Slashdot Mirror

just go with hannibal's writeup and the eetimes article.

• The desktops were (arguably) more in need of a processing power boost than the laptops
• Apple designs (with IBM's assistance) the motherboard controller chip for the G5s, with the following implications:
  
  • the design takes longer to complete than if IBM were designing it by themselves (sheer conjecture, but...)
  • the controller chip in the G5 is currently too hot to put in a laptop
  • Apple must wait for a die-shrink for the controller chip before it is suitable for laptops
  
  
• IBM is currently working on a shrink from 0.13 micron to 0.09 micron process for G5s
• Apple could choose to build a G5 laptop running at around 1.2-1.4 GHz, but the performance gains over a similarly-clocked G4 are not high enough to justify the extra expense, especially considering pre-existing contracts with Motorola

A more detailed article. IBM has been doing dual-core processors in it's flagship Power line for a few years now, although it appears higher numbers of cores per die will only be appearing in more experimental IBM projects. Except perhaps the PS3 Cell Processor, a collaboration of IBM and Sony. Since the Cell group is based in Austin, there's likely to be some collaboration between TRIPS and Cell. As a matter of fact, they sound very similar.

Have had it for a year, and learned a lot about them.

There are three hybrids being sold right now:

- Honda Civic Hybrid
- Honda Insight
- (2003) Toyota Prius

The Hondas use a gas engine as their primary engine with an electric engine (which doubles as a generator during braking) as a secondary source of acceleration. Think of this arrangement as a gas engine with a massive-battery-powered electric supercharger. If the electric engine fails, the gas engine will still get you there -- it just accelerates slower.

The Prius is the other way around -- its primary engine is electric, and a secondary gas engine gives it the acceleration. In the Prius, it is possible to drive (with very little gas pedal pressure) on electric alone -- something the Prius owners call "Zen driving." If the gas engine fails, you can still drive on the electric.

Reliability:
I know people who own both a Honda Civic Hybrid (HCH) and a 2003 Prius. I personally own the HCH. I would have no hesitation in buying one again. I average 54 MPG in the summer, and 45 MPG in the winter. Range is ~600 miles per tank; I drive 50 miles a day, and fill up twice a month. I have heard that it is possible to get ~700 miles/tank in a Prius, but have never got a first hand report of that.

The Honda had one early bug with deep, cold weather -- very occasionally, the electric system would shut down and not restart until the car was shut down and restarted. A flash of the computer firmware fixed it. Many people reported this problem. It also had an issue with a squeak in a support pillar, fixed by shimming with a business card. Maintenance visits are scheduled for every 10000 miles. Gas mileage is better on the highway (51 mpg) than the city (49 mpg).

The 2003 Prius has had more problems. Issues were with "highway wandering" -- it feels like the car wants to migrate around the road -- and shaking of the steering wheel at low speeds. Cause: the entire power steering rack needed to be replaced. There's also the gas engine failure called the "Big Hand" that's fixed with a similar car reboot, but it's caused by the Accelerator Pedal Assembly needing to be replaced. Many people report that they've had both of these problems. Maintenance is more frequent at 7,500 miles. Gas mileage is better in the city (51 mpg) than the highway (49 mpg).

I don't know anyone who drives a Honda Insight. Supposedly they still make ~1500 of them a year, but rumor has it that Honda is going to be dropping them.

Near-term and Farther-out Models:
The Prius has been out longer than the HCH, and is getting a technological refresh in 2004. The 2003 Prius and the HCH are both 4-seater compacts, but the 2004 Prius will be a mid-size, with better mileage (59 city, 51 highway) and better acceleration (0-60 in 10 instead of 12). There is currently a waiting list for them.

Cars/Trucks/SUVs due to be released as hybrids in 2004:

Honda Accord,
Honda CR-V,
Ford Escape

2004 or 2005:
Chevy Silverado
Saturn Vue
GMC Suburban

Here's an in-depth look at Hybrid technology from the Union of Concerned Scientists (Google HTML translation here).

Here's the best description of what it's like to drive one.

There is a federal tax deduction of $2000 for buying a new hybrid. Several states also offer their own tax deductions or credits -- check here to look them up. Some states (but not all) also let you drive them in their high occupancy lanes, even though you may only have one driver in it.

Oh, by the way -- I know that the 2003 Honda Civic Hybrids are

Got the same car. Love it. Not too long ago I read some article (in Ars Technica) about how driving the HCH changes the way you drive. It's absolutely true. I'm from a country where people drive with intent to kill. Stop signs are taken as suggestions. Always drove fast, recklessly and (my wife would argue) stupidly. I was brought up that way, sorry. Anyway, since I got the HCH my only goal while driving is to maximize the mileage. This model comes with an instantaneous mileage reading and a cumulative one. So now I rarely go over 68 mph. I'm getting 57~58 mpg on the road, and around 48 in the city (I have the manual shift one - another cultural hangup). There's just one thing that I started doing that is definitely moronic. I tend to lock on big semis and tailgate them to improve the mileage even more. Feel like Lance Armstrong. Except he's not retarded, I think. Anyway the technology on these guys is pretty awesome. You get to a stop sign, and the engine stops. Start rolling again and the engine starts as you press the accelerator. A thing of beauty. Plus it's really quiet, and if you choose to ignore it, you would never know that you are not driving just a regular Civic. Finally, you get an obnoxious smug feeling when the idiots on the SUV zoom by you. (Though you could hit 100 mph if you were not so compulsively trying to break the barrier of 60 mpg).

The Civic Hybrid is very nice. It looks and feels like a normal car. In fact, you may have seen some driving around and didn't notice them. They look just like the normal Civic except for a small Hybrid tag on the right rear. Ars Technica has a good review of the Civic Hybrid.

Ars Technica did a really good review of the Honda Insight. You might find it informative.

statistics can be very misleading. for example:

Common sense can cloud statistical results. For instance, a technology firm discovered that 40% of all sick days were taken on a Friday or a Monday. They immediately clamped down on sick leave before they realised their mistake. Forty per cent represents two days out of a five day working week and therefore is a normal spread, rather than a reflection of swathes of feckless opportunists trying to extend their weekends.

(preceding was taken from an ars technica article)

if 90% of servers are linux servers, then it makes sense that 90% of attacks should be against linux servers, right? im pretty sure linux is more than 67% of servers right now, so 67% is actually very low!

• Linux Desktop Basics
• Human Interface Consistency (including default Gnome Desktop)
• Gnome uses SVG plus web browser and file manager
• Desktop accessabiity and developer GConf stuff & streaming ability
• Pango Text Rendering
• Final Thoughts

• Linux Desktop Basics
• Human Interface Consistency (including default Gnome Desktop)
• Gnome uses SVG plus web browser and file manager
• Desktop accessabiity and developer GConf stuff & streaming ability
• Pango Text Rendering
• Final Thoughts

• Linux Desktop Basics
• Human Interface Consistency (including default Gnome Desktop)
• Gnome uses SVG plus web browser and file manager
• Desktop accessabiity and developer GConf stuff & streaming ability
• Pango Text Rendering
• Final Thoughts

• Linux Desktop Basics
• Human Interface Consistency (including default Gnome Desktop)
• Gnome uses SVG plus web browser and file manager
• Desktop accessabiity and developer GConf stuff & streaming ability
• Pango Text Rendering
• Final Thoughts

• Linux Desktop Basics
• Human Interface Consistency (including default Gnome Desktop)
• Gnome uses SVG plus web browser and file manager
• Desktop accessabiity and developer GConf stuff & streaming ability
• Pango Text Rendering
• Final Thoughts

• Linux Desktop Basics
• Human Interface Consistency (including default Gnome Desktop)
• Gnome uses SVG plus web browser and file manager
• Desktop accessabiity and developer GConf stuff & streaming ability
• Pango Text Rendering
• Final Thoughts

• Linux Desktop Basics
• Human Interface Consistency (including default Gnome Desktop)
• Gnome uses SVG plus web browser and file manager
• Desktop accessabiity and developer GConf stuff & streaming ability
• Pango Text Rendering
• Final Thoughts

• Linux Desktop Basics
• Human Interface Consistency (including default Gnome Desktop)
• Gnome uses SVG plus web browser and file manager
• Desktop accessabiity and developer GConf stuff & streaming ability
• Pango Text Rendering
• Final Thoughts

Here's an Ars Tecnica Review that answers your question:
http://arstechnica.com/reviews/003/software/gnome- 2.4/gnome2.4-1.html

Well, GNOME has already won awards for its accessibility work, and it has a mature framework for dealing with this (disclaimer: I've never tried it, just heard people praising it).

The article itself points this out in two places.

Well, GNOME has already won awards for its accessibility work, and it has a mature framework for dealing with this (disclaimer: I've never tried it, just heard people praising it).

The article itself points this out in two places.

Here's the other side of the story. Special Force, an fps " allegedly produced by the Lebanese terrorist group Hezbollah. The game's designers seek to "correct" the influence of western gaming concepts on the Middle East. In particular, they want to combat the view that the "oppressor" always defeats the "Arab."".

I found an article about it in arstechnica.

or so says Ars Technica. In addition most of the initial shipments will go to motherboard manufacturers for bundling with their boards. I really don't like the idea of that becoming common practice as that much purchasing power will mean tight pricing controlls. Read more Here.

To the CPU, the RAM looks like one long, thin line of storage cells, each with a unique address. If the CPU wants a piece of data from RAM, it first places the address of the location on the address bus. It then waits a few cycles and listens on the data bus for the requested information to show up. Check out the following conceptual picture of how RAM looks to a CPU.

DRAM accesses are addressed sequentially (just like a hardrive) as well. You have memory addresses that start at 0 and go to the last bit of memory. How is this not sequential, please explain to me how 0 to n is not sequential? But I digress, because I'm not talking about how the data is laid out on the device, I'm talking about how the device is accessed.

Oh, and having a seek time makes it not a random access device? I guess we better find a new name for DRAM, because there is a penalty for random memory accesses. How big is the bus to the memory on your CPU that has no penalty for random memory accesses?...64 bits?... 128 bits? To have no penalty and to meet your definition of RAM, a 64 bit computer would have 8 bytes of RAM. I don't think that will get you too far. For your definition of a machine with a reasonble amount of memory, one would need a bitline to every memory cell (to avoid those "seeks" caused by selecting columns and rows of memory), so a 256MB machine would need a 2,147,483,648 bit memory bus. I guess you could address more than one bit at a time, but even a 256 bit bus machine would have 1MB per line...sure sucks if you want to access just a few bytes. That would be your highest precision as far as reading from memory. So you'd have to stream that whole 1 MB across the bus for just 1 byte of data. Your truly random access machine without any seek penalty sure does suck!

You say the random access penalty is not that big? Hmm, then why don't we see memory bandwidth figures using a random access scheme? Probably because all the page misses would make the system run like a dog, relatively speaking. Streaming is so much faster since the system has a pretty good idea which pages to open next and can hide the latency of opening new pages. This is actually the same as a hard drive where you have lower latency if you read one contiguous chunk of data rather than having it scattered across the drive. In both cases there is a delay as the medium needs to "switch" to another address. For a hard drive, it moves the head, in a DRAM, electrons move around altering the path of electricity. In both cases there is physical movement and change (yes, electrons are actually moving around, it's mechanical on a small enough scale), which means an unavoidable delay.

The mechanics are different but I'm not arguing about the mechanics. I'm not arguing about the speed differences between mediums either. I'm also aware that a hard drive is orders of magnitudes slower than conventional DRAM, but that has no bearing on my assertion that a hard drive is capable of random access. I'm not even arguing that it is RAM...but that is really a context issue rather than some concrete qualification. Still confused? I guess I better paint a picture for you...

A truly sequential device would require you to read or skip sequentially around, not being able to avoid passing over part of the media on the way to another. Take a tape, for instance, where to get to another part, one needs to traverse all the tape between the two points. Certainly, it doesn't need to be read, but the head can't somehow skip this part of the media in between the two points, it will pass over the head whether the head is "listening" or not. That's in contrast to a hard drive/cd rom/zi

Forgotten Link

If you read Hannibal's appendium to his 970 articles he points out that the Altivec looks better than he initially thought. This isn't to say there isn't room for improvement. But to say that the G4 is the Altivec champ seems somewhat incorrect.

Hannibal's Followup

It is true that with G4 altivec code the G5 won't be as efficient as it could. Perhaps for some code the G4 might do better. But this is a simple matter to fix with a grep through your code. But initially altivec code won't show off the G5 until programmers do a simple recompile.

Here's a good .net primer.

1) Does this kid need to learn his lession in jail?
No, This kid is young. He's stupid.

I disagree. He does need to go to jail, if found guilty. Penalties for a crime are not only punishment for the convicted, but hopefully a deterrent to future potential criminals.

I'm sure he didn't do this realizing that he'd be headed to jail in a few months (if proven guilty).

So he's been living in a cave? (well...maybe his parents basement). With all the publicity going around, and a couple of high profile cases caught and jailed...I'm sure he knew. Probably just so arrogant that he thought he'd never be caught.

someone came along and leaned up on it wrong and it stopped working. I'd be pissed at the manufacturer, not so much the leaner (who is laying on the ground with a bloody nose by now).

Evidently you do advocate some punishment for the 'leaner/virus writer'.

Yesh. The PPC 970 bus uses something of a packet based protocol. Thus, there is an overhead that the simpler P4 bus does not have. The information is available right from the excellent ArsTechnica articles about the PPC970, specifically, this one.

Not quite true and not the full story. According to this article, the G5 dissipates about 42 watts at 1.8 Ghz. The dual processor systems would dissipate about 84 watts. The P4 2.8 Ghz chips dissipate 68.4 watts.

The heat is too high for a laptop though. The 1.2 Ghz 970s run around 20 watts. [source]

Now although all my senators and house members are owned by corporations, the justice system will find in favor of crusaders like me and Eric Scott Raymond.

Um, not necessarily, I'm afraid. Witness the DeCSS judgement.

And just to piss everyone off I'll even throw Apple under the bus and point out that the "New and Improved" finder under 10.3 really looks like crap.

That article got its facts wrong, as I pointed out over a month ago in another Slashdot thread. I tried to dig up the old article or the comment that I posted back then, but the dain-bramaged Slashdot search engine won't let me pull up older comments that I've left. (I can see the last 24 comments I wrote, but that's it.)

So, to reiterate YET AGAIN, the EE times article you are citing is IN ERROR. The combined power dissipation of two 2.0 GHz G5 chips is 97 watts. The fact checking for the article in question was very poor, and this has been debunked many places, not just here in Slashdot.

On a personal note, I'm shocked and dismayed that one piece of disinformation can persist for so long, even when so many articles (such as the excellent ArsTechnica series on the PowerPC 970, aka the G5) obviously contradict the one article you chose to cite as a reference. For instance, in the ArsTechnica article, the power dissipation of the 1.8 GHz G5 is given as 42 Watts. It doesn't take a genius to figure out that a 2.0 GHz G5 shouldn't dissipate 97 Watts, over twice what a 1.8 GHz part dissipates. An 11% increase in clock speed does not more than double the power requirements for a semiconductor!

So do us all a favor, and stop spreading disinformation based on one provably flawed piece of news. Of course, I e-mailed a complaint to the author of the article, but like most people, he doesn't have the intellectual fortitude to own up to a mistake and admit he didn't check his facts. I suppose I should have e-mailed the editor instead...

Personally, I find the whole thing highly obnoxious. Not that I think there shouldn't be a simple, intuitive user interface. That should be one of the main priorities of anyone trying to sell anything to consumers. I know I've quickly left web sites because they are cluttered and disorganized.

My problem is that I think the Windows interface has been counterintuitive from the beginning, and they have *NEVER* changed that clunky interface. They just pop up annoying little dialogue boxes or paperclips that try to guess what you are trying to do (most often unsuccessfully). Rather than designing an intuitive UI in the first place, they cover up poor design with annoying explanations.

If you want to know specifically what I'm talking about, I think Paul Siracusa's About the Finder... article at Ars Technica explains it well. While he is comparing MacOS 9 to MacOS X, much of the discussion easily applies to Windows.

Personally, I find the whole thing highly obnoxious. Not that I think there shouldn't be a simple, intuitive user interface. That should be one of the main priorities of anyone trying to sell anything to consumers. I know I've quickly left web sites because they are cluttered and disorganized.

My problem is that I think the Windows interface has been counterintuitive from the beginning, and they have *NEVER* changed that clunky interface. They just pop up annoying little dialogue boxes or paperclips that try to guess what you are trying to do (most often unsuccessfully). Rather than designing an intuitive UI in the first place, they cover up poor design with annoying explanations.

If you want to know specifically what I'm talking about, I think Paul Siracusa's About the Finder... article at Ars Technica explains it well. While he is comparing MacOS 9 to MacOS X, much of the discussion easily applies to Windows.

Update to self:

They're already getting close.

It was origionally found at Ars Technica

> They've been out awhile.

Well sure, if you want to an AIX box like the p630 Model 6E4 for a list price of $16k, or the IntelliStation POWER 275...

I think the above poster might have been interested in IBM 970 kit, as Big Blue's still shipping POWER3-II's in the blade "space".

Also, a 970 is not a POWER4 - 0.5 MB vs 1.5 MB L2 cache, 0 MB vs 8 MB L3, dual- vs single-core, no Altivec on Power4, 1.8 vs 1.4 GHz max (currently shipping) freq, 0.13 um vs 0.18 um, decreased MTBF on the 970 to make it faster... &c.

Ars had this story on 8-15. You guys are falling behind.

--jdan

I would want to be anonymous too if I were just speculating without engaging the brain. The GPU they use is called the emotion engine. It is quite impressive and the raw performance (in terms like polygons per second ;-) has been noteworthy.

"..the reporters, in theory, know much less about the science than those reviewing for the journal..."

If the reporters know so little about the subject in question, why do they publish an article like this acting as authorities on the subject?

Completly apart that the peer review process isn't completly foolproof (look at this story if you need convincing).

The journalists have a responsibility to the comunity to publish realistic scientific stories - There are plenty more interesting scientific stories they could publish, but from what I've seen they always go for fringe/crackpot ideas and studies like this.

Anyone with a HS education can get a job in sales at these places. Often it takes little more than simply having an A+ cert to be a hardware tech. If you have any actual skills or knowledge, you'd be working somewhere else for more money. (to top things off, most places won't let minors staff
registers; a combination of insurance issues and being able to actually press charges if somebody's robbing the till)

Maybe since you were young things have changed; there's nothing 'technical' about selling this stuff. It's all pre-packaged, pre-assembled crap that doesn't even come with real user manuals (instead they've got glossy 'quick-start' cards) that most people are buying from these places. What is there to learn by looking at the stats on an HP v. an eMachines system?

Then there's the whole bit about the cost of high-tech toys being so low that working any job can score you a decent system before too long (a kick-ass DIY system can be had for under $1500; something usable can be picked up, used, for a few hundred).

Not to mention that, while working minimum wage, relying on commissions to actually make money, anyone with a conscience will quickly slide in sales under the goons that sell $3k systems to ppl who just want to browse the web & read email.

I think you are referring to spooky action at a distance

Ars Technica story on the same thing.

However IBM's recent entry, the PPC970, has radically altered the desktop landscape. The new Apple computers powered with the PPC970 are genuine workstations sold as desktops. The ARS Technica article indicates that the SPEC2000 performance for the PPC970 is 937 and 1051 for integer applications and floating-point applications, respectively. The Athlon64 is a weaker version of the server chip, the Opteron. The PPC970 has about the same performance as the Opteron. (reference: SPEC performance list)Hence, the PPC970 is sure to beat Athlon64 across a broad range of applications.

What is particularly interesting about the PPC970 is that it was designed and built largely without H-1B employees. Both IBM and HP have a policy of not hiring anyone who does not have American citizenship or permanent residence unless that applicant has a Ph.D. Clearly, American companies do not need H-1B employees to produce awesome products.

"Now, a number of people were pretty amazed to see such high OS X numbers, and some even alleged that the ballot box was being stuffed--either from a third-party site encouraging Mac votes, or by "multiple voters" or some other evil scheme meant to over-represent what must surely be a dying breed *cough*."

Windows - 66.42%
(unknown) - 15.06%
Macintosh - 11.46%
Linux - 6.49%

"I suspect that a significant number of the "unknowns" are either Mac users, or Opera users on either platform."

"I do think it's pretty interesting that, when analyzed as combos of OS and browser, the next biggest combo after IE on Windows 2000 and XP is MacOS X and Safari, even ahead of Mozilla on Windows, or on the Mac."

Style and substance aren't totally separable, though. Is it part of the substance of her article that she makes such errors as the multiply-noted (by others) claim that males have an additional chromosome, or her very questionable reasoning as to games improving one's cognitive skills?

Her wording was wrong in regards to the "extra" chromosone. It should have been different and it has been noted. Her reasoning about the effects on cognitive skills was derived from the following article on Ars Technica.

It's true that there is the Mach kernel involved (here's a simplified cake-layer diagram), but from the context that Ritchie provides ("The way I use them, which is as a casual programmer, it doesn't matter--they are all the same...") none of that is really relevant. The APIs that he expects from a unix are there in the FreeBSD 4.4 code layer. (X11 is there too, by the way). He would have to be using the word "FreeBSD" very pedantically to mean that it's FreeBSD/PPC and not the FreeBSD 4.4 in OSX. It doesn't sound like he's being pedantic in this interview.

Maybe this is offtopic, but if you want really elegant language processing you should check this out. Basically, you look at the compressiblity of given text and can determine what language it's in, or even what author produced it. This works with as few as 20 words.

I realize this isn't translation, but cool nonetheless. For further reading see here and here.

I didn't intend to imply everyone needed to be making dvd's on their laptops ... but really, if a laptop is the only new computer you can afford, may as well enjoy the bells and whistles, right?

Indeed, I was just looking at this yesterday after taking over responsibility of a client's NT4 -> 2000 Active Directory migrated network, as the client machines are a mix of 98, 2000 Pro, and XP Pro.

This URL may be of some use?:
Active Directory Client Extensions for Windows 95/98 and Windows NT 4.0

I've yet to have time to check whether it's actually on the 2000 Server CD, but I hope so... (I still want to get rid of the 9x clients though)

Speaking of hardening Windows networks, I'd recommend checking out a few of the following:
Berkeley Labs Computer Protection Program: Windows Security (including guides on how to harden 2000 & XP)
Some interesting Windows password quirks
Ten Windows Password Myths
Securing Windows 2000: First Steps

That should be enough to get started :) Cheers,

Stef