Slashdot: News for Nerds


Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!



Linus Torvalds: "GCC 4.9.0 Seems To Be Terminally Broken"

evilviper Re:Oe noes! A compiler bug! (554 comments)

GCC is terminally broken for ever and ever and ever.

GCC is a mess that has been getting consistently worse since 3.0. It's so bad that compiling GCC with GCC, with any CPU optimizations enabled, produces a non-working compiler. It just keeps getting bigger and slower, and has a great many proprietary GCC-isms that open source developers keep using, not even realizing they're bugs.

The crappiness of GCC has driven tons of people away, and spurred the development of LLVM, tcc, and others.

I've seen fewer bugs in GCC than any other production compiler ever.

Either you're not looking (myopia is fun), or you have very little experience with other modern compilers.

End result, the GCC people will fix this bug in short order

With this much publicity, they might... But major bugs that get reported, but don't hit the /. front page, and often linger for year after year.

The success of Linux is 100% built off the success of GCC. There have been no other credible compilers for Linux throughout the majority of its existence

Only true if you drink rms' kool-aid... Otherwise, any of the proprietary compilers out there would have done the job just fine. Or Linux developers would have put some effort into getting another compiler up-to-par for their purposes if nothing had been available, kinda like they did with the kernel...

And now, the competition is just waiting to break-through and rid us of all the GCC nonsense.


Linus Torvalds: "GCC 4.9.0 Seems To Be Terminally Broken"

evilviper Strange censorship (554 comments)

"pure and utter sh*t," only with no asterisk

So he actually called it, "pure and utter sht"?


Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re: Alternative explanation (368 comments)

Your first paragraph is a pointlessly incredulous waste of time.

The rest is vastly exaggerating the difficulty... There are several very small open source multicast file distribution software packages out there that already do all of this, and work perfectly. It is not a major technical feat.


Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Could be a different route involved for the VPN (368 comments)

Verizon providing free access to Netflix is the logical conclusion of your attitude. You don't care about the technical details, and just want everything handed to you. That gives upstream services the upper hand to demand absolutely anything and everything from Verizon, including making them build-out their network as cheaper ISPs put more burden on Verizon, or even making Verizon pay for access, lest they lose customers.


Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Alternative explanation (368 comments)

There's never been any such exceptions to peering arrangements before. End of story. You think the way the internet has always worked should change, but don't want to admit it.

Consider it the other way around. Level-3 certainly wouldn't be jumping at the chance to offer free-peering with some tiny ISP. Is that a "fraudulent business practice" from Level-3?


Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Alternative explanation (368 comments)

the concept of peering traffic "parity" does not apply to local ISPs connecting to backbone providers.

It always has before.

And Verizon is a backbone provider in their own right, not just a local ISP.


Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re: Could be a different route involved for the V (368 comments)

Who cares about the imbalance?

Verizon and Level-3.

It's the congestion we want fixed.

Good, then complain to Netflix. They can fix it.


Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Could be a different route involved for the VPN (368 comments)

I'm trying, but this isn't the /. I remember. We've been overrun with AOLers, so of course my comment (with all those pesky little facts that get in the way of the anti-ISP ranting) gets modded-down.

I only hope SoylentNews turns out better...


Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Could be a different route involved for the VPN (368 comments)

Stupid? I'm the only one here who seems to know what peering agreements are, and how they've worked for the past several decades.

There's no question Verizon has plenty of bandwidth. The problem here is Level-3 breaking their peering agreement, and not wanting to renegotiate, so Verizon has ever right to disconnect Level-3 and Netflix from their customers. Instead, they let the peering point get congested, until a new agreement is worked out.

It's how peering has always worked. You're the one arguing we need to erase the history of the internet, and turn it into a receiver-pays model, where every site you visit gets a few cents from Verizon.


Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Could be a different route involved for the VPN (368 comments)

In the old days, when peering became imbalanced, ISPs would shut-off peering with each other, bifurcating the internet for weeks until one side agreed to pay another. These days, they just let the peering points get congested, and don't upgrade it.

I'm sure if Level-3 would agree to pay Verizon for the peering imbalance, Verizon would upgrade the peering points, but that would cost Level-3 more in the long-term. Level-3 pointing the finger at Verizon, and shaming them in the court of public opinion, is cheaper than fixing their peering imbalance.

2 days ago

Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Could be a different route involved for the VPN (368 comments)

So it's your assertion that Verizon should pay to run lines to the Netflix data center, and give them all the free bandwidth they can use?

2 days ago

Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Could be a different route involved for the VPN (368 comments)

Peering agreements have never depended on who requested what data. They're much simpler up/down traffic ratios.

Everybody seems to have enough hate of Verizon, and love of Netflix, that they just want to punish the former, in favor of the later, no matter the circumstances.

2 days ago

Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Could be a different route involved for the VPN (368 comments)

Verizon just changed all their FIOS plans to be symmetric up/down, so they're at least less guilty than other ISPs.

In addition, Verizon isn't just customers. They host sites, too. More Redbox streaming going over Level-3 would help to level things out.

Or Level-3 could offer cut-rates to online backup service providers, who recieve a lot of traffic from customers, and only a fraction as many requests. Or Netfliix could change their player to upload junk data to some random server all-the-time, which would help tremendously.

The point remains, no-fee peering has always required roughly equivalent up/down traffic, so the horrible imbalance Netflix causes, is going to cause peering disuptes, legitimately, without any evil conspiracy from Verizon and others. And that's not even getting started on Level-3's poorly concieved CDN, taking money from ISPs while futher imbalancing their peering arrangements.

2 days ago

Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Could be a different route involved for the VPN (368 comments)

That is a load of horeshit technobabble meant to obfuscate and mislead.

No, his explanation is spot-on. If "technobabble" means you didn't understand it, that's besides the point.

There is no reason that any link between it and another network should remain saturated if both sides are acting in good faith to serve their respective customers

Level-3 is acting in bad-faith in a couple different ways. Offering to give Verizon a small amount of money does not obviate the large ways in which they are acting badly, which will make them even more money.

it is Verizon's customers who are requesting and already paying for the content in the first place.

Customers pay both Verizon and Netflix. Both sides are supposed to pay their own costs of transit and bandwidth. Saying that Verizon should acquies to badly-behaving peers is a small step away from saying that Verizon should provide free internet services for every service their customers request.

Verizon is choosing to not upgrade its connections to shake down Netflix

Verizon is choosing not to upgrade it's peering points with Level-3 because they are no longer evenly sharing traffic up/down as all free peering arrangements have ALWAYS required, yet Level-3 doesn't want to pay for the imbalance, and Netflix doesn't want to shift some of their Verizon traffic to a different transit provider than Level-3.

2 days ago

Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Could be a different route involved for the VPN (368 comments)

Netflix offered to provide co-located CDNs, and all Verizon had to provide was electricity and space (both of which are negligible compared to the cost for Verizon to pay for bandwidth.) Verizon elected not to take the option that would save them money

If Netflix got a free CDN setup without paying ISPs anything, Verizon would quickly see all the other CDNs refusing to pay them, too.

On that same note, I know where you can get a FREE 40-hour/week job... You won't have to pay a penny for this FREE job.

2 days ago

Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Could be a different route involved for the VPN (368 comments)

Failing to have peerage agreements in place to honor your downstream sales commitments is a form of throttling - Or, I would daresay, a form of outright fraud.

You're one small step away from saying that Verizon should be giving free internet service to every service that has content their customers want to send/receive.

That's never been how internet service has worked, and forcing Verizon to acquies to misbehaving peers will result in a huge increase in your internet service prices, as everything suddenly becomes client-pays, instead of both sides mutually paying their half.

2 days ago

Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re:Could be a different route involved for the VPN (368 comments)

This doesn't change the fact that the customer paid for 75Mbps and got... a lot less.

1) That's not a news story. It's been happening for a long time.
2) ISPs always warn you that they aren't responsible for the speeds of 3rd party servers, and can't guarantee they'll be fast, too.
3) He paid both Verizon and Netflix. You can make the case for either one of them being at-fault.

2 days ago

Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re: Alternative explanation (368 comments)

Except that if you do that, there is functionally no difference between sending the traffic to a multicast destination and a unicast destination.

Nonsense. With a subscriber base as large as Netflix's, there will always be several people requesting the same video stream, within a minute or so of each other.

Well, Sub2 gets the show 10 minutes later, assuming he's joining the same multicast group as Sub1. Sub2 is not happy, he wanted to watch the entire show.

Actually, when Sub2 joins, his system can start buffering the show 10-minutes in, and while doing so, send a request to resend the first 10 minutes of the video. Netflix servers save bandwidth, and both users watch the video they wanted, when they wanted.

2 days ago

Enraged Verizon FiOS Customer Seemingly Demonstrates Netflix Throttling

evilviper Re: Alternative explanation (368 comments)

Netflix supplies a tower-pc sized box with all of netflix on it to ISPs for free

That's like saying your boss is willing to give you a full-time job for FREE. In other words, failing to pay you... But it's FREEEEE! Don't you feel special, getting something for FREE?

It's Netflix who's trying to get something very valuable for free, out of the deal. Other content delivery network providers, like Akamai, pay good money for the privilege of having their servers hosted by ISPs on their fat pipes near to customers.


2 days ago



Low Power Monochrome LCD screens

evilviper evilviper writes  |  about 5 years ago

evilviper (135110) writes "Ever since I picked up my Psion 5mx handheld many years ago, I was impressed all-around with the clarity and amazingly low power requirements of simple monochrome LCD screens. I didn't even find myself missing color. Now, I find myself tasked with replacing hundreds of dumb terminals with surplus PCs, doing nothing at all but running a simple terminal emulator, and wasting power. With the majority of power being drawn by the monitor, and requirements so modest that switching to a grayscale display would go practically unnoticed, naturally, my mind goes back to those inexpensive and power-sipping monochrome LCD screens. Sadly, all I've found in my searches are old CRTs, and very expensive high-end LCDs for medical imaging.

So where can decent-sized monochrome LCD screens (preferably with VGA connectors) for PC output be found? Even a text-only screen, ala. digital watches, might be acceptable. I'm sure I'm the only person to ever have this need. What do others in similar situations do to reduce the power consumption of largely idle PCs that are needed merely as a small upgrade from dumb terminals?"

Cellphone Signals Used to Track Traffic Patterns

evilviper evilviper writes  |  more than 6 years ago

evilviper (135110) writes "Engineers at the Indiana Department of Transportation have developed a method to track how long it takes vehicles and pedestrians to travel from one point to another. "Now we have a way to measure how slow traffic is on a given stretch of road or how long it's taking people to get through airport security at a given concourse and time of day." Data from such a system would provide not only information about short-term factors such as congestion from construction work zones, but also long-term trends requiring design changes.

Research findings will be detailed in a paper appearing in the June issue of the Institute of Transportation Engineers Journal. The researchers have filed a patent on the method, and the basic technology is available commercially to create the tracking system.

(apologies for the previous submission lacking URL)"

Cellphone Signals Used to Track Traffic Patterns

evilviper evilviper writes  |  more than 6 years ago

evilviper (135110) writes "Engineers at the Indiana Department of Transportation have developed a method to track how long it takes vehicles and pedestrians to travel from one point to another. "Now we have a way to measure how slow traffic is on a given stretch of road or how long it's taking people to get through airport security at a given concourse and time of day." Data from such a system would provide not only information about short-term factors such as congestion from construction work zones, but also long-term trends requiring design changes.

Research findings will be detailed in a paper appearing in the June issue of the Institute of Transportation Engineers Journal. The researchers have filed a patent on the method, and the basic technology is available commercially to create the tracking system."

evilviper evilviper writes  |  more than 7 years ago

evilviper (135110) writes "Everyone's favorite company (SCO) is now distributing everyone's favorite GPL'd Unix video program (MPlayer) for their UnixWare and OpenServer operating systems. With it, they are including numerous codecs in the form of Windows DLLs, from Microsoft, Apple, RealNetworks, and others. This brings up numerous questions about the legality of re-distributing freely downloadable files, and the enforceability of EULAs on the related software. It also raises the question of whether the GPL extends to closed-source, binary code being loaded by GPL'd programs at runtime, such as DLLs or Linux kernel modules. W714%20&%20OSR6-FINAL.pdf .php?rid=270 .php?rid=271"



FTA Satellite Reception in North America

evilviper evilviper writes  |  more than 6 years ago

If you live in the western US, more than 100 miles from a major city, over-the-air TV can be very limited, available only via a weak signal, or entirely unavailable. There, your only choice is satellite.

The question is, do you want to pay $50 a month to a direct broadcast satellite (DBS) service like DirecTV and Dish, or do you want to be free? DBS providers will give you a large number of channels, but I find 95%+ of my TV viewing is shows available on local broadcast channels anyhow. That leads us to free-to-air (FTA) satellite.

Over in Europe, Asia, Africa, there are a large number of channels broadcast unencrypted. The western US, however, is the absolute ass-end of the world... separated from the rest of the world by the two largest oceans in the world, and, with such a minuscule population density that practically nobody is interested in serving the area. See previous point about TV reception.

Point being, with FTA in North America, there isn't a big selection. As a rule of thumb, as an English speaker with a KU band receiver, you can really only expect to get PBS, news, and shop-at-home channels. Things get more interesting, however, in the C-band... There you have a much larger selection of channels, intended to be picked up by cable companies and the like, and include all the major broadcast networks. Even though just about all growth is in the KU band, it's clear there is still a lot of life left in the C band. I recommend a dual, C/KU LNBF. You really don't even need to worry about noise and gain figures for LNBFs these days, unless you're looking to pick up seriously fringe signals.

The second issue is size. I don't consider a dish a good purchase if you can't take it with you when you move elsewhere. The only way to guarantee your future land-lord/neighborhood association/etc. absolutely can't forbid you from installing a dish is to follow FCC rules and buy one that is no larger than 1 meter in diameter. Fortunately, this isn't 80s technology anymore, and 1 meter is perfect sufficient for C-band reception. Furthermore, you can get just a bit more signal if you go with an offset dish, rather than an old, standard, parabolic dish, as the LNBF is not located where it blocks signal reception. Since you're size-limited, it's better to buy an offset antenna, and get the maximum amount of signal possible. In most cases, there is effectively no price difference.

Analog has nearly gone the way of the dodo. 4DTV/VideoCipherII is popular in North America, and a good option if you want to get all the standard "cable" channels (for a lower price than cable or DBS, but otherwise, there's practically no FTA content, so it's quite safe to skip it. That just leaves one option, the one that has quickly taken over most satellite transmissions, worldwide: DVB-S... For about $40, you can get a nice PCI DVB-S card, and turn any old PC into a TV tuner, or DVR. But, if you're not very technologically competent, there are also $150 "FTA receiver" set-top-boxes that will do the job as well, if less flexibly.

Though I'm really not interested in it myself, I should at least mention that FTA is commonly used as a euphemism for illegally descrambling encrypted satellite TV signals. Now, DirecTV and 4DTV broadcasts are immune, partly because they are proprietary, and relatively few people have interest in. Dish Network, however, uses standard DVB-S equipment, and NagravisionII encryption, which can be decrypted with "softcam" programs that are available. Of course these things are difficult to find, due to the risk of lawsuits under the DMCA, but "softcam" software certainly exists both for computers (Windows and Linux) as well as "unofficial" firmware for damn near all "FTA" receivers. Yes, it is a bit of a shell-game, and no doubt the vast majority of FTA sales are based on this.

To see what satellite signals are freely available to you, visit lyngsat:

Just be sure to take that listing with a grain of salt... Sometimes, that list shows signals available that are actually from satellites on the opposite side of the planet (see: Playboy One).

You'll also need standard equipment like a dish rotator, that is strong enough to hold the weight of the dish, and communicates with the receiver in some standard they both understand, but that's pretty straight forward to check on. Not to mention good quality coax cable, and absolutely no splitters in-line, or else you'll cause some serious damage.

Buy a compass, and mount your dish so it has a perfectly clear view all across the southern horizon... If there's a tree or a building in the way, you've got almost no chance of getting any reception at all. You can mount it to your roof, or to the ground, but either way, make sure your mount can handle hundreds of pounds of force, without the dish shifting even a fraction of a inch, otherwise you'll have picture drop-outs when it's windy. Manage all of that and, barring heavy commercial air traffic, you should get a perfect signal, all the time. The dish is far larger than it needs to be for KU-bands, enough so "rain fade" shouldn't occur in even the worst storms.

Hopefully this was helpful for those of you considering purchasing a satellite dish. If you still have any questions, feel free to ask in the comment section.


HDTV Reception: Everything You Need to Know

evilviper evilviper writes  |  more than 6 years ago

Both ignorance and misinformation abounds about the coming switchover to HDTV in the US. I will cut right to the facts.

Right now all major TV stations have two transmitters, their old/original, and a new, secondary for their DTV broadcasts, which is always lower power, on UHF. As of February 2009, they will be shutting off one of the two, and broadcasting digital-only. The FCC has allowed them to select which of the two they wish to continue using. At this point, it's largely settled. VHF-low is going away.

Practically all broadcasters on VHF-low (channels 2-6) are dropping their existing channel assignments, and switching to a higher (UHF) one. There are very few exceptions. There are also (much lesser) signs of broadcasters leaving VHF-high (7-13), but for the time being, there are lots of VHF-high channels, so you will need an antenna that can receive channels 7-13.

Also important is the fact that ATSC doesn't handle multi-path interference (ghosts) very well. Unlike current analog NTSC, it won't just make the picture look worse, if it's significant, it will prevent you from getting a picture at all. If you are in a city with tall building or big hills/mountains around, this is very important in the selection of an antenna. Older and more common (rectangular) VHF antennas are less directional, and can do little to suppress multi-path interference.

Also important: The FCC is removing UHF channels 52-69. There will no longer be TV broadcast in that range.

STRONG SIGNAL (0-20 miles):
If you are just a few miles from all the TV stations you want to receive, and there are no obstructions in the way, you would probably do well to just get a TV-top antenna that does both VHF and UHF (rabbit ears for VHF and a loop or bow-tie for UHF).

MEDIUM SIGNAL (20-40 miles):
If you use an older rectangular VHF antennas, multi-path interference on VHF signals can pose a problem. Don't get an antenna that is large, and/or highly directional with high gain (db) or else it will be unnecessarily difficult to install and aim the antenna. Hopefully all the channels you want to watch are coming from the same general direction, otherwise, you'll need a (low-end) antenna rotator to get a good signal or possibly two antennas pointed in different directions. I suggest a Winegard 4400 (4-bay) for UHF, and possibly an AntennaCraft Y5-7-13 for VHF-hi. To connect them, DON'T USE A REGULAR SPLITTER or you'll ruin your reception for no good reason. You want a $2 UHF/VHF combiner, commonly made by Pico Macom/Blonder Tongue/Holland/etc.

The most popular type of UHF antennas, yagis with corner reflectors (they look like big metal arrows), design is such that they perform poorly at lower UHF channels, and best at the highest UHF channels. One of the antennas I was considering buying only starts to perform well at UHF channel 50, channels which won't be around in a year.

With that in mind, I recommend an 8-bay antenna instead... One of the top performers which is also far less expensive than other 8-bay antennas is the Winegard 8800. 8-bays perform quite well at the lower range of UHF frequencies from 14 to 50. A few 8-bay antennas claim to get good VHF performance, but it's all misinformation (which I also fell for). At best, they only barely outperform the most basic rabbit ear designs, If you're far enough from UHF stations that you need an 8-bay, you'll also be far from VHF stations, and need a GOOD VHF antenna to bring in the signal. A barely-capable-of-VHF 8-bay simply doesn't have enough gain to get decent VHF reception. A $20 VHF-high antenna will blow them all away. Use a VHF/UHF combiner, NOT a standard splitter to connect them.

VHF-hi needs much smaller and cheaper antennas, and VHF-low is almost gone, except in Alaska, where stations are keeping their low VHF stations due to terrain.

There's one other advantage that multi-bay antennas have over yagi/corner reflectors. If you are renting an apartment, you are allowed to install any antenna you want, provided that it resides entirely on your property. Being able to mount a long yagi/reflector without it sticking out (no longer your property) is pretty unlikely, unless you have a huge balcony. Multi-bay antennas are quite flat, and even just an outside window ledge could be used to mount it legally. Just make sure it's mounted securely, because they are heavy. If there's no location for installation outdoors, you could also install it on the inside of a window, facing out, perhaps hidden from view by your curtain... a yagi would have to stick out, halfway across the room.

There's no substitute for putting your TV antenna on a nice long pole, hooked up to your HDTV tuner, and walking around your house with it until you find a location that gets a strong signal on all the channels you care about. A simple procedure, but a lot of time-consuming work. An antenna can be mounted up to 12' above the top of your roof without requiring you to apply for any permits, and in general, the higher your UHF antenna is, the stronger the signal.
Stay far away from power lines. Be sure to ground both the antenna pole and the coax cable outside of your house to prevent lightning damage.
Large multi-bay antennas are very heavy, so be sure to get heavier gauge pipe for mounting it.
If you are in a fringe area, and your cable runs aren't pretty short, you'll need a mast-mounted signal preamplifier. Lower noise (2.0db) is better, and only slightly more expensive than the rest. Gain (eg 15db) will hardly be an issue UNLESS you have extremely long cable runs... Only then, a second amp might help.

I'm looking forward to HDTV, not for the resolution, but because 95% of the shows I care about are on broadcast TV, and signal quality (fringe area) is the only reason I consider cable/satellite. DVRs, Netflix and Hulu obsolete most of the need for cable, and the sudden and total deterioration of the quality of original programming on the major cable networks sealed pay-TV's fate, IMHO.

If you still have any questions about the basics of digital TV, feel free to leave a comment, and I'll get to it. Coming up next time (soon!) similar tips for satellite reception...

FCC DTV tentative frequency assigments:
Antenna performance comparison:

Products Linky:

UHF/VHF combiner: ... Pico Macom "Tru Spec" UVSJ

Winegard UHF 4-bay 4400:

Winegard UHF 8-bay 8800:

AntennaCraft VHF-hi y5-7-13:

AntennaCraft VHF-hi y10-7-13:

top run by trained monkeys...

evilviper evilviper writes  |  more than 8 years ago

It has to be said... It's clear is run by trained monkeys.

I was subscribed to the theora mailing list for a very long time. Recently, I tried to send a message to the list, only to find it rejected because my IP (on which I run my personal SMTP server) is dynamic. Don't ask why I don't use my ISP's SMTP server, that's an argument for another time. Suffice it to say that I'm stuck with this arrangement, and quite happy with it.

Finding that my posts were not welcome (and angry that Theora is at least 3+ years behind schedule) I decided to unsubscribe and wash my hands of it. Guess what? is subject to the same IP filtering... I guess they don't want their bots to waste time reading spam... So, of course I use the web-based interface, and unsubscribe.

Figuring someone at would want to know about this, I e-mail Monty. Sadly, Monty's address is subject to the same filtering as all other addresses at I do what any nerd would, look up the whois information for "Great honeybee of infinity" the contact info for the domain is

So, that is why I've come to the conclusion that is run by trained monkeys. These are the people we're supposed to depend on to give us the next-gen free audio/video standard? I still like Vorbis, but I've lost all faith in Theora.

Comments welcome.


Low-power CPUs

evilviper evilviper writes  |  more than 9 years ago

I've decided to split this issue into two parts... First: CPUs.

People like to complain... A LOT... that modern CPUs are too hot. While I agree to some extent, the fact is, you just need to be more knowledgable, and you can still get very high-end CPUs with very low power requirements.

Because people have varying needs, I've included processors of many different speeds, and varying power requirements. The only rule is that they have to stand-out as significantly lower power than similar processors.

I include both Intel and AMD chips. I do not include VIA processors, because their performance is quite terrible, and their power requirements not all that low. And most importantly, I am listing desktop chips only, not mobile processors.

Proc Model Max
K6-2-400AFR 16.9W
PIII-800 (6-B-x) 10.63W
Athlon 6-8-1 XP-1600+ 48.5W
Opteron-140 (1.15V) 30W
Athlon 64 (1.0GHz 1.1V) 22W
Sempron (1.0GHz 1.1V) 20W
Athlon 64 FX (1.2GHz) 25W
Xeon-1.6G LV 30W
Athlon 6-A-0 XP-2000+ 60.3W
Sempron 6-A-0 3000+ 62W

BTW, that was a very nice looking table before Slashcode chewed it up, and stripped important info for no particular reason.

Anyhow, processor selection is only about 1/3 of the battle. The even more important issues is power draw, and power management of your motherboard. Unfortunately, neither one is publicly documented, nor are the manufacturers willing to release ANY information at all when I requested that they do so. So, it's very tricky.


Cheap Computer Cooling / Silencing

evilviper evilviper writes  |  more than 9 years ago

Computers don't need to be LOUD at all, and you don't need to spend hundreds of dollars to get them running near-silent. You also don't need to sacrific cooling to get quiet.

The problem is a simple one. Computer manufacturers use the cheapest heatsinks and fans they can buy, that will move enough air to adequately cool the system. This results in a savings of about $2 for them, and a lot of noise for you. It's easy to fix this, although it will cost a bit more than $2 to do it after the fact.

The first thing to do is to replace your heatsink. It's probably 60mm in size, with ultra-thin fins, thermal tape, and a tiny bit of space between the fins, such as this one. There are numerous alternatives. The main things you need is plenty of space between the fins, and the ability to accomodate an 80mm fan. You can find one for under $5. I would personally recomend the Thermalright ALX-800, which is easily found for $16 or less.

Install this new heatsink, but for the love of all that is holy, do not use thermal tape. It is unbelievably terrible. Using thermal paste (eg. Arctic Silver 5) will keep your processor a good 20 degrees cooler, with no other changes made.

Alternatively to a new heatsink, if your 60mm one is pretty good, you could just grab an 80mm to 60mm 'fan adapter', for about $2. I recomend buying a larger heatsink in almost all cases though.

Now, there is one fan that stands above all the rest... The Enermax 80mm. It is thermo-controlled, so it will be incredibly quite when your system is cool, and only slightly noisy when your system is very hot. Plus it's just a very quiet fan, that is cheaper than any out there. They are as cheap as $3.50 each (free s+h) if you buy 10 or more: from If you want less than 10 (why?), newegg's shipping prices are outrageous, so you'd be better off buying from: or you could use, and maybe find them cheaper (not likely though).

These fans need to replace all the others in your system. Your CPU fan, your case fans (if any), and most importantly, your power supply fan. Unplug your power supply from everything before opening it up. You'll have to cut the wires going to the original fan, and leave them several inches long. Cut off the connector on the new Enermax fan, strip both pairs of wires about an inch from the end, and twist them together. Using a few rounds of electric tape will be enough to hold them together... Just make sure you route the wires so the tape won't be resting on the heatsink. It won't likely melt through the tape, but it's a chance I do not want to take.

If your soldering skills are good, I would suggest getting a 2 or 3 prong connector, and soldering it to the power supply. You can just use the connector on the molex adapter that comes with each of the enermax fans. Desolder the two leads going to the original fan, but mark which is positive/negative, and solder the 3-pin connector in there, with the correct polarity. Just cut the 3rd pin, since it's of no use.

These fans are almost the exact ones used by Enermax in it's own power supplies. Enermax has a reputation of having high quality and quiet power supplies, and now your cheap power supply will be just as quiet as an Enermax.

Connect all the fans, and turn on the system. Suddenly, the noisiest component in your system will be the hard drive. You'll still be able to hear the fans--no fan can be completely inaudible--but I can assure you they will be just as quiet as you've dream they would be.

I do not buy ANYTHING except thermally-controlled fans, for very good reason. Not because they are quiet when your system is running cool, but because it almost guarantees your system will never overheat. I live in the desert where tempuratures reach 130F degrees, and without any airconditioning at all, my systems all remain up and running, all day, every day. A half-dozen computers is a huge source of heat, but the CPUs in my systems still never get over 120F degrees (indoor tempuratures don't normally exceed 95). Not only is my solution very quiet, and very inexpensive, but it is also far more heavy-duty than the cooling system you have now. It should keep even the hottest CPUs from overheating, in abient tempuratures higher than the hottest recorded tempurature on the face of the Earth. Just as long as your system isn't in direct sunlight, it will stay cool and quiet.

This solution has more than pacified me in my search for silence. The noise from my CPU/PowerSupply/Case -fans is no longer an issue. Hard drives are getting less noisy every generation, so they won't be an issue once I upgrade my oldest (40GB) drives. The incredible noise from my DVD-ROM is another story!!!

My next journal entry will be on cool-running, yet fast CPUs and motherboards. Somewhat related to my former story on AMD power management/S2K. Expect it to appear within a week for your reading pleasure.


What you need to know.

evilviper evilviper writes  |  more than 9 years ago

For many years, I've followed a lot of stories about law enforcement, privacy issues, and your rights. After seeing how law enforcement in the USA works, I think it's quite important for EVERYONE to know just one thing... From years and years of my own experiences, I can tell you it's the only way to operate.

Do everything by the book. Go through official channels. Volunteer absolutely nothing.

I know it may sound like something you'd hear from a career criminal or a crazy privocy advocate, but I'm neither. It's just how things work in the real world.

For instance... All the time I hear stories of someone who gave their boss an ultimatum. Perhaps their boss is doing something illegial, immoral, or just in conflict with company policy. What inevitably happens, is that their boss swears he will repent, and change his evil ways. This only serves to placate the complaining employee, and keep them from complaining through official channels. While that's happening, their boss is very busy, scheming to counter the threat. What they do is what the employee should have done... The boss goes through official channels, filing complaints against the employee, claiming anything from harassment to incompotence. In one instance, I've even heard of a boss later telling the employee they can have some time off, but never filing the proper paperwork, and denying that the conversation ever took place. This all culminates in firing the employee, and setting yourself up with a made-up reason the employee might hate you, and be willing to make-up a story to hurt their (former) boss. This obviously could have been avoided if the employee wasn't quite so considerate, and sheepish. If you go through offical channels, this will not happen. Another good lesson in this is that you should not do anything significant, strange, etc., without having it in writing. If you are getting extra time off, make damn sure you have a copy of the request form, or whatever your company happens to use.

What police do really concerns me the most. Whenever they suspect you of doing something illegial, but don't have enough for a warrant, they will try to con you into letting them search. Most commonly they say they'll be nice to you, if you are nice to them. Sounds good in theory, but once they get what they want, you can bet they won't be doing you any favors at all. Less often, I hear about police that ask to search your home or vehicle, saying they don't care about anything illegial you might have, they only want to see if there's someone or something they're looking for. Of course, once you give them permission, and they stumble on to something illegial, they'll take you in. It's probably on their vehicle cameras, but even if it's brought-up in court, the fact is, they can lie their asses off with no consequences. Once you try to help the cops out a little, and volunteer to anything, they've got free reign, and no matter what they told you before hand, they are going to use it.

In addition, there may be built-in privacy or other types of limits in the laws, that don't apply when you are scammed into volunteering. Going back to the house search senario... If you volunteer to let them look through your house for, let's say, a suspect, and they get nosey and decide to look through your dresser drawers, where they find drugs, or perhaps a gun that is not registered in your name (they check gun regs. before they enter your house, of course) then they will take you in on those charges, and everything will be admisible in court. If you did everything by the book, they would have had to get a warrant, and it would have had to specify that they weren't looking for weapons nor drugs, then they couldn't pull tricks like that on you.

Let's cover this by a much different senario. DNA and privacy. Lots of DNA dragnets are happening, so it's a very relevant subject. Let's say you are asked to volunteer DNA, and they threaten to arrest you, or just give you a really hard time if you don't cooperate. Well, the rules are completely different if you volunteer. If they go through the trouble of getting a warrant, then force you to give DNA, there may be rules in your locale that don't allow them to check those results against any case other than the one specified, and prevent them from revealing that DNA info to anyone. Lots of people are concerned that their insurance company will find out that they are pre-disposed to develop a certain disease, and silently revoke their coverage. It's a very real threat, since there have been cases brought to court of it happening.

Protesters are being forced to get smarter about their rights, because police have been very sneaky. They can ask you for your name, or more often ask for ID, and then find out if they can put together any reason to arrest you. You have no reason to give them your name, unless you were pulled-over while driving, or after you've been arrested. Before then, just say no, or even make-up a name... It is your right. I've never been a protestor, but around here, it's not too hard to be in a bad place, and look suspicious. I've had cops ask me for ID. I happily shake their hand, and introduce myself, by first name only. They can be persistant, but you don't have to give them anything. Even if you give them your ID only because they said they'd arrest you otherwise, they were lying, and you just volunteered information. Now, you migh be arrest for something as simple as an unpaid speeding ticket.

Volunteering for anything only does bad things to you, and never anything good. No matter what anyone promises you, don't believe them. Go by the book, don't volunteer anything.

Even if you aren't a protestor, nor a minority in a white neighborhood, nor look homeless, these are important rules to live by, and you'll get screwed over every time you don't strictly follow them.

Although I have not experienced most any of these senarios, I do directly know people who have, and I've been in less serious situations where I got "the short end of the stick" as they say.

Now, since you might not see how it applies to the situation you are in, let's expand on the rules to live by.

Always go by the book, and follow procedures in-order, and don't skip anything.

Volunteer no information to anyone. Looking or sounding suspicious is no crime.

Get everything REMOTELY important, in writing.

If admitting guilt, be VERY specific about what you did, and what you don't admit doing.

Don't sign anything that you have not read fully and very carefully. Be very supicious of pressure.

There is no such thing as a "standard form", you are going to get screwed if you just sign.

Even if someone sounds like they are trying to be nice, be VERY skeptical of their good intentions.

I welcome everyone to add their own comments here. Either if you'd like to add to these rules, or have your own experience to share.


AMD CPUs Power Management Problem...

evilviper evilviper writes  |  more than 10 years ago

A person can only hear "bullshit" so many times before getting annoyed, and doing something about it.

I found, from the replies to my recent post that there is a startling lack of knowledge about AMD's power management issues with their CPUs. Since there is such a lack of knowledge, and even more a lack of belief, I thought I go through the issue, in good detail, right here.

With all PCs, your OS sends a HLT instruction when it doesn't require all the processing power your CPU has to offer. On an Intel chip, this will quite simply cause the CPU to draw less power, and put out less heat. As a matter of fact, the difference in tempurature and power usage is very significant. Unfortunately, this does not happen on AMD processors.

The problem with AMD, is that the processor will not do anything with those HLTs, unless the northbridge disconnects. The northbridge, unfortunately, won't disconnect. So, your AMD CPU, even when idle, will not lower it's tempurature, or power usage.

A common rant I heard in several replies was people saying that their CPU tempurature did drop when it was idle. Well, yes, that's true. Even an AMD CPU will use slightly less power, and slightly drop in tempurature when it isn't being fully utilized, but the change is seriously insignificant. People have just gotten used to the fact that fast CPUs are incredibly hot, and don't realize that they should be seeing an incredible drop in tempurature and power usage, not the tiny change they currently get.

So, as I was saying, the Northbridge doesn't disconnect, and the CPU doesn't idle itself because of it. Well, there is a register (in the Southbridge) that will cause the northbridge to disconnect. It's known as the "S2K bus disconnect" bit (no relation to Y2K). What AMD expected was that motherboard manufacturers would enable that register in the BIOS, and everything would be good. However, practically none of the manufacturers paid any attention, and didn't set the register. AMD wasn't all that bright, in that they registered hundreds of chipset for use with AMD processors, even though they didn't set the S2K bit, and hence caused the processor to run much hotter, and waste lots more power.

Well, with processors running even hotter now, and using up even more power, AMD announced that they won't certify any chipsets as Barton-compatible, unless the BIOS sets the S2K bit. So, should you buy a shiny new motherboard to significantly lower your heat output and power draw? No, you probably don't need to... This can be done by software.

FVcool is a very good program for Linux and the BSDs. It does not work with every system on the planet, but if it does work, you'll see serious power savings. Be sure you apply the patch on the same page to FVCool v1.03, otherwise you're likely to have audio problems. You really should Download and try it. If you run a Windows machine, you can use VCool, although the power savings, even in an idle loop, won't be as significant as with fvcool on Unix.

Thanks to /.er gid (who also didn't believe me initally) we have a graph of CPU tempurature to prove the point. You can see that he started using fvcool on his system on Tuesday the 4th, just after 20:00 (8:00pm). There is a very very stark contrast before and after he started using it. And as significant as the difference in tempurature looks, the difference in power usage is even more impressive... My 90watt system drops to under 50watts when running FVcool. That is a serious power savings, and the drop in tempurature is more significant here in the desert where I live. The cost of powering a system that doesn't idle properly, and the cost of air conditioning to cool down the system that is running too hot, gets to be a very expensive proposition.

Now that I'm done praising the advantages of idling your processor, I have to spend some time complaining about AMD being morons. They did not require chipsets to enable S2K bus disconnect, wasting untold ammounts of power for many years. And even though they've done this now, their method of idling the processor is really quite bad... Intel chips do not require the northbridge to disconnect, and I must say, that's the way to go. When the northbridge has to disconnect, you certainly run the risk of having all sorts of things going wrong with your system. Corrupting or distoring video and audio is a common problem. Network connections being disabled is a very common problem. There is also the potential for systems freezing, or ATA/SCSI controllers failing in obvious or subtle ways.

I don't want to scare you away from using fvcool... The chances that you will have a serious problem with it are slim. However, I just have to point out that all these problems are caused by AMD simply being moronic about processor design. If they had just let the CPU idle down when it recieves a HLT, regardless of the state of the northbridge, AMD systems would have been running far cooler for many years, all without a single one of the problems described above.

If you'd like to read more about the issue, the writer of VCool has provided a very good page detailing it.


WARNING! Javascript is the root of all evil! ALERT!

evilviper evilviper writes  |  more than 10 years ago

You are probably here because you clicked on my sig.
You want to know why javascript is awful and should be disabled by everyone.

The link below is to a page that uses some very simple javascript code to repeatedly open up an intifinte number of windows. In those windows are some disgusting images, which you will want to close as quickly as possible. Javascript has taken control of away from you, and will not allow you to leave without a fight.

So, if you are smart, you will disable javascript on all web browsers you have access too, and tell all the websites you visit that you are unable to use their javascript features, which are always pointless, and would be implimented much better without the use of javascript.

Very well... You have been warned. So, without further delay, here is THE LINK!!!


According to O'Reilly, My Name is "Some"

evilviper evilviper writes  |  more than 11 years ago

Hmm, after reading an interview with Alan Kay, I made a couple comments that criticized him (and others) for complaining about the current state of computing , but not saying specifically what it is that we should have:

Today, I read an O'Reilly article which said (and I blockquote):

Some have criticized this interview because Kay points out the things that are wrong, but doesn't really offer solutions for correcting them.

I don't find anything wrong (or right for that matter) with the article, I just find it to be another interesting note on how far words on slashdot can carry.

This is the second event that belongs in my personal slashdot scrapbook of world impact (well, second under this username). The first being my solutions to the problem of unreachable DNS servers (which Sam later implimented into MaraDNS)

I find the first to be much more impactful on the world (I'm convinced it will end up in other DNS servers, and eventually into an RFC), but for some reason, I find this recent one to be much more curious.

Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Create a Slashdot Account