From following that link, its amazing how cheap this thing is (grape 5).
$40K including an Alpha host and software. Only $10K for the actual superCruncher. Plus its small, so it shouldn't suck up that much power. This is much more powerful than a cluster of 5-7 linux pcs
There's a few things they've done that make Java look better than than real world situations.
1. Medium to big Java apps need 128mb-256mb system RAM to be useable. HotSpot increases memory footprint (uses memory for compiler + bytecode and native code is in memory), but does not enhance every type of app. HotSpot looks great on many benchmarks (small loop intensive apps tested on systems with much memory), but for many apps it slows things down.
2. By pre-running the code for 1 second to get how many iterations to use for 10 seconds, you're making sure that hotspot and JITs fully kick in, without countin any of their execution overhead.
3. Contrary to what you might expect, there's no UI in the game of Life benchmark.
4. The benchmarks are set up to favour run-time optimizations by having function parameters that are constant for long periods of time (ie matrix size)
Java is just fine when you have tons of memory, but if your users have 64mb or less, go with vb or cpp.
The benchmarks in the article have completely avoided any JVM/hotspot initialization overhead, as well as sticking to things JIT compilers are good at.
Anyone can stop any sale on Ebay, by saying that an item infringes any law. If they are mistaken about the item infringing their rights or the law, then they have commited purgury, which is a criminal offense.
Anyone who's item is pulled can resubmit it. Then its up to the first guy to sue the second guy if he still thinks there's a violation.
Considering that the random number sources for all these applications is deterministic based on your computer's clock (ie. no matter how hard you wish, the same specific random number stream is generated), any observed relationship between your wishes and results, is a case of you modifying your intended wishes after seeing the results.
On another note, there was some technique using electrodes to measure alpha and beta waves in the user's brain that claimed people were able to control a mouse using brainwaves. Don't know what became of that project.
The discovery.ca piece oogled the science with a bright eyed moron anchor dork that the show uses.
The piece was pretty much an infomercial for PEAR.
on your advise, I went to PEAR's site and looked at what pdf's they had available (1). Of course this is wasting my time, since their paper like most others, is vague enough not to provide any holes that the reader can simply shoot down the results with.
If I really wanted to, I could run tests until I had some results that showed there was no ESP effect. Wanting the results and patience is enough to generate data that supports the results within a reasonable time in most cases, and so self created statistical validation procedures yields no useful information to outsiders.
The other issue in the context of the overall article, is that even if they have discovered a valid ESP phenomena, its discovery is currently useless to computer input. Even if the interaction of the mind is able to affect a random binary phenomena 55% of the time, that 45% error rate is too high to be useful. Moving a mouse in one of 8 directions (3 bits) would have an 85%+ error rate.
I saw a dicovery channel magazine segment on these guys, and wasn't impressed.
I believe their results are biased.
Its based on a pretty basic statistical phenomenon that unfortunately is not documented very much in research.
If you generate 20 hypotheses (hypothesising structure) on random phenomenon, odds are 1 of them will falsely show structure within the common 19/20 (95%) confidence intervals.
when you combine this with the selective labelling of people as talented and untalented (in ESP), and throwing out of data when the subjects feel tired, or are otherwise displeased with the results, you get the self fullfilling results used to obtain increased funding from the stupid.
icee already knew about sympatico.ca (ISP from Canada--duh). And unless he quickly BS'd the excuse about why it was taking a long time to make an arrest (Cross border evidence) He already knew about Canada too.
I don't see how 2600 told him anything he didn't know, and agree with the original poster.
However, I would say that AI in general has taken great strides in many areas. This is especially true for games.
...I'll use this specific example: The Heroes of Might & Magic... blah blah blah
You stupid fucking cunt! Just because you read "AI" on the back of your stupid game box, doesn't mean your stupid game is doing anything other than Artificial Stupidity, you stupid fucking cunt.
Although precicely categorizing what is and what isn't AI remains elusive and is always shifting, How "fun" publishers choose to tune computer player control, has no bearing on the field.
Of all the 70's over hyping of AI, the only mild widespread success is voice dictation.
I agree with you, and the rsa example is a good one.
Another example are dorks like me, who by the time they're awarded a patent, would be spending 2-3 years forming a company, getting venture capital, applying the software into an asic, and generally trying to get off the ground.
What's a good solution to prevent amazon.com from polluting the world with seemingly unenforceable patents, hurts the small guys.
I think the real solution is patent classes, where some classes have shorter protections than the other. It wouldn't necessarily be all software patents must be in one class.
rather consideration for development time/resources taken to discover and prove the invention, and well as effort required to implement the invention should be taken into consideration when assigning a patent class.
simple ideas like one-click where seminal idea to implementation can be accomplished in less than a week do not merit receiving a patent, but at worst should be awarded the lowest possible patent class.
It will take off quite slowly if its ever successful.
Except for the fridge, I wouldn't spend more than an extra $5 for the net enabled version of the device.
For the fridge, I don't want to upgrade it every 2 years. I can just picture myself spending 2 weeks comparison shopping on the cpus used in each competing fridge. Or postponning the purchase because sony announced that their model that will ship in 6 months has gigabit ethernet, and netscape v7.01.
I guess my wife will see one at the Jones'es, and then that will make us all of a sudden need one.
I don't think you can get both speed/jumping and strength/payload capacity simultaneously.
or its at least a really tough design...
significant increase in strength and payload capabilities means that it has to be self supporting almost.
Having shock absorbtion, ability to turn hips and shoulders, and to just keep your balance in a bipedal system all get compromised.
you almost have to go with caterpilar chains for movement, and so you might as well make a mini tank. Since a mini tank can't go that many more places where a big tank can go, you might as well keep your big tank.
In a lightweight frame approach designed to stress balance for some minimal strength inprovement, anything that enhances your strength, is going to hurt your flexibility. In a quake type combat environment, you need to be able to turn quickly (hip and shoulder flexibility) and aim/fire.
They should be designing boots that make you faster, jump higher in one suit that helps combat situations where nimbleness is needed, and a different strength suit for the guys who have lug around the rocket launchers.
Slashdot Mirror
From following that link, its amazing how cheap this thing is (grape 5).
$40K including an Alpha host and software. Only $10K for the actual superCruncher. Plus its small, so it shouldn't suck up that much power. This is much more powerful than a cluster of 5-7 linux pcs
There's a few things they've done that make Java look better than than real world situations.
1. Medium to big Java apps need 128mb-256mb system RAM to be useable. HotSpot increases memory footprint (uses memory for compiler + bytecode and native code is in memory), but does not enhance every type of app. HotSpot looks great on many benchmarks (small loop intensive apps tested on systems with much memory), but for many apps it slows things down.
2. By pre-running the code for 1 second to get how many iterations to use for 10 seconds, you're making sure that hotspot and JITs fully kick in, without countin any of their execution overhead.
3. Contrary to what you might expect, there's no UI in the game of Life benchmark.
4. The benchmarks are set up to favour run-time optimizations by having function parameters that are constant for long periods of time (ie matrix size)
Java is just fine when you have tons of memory, but if your users have 64mb or less, go with vb or cpp.
The benchmarks in the article have completely avoided any JVM/hotspot initialization overhead, as well as sticking to things JIT compilers are good at.
Java 2 added some accurate cross platform math classes to address these issues.
This works out to 6M/minute, or 100K/sec.
is this 640x480x24bit @ 30fps?
How much of a compression difference is it over MPEG-2?
Anyone can stop any sale on Ebay, by saying that an item infringes any law. If they are mistaken about the item infringing their rights or the law, then they have commited purgury, which is a criminal offense.
Anyone who's item is pulled can resubmit it. Then its up to the first guy to sue the second guy if he still thinks there's a violation.
ru the same guy that did the 'I will choose free bear' (freewill parody) in that recent poll?
:)
entertaining stuff!
Considering that the random number sources for all these applications is deterministic based on your computer's clock (ie. no matter how hard you wish, the same specific random number stream is generated), any observed relationship between your wishes and results, is a case of you modifying your intended wishes after seeing the results.
On another note, there was some technique using electrodes to measure alpha and beta waves in the user's brain that claimed people were able to control a mouse using brainwaves. Don't know what became of that project.
Subspace/Asteroids has explored newtonian physics dogfighting quite a bit. Its plenty fun
Sounds more like a 3d version of subspace, which is my all time fave action game.
http://www.subspacehq.com/
The discovery.ca piece oogled the science with a bright eyed moron anchor dork that the show uses.
The piece was pretty much an infomercial for PEAR.
on your advise, I went to PEAR's site and looked at what pdf's they had available (1). Of course this is wasting my time, since their paper like most others, is vague enough not to provide any holes that the reader can simply shoot down the results with.
If I really wanted to, I could run tests until I had some results that showed there was no ESP effect. Wanting the results and patience is enough to generate data that supports the results within a reasonable time in most cases, and so self created statistical validation procedures yields no useful information to outsiders.
The other issue in the context of the overall article, is that even if they have discovered a valid ESP phenomena, its discovery is currently useless to computer input. Even if the interaction of the mind is able to affect a random binary phenomena 55% of the time, that 45% error rate is too high to be useful. Moving a mouse in one of 8 directions (3 bits) would have an 85%+ error rate.
I saw a dicovery channel magazine segment on these guys, and wasn't impressed.
I believe their results are biased.
Its based on a pretty basic statistical phenomenon that unfortunately is not documented very much in research.
If you generate 20 hypotheses (hypothesising structure) on random phenomenon, odds are 1 of them will falsely show structure within the common 19/20 (95%) confidence intervals.
when you combine this with the selective labelling of people as talented and untalented (in ESP), and throwing out of data when the subjects feel tired, or are otherwise displeased with the results, you get the self fullfilling results used to obtain increased funding from the stupid.
Actually,
icee already knew about sympatico.ca (ISP from Canada--duh).
And unless he quickly BS'd the excuse about why it was taking a long time to make an arrest (Cross border evidence) He already knew about Canada too.
I don't see how 2600 told him anything he didn't know, and agree with the original poster.
Datacasting will be far more useful than HDTV or even SDTV. There seem to be lots more economically useful applications for low definition tv.
You can fit 194 100Kb compressed video streams into one channel.
Porn and action films would have a high demand for higher quality broadcasts. Other stuff would have higher demand if it were cheaper.
The one thing I don't get about datacasting is how does it deal with transmission errors
You stupid fucking cunt! Just because you read "AI" on the back of your stupid game box, doesn't mean your stupid game is doing anything other than Artificial Stupidity, you stupid fucking cunt.
Although precicely categorizing what is and what isn't AI remains elusive and is always shifting, How "fun" publishers choose to tune computer player control, has no bearing on the field.
Of all the 70's over hyping of AI, the only mild widespread success is voice dictation.
There will always be low end and high end appliances.
You can still get a fridge without a water and ice dispenser.
You should always ask yourself why something is news. 99%+ of fraud and security breaches are leaked by competitors.
well corruption would be one plausible explaination how the one-click patent could have been awarded.
I agree with you, and the rsa example is a good one.
Another example are dorks like me, who by the time they're awarded a patent, would be spending 2-3 years forming a company, getting venture capital, applying the software into an asic, and generally trying to get off the ground.
What's a good solution to prevent amazon.com from polluting the world with seemingly unenforceable patents, hurts the small guys.
I think the real solution is patent classes, where some classes have shorter protections than the other. It wouldn't necessarily be all software patents must be in one class.
rather consideration for development time/resources taken to discover and prove the invention, and well as effort required to implement the invention should be taken into consideration when assigning a patent class.
simple ideas like one-click where seminal idea to implementation can be accomplished in less than a week do not merit receiving a patent, but at worst should be awarded the lowest possible patent class.
a STL or java vector is something that's only slightly related to vector processing.
an STL vector is basically an array that can grow dynamically.
vector processing is when you want to do the same operation to each member of an array.
How easy are these c and C++ libraries to use?
Are they saying any vector type processing can be easilly rewritten, and so lots of aps can be enhanced?
I vaguely know altivec is cleaner than the x86 simd stuff, but can the same thing be applied to mmx, 3dnow etc... ?
what parts of their kernel gain performance?
This thing has about a 45 mile max range. That's going at max speed.
So you think you can hop from gas station to gas station? (btw, 4.5mpg is not bad).
It can make an emergency landing with just 2 engines, but what happens when all 4 engines run out of gas?
There marketing this as an RV though, not a commuter vehicle, so I guess they thought of all this. Still is pretty cheap.
It will take off quite slowly if its ever successful.
Except for the fridge, I wouldn't spend more than an extra $5 for the net enabled version of the device.
For the fridge, I don't want to upgrade it every 2 years. I can just picture myself spending 2 weeks comparison shopping on the cpus used in each competing fridge. Or postponning the purchase because sony announced that their model that will ship in 6 months has gigabit ethernet, and netscape v7.01.
I guess my wife will see one at the Jones'es, and then that will make us all of a sudden need one.
I don't think you can get both speed/jumping and strength/payload capacity simultaneously.
or its at least a really tough design...
significant increase in strength and payload capabilities means that it has to be self supporting almost.
Having shock absorbtion, ability to turn hips and shoulders, and to just keep your balance in a bipedal system all get compromised.
you almost have to go with caterpilar chains for movement, and so you might as well make a mini tank. Since a mini tank can't go that many more places where a big tank can go, you might as well keep your big tank.
In a lightweight frame approach designed to stress balance for some minimal strength inprovement, anything that enhances your strength, is going to hurt your flexibility. In a quake type combat environment, you need to be able to turn quickly (hip and shoulder flexibility) and aim/fire.
They should be designing boots that make you faster, jump higher in one suit that helps combat situations where nimbleness is needed, and a different strength suit for the guys who have lug around the rocket launchers.