I hope that this will put some fire into TI to do a better job with their development tools. Primarily I'm pissed about the prices for their compilers and evaluation environments. Their market share is all big customers. All the small shops that I know about are using other products.
As someone who is employed to write an OS, when I read this last Fri I was impressed. This definition leaves little to be argued with. He has defined what an OS is better than half the computer books available. Personally I think that an OS is also the set of utilities used to give the user control over the resource allocation, but this is a weak statement and only applies in certain instances. I define the OS as the kernel+ basic utilities+ libs. These utils can be file management utilities, and utilities to control how much RAM, CPU, disk space, IO, etc an application has access to.
I give M$ room on what is considered a basic utility. Things like regedit are required for real system administration and as long as windows continues to use flat text files to store some configuration information (boot.ini file or whatever its called now) I consider notepad/edit a necessary feature. On the other hand things like wordpad, wang imaging etc, are not part of the OS and M$ really should have moved them into that crappy little bonus pack they try to sell for more than its worth.
My take on the IE issue is that as soon as people began to use HTML as a help format a HTML viewer should have been included somewhere in the libs. This boils down to whether or not you believe that the old winhelp.exe should be considered part of the OS. Is the help function prevalent enough to have a consistent API/Format for applications to create consistent looking help? The real solution probably would have been much like the KDE one. Install a 'light' browser by default and then if people want a better browser they have to get it separately.
The trick is to buffer the data yourself and feed it to the output plugin closer to the actual output play time instead of just filling the buffers till they indicate they are full.
I don't imaging that this is any more difficult than what I had to do to get around the buffering provided by the hardware/driver that I was talking to for output purposes. It was really quite easy, although I dumped the whole scheme and worked around a couple of bugs in the driver to just provide an interface to its buffering scheme. The end result is small and seems to be free of bugs.
ISA is one of the best buses for quick and dirty HW prototyping. It has zero protocol overhead to deal with to get an OS/BIOS to detect a card, its easily accessible from software (PIO and well defined programming specs for setting up IRQ channels), and its easy to interface/work with (I created my first ISA card my freshman year in high school) from a hardware perspective.
Besides its still an order of magnitude faster than existing USB, and all my ISA cards work under every significant PC os while USB/Firewire still don't really work under anything but win98 (I'm talking full support here). My $5 NE2000 clone and SB16 are my fall back hardware when my 100base PCI nic's and funny 3D sound cards don't work under Beos, NT, Linux, xxxBSD, QNX etc... Last thing I need is for my keyboard/mouse/modem/floppy etc to be incompatible too. It really bothered me that M$/Intel didn't force an ISA ver4 on the market when they came out with ISA PNP. They could have fixed things like optional slot select line (fixes PNP), level triggered interrupts, etc. They could have probably done it all with one of the extra pins. A v.4 pin when pulled hi indicates that the card is ver 4 compatible.
Of course what do you expect from a mob of corporations on a planet in a galactic backwater?
On the other hand NT won't support that large of a boot partition (4g if I remember correctly) and getting it to recognize big IDE (>8gig) is a real pain in the ass during install. Once you get it installed though just make a big ass data partition. This backfires though. A lot of apps want to do crap in the \winnt directory and the boot drive. I've been screwed by running out of space on that primary partition.
I imagine that the winner of this contest will have a box that doesn't reboot. Its not hard to belive that there are things that can (and have) happened that keep a machine from shuting down. I have personaly seen shutdown times >45 mins.
BTW: I miss that perfmon tool too and the original poster is correct about lack of support for win98 things in NT. USB is the one that pisses me off the most.
Every article I have read on that site has at least 1 point that gets on my nerves due to a low tech understanding of the issues. He claims that its a compiler difference that causes the big jump in performance on the Coppermine with the Q3 benchmark.
Lets think about this for a minute! If Intel didn't change the SSE core then why would a compiler with SSE changes produce a binary that ran better on the same SSE core? The answer: Because its not the compiler! A better answer would be that the changes intel made to the cache result in significant performance gains in some situations. Tom dumps their tech doc's on what they did:
1 they increased the associatively of the cache 2 they widened the L2 data path to the CPU 3 they decreased the latency 4 Lastly they decreased the size by 1/2.
In general 1 and 4 tend to cancel themselves to give similar performance (pick up an architecture book and read about caches if you don't know what I'm talking about here) So we are left with 2 and 3. Now 2 and 3 tend to allow you to get to the cache faster and get more data per cycle. Now quake is really an tiny engine (significant amount of the time it supposedly fits in L1) accessing a massive amount of data. Now lets assume that quake is so tight that it manages to fetch its data out of L2 cache a very large percentage of its time (as opposed to windows just randomly switching tasks, and using main memory like a big disk cache) now if suddenly your data loads which were always in the cache get to the processor faster keeping it from stalling a pipeline for 5 or 6 cycles what happens?
Memory architecture is a __BIG__ deal with modern CPU's. A very large percentage of time on modern CPU design is spent trying to optimize data accesses. The intel engineers have done their homework. The PC market now considers games the standard benchmark (Quake being the main one, Celery-vs-K6! When was the last time you out typed Word? On the other hand when was the last time your Celery helped you kick that poor K6 owners ass because you were getting an extra 20fps?) so they discovered a way to help quake out while maintaining decent performance with data sets that were more sensitive to cache size rather than access times.
A couple of weeks ago there was a bit of news about South Africa (if I remember correctly) completely ignoring the patents on certain drugs and manufacturing them for their population. Their excuse was that the inflated costs the drug companies charged to cover R/D and make a profit effectively kept large portions of their population from using the drugs. Apparently they formed some kind of organization to attempt get countries to allow the manufacture and sale of at-manufacture-cost drugs for people who were unable to pay full price. This organization is apparently getting a lot of support.
How is it that Katz can mess up such a good topic in such a bad way? I've been saying the last few months that the USA is the only police state in the world where the citizens aren't aware they are living in one. For instance and from the top of my head..
The USA imprisons a larger percentage of its population than any other country in the world.
After you have completed your punishment you still do not have any rights. Felons cannot vote in many states and in others they need special treatment to be allowed to vote. Former felons cannot own firearms. There are others. What really bothers me is the class of crimes that are considered felonies is getting larger every year. For example in FL (where i used to live) it is a felony to trespass on a construction site.
The "zero tolerance" laws that seem to be getting more and more popular. I had a friend who every night when he left work at 12:30 was pulled over and searched because he was 'suspicious' looking.
The fact that every year you need another 'license' to do something that had been done for years without one. License is a nice word for 'permission from the government'.
The police have a 'right' to any information you can provide them on any subject they feel that they need information (otherwise you are obstructing justice) on. If they think you might be guilty of something the police can 'steal' this information from you (read wiretapping).
The government maintains extensive databases of personal information on very large parts of the population (read fingerprint, social security, etc.) and has access to large databases maintained by other sources (note above).
People are 'watched' on a daily basis. Where I live I pass at least a dozen cameras a day that are operated by the local government. At work there are cameras to monitor everything.
The current attitude that 'rights' not explicitly given by the constitution are 'privileges' and therefore you can be forced to either give up the privilege or sign away a right. Does anyone actually believe that driving is a 'privilege' and you don't need that privilege? What kind of life can you maintain in most areas of the country if you don't have access to an automobile?
Many activities while harmless in and of themselves are considered serious crimes. Every year this list keeps growing. Take owning certain substances (drugs or otherwise), building certain devices etc.
People actually take laws that are obviously against the constitution seriously. National news recently pointed to a MI man who was charged and found guilty of public profanity. I should point out that the law they convicted him on was quite old and the ACLU is fighting his sentence.
This is a good question that is often discussed. There are a couple school of thought on it.
1. Larger instructions (read more bits to do the same thing) require higher memory bandwidth. This is a bad thing because (esp on SMP) you need all the bandwidth you can get. More allocated to the instruction stream is less for the data stream. Of course the i-prefetch keeps any memory latency issues from stalling the instruction stream.
2. There is a small group of people who point out that the x86 doesn't really appear to suffer much of a performance loss (note the SPEC numbers) due to its small register set. They point out the fact that the CISC instructions with direct memory reads allow a subset of the memory to be cached in the L1 which effectively gives you access to a small portion of memory at near register access speeds. Given proper register renaming this turns out to be an interesting proposition.
On a side note the PIII vs celery cache size issue is similar to #2. Do you go for a larger slow register set (cache things in registers that are used a lot and use memory operations to L1 for general data manipulation) or a smaller register set that is much faster.
I personally have this theory about the amount of temporary data storage (registers) needed for a process is less than the number provided by most RISC processors. What I find is that I end up using a subset of registers as index and accumulators, and the rest are left unused for long periods of time (note PPC subroutine calling conventions) or used to rotate data from memory or to memory while performing some operation while it is in the register.
I should also say that hard core RISC people are fond of pointing out that it takes the x86 significantly more silicon to achieve the same performance as a RISC machine.
Well the problem with this is the '586' was the pentium and MMX pentium, the Ppro,PII,PIII,celery,Xeon,etc.. are the 686.
Re:Drunk Driving (off topic)
on
Dear Mr. Straw
·
· Score: 1
Actually, In both states I have lived in you can refuse the tests. The state then has a _MUCH_ weaker case. If you refuse the test then it is an instant suspension of your license for 6 months (if I remember correctly.) The bottom line is if you are going to fail the test then don't take it!!! The evidence then becomes stuff like 'the officer found empty bottles in the back of your car and smelled alcohol on your breath after pulling you over for driving funny' which is much easier to avoid (put the bottles in the trash and eat a lot of really strong breath mints). Of course if you refuse the test then you automatically get a night or two in jail along with your suspension, which you would get if you failed too.
This isn't legal advice either and may vary in other states.
Re:Civil Rights in UK (locks on sheds vs crypto)
on
Dear Mr. Straw
·
· Score: 1
This is not legal advice... But I don't think you have to give them the combination to your shed here either. The difference is it may be in your best interest to give them the shed combination even if your are guilty. On the other hand you probably are better off fighting a 'obstruction of justice' case than a full criminal charge if the evidence they need to convict you is in the email. With the shed your not going to stop them from getting the information and the last thing you need is another charge but with the email they may not be able to get the critical info if you don't help them.
If they are to buy anything, the first thing they should buy is Troll Tech! Then they can release Qt under the GPL which would solve a lot of issues with the commercial use of KDE.
It is entirely possible that the problem set is a better fit for the xeons and therfore gains significant performace increases (above the standard xeon 3%-5%) with them over the standard PII/PIII. This could be due to:
1. A piece of code that needs access to >512 megs ram.
2. A piece of code that performs significatly better with >512k of L2 cache.
Although I am not a big fan of saying xeon=server there are cases where the xeon solves a problem MUCH better than the standard PIII. I hope that the appropriate research has been done in this case as apposed to the salesman walking in and saying "you need..."
Nah... People will buy shit, even when they know its shit, especially from M$! '98 crashed during a big demo right before its release. Did that slow down sales?
I agree that the palm will be dead in a couple years if they keep this crap up.. I don't particularly feel the need for a color display,it might be neat but I would prefer the battery life. What pisses me off is that I can get all these goodies (color screen, 32 megs, 133mhz processor, audio record and playback abilities etc.) on a wince device for EXACTLY the same price as a Palm V!!!!! Playing around with them at the local office supply store an uneducated customer isn't even going to look twice at the palm. If they were more reasonably priced (less than $100 for this shit models like the palm III) they might have a fighting chance.
Basically what I'm saying are the people in the 3com marketing department are as stupid as the people at Iomega (just think we could all have zip drives now if they hadn't decided to sell the disks for $10). What palm needs are cheap prices on the BW models and a couple of color/goodies models to satisfy those users who feel the need to keep up with their wince touting buddies.
In the not so distant past, before x11amp became xmms, I made a xmms port to AIX. Then before I had a chance to roll the changes back into the common source 4front took over and closed the CVS server. Just recently they made an OSS port to AIX. Now, there isn't any 'need' for a direct output plugin since you just use the OSS plugin in xmms. The only problem is that there seems to be a conflict of interest on the part of 4front because they are trying to sell copies of OSS for $30. I get the feeling from the complete lack of response from them that they are not interested in the direct output plug-in because it could cause competition for their proprietary (for sale) OSS driver. Look at the message you get when you try to download a cripple ware OSS driver. Please fill out this form before downloading Open Sound System The evaluation copy of OSS comes with a FREE limited time evaluation license which is activated on the day you install OSS. The software will only work for a limited time and has to be unloaded and reloaded into the operating system periodically. You may contact 4Front Technologies for technical support.
Without verifying this I seem to remember that VIA is using direct mapped L2 (L3 on AMD K6-III:>) as opposed to the set associative caches Intel is using. This goes hand in hand with a study I remember where a direct mapped cache in general needed to be twice as large as a 2 way set associative cache to maintain a similar hit rate (read performance). So basically your 1meg (is equal to the 512k from intel hit wise) Of course there are latency issues here too that other readers have commented on.
This discussion is exactly idea I had a couple months ago. I don't think i've used the term 'hacker' in about 2 years because of its misunderstood connotations. So I spent some time thinking about a replacement word/idea. My thoughts soon went on to the Techno-mages in B5 who practice esoteric manipulations of technology that the common man can see only as magic. I found the imagery to be nearly what I wanted. The only problem was the elitist society practiced by the Techn-o-mage. So I decided that I might start using the term Tech-Wizard to describe what I might have normally refereed to as 'hacker'. The term is readily apparent to the non computer user and maintains a link to the old 'Wizard' without the sysadmin/UNIX connotation of the the term. The only real problem is that it is pretty clumsy sounding.
"In response to questions about privacy concerns, Bank United said the iris pictures will not be distributed to anyone outside the bank." Until the government thinks they need the information to help solve some 'crime' and forces the bank to give it up.. Or some company offers the bank a shitload of money for the information. Is there a written contract saying the bank won't change its stance on this in a couple years? This iris scanning stuff is 'cool' technology that has been begging for an application the last 8 years or so IMHO.
Oh, and what is the statistical occurrence of ATM fraud? 1 in 10 million (guessing here) transactions? Is this going to lower it? come on.. Its gee wiz stuff with a real privacy issue associated with it.
This yanks my chain about as much as the finger printing being done in elementary schools for 'child safety' reasons.
Actually, I have a lot of respect for the 6x86 series from cyrix. The problem with the thing is that cyrix insists on that silly PR rating because they cant get the clock rates up (I don't think fabbing them with 486 type technology helps much, but the alpha is fabbed with 486 tech too... he he he) When you compare the cyrix clock to clock with the offerings from intel and AMD the cyrix looks pretty good. Admittedly the math isn't up to intel spec's (which is almost always an unfair comparison because of pentium math pipeline optimizations) but it pretty much matches the AMD at same clocks again. If cyrix could get these things up to 500mhz tomorrow then you would see cyrix giving intel a run for their money again (lest you forget the reason the celery is around in the first place, it was because cyrix was selling a 133mhz cpu at intel's 133mhz price point but calling it a 166 because it got a little better than pentium 166 performance in the benchmark of the day (which was winbench)). Of course cyrix would probably insist on calling it a PR600 which would piss a bunch of quake heads off because they bought a cheap CPU only to discover it didn't do as well in their favorite game as the similarly priced Intel. On the other hand the tiny minority of people who use win '95 for development, surfing the web, or writing papers for school (and later turn it into a linux box) but don't play quake would think it a great cpu because they paid for a 133mhz cpu that gets a lot more performance than anyone else's 133mhz cpu.
BTW has anyone noticed that a PIII-450 on pricewatch is cheaper than a PII-450 or a K6III-450?
About 10 years ago I missed, by 4 blocks, being zoned into a high school where two students and a teacher were killed by a student on a rampage with a gun. This type of stuff has been going on for years.. Doom wasn't around to influence them and I doubt they had an Internet connection.
I think that the 'capacitor' theory is a good one. IMHO censor ship, gun control, safety measures will do little to alleviate this problem. Just about any idiot will tell you that these two were miswired but the question is what in the environment caused them to be miswired or irritated the problem to the point where it reached this magnitude.
I can think of two possible solutions to remove part of the capacitor charging. Make high schools smaller to foster a sense of personal belonging in a community as well as to offer a better learning environment. To small might be bad but I'd be willing to bet that a study of extreme violence verses school size might have some interesting results. Does anyone know of any studies like this? My second suggestions isn't going to be a popular one with most of america either. That is get rid of the students who don't want to be there. I can tell you that most of the 'real' problems when I went to HS were caused by people who didn't want to sit through 6 hours of teaching. That included most of the 'jocks' who pretty much ignored the teachers and did their own thing in class including bugging the hell out of those of us who were actually there for a reason.
Slashdot Mirror
I hope that this will put some fire into TI to do a better job with their development tools. Primarily I'm pissed about the prices for their compilers and evaluation environments. Their market share is all big customers. All the small shops that I know about are using other products.
As someone who is employed to write an OS, when I read this last Fri I was impressed. This definition leaves little to be argued with. He has defined what an OS is better than half the computer books available. Personally I think that an OS is also the set of utilities used to give the user control over the resource allocation, but this is a weak statement and only applies in certain instances. I define the OS as the kernel+ basic utilities+ libs. These utils can be file management utilities, and utilities to control how much RAM, CPU, disk space, IO, etc an application has access to.
I give M$ room on what is considered a basic utility. Things like regedit are required for real system administration and as long as windows continues to use flat text files to store some configuration information (boot.ini file or whatever its called now) I consider notepad/edit a necessary feature. On the other hand things like wordpad, wang imaging etc, are not part of the OS and M$ really should have moved them into that crappy little bonus pack they try to sell for more than its worth.
My take on the IE issue is that as soon as people began to use HTML as a help format a HTML viewer should have been included somewhere in the libs. This boils down to whether or not you believe that the old winhelp.exe should be considered part of the OS. Is the help function prevalent enough to have a consistent API/Format for applications to create consistent looking help? The real solution probably would have been much like the KDE one. Install a 'light' browser by default and then if people want a better browser they have to get it separately.
The trick is to buffer the data yourself and feed it to the output plugin closer to the actual output play time instead of just filling the buffers till they indicate they are full.
I don't imaging that this is any more difficult than what I had to do to get around the buffering provided by the hardware/driver that I was talking to for output purposes. It was really quite easy, although I dumped the whole scheme and worked around a couple of bugs in the driver to just provide an interface to its buffering scheme. The end result is small and seems to be free of bugs.
ISA is one of the best buses for quick and dirty HW prototyping. It has zero protocol overhead to deal with to get an OS/BIOS to detect a card, its easily accessible from software (PIO and well defined programming specs for setting up IRQ channels), and its easy to interface/work with (I created my first ISA card my freshman year in high school) from a hardware perspective.
Besides its still an order of magnitude faster than existing USB, and all my ISA cards work under every significant PC os while USB/Firewire still don't really work under anything but win98 (I'm talking full support here). My $5 NE2000 clone and SB16 are my fall back hardware when my 100base PCI nic's and funny 3D sound cards don't work under Beos, NT, Linux, xxxBSD, QNX etc... Last thing I need is for my keyboard/mouse/modem/floppy etc to be incompatible too. It really bothered me that M$/Intel didn't force an ISA ver4 on the market when they came out with ISA PNP. They could have fixed things like optional slot select line (fixes PNP), level triggered interrupts, etc. They could have probably done it all with one of the extra pins. A v.4 pin when pulled hi indicates that the card is ver 4 compatible.
Of course what do you expect from a mob of corporations on a planet in a galactic backwater?
Yes it should have had a 'funny' rating but..
On the other hand NT won't support that large of a boot partition (4g if I remember correctly) and getting it to recognize big IDE (>8gig) is a real pain in the ass during install. Once you get it installed though just make a big ass data partition. This backfires though. A lot of apps want to do crap in the \winnt directory and the boot drive. I've been screwed by running out of space on that primary partition.
I imagine that the winner of this contest will have a box that doesn't reboot. Its not hard to belive that there are things that can (and have) happened that keep a machine from shuting down. I have personaly seen shutdown times >45 mins.
BTW: I miss that perfmon tool too and the original poster is correct about lack of support for win98 things in NT. USB is the one that pisses me off the most.
Every article I have read on that site has at least 1 point that gets on my nerves due to a low tech understanding of the issues. He claims that its a compiler difference that causes the big jump in performance on the Coppermine with the Q3 benchmark.
Lets think about this for a minute! If Intel didn't change the SSE core then why would a compiler with SSE changes produce a binary that ran better on the same SSE core? The answer: Because its not the compiler! A better answer would be that the changes intel made to the cache result in significant performance gains in some situations. Tom dumps their tech doc's on what they did:
1 they increased the associatively of the cache
2 they widened the L2 data path to the CPU
3 they decreased the latency
4 Lastly they decreased the size by 1/2.
In general 1 and 4 tend to cancel themselves to give similar performance (pick up an architecture book and read about caches if you don't know what I'm talking about here) So we are left with 2 and 3. Now 2 and 3 tend to allow you to get to the cache faster and get more data per cycle. Now quake is really an tiny engine (significant amount of the time it supposedly fits in L1) accessing a massive amount of data. Now lets assume that quake is so tight that it manages to fetch its data out of L2 cache a very large percentage of its time (as opposed to windows just randomly switching tasks, and using main memory like a big disk cache) now if suddenly your data loads which were always in the cache get to the processor faster keeping it from stalling a pipeline for 5 or 6 cycles what happens?
Memory architecture is a __BIG__ deal with modern CPU's. A very large percentage of time on modern CPU design is spent trying to optimize data accesses. The intel engineers have done their homework. The PC market now considers games the standard benchmark (Quake being the main one, Celery-vs-K6! When was the last time you out typed Word? On the other hand when was the last time your Celery helped you kick that poor K6 owners ass because you were getting an extra 20fps?) so they discovered a way to help quake out while maintaining decent performance with data sets that were more sensitive to cache size rather than access times.
A couple of weeks ago there was a bit of news about South Africa (if I remember correctly) completely ignoring the patents on certain drugs and manufacturing them for their population. Their excuse was that the inflated costs the drug companies charged to cover R/D and make a profit effectively kept large portions of their population from using the drugs. Apparently they formed some kind of organization to attempt get countries to allow the manufacture and sale of at-manufacture-cost drugs for people who were unable to pay full price. This organization is apparently getting a lot of support.
How is it that Katz can mess up such a good topic in such a bad way? I've been saying the last few months that the USA is the only police state in the world where the citizens aren't aware they are living in one. For instance and from the top of my head..
The USA imprisons a larger percentage of its population than any other country in the world.
After you have completed your punishment you still do not have any rights. Felons cannot vote in many states and in others they need special treatment to be allowed to vote. Former felons cannot own firearms. There are others. What really bothers me is the class of crimes that are considered felonies is getting larger every year. For example in FL (where i used to live) it is a felony to trespass on a construction site.
The "zero tolerance" laws that seem to be getting more and more popular. I had a friend who every night when he left work at 12:30 was pulled over and searched because he was 'suspicious' looking.
The fact that every year you need another 'license' to do something that had been done for years without one. License is a nice word for 'permission from the government'.
The police have a 'right' to any information you can provide them on any subject they feel that they need information (otherwise you are obstructing justice) on. If they think you might be guilty of something the police can 'steal' this information from you (read wiretapping).
The government maintains extensive databases of personal information on very large parts of the population (read fingerprint, social security, etc.) and has access to large databases maintained by other sources (note above).
People are 'watched' on a daily basis. Where I live I pass at least a dozen cameras a day that are operated by the local government. At work there are cameras to monitor everything.
The current attitude that 'rights' not explicitly given by the constitution are 'privileges' and therefore you can be forced to either give up the privilege or sign away a right. Does anyone actually believe that driving is a 'privilege' and you don't need that privilege? What kind of life can you maintain in most areas of the country if you don't have access to an automobile?
Many activities while harmless in and of themselves are considered serious crimes. Every year this list keeps growing. Take owning certain substances (drugs or otherwise), building certain devices etc.
People actually take laws that are obviously against the constitution seriously. National news recently pointed to a MI man who was charged and found guilty of public profanity. I should point out that the law they convicted him on was quite old and the ACLU is fighting his sentence.
This is a good question that is often discussed. There are a couple school of thought on it.
1. Larger instructions (read more bits to do the same thing) require higher memory bandwidth. This is a bad thing because (esp on SMP) you need all the bandwidth you can get. More allocated to the instruction stream is less for the data stream. Of course the i-prefetch keeps any memory latency issues from stalling the instruction stream.
2. There is a small group of people who point out that the x86 doesn't really appear to suffer much of a performance loss (note the SPEC numbers) due to its small register set. They point out the fact that the CISC instructions with direct memory reads allow a subset of the memory to be cached in the L1 which effectively gives you access to a small portion of memory at near register access speeds. Given proper register renaming this turns out to be an interesting proposition.
On a side note the PIII vs celery cache size issue is similar to #2. Do you go for a larger slow register set (cache things in registers that are used a lot and use memory operations to L1 for general data manipulation) or a smaller register set that is much faster.
I personally have this theory about the amount of temporary data storage (registers) needed for a process is less than the number provided by most RISC processors. What I find is that I end up using a subset of registers as index and accumulators, and the rest are left unused for long periods of time (note PPC subroutine calling conventions) or used to rotate data from memory or to memory while performing some operation while it is in the register.
I should also say that hard core RISC people are fond of pointing out that it takes the x86 significantly more silicon to achieve the same performance as a RISC machine.
Well the problem with this is the '586' was the pentium and MMX pentium, the Ppro,PII,PIII,celery,Xeon,etc.. are the 686.
Actually, In both states I have lived in you can refuse the tests. The state then has a _MUCH_ weaker case. If you refuse the test then it is an instant suspension of your license for 6 months (if I remember correctly.) The bottom line is if you are going to fail the test then don't take it!!! The evidence then becomes stuff like 'the officer found empty bottles in the back of your car and smelled alcohol on your breath after pulling you over for driving funny' which is much easier to avoid (put the bottles in the trash and eat a lot of really strong breath mints). Of course if you refuse the test then you automatically get a night or two in jail along with your suspension, which you would get if you failed too.
This isn't legal advice either and may vary in other states.
This is not legal advice... But I don't think you have to give them the combination to your shed here either. The difference is it may be in your best interest to give them the shed combination even if your are guilty. On the other hand you probably are better off fighting a 'obstruction of justice' case than a full criminal charge if the evidence they need to convict you is in the email. With the shed your not going to stop them from getting the information and the last thing you need is another charge but with the email they may not be able to get the critical info if you don't help them.
If they are to buy anything, the first thing they should buy is Troll Tech! Then they can release Qt under the GPL which would solve a lot of issues with the commercial use of KDE.
It is entirely possible that the problem set is a better fit for the xeons and therfore gains significant performace increases (above the standard xeon 3%-5%) with them over the standard PII/PIII. This could be due to:
1. A piece of code that needs access to >512 megs ram.
2. A piece of code that performs significatly better with >512k of L2 cache.
Although I am not a big fan of saying xeon=server there are cases where the xeon solves a problem MUCH better than the standard PIII. I hope that the appropriate research has been done in this case as apposed to the salesman walking in and saying "you need..."
Nah... People will buy shit, even when they know its shit, especially from M$! '98 crashed during a big demo right before its release. Did that slow down sales?
How about the important link... :> LP's statement against the Digital Telephony and Communications Privacy Improvement Act of 1994. "Therefore, BE IT RESOLVED that the Libertarian National Committee opposes enactment of the proposed Digital Telephony Act as a serious infringement of civil liberties and a gross violation of property rights."
I agree that the palm will be dead in a couple years if they keep this crap up.. I don't particularly feel the need for a color display ,it might be neat but I would prefer the battery life. What pisses me off is that I can get all these goodies (color screen, 32 megs, 133mhz processor, audio record and playback abilities etc.) on a wince device for EXACTLY the same price as a Palm V!!!!! Playing around with them at the local office supply store an uneducated customer isn't even going to look twice at the palm. If they were more reasonably priced (less than $100 for this shit models like the palm III) they might have a fighting chance.
Basically what I'm saying are the people in the 3com marketing department are as stupid as the people at Iomega (just think we could all have zip drives now if they hadn't decided to sell the disks for $10). What palm needs are cheap prices on the BW models and a couple of color/goodies models to satisfy those users who feel the need to keep up with their wince touting buddies.
In the not so distant past, before x11amp became xmms, I made a xmms port to AIX. Then before I had a chance to roll the changes back into the common source 4front took over and closed the CVS server. Just recently they made an OSS port to AIX. Now, there isn't any 'need' for a direct output plugin since you just use the OSS plugin in xmms. The only problem is that there seems to be a conflict of interest on the part of 4front because they are trying to sell copies of OSS for $30. I get the feeling from the complete lack of response from them that they are not interested in the direct output plug-in because it could cause competition for their proprietary (for sale) OSS driver. Look at the message you get when you try to download a cripple ware OSS driver.
Please fill out this form before downloading Open Sound System The evaluation copy of OSS comes with a FREE limited time evaluation license which is activated on the day you install OSS. The software will only work for a limited time and has to be unloaded and reloaded into the operating system periodically. You may contact 4Front Technologies for technical support.
Without verifying this I seem to remember that VIA is using direct mapped L2 (L3 on AMD K6-III:>) as opposed to the set associative caches Intel is using. This goes hand in hand with a study I remember where a direct mapped cache in general needed to be twice as large as a 2 way set associative cache to maintain a similar hit rate (read performance). So basically your 1meg (is equal to the 512k from intel hit wise) Of course there are latency issues here too that other readers have commented on.
This discussion is exactly idea I had a couple months ago. I don't think i've used the term 'hacker' in about 2 years because of its misunderstood connotations.
So I spent some time thinking about a replacement word/idea. My thoughts soon went on to the Techno-mages in B5 who practice esoteric manipulations of technology that the common man can see only as magic. I found the imagery to be nearly what I wanted. The only problem was the elitist society practiced by the Techn-o-mage. So I decided that I might start using the term Tech-Wizard to describe what I might have normally refereed to as 'hacker'. The term is readily apparent to the non computer user and maintains a link to the old 'Wizard' without the sysadmin/UNIX connotation of the the term. The only real problem is that it is pretty clumsy sounding.
"In response to questions about privacy concerns, Bank United said the iris pictures will not be distributed to anyone outside the bank."
Until the government thinks they need the information to help solve some 'crime' and forces the bank to give it up.. Or some company offers the bank a shitload of money for the information. Is there a written contract saying the bank won't change its stance on this in a couple years? This iris scanning stuff is 'cool' technology that has been begging for an application the last 8 years or so IMHO.
Oh, and what is the statistical occurrence of ATM fraud? 1 in 10 million (guessing here) transactions? Is this going to lower it? come on.. Its gee wiz stuff with a real privacy issue associated with it.
This yanks my chain about as much as the finger printing being done in elementary schools for 'child safety' reasons.
Actually, I have a lot of respect for the 6x86 series from cyrix. The problem with the thing is that cyrix insists on that silly PR rating because they cant get the clock rates up (I don't think fabbing them with 486 type technology helps much, but the alpha is fabbed with 486 tech too... he he he) When you compare the cyrix clock to clock with the offerings from intel and AMD the cyrix looks pretty good. Admittedly the math isn't up to intel spec's (which is almost always an unfair comparison because of pentium math pipeline optimizations) but it pretty much matches the AMD at same clocks again. If cyrix could get these things up to 500mhz tomorrow then you would see cyrix giving intel a run for their money again (lest you forget the reason the celery is around in the first place, it was because cyrix was selling a 133mhz cpu at intel's 133mhz price point but calling it a 166 because it got a little better than pentium 166 performance in the benchmark of the day (which was winbench)). Of course cyrix would probably insist on calling it a PR600 which would piss a bunch of quake heads off because they bought a cheap CPU only to discover it didn't do as well in their favorite game as the similarly priced Intel. On the other hand the tiny minority of people who use win '95 for development, surfing the web, or writing papers for school (and later turn it into a linux box) but don't play quake would think it a great cpu because they paid for a 133mhz cpu that gets a lot more performance than anyone else's 133mhz cpu.
BTW has anyone noticed that a PIII-450 on pricewatch is cheaper than a PII-450 or a K6III-450?
For the most part the disks spin at the same rate on the inner tracks as the outer ones but the outer tracks have more sectors.
About 10 years ago I missed, by 4 blocks, being zoned into a high school where two students and a teacher were killed by a student on a rampage with a gun. This type of stuff has been going on for years.. Doom wasn't around to influence them and I doubt they had an Internet connection.
I think that the 'capacitor' theory is a good one. IMHO censor ship, gun control, safety measures will do little to alleviate this problem. Just about any idiot will tell you that these two were miswired but the question is what in the environment caused them to be miswired or irritated the problem to the point where it reached this magnitude.
I can think of two possible solutions to remove part of the capacitor charging. Make high schools smaller to foster a sense of personal belonging in a community as well as to offer a better learning environment. To small might be bad but I'd be willing to bet that a study of extreme violence verses school size might have some interesting results. Does anyone know of any studies like this? My second suggestions isn't going to be a popular one with most of america either. That is get rid of the students who don't want to be there. I can tell you that most of the 'real' problems when I went to HS were caused by people who didn't want to sit through 6 hours of teaching. That included most of the 'jocks' who pretty much ignored the teachers and did their own thing in class including bugging the hell out of those of us who were actually there for a reason.