Slashback: Dataplay, XviD, PPC 183
Pins and needles, pawns and bishops. s20451 writes "It looked grim earlier in the week, but following a fifth game meltdown by Kramnik and a brilliant game 6 by Fritz, the computer has tied the match 3-3. Betting on the computer in game 6 would have brought you a 7-1 return! I'll be on the phone to Vegas."
The new, new, new economy has room for camels. SwiftOne writes "According to their website, The Perl Journal has gotten enough subscriptions to begin online release (the planning of which was previously covered, along with the concerns about not reaching their goal. The first (next) issue is expected in early November."
Maybe it was the 15th-mover disadvantage. melt writes "Dataplay, the Boulder-based manufacturer of quarter-sized recordable discs and drives, finally called it quits on Friday, October 11, 2002. The remaining 120 employees (who have been on furlough for the past few weeks) have been let go and the company has closed shop. They are looking for a buyer for the remaining pieces. Full story at the Rocky Mtn News web site."
Zoom in until you see little canyons ... Twirlip of the Mists writes "IBM's chief scientist for their iSeries family of servers (a.k.a. the AS/400 family) has an article on iseriesnetwork.com describing the somewhat confusing history of the POWER4 microprocessor. In light of recent speculation about a possible relationship between IBM and Apple, this article is of particular interest. It clears up-- at least partially-- some of the complex, incestuous relationships between the PowerPC architecture, the PowerPC processor family, and the POWER4 processor. As an added bonus, there's some talk about the upcoming POWER5 and POWER6 processors near the end. The key phrase (and disclaimer): 'expected to appear in 2004.'"
Shame on Sigma.
Gruturo writes "After almost 3 months the XviD project and website have reopened, though Sigma Designs has not complied yet with all their requests (they still carry their copyright on many modified sources). In these last 2 1/2 months the project still went underway, although unofficially:
B-frames are practically ready, motion estimation algorithms have been improved, work started for Qpel implementation."
Please stop teasing us. If you liked Cryptonomicon, you've probably been impatiently watching the announcements of when the next Stephenson book would appear. wka writes "Previous false starts notwithstanding, Amazon says Neal Stephenson's new novel Quicksilver will be published in January."
And next week, building box-girder bridges. scubacuda writes "Lawmeme has released Part III to their Law School in a Nutshell series (Part I and Part II were previously featured on /.)"
Quarter-sized storage disks (Score:1, Interesting)
Re:Quarter-sized storage disks (Score:2, Interesting)
I don't know why, (Score:5, Funny)
Re:I don't know why, (Score:4, Insightful)
If I were on the programming team I'd take that as an expert suggestion to add a negative adjustment to the evaluation of queen exchanges.
-
Damn! Now I need a new travel book... (Score:4, Interesting)
Anyone have any good recommendations on geek books suitable for 26+ hours of flying (and a few couple-hour jetlag-induced insomic sessions)? Besides the Slashdot book review section, I mean. Novels and such...
-B
Knuth! (Score:5, Funny)
How about the 3 volume 'Art Of Computer Programming' [amazon.com] by Knuth.
Not quite a novel, but meaty enough to give you loads of info, or technical enough to send you to sleep (depending on your mood at the time)... or if you still need sleep, you could try hitting your head with the books, I know how the engine noise and people can just keep you completely wired during a flight.
Enjoy your flight.
Re:Knuth! (Score:4, Funny)
Re:Damn! Now I need a new travel book... (Score:4, Informative)
It's also an excellent book in its own right -- it won the National Book Award in 1974, and it would have been awarded the Pulitzer Prize had the board not considered it obscene and overriden the judges' decision.
Re:Damn! Now I need a new travel book... (Score:4, Interesting)
It is obscene; that judgement is correct.
But it is also divine.
In fact, you can pick an arbitrary pair of opposite, highly charged adjectives, and both will likely be a fair description of Gravity's Rainbow. So as well as being tedious and boring, it is also challenging and endlessly fascinating. Not to mention deadly serious and deadly humorous. I can't think of a novel that has more influenced my worldview than this one.
A-and how can you say no to a book that has lame calculus humor in grafitti, or a bunch of drunk Army engineers chasing the protagonist, singing limericks about Doing It with the German V2 rocket hardware?
(Hints for the first-time reader of GR: you don't have to understand it the first time through. Hell, you can just skim it. It's still funny and interesting. Also, gin helps a lot.)Re:Damn! Now I need a new travel book... (Score:2)
GEB! (Score:4, Informative)
Here's an Amazon Link [amazon.com]
Re:GEB! (Score:2)
"I know, I'll write a piece about a recording that fails to resolve, and my essay will also fail to resolve!"
"I know, I'll write a vignette that mentions acrostics, and it will itself be an acrostic!"
"I know, I'll write an essay about a mirror fugue, and the essay will itself read the same forwards and backwards!"
"I know, I'll write an chapter about recursion, and that chapter will itself be recursive!"
Gah. That little trick got old after its first use.
Re:Damn! Now I need a new travel book... (Score:2)
26+ hours? Are you coming from Oz or New Zealand?
I'd suggest-- as you might guess from my nickname-- two of Vernor Vinge's novels: A Fire Upon the Deep [barnesandnoble.com] and A Deepness in the Sky [barnesandnoble.com]. The two are technically a novel and its prequel, but they're really only linked thematically. And they're both outstanding works of science fiction with particular appeal to computer geeks. Both are available in paperback, but they're long enough by far to keep you occupied while you're en route.
Eighteen months ago, my official org-charted job title was "Programmer-at-Arms," inspired by Deepness. These are two really cool books.
Re:Damn! Now I need a new travel book... (Score:2)
It's 13 hours each way, and I'll be damned if I'll pay money to fly books all the way to Europe only to sit about reading them while there. Except for those odd in-between hours that first night/morning, I plan on looking at things which are part of a scenic view, not imagining them.
I'd suggest-- as you might guess from my nickname-- two of Vernor Vinge's novels: A Fire Upon the Deep and A Deepness in the Sky.
Way cool. I'll definitely check them out. Thanks for the suggestions!
-B
Re:Damn! Now I need a new travel book... (Score:2)
Earlier this year I got roped into a last-minute business trip from to Sydney. (I live in the US.) Thing is-- and I know this makes me sound like a nerd-- I was just getting started on the fourth Harry Potter book, and I was really digging it. So, yeah, I lugged a giant hardcover book from the US to Australia-- about 24 hours from door to door, because I don't live in Los Angeles-- and back.
Was it worth it? Shit, yeah. The in-flight movies were terrible. I've blocked most of 'em out, but I remember turning off "Zoolander." I actually chose to sit there and stare at the back of the seat in front of me rather than watch that steaming pile of crap. If I hadn't had my book, I would have gone quietly nuts.
Come on, teleportation.
Re:Damn! Now I need a new travel book... (Score:2)
The last time I flew to England it was on Virgin, and I didn't have a book, laptop, palm pilot, anything. Bad idea. You're dead right: the movies were absolutely horrible. I actually found myself playing Nintendo (think Super Mario-era Nintendo) during the flight the movies were so bad. Hell, there were little kids who wouldn't even play it. But it was better than the movies. I don't even remember what was playing and I don't think I want to see "Undercover Brother" this time, either. Which is why I asked for book recommendations. I honestly thought that Quicksilver would be out in time and hadn't even though about what to get.
Come on, teleportation.
The world is an incredibly small place now; I shudder to think what teleportation will do to it. I remember in the early 80's my older brother went to England and we were beside ourselves because he brought back Dr Who stuff, books (I wanted a British dictionary), records (vinyl, kids: Two Tone stuff, The Clash, The Damned, Buzzcocks, etc, etc) and creepers/Doc Martens because we couldn't get them anywhere in Phoenix. We had to make roadtrips to LA for the music and such and even then we didn't get everything we wanted (although we usually got some things we didn't). Remember "imports"? Man, that was the shit when the copy you ordered came in like 19 weeks later. I still have a blue vinyl Captain Sensible Birthday EP a guy carried back in a suitcase for me. Got a red vinyl copy of Strawberries, signed Madness LPs, some Police B-sides, and a bunch of other junk I have to rip someday, too. But now you can get it all off the Net. Or at Sam Goody. There's about four corporations who own everything, and it's the same stuff no matter where you go.
It used to mean going to England or France or Pakistan or where ever meant that you were going to some place that was different than where you were form. You got and tasted and smelled stuff you couldn't get back home. The most popular restaurant I saw when I was last in London was TGI Fridays and the most popular beer was Budweiser. You could hardly get away from Budweiser. One barkeep was telling me that he ran out of it weekly. The Virgin Megastore there had everything I can get here. I did wind up buying a driving game, though. It was right-hand drive. Now, going there is almost like going to Seattle, except everyone sounds funny. I'l have to remember to stay in th epubs and on the back streets.
Once teleportation hits, the world really will be flat.
-B
Re:Damn! Now I need a new travel book... (Score:2)
Without getting into specialist ales, you should try some bitter like Directors, or Youngs Ordinary. If you have to go for a lager, the most consistently quality is good old Stella, affectionately known as Wifebeater due to its sometimes excessive behaviour altering side effects. Caffreys is good for an almost hangover free 10 to 14 pint pub crawl (may take additional training).
And really, calling TGIs a restaurant is pushing it, eh?
Re:Damn! Now I need a new travel book... (Score:2)
I find it amazing people anywhere drink it. I know a joke which applies here:
Q: Why is drinking Budweiser like having sex in a canoe?
A: Because it's fscking close to water.
Thank you, thank you. I'm here all week...
Without getting into specialist ales, you should try some bitter like Directors, or Youngs Ordinary. If you have to go for a lager, the most consistently quality is good old Stella, affectionately known as Wifebeater due to its sometimes excessive behaviour altering side effects. Caffreys is good for an almost hangover free 10 to 14 pint pub crawl (may take additional training).
I'll give the bitters a go, for certain. And I hadn't noticed the altering effects of Stella. But I'm normally fairly even tempered even in the face of the worst alcoholic adversity. If you're ever in the US, try King Cobra (or any other "beer" which comes in a 40 ounce bottle) if you want a sample of American riot beer.
I really appreciate the pub crawl advice. We're planning on doing a Soho [fancyapint.com] crawl. I'll defintely keep the Caffreys advice in mind (as long as I'm able to, at least).
And really, calling TGIs a restaurant is pushing it, eh?
Yeah, that's certainly giving it something it shouldn't have. I just didn't know the word for "pretentious meat market-ish place which serves awful food and is filled with idiotic, cologne-drenched beeper salesmen either yelling at some inane sporting event on a loud TV or trying to tag the nearest waitress".
-B
Re:Damn! Now I need a new travel book... (Score:3, Interesting)
Smirnoff Ice?! It has a chemical in it which makes you crave it fortnightly... :-)
I drank Caffrey's and Harp and Bass and many "regional" beers/ales often given to Yanks to make them feel they'd had their money's worth. I also enjoyed Stella Artois (ordered simply as "Stella") because it reminded me of what pedestrian American beer could have been. My favorite beer I had while there was a couple of pints I had in a pub called the Leinster Arms [fancyapint.com]. I have no idea what it was called, but it was good, and went well with a generic pub lunch. That I'll probably never have it again is just as well, I suppose, since it was a bit of culture which would likely be cheapened if enjoyed anywhere but then and there.
-B
Re:Have things changed *that* much? (Score:2)
Yeah, I think so. We just did like you did. Or if I saw a tap handle that had an unknown name on it, I'd ask about it.
It was just really hard to get away from Bud advertising and feel like we were in a "real British pub" and not some US imitation.
-B
Re:Damn! Now I need a new travel book... (Score:2)
Re:Damn! Now I need a new travel book... (Score:2)
Geez... I had never even thought to go back to the classics...
-B
Re:Damn! Now I need a new travel book... (Score:2)
Going, and gone. (Score:4, Insightful)
Well, can't say I'm surprised. While there are a VERY few uses for drives this small, the demand is pretty dang small. For what I'm sure is much less, you can get much more storage at a resonable size. So while the technology is very, very cool... it realy doesn't have enough people to support it. (Unless it was made by a big manufacturer like IBM that also did many other things...)
Quarter sized disks (Score:1)
For the sake of us outside the US, just how big is a quarter, anyway?
Re:Quarter sized disks (Score:5, Funny)
Bigger than a nickle, smaller than a half dollar. Glad to help.
About half the size of a dataplay disk. (Score:2)
They are about half the size of a dataplay disk, actualy.
Re:Going, and gone. (Score:4, Interesting)
mp3 players, digital cameras, digital books... all of these want small, cheap, media in the 500MB range.
It just wouldn't fly for prerecorded music. (And not only because of their crappy DRM scheme) Music media is dead. I predict that CDs will be the last major music media. SACD and DVD-Audio may have a breif blossom, but for the most part, the current generation of adults are happy with CDs, and the current adolescent generation will wonder why people care about the physical media when you transfer files from your mp3 server to your ipod.
Unfortunately, prerecorded music was exactly the niche Dataplay aimed at.
I was going to say that this would be a perfect match for the 10MP cameras coming out, but I expect those to eat batteries nearly as fast as they eat RAM so they'll be be used mostly in a tethered environment.
Would not have worked for that either (Score:3, Interesting)
CompactFlash and SD cards hold hostage the storage range from 64Mb to 1GB. There was no way another format was going to come in to unseat them, even one somwhat cheaper.
What is needed is something with about a 100gb storage space in the size of a CF card or smaller, for future digital cameras. That would be enough of a lead to unseat CF and SD, and provide enough room for cheap video cameras as well.
Re:Would not have worked for that either (Score:2)
Re:Would not have worked for that either (Score:2)
As for what will need it, I'm not totally sure what it will be but I know the need will be there. Along the lines the other poster mentioend, it might be HDTV video. Perhaps it's 3D video with four different perspectives encoded. Perhaps it's 50 megapixel images with more depth and dynamic range than film (which is the use I'd have for it). I'm not sure but I do know that even now the 1GB cards are looking small when you are dumping TIFF images instead of JPG from your camera.
Right now I take between 15 to 20 40-exposure rolls of film during a trip. The film is a little bulky but not as annoying and expensive as whatever digital solution gives me that number of pictures at film resolution. Although I can bring a laptop (and possibly a spare HD) with me, I think it would be a lot nicer to just bring a camera and one or two cards with me.
Quicksilver (Score:1)
Re:Quicksilver (Score:3, Informative)
Sigma (Score:5, Interesting)
Shame on Sigma
More like shame on me for buying the damn card, what a piece of crap that is... my friend in the DV industry told me his department tested the card and the conclusion was (apart from being crap) that they had really skimped on the hardware acceleration/decoder processor (just so it could do the bare minimum)... looks like they skimped on the 'development' (read. stolen) of the DivX (cough, XVid) implementation and the complete joke they called a 'player'.
The other thing was I had to look when buying it, 'cos NewEgg didn't sell it... only Sigma themselves and CDW sold it (the latter where I got it from). Hmm, NewEgg rules... I must learn to trust my instincts now and think twice to the thought 'Why aren't NewEgg selling it?'. DOH, DOH and thrice DOH! (Shame on me).
Re:Sigma (Score:2)
Sigma's tech support gave me a cookie cutter response which did not help me in my situation.
And now they're violating the GPL? Last card I buy from them.
Quarter size might be too small (Score:5, Insightful)
Re:Quarter size might be too small (Score:5, Funny)
If you routinely get quarter-sized objects lost in your hair, maybe it's time to pay a visit to the barbershop. Or at least invest in a comb...
Re:Quarter size might be too small (Score:1, Funny)
Re:Quarter size might be too small (Score:2, Informative)
They came out roughly the same time as Zip's did. Zips were 100mb disks were $25, the drives were $299 and included a scsi card. In contrast MD-Data disks 120mb were $50, the drives wer $800+ and did not come with scsi cards, and were frequently UW SCSI only.
SONY marketed them towards the professional music and broadcast industries almost exclusivly and never made it accessible to the average computer user.
Re:Quarter size might be too small (Score:1)
Lossless audio on a 3" DVD disc (Score:2)
if only there was a standard for lossless audio on 3-inch DVDs.....
Can MLP be run at 44 kHz 16-bit stereo? If so, use the DV/DA standard. If not, use DVD-Video with blank video and PCM audio.
Now just wait for the 3-inch DVD-R blanks to come out.
Re:Quarter size might be too small (Score:2)
You can right now go and buy 2.3GB 3.5" magneto-optical media and the drives they work with. Their dimensions are very similar to that of 3.5" floppies, but look much cooler.
On the downside, complete lack of mass marketing means that they're still expensive.
Have a look at Fujitsu's magneto-optical products [fujitsu.com] for more info (it's in Japanese.)
I would love to see these take off and replace floppy drives. MO has a long reputation of speed, stability and robustness. Certainly they are an order of magnitude faster and more reliable than Zip disks, not to mention storing over two gigabytes per disk.
what? (Score:2)
These folks missed the boat. Not even IBM can make a go of microdrives, that's why they sold the division. I want one of those 1G models. If these folks can make a compact flash drive that fits their disks, that would be cool. They have a big race to beat falling compact flash prices.
Re:what? (Score:2)
Proof of DRM's failure (Score:4, Interesting)
Quicksilver in January? (Score:5, Informative)
Re:Quicksilver in January? (Score:2)
A 6-month publication cycle would be very unusual.
return (triumphant?) of XviD? (Score:2, Funny)
Talk like Yoda, Timothy is beginning. =P
Megalomania at DataPlay (Score:2, Flamebait)
I, however, will just laugh. If, indeed, "everyone" at this company believed in this format, I'm glad to see them tank. There was some sick shit going on there; here's hoping someone incompetent buys them up and kills their wonderful "technology".
Power5 and Power6 w/ App specific instructions (Score:5, Insightful)
Instead of using a sequence of instructions to perform a common function, the operating system will use a single instruction that causes the entire function to be performed by the POWER5 microprocessor hardware. Examples of these common functions include TCP/IP processing, communications message-passing operations, and virtual memory subsystem operations, to name a few. The interfaces to all of these silicon accelerators will be open so that other operating systems, for example Linux, can take advantage of them.
and
When POWER6 arrives in 2006, it is expected to extend the Fast Path idea to even higher-level software such as DB2 and WebSphere processing. Again, all of the silicon accelerator interfaces will be open, so other software developers wil be able to take advantage of the improved performance.
Bill Gates once said that when a given bit of functionality is sufficiently standardized, it should be part of the OS.
No we will make it part of the CPU.
Re:Power5 and Power6 w/ App specific instructions (Score:2)
It's kind of ironic that something starting out as a RISCish processor is going back to the older mainframe & VAX architectures where single intructions execute what would be lines of code in a high level language.
Not that it would hurt any to have an op that does TCP checksumming in silicon.
Re:Power5 and Power6 w/ App specific instructions (Score:5, Insightful)
Indeed. After contemplating my original post, and your response, I am now thinking App specific stuff in the CPU is BAD.
Back in the good old days Crays had Front End Processors which did all the IO for them. This concept is being resurrected (somewhat) by intelligent NIC's and caching disk controllers.
It would seem counterproductive to me to do TCP checksumming in the cpu. Why not just put some more smarts in the NIC, with some embedded ram, and just make the API stream based. IE Heres a web page, send it, and worry about all the ACK's, etc, for me. Doing all that memory accessing in the CPU would thrash the cache a lot, no? Especially with gigabit getting more common on motherboards (BTW Tyan has a new dual athlon with gigabit, and U320). Not to mention 10G ethernet coming soon. And then there is encryption! We are going to need a dedicated CPU to handle that, or you will not get much work done on your app!
Oh, and you could use n-2 gen chip fabs for these, since they would be smaller cheapers chips with regard to modern cpu's..
No kidding. (Score:2)
Re:No kidding. (Score:3)
Not at all. The problem is that sure it looks good today, but commodity cpus increase in performance much faster than any other kind of processor. So, today putting the network stack in the NIC might look good, but 6 months from now it won't be so good and 1 year from now it will be a bottle-neck.
The industry has been swapping back-and-forth between smart I/O peripherals and dumb ones for at least 20 years now, each time they get to ~100% dumb I/O somebody decides that it would be a good idea to re-invent the past and go for a smart I/O controller, other companies follow suit we get a surge of smart I/O and after a couple of years people start to realize what a drag all these 1-2 year old I/O controllers are on modern systems and they swing back to dumb ones - rinse, lather, repeat.
Also, in the specific case of network stacks there are some smart asm tricks that will do most of grunt work (i.e. checksumming) in the cpu for the same cost as the data-copy itself. So, unless you do dma directly to the user buffers, the hardest part of the network stack is basically free.
Re:No kidding. (Score:2)
Have you been watching the 3D scene? In a fairly short time, we've gone from Voodoo and TNT cards to shader-based Radeon and GeForce cards. For some applications, custom procssors are just what the doctor ordered.
CISC at its finest. (Score:2)
Because you need much more die space for decoding of instructions, it becomes harder to ramp CISC up to higher clock speeds. That's why RISC was introduced.
Now, unless you've been asleep for 12 years, you know that modern x86 CPUs are a combination: CISC instruction set (and benefits thereof) with a fast-path decoder for most commonly used instructions, with a slower conversion for more complex/less used x86 instructions, all of which are crunched through a RISC core which has more registers and other bits to aid parallel pipelining of instructions. So far this has proven to be really great. Transmeta's even taking it a step further by introducing codemorphing, which lets the entire CPU just be a JIT x86 environment running on a VLIW core.
Why are they going this way? It doesn't really seem to make sense compared to the traditional trends in computer processor design.
Re:Power5 and Power6 w/ App specific instructions (Score:2)
I recall the Burroughs B6500/6700/etc machines (late 1960s original design) had a linked-list-lookup instruction (which among other things was used in memory allocation), also some instructions for doing edits on strings. The VAX of course had instructions for calculating polynomials and do cyclic redundancy checks.
So now we get opcodes to do 'send TCP/IP packet' and such? Cool, I'm all for moving functionality to hardware if it's standardized and if it speeds things up. (Which is why we all like real modems over winmodems, right?)
Writing compilers for that stuff gets interesting, though.
Re:Power5 and Power6 w/ App specific instructions (Score:2)
So now we get opcodes to do 'send TCP/IP packet' and such?
I can't wait for the first TCP/IP exploit patch for Linux on this new processor:
1. Power down machine
2. Attach anti-static strap to wrist
3. Remove old CPU
4. Install new CPU
5. Power-up machine
Re:Power5 and Power6 w/ App specific instructions (Score:2)
IBM have a long history of building super-CISC processors.
The S/38 - AS/400 was one good example. The CPU had instructions like 'Create Database'.
But appearances deceive. Such instructions are implemented in software / microcode.
And they are designed to form part of the noose that keeps customers roped in.
It is difficult to write portable software when you have this kind of dirty separation between hardware, OS, and application.
The hint that Websphere and DB2 might use such functions should give room for pause.
Moving OS functions into the CPU is unnecessary, designed to kill portability, and create lock-in.
Computer comeback! (Score:4, Interesting)
IMO, this is only because of Fritz being allowed to make changes to itself to edit its openings. Previous matches were usually charachterized by the games falling into a predictable queen sacrifice. But it looks like by changing the openings around, Fritz is preventing Kramnik from forcing the game into a defensive draw.
Also, The last 2 games have been charachterized by risky Kramnik moves that might be very beneficial against humans, but Fritz is able to play essentially perfect defense. To me, it seems like Kramnik has thrown out his very defensive strategy that gave him a 2 game lead in favor of a more attacking strategy.
5 bucks on Fritz.
sacrifice? (Score:3, Informative)
Amazon's info about QuickSilver (Score:3, Informative)
Pining for Stephenson? (Score:3, Interesting)
His Saddam Hussein Germ Warfare [amazon.com] novel is inexplicably out of print.
Quarter-sized? (Score:2)
Dataplay, the Boulder-based manufacturer of quarter-sized recordable discs and drives...
Does this mean the discs are 25% of the diameter of a regular compact disc? Does it mean they are 25% of the area of a compact disc? Does this mean they are the size of a US 25 cent coin?
Even the Dataplay web site acts coy about the actual size, saying things like: "DataPlay is a miniature media that can be used to play, record and store anything digital." (from the FAQ)
Does anyone know how big these actually are (or were)?
Re:Quarter-sized? (Score:4, Informative)
Re:Quarter-sized? (Score:2)
I'll still be using my MD recorder. :)
Re:Quarter-sized? (Score:2)
Deep Blue creator remarks on Deep Fritz (Score:5, Interesting)
Either way, I don't think this match has anything like the quality of the Kasparov-DB2 match.
Re:Deep Blue creator remarks on Deep Fritz (Score:2)
Hardware speed isn't everything. Why else would programs running on the exact same hardware show such great variation in ability? Fritz might not be able to evaluate as many positions per second as Deep Blue, but it evaluates them better. Kramnik and Kasparov are fairly evenly matched. Fritz seems fairly well matched with Kramnik, and Deep Blue with Kasparov. It doesn't exactly take an advanced degree in math or logic to figure out the transitive relationships and conclude that Fritz and Deep Blue are a lot closer in strength than the raw hardware numbers would indicate.
So why doesn't he? Talk is cheap. Despite all of its raw hardware speed, Deep Blue would not have beaten Kasparov had it not been for Joel Benjamin spoon-feeding it tips on how to beat one specific player. Kramnik, Anand, or any of a half-dozen other top grandmasters would have kicked its ass because it was not tuned to their styles. Fritz, by contrast, is not so reliant on tuning and would probably do better in a tournament setting against multiple top-level opponents. When Hsu can write a program that's even IM level, without having a GM hold his hand, his claim will have some credibility.
Re:Deep Blue creator remarks on Deep Fritz (Score:2)
2) Deep Thought was playing at IM level (maybe even GM level) long before Benjamin got involved. And there's no reason at all, none, zip, zero, to think that Deep Fritz is doing a better evaluation at each node than DB did. Quite the opposite. Fritz has to evaluate each node by running some series of Pentium instructions and it has to carefully balance cycles-per-node against available cycles. Adding more terms slows the evaluation down and limits search depth. Deep Blue used a hardware evaluator and if they wanted more evaluation terms they just added more hardware, keeping the nodes/sec constant (within reason).
3) Deep Thought (and maybe the early Deep Blue) played on ICC as "scratchy" for a long time and creamed many GM's and IM's. I don't remember whether it was DT or Hitech that beat GM Bent Larsen in a tournament game, but DT was certainly stronger than Hitech. DT/DB also played a number of "training" games against GM's and apparently did extremely well. GM Robert Byrne lost a two-game match in Hsu's lab, for example, which he wrote about in the NYT.
4) There's no question that GM involvement (not just from Benjamin, but also Dlugy and others) helped Deep Blue. Similarly, the Fritz developers are also getting plenty of GM assistance. What kind of stupid snipe is that, expecting Hsu (who cheerfully admits not being a good chessplayer) to not have strong players helping him? He's entitled to the same kind of help that the Fritz developers get. Plus don't forget, he has the Deep Blue code to work from (he acquired the rights to it when he left IBM).
5) You are deluded if you think preparing against a specific player makes a big difference in performance against that player. It makes a small difference. Opponents (including DB vs Kasparov) prepare against each other because when they're almost evenly matched, they must gain and use any advantage they can, even small ones. However, when they're not evenly matched to start with, preparation doesn't help. That's why Kasparov regularly crushes strong masters in simuls even though they've prepared against him and he's never heard of them. If you think a weak player can beat a strong one by preparing against the strong player, try preparing against Kasparov (or any other GM) yourself sometime, issue a challenge with enough cash behind it to make it worth the GM's while, and see how well you do.
Just imagine.... (Score:1)
PPCs? Ooh! Where?! (Score:4, Funny)
Dataplay disks dead? (Score:4, Funny)
Don't you wish someone would dream up and create a music player, maybe with a little tiny hard drive in it, something like 20 gigs say, and have a nice big screen and make it light and small? Oh, and add a great user interface and a simple wheel to make it work very easy. And sync it with a desktop. And let me put whatever songs I want on it without having to help pay for Valenti's new pool lining.
If only...
XviD community (Score:4, Informative)
Obligatory Python Quote (Score:3, Funny)
For those of you wondering where the 'box girder bridge' reference originates...
"How To Do It"The cast:
ALAN
John Cleese
NOEL
Graham Chapman
JACKIE
Eric Idle
The sketch:
(Cut to a sign saying 'How to do it'. Music. Pull out to reveal a 'Blue Peter' type set. Sitting casually on the edge of a dais an three presenters in sweaters - Noel, Jackie and Alan - plus a large bloodhound.)
Alan: Hello.
Noel: Hello.
Alan: Well, last week we showed you how to become a gynaecologist. And this week on 'How to do it' we're going to show you how to play the flute, how to split an atom, how to construct a box girder bridge, how to irrigate the Sahara Desert and make vast new areas of land cultivatable, but first, here's Jackie to tell you all how to rid the world of all known diseases.
Jackie: Hello, Alan.
Alan: Hello, Jackie.
Jackie: Well, first of all become a doctor and discover a marvellous cure for something, and then, when the medical profession really starts to take notice of you, you can jolly well tell them what to do and make sure they get everything right so there'll never be any diseases ever again.
Alan: Thanks, Jackie. Great idea. How to play the flute. (picking up a flute) Well here we are. You blow there and you move your fingers up and down here.
Noel: Great, great, Alan. Well, next week we'Ll be showing you how black and white people can live together in peace and harmony, and Alan will be over in Moscow showing us how to reconcile the Russians and the Chinese. So, until next week, cheerio.
Alan: Bye.
Jackie: Bye.
(Children's music.)
Screw Dataplay...Phillips does it better! (Score:2, Informative)
Phillips now has a bluelaser system, working prototype, the size of a two euro coin which holds one gig of data
Re:Screw Dataplay...Phillips does it better! (Score:2)
Welcome back XviD, good riddance DataPlay (Score:3, Interesting)
As for DataPlay... People are pretty happy with compact discs as they stand now. I've never heard any of my friends or their friends or anyone I've met in real life ever tell me they had complaints about the audio quality of CDs. Mostly, people seem to think CDs are just too expensive and a few agree they're too easily scratched. I don't know what kind of crack the inventor of these DataPlay discs was on, but "smaller" is not a good primary selling point. For me, I want as much music available as possible at my fingertips and it was a hard drive based player that provided that. Shame DataPlay wasn't into those, the name would have worked.
Re:Xvid is depricated. (Score:5, Informative)
Re:Xvid is depricated. (Score:5, Funny)
Re:Xvid is depricated. (Score:4, Funny)
Q #131: why does mplayer sucks? :)
A: why not?
Q #132: I can't see any picture, only hear the sound
A: you are blind
Q: #133: I have configured and compiled mplayer, how do I use it?
A: try sticking it up your ass.
Q: mplayer crashing with every files. i've attached output of ls -la /etc/shadow
A: we need also output of cat
Re:Xvid is depricated. (Score:5, Informative)
PS: The word is deprecated. A java compiler would hate you.
xvid is most importantly an mpeg4 encoder (Score:2, Informative)
mplayer is not a video codec.
Mod DOWN. (Score:2, Informative)
MPlayer is a very good media player for Unixes.
Xvid is an open-source mpeg4 video codec. MPlayer competes with Xine. Xvid competes with Divx.
Re:The Java experts don't agree with you (Score:1, Redundant)
So are you, apparently. Perhaps you should try learning to spell from a fucking dictionary [m-w.com] rather than programmers, who are notoriously bad at it.
Re:The Java experts don't agree with you (Score:1, Informative)
The morons at "experts exchange" don't.
Re:The Java experts don't agree with you (Score:2)
Remember deprecation is what happens to old code, and depreciation is what happens to old computers.
And depricated? That's the condition of a BMW after you remove all the lawyers.
deprecate
Pronunciation: 'de-pri-"kAt
Function: transitive verb
Inflected Form(s): -cated; -cating
Etymology: Latin deprecatus, past participle of deprecari to avert by prayer, from de- + precari to pray
Date: 1628
1 a archaic
people -- Tobias Smollett>
2
3 a : PLAY DOWN
b: BELITTLE, DISPARAGE
<the most reluctantly admired and least easily deprecated of
depreciate
Pronunciation: di-'prE-shE-"At
Function: verb
Inflected Form(s): -ated; -ating
Etymology: Late Latin depretiatus, past participle of depretiare, from Latin de- + pretium price
Date: 15th century
transitive senses
1
2
intransitive senses
synonym see DECRY
- depreciable
adjective
- depreciatingly/-shE-"A-ti[ng]-lE/
adverb
- depreciation
noun
- depreciative
-shE-"A-tiv/ adjective
- depreciator
noun
- depreciatory
-"tor-/ adjective
Re:The Java experts don't agree with you (Score:2)
The arguement is whether "deprecate" is spelled "depricate" or "depracate", both of which are wrong.
Re:The Java experts don't agree with you (Score:2)
It's interesting that, of all the moderations applied to your comment, slashdot picks "Offtopic" as the one to display.
Re:The Java experts don't agree with you (Score:2)
Well, I've been planning to mess with slashcode after this semester is over, I guess that's something to look for.
Re:Krahulik.. chess etc.. (Score:3, Interesting)
For Go you need about 10 times as much storage - so Moores law says storing all the moves for about 5 more years.
Re:Krahulik.. chess etc.. (Score:5, Insightful)
I don't understand what you mean by saying that the hard part is writing a random number generator. Random number generation it itself doesn't have anything to do with much. The question is the algorithms used to find the more likely moves.
As for Go needing 10 times as much storage, you are so far off that I worry that you don't know the meaning of the number 10! Each chess move gives about 35 legal options. A player in a Go move has about 200 possible moves on average (the number starts at 361 and mostly goes down from there). After five moves from each player there are about 1.8 billion possible positions in chess -- and 64 trillion for Go. That's a factor of 32,000 more positions, and that's only five moves in. Go games usually have more moves than chess games.
It is really laughable to even suggest that all the possible moves in Go will be stored in a computer within the next 500 years. Though that isn't necessary to beat a human shodan (as I mentioned, chess programs don't evaluate ALL of the possible positions). What's really necessary to beat a human master at Go is to be able to make some judgement on the relative value of different positions. Computers can't currently do that properly, so while a chess computer searches for that perfect move that forces checkmate, the computer playing Go has a hard time understanding what it's supposed to be searching for.
A good article I found and got some numbers from is http://www.anusha.com/times-go.htm [anusha.com].
Programs, not computers. (Score:2)
Well, this really says a lot more about the people writing the Go playing code then it does about the computers themselves.
Re:Programs, not computers. (Score:2)
Re:Krahulik.. chess etc.. (Score:4, Informative)
I've done a lot of research into Combinatorial AI design for my program, Gamazons, which has a branching factor that makes chess look like childs play (the opening move has 3-4k different possibilities). For these types of programs, you've got a few primary factors that make a big difference in how your game plays. You need to search fast & deep, prune out as many paths as you can that won't produce worthwhile results. But then you need really good heuristics that give a value to a board state. You have to be careful that your heuristics are incredibly efficient, because your heuristic function will be run on every node (board state) in the search tree. However, from what I've noticed in my program, the quality of the heuristics is the most critical part of the whole game. It doesn't matter much how deep fast and deep you can search if your searching gives inaccurate values to the board states.
It's the same with chess games. Sure, having lots of big beefy hardware was a nice factor for Big Blue, but the defining factor that really made it shine was the quality of its heuristics. They had on staff a number of grand chessmasters as well as a database of all the big games to develop a good opening book (real chess programs don't start searching & evaluating board states until a good 10-15 moves into the game. They search these out ahead of time and store them on disk as the opening book).
Re:Krahulik.. chess etc.. (Score:2)
Re:Krahulik.. chess etc.. (Score:3, Insightful)
http://www.kuro5hin.org/story/2002/7/11/62356/9
If you don't, a computer sucks at Go because of the exponentially larger solution sets involved.
Re:Krahulik.. chess etc.. (Score:2, Insightful)
Actually, the exponentially larger sets involved are not the main issue. Sure, it makes finding a good move a longer task, but remember Moore's law: computer power grows exponentially too. So, if that was the only problem, we could guaranty that computers would eventually become go experts.
However, the main problem is deeper. It is at the root of the combinatorial game theory that sits behind Fritz' victories: the evaluation function. As mentioned in the article you link to:
Computers are just number crunching machines. If you can't feed them with numbers, them will do you no good. So that's what this evaluation function does: it takes a given board situation, and evaluates its value to the player. These values are computed for all the leaves of the explored moves tree, and are used to prune this tree down. Finally, you are left with the best next move. Therefore, 'best' depends not just on the size of the tree, but also on the quality of the evaluation function.
The bottom line is:
And that is exactly what the problem is. While it is pretty easy to say (compute!) who has an edge on a chess board, we still haven't found a good computational way to do judge the value of a go board situation. Of course experts go players DO have a feeling of a better situation, but that's exactly it: a matter of feelings, not of equations. Until we have put these go feelings into equations, whatever the processing power, computers will just suck at go.
Re:Krahulik.. chess etc.. (Score:2)
More accurately, computers suck at Go because we don't know how the brain works well enough to program computers to think like people (to intuitively see the gestalt of a game position), therefore we must resort to modified brute force heuristics, and the exponentially larger solution set of Go makes it poorly adaptable to this method of programming.
However, if we did know how to program computers to work like brains, it's theoretically possible that even antiquated 386 level hardware would be adequate to beat us at Go, chess or any other sort of board game.
In a nutshell we are talking about the Holy Grail of computing: A.I.
Re:Why the Perl Journal will eventually die (Score:5, Informative)