AGP 1x and 2x were 3.3V, 4x was 1.5V and 8x was 0.8V.
Afaict virtually all stuff that supported 0.8V supported 1.5V as well. So that left 1.5V/0.8V vs 3.3V as the main compatibility issue. There was a notching system that was supposed to indicate whether a card/motherboard supported just 3.3V, just 1.5V/0.8V or both and prevent incompatible combinations from mating. Unfortunately some manufacturers miskeyed thier products.
BTW PCI also had two voltages though the lower voltage was generally only seen on pretty high end boards (all the higher speed variants of PCI needed the lower voltage).
And there are lots of different ram types arround too though maybe not as many as there were back in the day, unregistered/registered/fully bufferd, different DDR versions etc. Then there are some machines that are very fussy about ram (this is apparently a big issue with the EEE 900)
True but the main difference afaict is the length of the hydrocarbon chains and afaict thier main problem is chains coming out shorter than desired. So I imagine it would be too hard to adapt the process to make petrol.
1 - capture the Methane, and bottle it. This is called "natural gas", and is a great fuel, assuming you don't mind it being a Gas, not a liquid. I'm sure the Navy can figure out how to power fighter planes with this stuff directly,... if they really have to. IIRC the trouble with methane is it's small/light molecules make it a PITA to liquify and store liquid (which are nessacery for high density storage).
It can be done on a ship as evidenced by the LNG tankers that are now carrying methane arround the world (maybe they could use it to supply the carriers support ships) but I suspect it would be weight prohibitive for aircraft.
The US produces quite a lot of oil domestically (not as much as they used to nor as much as they consume but still a lot), probablly enough to supply their military.
The deep rationing would suck for civilians in the US though.
Nonono, you don't get it. 2.5.9 contains the bug fix for people/distros who don't want to move to a new major release. Do you have anything to back up that claim or is it just a guess?
are there any URL shorteners that only use lower case letters and numbers? It would make the shortened urls a little bit longer but MUCH easier to read out.
Here is a recipie to build a set of 2.6.1 packages for debian lenny based on the packaging ari has done for sid (but not uploaded yet hence the download from svn.debian.org).
wget http://sourceforge.net/projects/pidgin/files/Pidgin/pidgin-2.6.1.tar.bz2 bunzip2 pidgin-2.6.1.tar.bz2 tar -xf pidgin-2.6.1.tar gzip pidgin-2.6.1.tar mv pidgin-2.6.1.tar.gz pidgin_2.6.1.orig.tar.gz cd pidgin-2.6.1 svn export -r 14052 svn://svn.debian.org/svn/collab-maint/deb-maint/pidgin/trunk/debian sed -i s/tcl8.6-dev/tcl8.5-dev/ debian/control sed -i s/tk8.6-dev/tk8.5-dev/ debian/control sed -i 's/libgstfarsight0.10-dev (>= 0.0.9),//' debian/control sed -i 's/(>= 0.4.53)//' debian/control sed -i 's/(>= 1.1.1)//' debian/control sed -i 's/--enable-vv/--disable-vv/' debian/rules dpkg-buildpackage
if it complains about missing build-depends install them and run dpkg-buildpackage again
note: I had to disable video/voice because libgstfarsight is not available in lenny.
Sadly most non-technical users here in the UK do and most of them are very difficult to persude to either use a multiprotocol client or switch entirely.
There are debianised source packages for 2.6.1 on getdeb (you have to follow the link for a particular distro release and then there is a source link there), dunno how well made they are.
2.5.9 is availible in debian sid and at least up until now i've found sid's pidgin packages compile fine on lenny.
Infinitely more expensive than doing a simple free bank to bank payment online. Depends on where you live and where the recipiant lives. I looked into transferring money from my UK bank account to an ebay seller in germany that didn't take paypal (yes I know I should check before bidding and usually I do but when you are dealing with multiple auctions of the same product it's easy to get confused). It would have cost me more than the value of the transaction. Ended up sending cash through the post to settle the transaction .
Yes paypals fees are annoying but for small international transactions what better alternatives are there?
Also I worry about giving my bank details out, sure there are supposed to be safegaurds on direct debits but i'd still rather not have to deal with a fraudulant one if I can help it (you could get arround this to some extent though by using a savings account that doesn't allow direct debit)
email isn't exactly great for timely delivery, some places have pretty slow email systems and then there is the issue of greylisting. Plus some peoples email addresses are almost as bad as a lot of urls.
IM can be an option but a lot of people are banned from using that at work.
The big question is how will this affect "office ready" and/or volume license users. IIRC with the java issue MS stopped shipping media for versions with the MSJVM included but customers who already had media were still allowed to install them under the downgrade rights of newly purchased volume licenses (and third parties shipping software made with MS technology could still offer the redistributable).
Will that be the case here or are we likely to end up with a situation where IT departments have to carefully manage the two versions of word with seperate licensing.
Under what license terms though? I remember MSDN lets you install whatever versions you want on as many computers as you want (though with products that require activation you have to convince MS that you are following the other requirements to get more than 10 machines activated) but only for development and testing purposes and (with a few limited exceptions for user-acceptance testing) for use by yourself or someone else with a MSDN subscription.
But does it come in AMERICAN English? I don't think MS bothers making seperate british english and american english versions of software so the answer is probablly yes.
Ethene is not the same as ethane and propene is not the same as propane.
It's probablly possible to convert short chain alkanes (no double bonds) into short chain alkenes (one double bond) but i've no idea how difficult it is.
Afaict most plastics are made from small unsaturated hydrocarbons like ethene and propene which are then polymerised. Theese hydrocarbons are made by cracking bits off the less valuable hydrocarbons in crude oil (e.g. you take stuff that's a bit too heavy to be petrol, crack bits off and get petrol and ethene/propene).
There have been some plastic-like substances made from biologically derived materials and i'm pretty sure other sources for unsaturated short chain hydrocarbons could be found too (they'd probablly just be a lot more expensive than cracking crude)
ets say the gaming takes 10kB/s and the cmts is still 1000kB/s. If 100 users max out their lines at once then each user gets 10kB/s and your ping if fine. However, if there was 150 users maxing out their line at once you'd have between 6-7kB/s and since the game is 10kB/s your ping would jump up. Thats not generally how packet switched networks work, typically the switches and routers don't care what user a packet came from just when it arrived and which link it is to be sent out on. It is possible to make them care about the source but this adds a LOT of complexity.
If a link is maxed out then the queue at the switch/router will fill up and latency for everyone using that link will increase by the length of said queue. This will happen regardless of whether the maxing out is caused by a load of users or by a handful. You can reduce this by using shorter queues but that increases the chance of packets getting lost if there is a sudden spike of packets.
It also means what bandwidth a user gets will depend on how agressive they are at taking it. This is one of the big issues with both bittorrent and download managers, both get a larger than fair slice of the bandwidth by using multiple TCP connections in paralell.
From most utilities we first world citizens expect good service more than just "most of the time". The power companies don't make judgements that your computers are more or less important than my electric shower. Despite this loss of power due to overloading is extremely rare in most first world countries because the power company has a good idea of the demand at a given price point and invests in sufficiant infrastructure to cover that demand.
The real problem is that the ISPs have advertised and sold unlimited internet at prices that don't come close to paying for someone to really use their link at maximum a significant proportion of the time.
You will notice that roads have a similar problem for similar reasons. The roads at rush hour get congested or even gridlocked because rather than selling the limited space to those who can afford it they make it a free for all.
If you are going to have traffic shaping it should be up to the customer which of thier traffic they want to pay high priority prices for and which of thier traffic they want to send on the slack capacity, not based on a value judgement of thier content by the local monopoly/duopoly communication providers.
Most other people are taking Checks, Visa, and Debit Cards as primary sources of transactions, leaving Cash a fourth level barely used. Maybe this varies by location but here in the UK for small transactions (like say buying some food in a takeaway or conviniance store) cash is by far the dominant form of payment. Some such places will grudgingly take cheques or cards but they really would rather not due to the fees (and often they set minimum transaction ammounts for card use). It is also more time consuming to take cards than cash afaict.
At the supermarket most people seem to use either cash or cards, not really sure which is more common. Some supermarkets have even started refusing to take cheques.
Cheques only seem to be used for large transactions with small buisnesses (e.g. paying a plumber) and as a way for buisnesses to pay off thier invoices from each other.
Busses tend to move pretty slowly, they also tend to be pretty big and sturdily built. When a bus crashes with a car the bus and it's passengers tend to come off much better than the car and it's passengers. Coaches move quicker but still slower than cars and still tend to come off best in a collision.
Still it does surprise me that they quote busses as safer than trains per-mile (per-trip and per-hour can be obviously attributed to short slow trips) the tables in wikipedia don't seem to have much info on the gathering of the data (the linked source just says it was from a "DETR study" whatever that is (maybe a reference to the UK department of transport given the references to railtrack and the fact that iirc dial.pipex.com is a british ISP).
The source also claims that a lot of rail deaths were due to things other than actual train crashes (people trespassing on the tracks falling off platforms and similar) . It does not state whether these are included in the figures in the table and whether similar deaths related to road transport use were included.
this post is NOT about blocking calls, it is about protecting kids from electromagnetic radiation. Mobile phones use power control so if you try and block the signal path you just end up forcing them to transmit at higher power and/or losing communication completely.
Slashdot Mirror
AGP 1x and 2x were 3.3V, 4x was 1.5V and 8x was 0.8V.
Afaict virtually all stuff that supported 0.8V supported 1.5V as well. So that left 1.5V/0.8V vs 3.3V as the main compatibility issue. There was a notching system that was supposed to indicate whether a card/motherboard supported just 3.3V, just 1.5V/0.8V or both and prevent incompatible combinations from mating. Unfortunately some manufacturers miskeyed thier products.
BTW PCI also had two voltages though the lower voltage was generally only seen on pretty high end boards (all the higher speed variants of PCI needed the lower voltage).
And there are lots of different ram types arround too though maybe not as many as there were back in the day, unregistered/registered/fully bufferd, different DDR versions etc. Then there are some machines that are very fussy about ram (this is apparently a big issue with the EEE 900)
True but the main difference afaict is the length of the hydrocarbon chains and afaict thier main problem is chains coming out shorter than desired. So I imagine it would be too hard to adapt the process to make petrol.
1 - capture the Methane, and bottle it. This is called "natural gas", and is a great fuel, assuming you don't mind it being a Gas, not a liquid. I'm sure the Navy can figure out how to power fighter planes with this stuff directly, ... if they really have to.
IIRC the trouble with methane is it's small/light molecules make it a PITA to liquify and store liquid (which are nessacery for high density storage).
It can be done on a ship as evidenced by the LNG tankers that are now carrying methane arround the world (maybe they could use it to supply the carriers support ships) but I suspect it would be weight prohibitive for aircraft.
2 - use Steam Reforming ( http://en.wikipedia.org/wiki/Steam_reforming ) to convert the methane to SynGas, then to Synfuel. Easy. Just ask Mobil. http://en.wikipedia.org/wiki/Synfuel
But that would add yet more inefficiancy and complexity to the process.
The US produces quite a lot of oil domestically (not as much as they used to nor as much as they consume but still a lot), probablly enough to supply their military.
The deep rationing would suck for civilians in the US though.
Didn't the US military a while back adapt thier diesel engined stuff to run on jet fuel as part of a "single fuel initative".
Nonono, you don't get it. 2.5.9 contains the bug fix for people/distros who don't want to move to a new major release.
Do you have anything to back up that claim or is it just a guess?
are there any URL shorteners that only use lower case letters and numbers? It would make the shortened urls a little bit longer but MUCH easier to read out.
Here is a recipie to build a set of 2.6.1 packages for debian lenny based on the packaging ari has done for sid (but not uploaded yet hence the download from svn.debian.org).
wget http://sourceforge.net/projects/pidgin/files/Pidgin/pidgin-2.6.1.tar.bz2
bunzip2 pidgin-2.6.1.tar.bz2
tar -xf pidgin-2.6.1.tar
gzip pidgin-2.6.1.tar
mv pidgin-2.6.1.tar.gz pidgin_2.6.1.orig.tar.gz
cd pidgin-2.6.1
svn export -r 14052 svn://svn.debian.org/svn/collab-maint/deb-maint/pidgin/trunk/debian
sed -i s/tcl8.6-dev/tcl8.5-dev/ debian/control
sed -i s/tk8.6-dev/tk8.5-dev/ debian/control
sed -i 's/libgstfarsight0.10-dev (>= 0.0.9),//' debian/control
sed -i 's/(>= 0.4.53)//' debian/control
sed -i 's/(>= 1.1.1)//' debian/control
sed -i 's/--enable-vv/--disable-vv/' debian/rules
dpkg-buildpackage
if it complains about missing build-depends install them and run dpkg-buildpackage again
note: I had to disable video/voice because libgstfarsight is not available in lenny.
I would agree with you if it wasn't for the fact that 2.5.9, 2.6.0 and 2.6.1 were released on the same day.
So unless you were very agressive with your updating you would most likely still be running an affected version.
Sadly most non-technical users here in the UK do and most of them are very difficult to persude to either use a multiprotocol client or switch entirely.
There are debianised source packages for 2.6.1 on getdeb (you have to follow the link for a particular distro release and then there is a source link there), dunno how well made they are.
2.5.9 is availible in debian sid and at least up until now i've found sid's pidgin packages compile fine on lenny.
Infinitely more expensive than doing a simple free bank to bank payment online.
Depends on where you live and where the recipiant lives. I looked into transferring money from my UK bank account to an ebay seller in germany that didn't take paypal (yes I know I should check before bidding and usually I do but when you are dealing with multiple auctions of the same product it's easy to get confused). It would have cost me more than the value of the transaction. Ended up sending cash through the post to settle the transaction .
Yes paypals fees are annoying but for small international transactions what better alternatives are there?
Also I worry about giving my bank details out, sure there are supposed to be safegaurds on direct debits but i'd still rather not have to deal with a fraudulant one if I can help it (you could get arround this to some extent though by using a savings account that doesn't allow direct debit)
Wikipedia does keep full history so you can cite a particular revision should you wish.
Not that I'd particulally reccomend citing either of them as anything more than sources of background reading.
email isn't exactly great for timely delivery, some places have pretty slow email systems and then there is the issue of greylisting. Plus some peoples email addresses are almost as bad as a lot of urls.
IM can be an option but a lot of people are banned from using that at work.
The big question is how will this affect "office ready" and/or volume license users. IIRC with the java issue MS stopped shipping media for versions with the MSJVM included but customers who already had media were still allowed to install them under the downgrade rights of newly purchased volume licenses (and third parties shipping software made with MS technology could still offer the redistributable).
Will that be the case here or are we likely to end up with a situation where IT departments have to carefully manage the two versions of word with seperate licensing.
Under what license terms though? I remember MSDN lets you install whatever versions you want on as many computers as you want (though with products that require activation you have to convince MS that you are following the other requirements to get more than 10 machines activated) but only for development and testing purposes and (with a few limited exceptions for user-acceptance testing) for use by yourself or someone else with a MSDN subscription.
But does it come in AMERICAN English?
I don't think MS bothers making seperate british english and american english versions of software so the answer is probablly yes.
Ethene is not the same as ethane and propene is not the same as propane.
It's probablly possible to convert short chain alkanes (no double bonds) into short chain alkenes (one double bond) but i've no idea how difficult it is.
Afaict most plastics are made from small unsaturated hydrocarbons like ethene and propene which are then polymerised. Theese hydrocarbons are made by cracking bits off the less valuable hydrocarbons in crude oil (e.g. you take stuff that's a bit too heavy to be petrol, crack bits off and get petrol and ethene/propene).
There have been some plastic-like substances made from biologically derived materials and i'm pretty sure other sources for unsaturated short chain hydrocarbons could be found too (they'd probablly just be a lot more expensive than cracking crude)
So use component for one of them. At HDTV resoloutions and refresh rates that should be more than sufficiant for a good picture.
ets say the gaming takes 10kB/s and the cmts is still 1000kB/s. If 100 users max out their lines at once then each user gets 10kB/s and your ping if fine. However, if there was 150 users maxing out their line at once you'd have between 6-7kB/s and since the game is 10kB/s your ping would jump up.
Thats not generally how packet switched networks work, typically the switches and routers don't care what user a packet came from just when it arrived and which link it is to be sent out on. It is possible to make them care about the source but this adds a LOT of complexity.
If a link is maxed out then the queue at the switch/router will fill up and latency for everyone using that link will increase by the length of said queue. This will happen regardless of whether the maxing out is caused by a load of users or by a handful. You can reduce this by using shorter queues but that increases the chance of packets getting lost if there is a sudden spike of packets.
It also means what bandwidth a user gets will depend on how agressive they are at taking it. This is one of the big issues with both bittorrent and download managers, both get a larger than fair slice of the bandwidth by using multiple TCP connections in paralell.
From most utilities we first world citizens expect good service more than just "most of the time". The power companies don't make judgements that your computers are more or less important than my electric shower. Despite this loss of power due to overloading is extremely rare in most first world countries because the power company has a good idea of the demand at a given price point and invests in sufficiant infrastructure to cover that demand.
The real problem is that the ISPs have advertised and sold unlimited internet at prices that don't come close to paying for someone to really use their link at maximum a significant proportion of the time.
You will notice that roads have a similar problem for similar reasons. The roads at rush hour get congested or even gridlocked because rather than selling the limited space to those who can afford it they make it a free for all.
If you are going to have traffic shaping it should be up to the customer which of thier traffic they want to pay high priority prices for and which of thier traffic they want to send on the slack capacity, not based on a value judgement of thier content by the local monopoly/duopoly communication providers.
Most other people are taking Checks, Visa, and Debit Cards as primary sources of transactions, leaving Cash a fourth level barely used.
Maybe this varies by location but here in the UK for small transactions (like say buying some food in a takeaway or conviniance store) cash is by far the dominant form of payment. Some such places will grudgingly take cheques or cards but they really would rather not due to the fees (and often they set minimum transaction ammounts for card use). It is also more time consuming to take cards than cash afaict.
At the supermarket most people seem to use either cash or cards, not really sure which is more common. Some supermarkets have even started refusing to take cheques.
Cheques only seem to be used for large transactions with small buisnesses (e.g. paying a plumber) and as a way for buisnesses to pay off thier invoices from each other.
Busses tend to move pretty slowly, they also tend to be pretty big and sturdily built. When a bus crashes with a car the bus and it's passengers tend to come off much better than the car and it's passengers. Coaches move quicker but still slower than cars and still tend to come off best in a collision.
Still it does surprise me that they quote busses as safer than trains per-mile (per-trip and per-hour can be obviously attributed to short slow trips) the tables in wikipedia don't seem to have much info on the gathering of the data (the linked source just says it was from a "DETR study" whatever that is (maybe a reference to the UK department of transport given the references to railtrack and the fact that iirc dial.pipex.com is a british ISP).
The source also claims that a lot of rail deaths were due to things other than actual train crashes (people trespassing on the tracks falling off platforms and similar) . It does not state whether these are included in the figures in the table and whether similar deaths related to road transport use were included.
this post is NOT about blocking calls, it is about protecting kids from electromagnetic radiation.
Mobile phones use power control so if you try and block the signal path you just end up forcing them to transmit at higher power and/or losing communication completely.