideline is: under 13 years old).

In summary: nobody wants censorship but if this optional, industry managed, minimalist effort dissuades the Australian government from introducing a mandatory, heavy-handed, secretive, broader than "illegal", no-appeal censorship scheme, then it might actually be a good thing.

Comment: Re:Fixed IP addresses? (Score 1) 159

by gilgoomesh (#34192194) Attached to: Dutch ISP Demos Symmetric 100Mbps DOCSIS3

Most ISPs around the world are starting to keep fixed IP addresses as an "added extra". It's nothing to do with cable/DSL/fiber.

There's two good reasons: people will pay more for fixed IP addresses and IPv4 addresses are starting to get expensive because they're running out (dynamic IP addresses can let you cram 10%--50% more users into the same address space).

Get a dynamic name instead -- you don't want to enter a number anyway.

Comment: Re:Why so short bursts? (Score 1) 326

by gilgoomesh (#32370576) Attached to: USAF Scramjet Hits Mach 6, Sets Record

A Mach 6, wind resistance alone is like detonating explosives in the air intakes. Wear is certainly a serious issue.

But that's not what they're looking at in this case -- they're looking at getting the combustion to be stable and controlled. Wear and tear is something to sort out once the vehicle is less experimental.

Comment: Re:I don't want to feed the trolls but... (Score 5, Informative) 353

by gilgoomesh (#29726723) Attached to: Major Snow Leopard Bug Said To Delete User Data

As far as I can tell, from reading this on other sites, the reproduction involves:

* Machine that was upgraded from Leopard to Snow Leopard
* Already had the Guest account enabled on Leopard.
* Logs into Guest account (not a remote login but a local, physical login)
* Is hard-booted (after crash, power failure, or power button) from Guest account back into Admin account.

Despite a combination of these steps, people are finding it hard to reproduce. So it's the sort of issue that could fall through the QA cracks.

With your bare hands?!?