I haven't had any problem so far on my new PC. It must be driver related problem. If driver update doesn't help then there is no other option but to upgrade hardware ( remember Vista ), which might not be an option for older PCs.
Funny thing is that Microsoft made HAL workaround to extend memory address space, kinda fixed driver DMA related problems, but they haven't provided fix for their own system reporting tool.
Example: My old PC has 1 GB of RAM and System Proprieties reported 1.00 GBs.
My new PC has 4GBs of RAM and and System Proprieties reported 3.12 GBs. That is quite strange, considering that both PCs have the same version of Windows.
There is another switch/3GB which offsets memory ratio allowing user application to use 3GB of RAM. I haven't tried to disable swap, but I have feeling that it won't help either.
I have tested that. Windows XP SP2 uses 4GB of RAM with or without PAE. With PAE it can swap and use all 4GB of RAM. BTW I never update Windows after installation, usually that only makes Windows unstable after some time.
The problem is that unpatched Windows XP SP2 Professional reports 3.something GBs of RAM even if it has 4GB installed.
As Your quote shows, the bug shows only in drivers that have improper implementation of double buffering. For an 32-bit application that is PAE unaware, memory address range is 4GB whatever page it uses, so *if* BSOD doesn't occur in a short time, it probably will not occur at all. OK, that is hard one to say on windows but I haven't got any problem at all so far.
In boot.ini file add switch/PAE at the end of multi(0)disk.. line
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="Microsoft Windows XP Professional"/noexecute=optin/fastdetect/usepmtimer/PAE
It will not solve Your PR problem nor will solve the problem with incorrect reporting of available RAM, but will allow 32-bit Windows XP Professional to use all of it.
In my experience, most programs / games can't use all 4GB of RAM, but if user is running more than one RAM hungry application (multitasks), 4GBs becomes useful.
Also we have to think about future Vista service packs so, 4GB is must have:)
But, 96 KHz sampling? You do know the Nyquist theorem, don't you? To preserve sample phase, and correct levels at high frequencies, sampling rate must be at least four times higher than highest frequency in sample. That is an audiophile rule and it is called oversampling.
Example 1: Try to sample signal that contains two frequencies of equal volume slightly below Nyquist frequency. Sample will contain both frequencies but with changed levels, and offset phases.
Example 2: Sample signal with constant amplitude that sweeps form one high frequency to another, both of them just below Nyquist frequency. Sample will contain signal that varies in amplitude level relative to frequency in sample.
That 48 KHz, even with 24-bit precision, should take care of all sounds possible for the human to hear? 48 KHz doesn't take care enough. It is the precision of phases and levels of harmonics that ears are sensitive about. If everything is in phase and with correct levels, it's a tone, if it is out of phase then it's just another noise.
When CD audio was introduced some compromises had to be made. One of them was medium size vs play time. In the end sampling rate suffered to double play time. The result of lower sampling rate is that noises are reproduced with reduced quality as they are not continuous time signals at that sampling rate and are not oversampled.
24 bit resolution does nothing better to sound quality as it is overkill. Good Hi-Fi equipment usually can't output dynamic range more than 80dB, common 5.1 systems are even worse than that as they are high on distortion and can't provide proper signal to noise ratio. Portable audio equipment is just slightly better than home theatre systems. The main problem is in digital to analog conversion, amplification of signal and speakers / headphones ( jitter, noise, slewing, non-linear mapping of input value to output voltage, clipping, saturation, etc.
16 bit ( 98dB ) is good enough, I hope 96 KHz sampling rate will become standard in consumer lossless audio.
I did some experiments with acetone on two stroke lawnmower and chainsaw. I used 3ml per liter of gasoline with 2% of two-stroke oil. While I couldn't measure "mpg" I did notice more power.
It seems that acetone does something to surface tension of gasoline, maybe affects the way that carburetors work allowing smaller fuel droplets to form.
I don't see why 35 mpg is big deal...
I had water in the fuel tank of my car once. I wanted to get rid of water without actually removing fuel tank. Instead of removing the fuel tank and cleaning it, I did pour in some anhydrous ethanol. I'd say that I poured in 0.5 liters ethanol on 6 liters of fuel. That fixed problem with engine pops and choking but totally f***ed up engine power and made another problem. It did something to the fuel efficiency. For next 200 kilometers car was using much more fuel than it used before I added ethanol! I think it used more than 18 L/100km.
One thing is sure, I'll never ever add another drop of ethanol in gasoline again!!!
btw, hate bring bad news, ethanol will prevent water to freeze.
I agree absolutely.
I did some experiments with flat 15 x 15 cm electrostatic mid-range / tweeter speaker more than 15 years ago. I found out that high frequency range or anything above 10KHz travels trough the air almost with no dispersion. Also I got bad ear ringing form just 2W of sound power. OK, I was 3 meters away from the speaker and in a closed room, but I keep wondering how safe is this and what will come next.
I guess that holographic advertising beaming into eyes and perfume carrying microcapsule shooted at pedestrians are to come in next 10 years.
every time I open Firefox or Winamp, I see a clear jump in my RAM memory? What I mean is...why is it that Vista sit at idle at around ~500mb, and the moment I open Firefox (which I have been using on a daily basis for months), it jump to around 650-700mb? If it is already cached, shouldn't it stay relatively constant? It seems that Vista is using a portion of RAM to pre-load most frequently used executables. This was reserved for privileged applications but on windows Vista it is a feature, and it is called superfetch. It is essentially self tuning application pre-loading disk cache.
The problem is that executable can not be executed while in superfetch buffer, they need to be loaded to RAM. Even if most of FireFox is in superfetch buffer it must load to RAM in order to run. If You are viewing pages with the java applets, then java virtual machine is in disk cache and will be loaded to RAM too. Also since You are using WinAmp and other programs, they too are loaded into superfetch buffer taking even more RAM.
That clears out why RAM usage jumps every time You start FireFox or WinAmp.
I remember slightly better way to prefetch applications. In the mid '80s hard drives and RAM were expensive. Home computers had 1 megabyte of RAM or less. Most home computers had no hard drive at all, instead they had one or two floppy drives. Floppy drives are generally slow, and couldn't top more than 20 kilobytes per second. To speed up DOS scripting Commodore corporation promoted residence. Once resident executable (an program or dos command) takes slightly more RAM than it uses on disk, and can be executed as many times as it is needed without any need for (floppy) disk and launches instantaneously. Also resident executable could be removed to free RAM. Unlike superfetch, resident system took only slightly more RAM than executable would need for an operation, and disk cache or RAM could bee freed.
I take it these changes under the hood involved putting in a lot of 'wait' commands Windows 3.0 to Windows 3.11 transition and enhancement was mostly done by removing NOP ( wait ) commands from kernel. I did some assembler coding back then, so I was interested in the way that *they* are doing it.
I'm expecting future windows XP updates to do quite opposite, to bloat XP kernel until they make it slower than any Vista out there. At some point, Vista will become valuable upgrade to anyone willing to stay on Microsoft's rails.
On C=64 any I/O is slow without turbo / FastLoad. I did not mind back then. After all, it only had 64KB of RAM, it never needed a lot time to load, at least not with properly cracked games as all of them used some kind of turbo / FastLoad.
Anybody remember SYS64738? Gotta love it. I do. I use it as a password.
Oh, s**t.
So if my experience is anything to go by, you'ld have to be a real enthusiast and pretty handy with a soldering gun to have one still working after all this time. I have one still working.
I remember I had to hammer the potting compound of the power supply to free the transformer and the PCB.
Tape drive was broken beyond repair, so I made interface to standard Hi-Fi tape deck. It was simple smith-trigger made from CMOS gate IC and few other components. With such interface any tape drive or line-level sound source is good enough.
I use PC / MP3 player to load the games and PC's sound card line input to save them.
I am not quite certain why the internal fuse blows up but it might occur because of short circuit in the tape drive cable.
Another solution, would be to open up more bandwidth to the 802.11x protocols. Looks like 802.11n is the new one that maybe could satisfy the need for the bandwidth, and maybe slightly better range.
However that won't solve the main problem, interference effects, as parent noted.
In an effort to appease the same treehuggers we were periodically stuck with forced solar panels on rovers instead of nuclear power, which among other things forces our Mars rovers to hibernate through the winter instead of working as usual. RTG's (nuclear powered thermal generators) are:
1. heavy: the size of one that would be of use is so great that rover must be made huge, and expensive / impossible to launch to Mars. NASA's choice was solar, I guess they know better.
2. dangerous: in case of bad launch someone has to find damn thing, or its peaces. Solar panels are safe to be left where they are...
Earth/Sun Lagrange points 4 and 5 might be bit dusty. While radio waves wouldn't have lot trouble with the dust, it might be fatal for the laser beam data link.
Cylinders and pistons and piston rings and valves are not designed to be hit with water while at operating temps. They could be hit with water spray. That will solve some problems. Ford has some solution with water injection cooling.
A less catastrophic way to harness this wasted heat would be to fill the cooling system with plain water and let it boil, then use that steam to run a steam engine or turbine. It is called turbo steamer and it's made by BMW. Google for it. Gizmag has an article with great details on it.
Air car could solve some big problems in automobiles: engine overheating, wimpy torque at low RPM, pollution in big cities, need for expensive fuels and lubricants...
Unlike the ICE, air car has no problems with overheating, in fact it will use the heat as an bonus to extract more power form compressed air...
The possible disadvantages would be low specific power of engine, and constant need for piston heating in cold weather / climate... so I guess we'll never see them in F1 cars...
... low yield EMP cannon is dirt cheap to make, and if it is tuned to cell phone receiving frequency it will fry the receiving circuit silently, leaving the phone without network.
However such EMP cannon will probably kill nearby pacemakers...
An Faraday cage is great idea, and is easy to control what comes in it or out of it.
That it is much better solution than jammer or EMP cannon, and certainly will not hurt anyone.
Help your friends and family upgrade or patch Windows. I am running unpatched windows and it is clean. Worm is spreading using so called "advanced technologies". As long as user/s use their browsers in HTML only mode, and "advanced technologies" aka services are disabled there is no way that anything can infect their system. OK, I admit I am using freeware software for browsing / e-mail / protection and I do hide behind NAT / masquerade.
It's the damn entangled-dll-jungle-hell. It's so hard to make all functional and open yet secured and reliable at the same time.
Well old news it is. I used to have RAM DISK for keeping environment variables and temp files on it. It worked much faster than swapping on hard drive. Yes I did use the bootable RAM disks to boot games and they did boot and worked much faster.
That was 15 years ago. I had 2.5 megs of RAM.
Now people think that their PCI, AGP or PCIe video cards will swap slower than their lame hard drives. How lame is that?
--
Don't mod me down if You disagree or don't understand my post. Think trough and use moderating options reasonably.
The Biosphere 2 looked like romantic, religious recreation of Noah's Ark more than it looked like a scientific experiment. It was an attempt to establish high biodiversity ecosystem on a small surface area. I was doomed to fail because small surface area ecosystems can not sustain high species richness. As someone already said if they went for 100% utilitarian O2 producing plants and food crops they could have more success with it.
Self-sufficient space habitat is different because it will actually have closed loop atmosphere and water recycling. It will use all available technologies, physical, chemical and biological processes to recycle water and atmosphere instead of relaying just on mother nature to do the job. They intend to grow their food. They will actually try to live in there up to three years using only resources that habitat can provide.
Then slightly different approach might yield better results. Machinery that melts rock instead of pulverizing it will probably make smooth walls that are easy to seal and hard enough to support itself. RTGs could power melting tool.
The problem with modern antivirus software is that they relay mostly on definitions.
Definitions are by definition either obsolete, ether will be updated next time when user hits update button. In both cases antivirus software relays completely on heuristic detection, by default turned on to lightest level of detection to reduce resource hoging and nag-like behaviour.
I can remember the viruses of late 90s. Most of them were quite small, had exceptional hiding capabilities and had used BIOS functions directly, avoiding detection easily. They did nasty stuff, mike256, WinWierd32 anyone ?
It is true, Windows XP SP2 can be "tuned" to be passively safe but even then You need something to immunize and check your system.
I ma using SpybotSD for a long time, without tea-timer feature, to immunize the system.
Using a browser that uses security ratings on sites, and can open unrated sites with disabled scripting can help a lot.
I use Netscape 8.1.3, as it can open sites in 3 different security levels, + one custom.
Common sense helps allot on the Internet ( downloading cracks anyone ? ) so little paranoia makes healthy Windoze. I guess that FireFox can be secured using noscript and similar plugins, but I have never tried it.
BTW I don't use IE / Outlook since 1998, and did my Windows install more than 2 years ago, so much for the security record.
You got all Your facts right.
Windows users can be dumb and work on bot camp or smart and never have any virus / worm / trojan even with no antivirus software installed at all.
So, they're taking an inherently flawed product (an OS that can be compromised in minutes)... Yes.
And as for being immediately compromised, doesn't anyone use a router with their broadband that has a firewall in it? Not everyone uses external firewall.
Firewall will not help against true viruses, but if is good can help against worms and trojans by preventing them to spread out from the infected machine. The problem is that firewall does not prevent infection, only can block port / protocol.
Slashdot Mirror
I haven't had any problem so far on my new PC. It must be driver related problem. If driver update doesn't help then there is no other option but to upgrade hardware ( remember Vista ), which might not be an option for older PCs. /3GB which offsets memory ratio allowing user application to use 3GB of RAM. I haven't tried to disable swap, but I have feeling that it won't help either.
Funny thing is that Microsoft made HAL workaround to extend memory address space, kinda fixed driver DMA related problems, but they haven't provided fix for their own system reporting tool.
Example: My old PC has 1 GB of RAM and System Proprieties reported 1.00 GBs.
My new PC has 4GBs of RAM and and System Proprieties reported 3.12 GBs. That is quite strange, considering that both PCs have the same version of Windows.
There is another switch
I have tested that. Windows XP SP2 uses 4GB of RAM with or without PAE. With PAE it can swap and use all 4GB of RAM. BTW I never update Windows after installation, usually that only makes Windows unstable after some time.
The problem is that unpatched Windows XP SP2 Professional reports 3.something GBs of RAM even if it has 4GB installed.
As Your quote shows, the bug shows only in drivers that have improper implementation of double buffering. For an 32-bit application that is PAE unaware, memory address range is 4GB whatever page it uses, so *if* BSOD doesn't occur in a short time, it probably will not occur at all. OK, that is hard one to say on windows but I haven't got any problem at all so far.
In boot.ini file add switch /PAE at the end of multi(0)disk.. line
/noexecute=optin /fastdetect /usepmtimer /PAE
:)
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="Microsoft Windows XP Professional"
It will not solve Your PR problem nor will solve the problem with incorrect reporting of available RAM, but will allow 32-bit Windows XP Professional to use all of it. In my experience, most programs / games can't use all 4GB of RAM, but if user is running more than one RAM hungry application (multitasks), 4GBs becomes useful.
Also we have to think about future Vista service packs so, 4GB is must have
Example 1: Try to sample signal that contains two frequencies of equal volume slightly below Nyquist frequency. Sample will contain both frequencies but with changed levels, and offset phases.
Example 2: Sample signal with constant amplitude that sweeps form one high frequency to another, both of them just below Nyquist frequency. Sample will contain signal that varies in amplitude level relative to frequency in sample.
More details on that in aliasing, and critical frequency.
That 48 KHz, even with 24-bit precision, should take care of all sounds possible for the human to hear? 48 KHz doesn't take care enough. It is the precision of phases and levels of harmonics that ears are sensitive about. If everything is in phase and with correct levels, it's a tone, if it is out of phase then it's just another noise.
When CD audio was introduced some compromises had to be made. One of them was medium size vs play time. In the end sampling rate suffered to double play time. The result of lower sampling rate is that noises are reproduced with reduced quality as they are not continuous time signals at that sampling rate and are not oversampled.
24 bit resolution does nothing better to sound quality as it is overkill. Good Hi-Fi equipment usually can't output dynamic range more than 80dB, common 5.1 systems are even worse than that as they are high on distortion and can't provide proper signal to noise ratio. Portable audio equipment is just slightly better than home theatre systems. The main problem is in digital to analog conversion, amplification of signal and speakers / headphones ( jitter, noise, slewing, non-linear mapping of input value to output voltage, clipping, saturation, etc.
16 bit ( 98dB ) is good enough, I hope 96 KHz sampling rate will become standard in consumer lossless audio.
I did some experiments with acetone on two stroke lawnmower and chainsaw. I used 3ml per liter of gasoline with 2% of two-stroke oil. While I couldn't measure "mpg" I did notice more power.
It seems that acetone does something to surface tension of gasoline, maybe affects the way that carburetors work allowing smaller fuel droplets to form.
I don't see why 35 mpg is big deal...
I had water in the fuel tank of my car once. I wanted to get rid of water without actually removing fuel tank. Instead of removing the fuel tank and cleaning it, I did pour in some anhydrous ethanol. I'd say that I poured in 0.5 liters ethanol on 6 liters of fuel. That fixed problem with engine pops and choking but totally f***ed up engine power and made another problem. It did something to the fuel efficiency. For next 200 kilometers car was using much more fuel than it used before I added ethanol! I think it used more than 18 L/100km.
One thing is sure, I'll never ever add another drop of ethanol in gasoline again!!!
btw, hate bring bad news, ethanol will prevent water to freeze.
I agree absolutely.
I did some experiments with flat 15 x 15 cm electrostatic mid-range / tweeter speaker more than 15 years ago. I found out that high frequency range or anything above 10KHz travels trough the air almost with no dispersion. Also I got bad ear ringing form just 2W of sound power. OK, I was 3 meters away from the speaker and in a closed room, but I keep wondering how safe is this and what will come next.
I guess that holographic advertising beaming into eyes and perfume carrying microcapsule shooted at pedestrians are to come in next 10 years.
There is a simple workaround for that... just send them a link to update to Firefox or Opera... The CSS will work like a charm for them too...
The problem is that executable can not be executed while in superfetch buffer, they need to be loaded to RAM. Even if most of FireFox is in superfetch buffer it must load to RAM in order to run. If You are viewing pages with the java applets, then java virtual machine is in disk cache and will be loaded to RAM too. Also since You are using WinAmp and other programs, they too are loaded into superfetch buffer taking even more RAM.
That clears out why RAM usage jumps every time You start FireFox or WinAmp.
I remember slightly better way to prefetch applications. In the mid '80s hard drives and RAM were expensive. Home computers had 1 megabyte of RAM or less. Most home computers had no hard drive at all, instead they had one or two floppy drives. Floppy drives are generally slow, and couldn't top more than 20 kilobytes per second. To speed up DOS scripting Commodore corporation promoted residence. Once resident executable (an program or dos command) takes slightly more RAM than it uses on disk, and can be executed as many times as it is needed without any need for (floppy) disk and launches instantaneously. Also resident executable could be removed to free RAM. Unlike superfetch, resident system took only slightly more RAM than executable would need for an operation, and disk cache or RAM could bee freed.
I'm expecting future windows XP updates to do quite opposite, to bloat XP kernel until they make it slower than any Vista out there. At some point, Vista will become valuable upgrade to anyone willing to stay on Microsoft's rails.
Oh, s**t.
I remember I had to hammer the potting compound of the power supply to free the transformer and the PCB. Tape drive was broken beyond repair, so I made interface to standard Hi-Fi tape deck. It was simple smith-trigger made from CMOS gate IC and few other components. With such interface any tape drive or line-level sound source is good enough.
I use PC / MP3 player to load the games and PC's sound card line input to save them.
I am not quite certain why the internal fuse blows up but it might occur because of short circuit in the tape drive cable.
However that won't solve the main problem, interference effects, as parent noted.
1. heavy: the size of one that would be of use is so great that rover must be made huge, and expensive / impossible to launch to Mars. NASA's choice was solar, I guess they know better.
2. dangerous: in case of bad launch someone has to find damn thing, or its peaces. Solar panels are safe to be left where they are...
Earth/Sun Lagrange points 4 and 5 might be bit dusty. While radio waves wouldn't have lot trouble with the dust, it might be fatal for the laser beam data link.
Air car could solve some big problems in automobiles: engine overheating, wimpy torque at low RPM, pollution in big cities, need for expensive fuels and lubricants...
Unlike the ICE, air car has no problems with overheating, in fact it will use the heat as an bonus to extract more power form compressed air...
The possible disadvantages would be low specific power of engine, and constant need for piston heating in cold weather / climate... so I guess we'll never see them in F1 cars
... low yield EMP cannon is dirt cheap to make, and if it is tuned to cell phone receiving frequency it will fry the receiving circuit silently, leaving the phone without network.
However such EMP cannon will probably kill nearby pacemakers...
An Faraday cage is great idea, and is easy to control what comes in it or out of it.
That it is much better solution than jammer or EMP cannon, and certainly will not hurt anyone.
It's the damn entangled-dll-jungle-hell. It's so hard to make all functional and open yet secured and reliable at the same time.
Well old news it is. I used to have RAM DISK for keeping environment variables and temp files on it. It worked much faster than swapping on hard drive. Yes I did use the bootable RAM disks to boot games and they did boot and worked much faster.
That was 15 years ago. I had 2.5 megs of RAM.
Now people think that their PCI, AGP or PCIe video cards will swap slower than their lame hard drives. How lame is that?
--
Don't mod me down if You disagree or don't understand my post. Think trough and use moderating options reasonably.
The Biosphere 2 looked like romantic, religious recreation of Noah's Ark more than it looked like a scientific experiment. It was an attempt to establish high biodiversity ecosystem on a small surface area. I was doomed to fail because small surface area ecosystems can not sustain high species richness. As someone already said if they went for 100% utilitarian O2 producing plants and food crops they could have more success with it.
Self-sufficient space habitat is different because it will actually have closed loop atmosphere and water recycling. It will use all available technologies, physical, chemical and biological processes to recycle water and atmosphere instead of relaying just on mother nature to do the job. They intend to grow their food. They will actually try to live in there up to three years using only resources that habitat can provide.
Then slightly different approach might yield better results. Machinery that melts rock instead of pulverizing it will probably make smooth walls that are easy to seal and hard enough to support itself. RTGs could power melting tool.
The problem with modern antivirus software is that they relay mostly on definitions.
Definitions are by definition either obsolete, ether will be updated next time when user hits update button. In both cases antivirus software relays completely on heuristic detection, by default turned on to lightest level of detection to reduce resource hoging and nag-like behaviour.
I can remember the viruses of late 90s. Most of them were quite small, had exceptional hiding capabilities and had used BIOS functions directly, avoiding detection easily. They did nasty stuff, mike256, WinWierd32 anyone ?
It is true, Windows XP SP2 can be "tuned" to be passively safe but even then You need something to immunize and check your system.
I ma using SpybotSD for a long time, without tea-timer feature, to immunize the system.
Using a browser that uses security ratings on sites, and can open unrated sites with disabled scripting can help a lot.
I use Netscape 8.1.3, as it can open sites in 3 different security levels, + one custom.
Common sense helps allot on the Internet ( downloading cracks anyone ? ) so little paranoia makes healthy Windoze. I guess that FireFox can be secured using noscript and similar plugins, but I have never tried it.
BTW I don't use IE / Outlook since 1998, and did my Windows install more than 2 years ago, so much for the security record.
You got all Your facts right.
Windows users can be dumb and work on bot camp or smart and never have any virus / worm / trojan even with no antivirus software installed at all.
Firewall will not help against true viruses, but if is good can help against worms and trojans by preventing them to spread out from the infected machine. The problem is that firewall does not prevent infection, only can block port / protocol.