10Base-2 (thinnet) cards use 50 ohm connectors. 10BASE-2 cards are connected to the cable via a straight T connector with no resistance, that is going to be a discontinuity regardless of the supposed impedances of the connector on the card. Lukilly the run from T-Peice to card is WAY shorter than the wavelength (wavelength at 10MHz is 10m) so it doesn't matter much.
When the people you are aware of tried this did they make sure the terminators matched the coax? (using 50 ohm terminators and 75 ohm coax is going to cause problems) and did they use 75 ohm coax for the entire run?
It's a bitch to try to crimp Ethernet without the right tools, and those will set you back a couple of hundred dollars (for the good ones). Never had a problem using a cheap crimp tool, yeah it's slighly more effort because of the lower leverage but the time and effort to do the actual crimping is tiny compared to that spent getting the wires into the plug right.
Frankly though if you are doing a proper permanant install you should be using wall ports and a patch panel anyway and therefore there is no need to crimp plugs to cable.
A cable's impedance varies directly with its length; if your cable has a characteristic impedance (e.g. ohms/meter) 3/2 higher than the standard, then simply make sure your runs are less than 2/3 of the standard's maximum lengths (185 m for 10Base2). BULLSHIT
But ultimately, I think you got that "50 ohm" number from the required termination at the ends of the cable, not the characteristics of the cable itself 50 ohm is the "characteristic impedance" of the cable. Characteristic impedance is essentially the impedance of an infinite
By matching the impedance of the terminators to the characteristic impedance of the cable you avoid getting reflections from the end of the cable.
P.S. I strongly suspect that with 75 ohm terminators you would have no problem running 10BASE2 on TV coax.
How is it any different than a watch store that advertises business hours of 9am-9pm? You know that noone is there... You know what they have inside The watch store owner isn't lying to encourage burglers in. He is merely stating facts about his buisness.
I don't know what if anything the law in various US states has to say on the matter but somehow I think the courts would take a dim view of deliberately luring someone into a deathtrap.
I wonder if 10base2 would work on 75 ohm coax with the right terminators. Matching the terminators to the coax is important because if you don't you will get reflections (and reflections are BAD for a system like ethernet) but given that the terminators and coax are matched I don't see any reason the end stations would be particulally sensitive to the impedance of the system.
Unfortunately I don't have any 10base2 gear handy to try it out.
Lots of things people do on thier computers cause a sequence of short operations on the drive at different locations (one example would be scanning a large folder full of files to generate thumbnails or otherwise assess some property of the files). HDDs are pretty constrained in how many operations they can perform per second due to the mechanics of the drive (at 7200RPM rotational latency alone is going to eat about 8ms per operation, 15000 RPM reduces that to about 4ms).
It also means backing up a HDD full of small files on a file level basis (yes I know you can image the whole drive but that eats a LOT of space) can be very slow.
Afaict the RIAA campaign is not about the money made or lost in the cases that actually go to trial, it's about 1: scaring people into not filesharing in the first place and 2: scaring those who are caught filesharing into settling rather than taking it to court.
If the judgements in court end up capped at a level not much higher than the settlements the RIAA currently offers that would likely result in a lot more people daring to go to trial which would make the litigation campaign much more expensive for the RIAA.
Anyway, it's called a beta for a reason. I'm surprised anyone would even use it in production at this point. What do you consider to be "in production"? from your description it sounds like only a handfull of users in a given group (whether internal or external) using this software can cause a heck of a lot of overbloated mails.
Microsofts proposed soloution seems to boil down to "just delete the affected mails" which is just NOT going to be acceptable in many cases.
Taking a look myself the ratings of these cards seem to vary from awful to about par for the course for newegg reviews (remember, people with a bad experiance are far more likely to review than those with a good experiance)
Afaict there are a few reasons for buying a lowish end soundcard
1: a requirement for more channels than the onboard audio gives (either for surround sound or for a multiroom setup) 2: broken onboard audio 3: a lot of onboard audio either lacks a mic input (having said that from one of the reviews these cheap cards don't look brilliant in that regard either, at least one poster is complaining of a disfunctional mic input)
After 2000, it got to the point where hardly anyone even buys a sound card, and only specialized IO cards exist. While i'm sure there is less of a market for many card types than there used to be the cards remain available and there are even PCIe models out there nowadays though the PCIe cards often seem to command a significant premium over PCI ones (PCIe sound cards seem to be particulally expensive).
While there are certainly cases where you are better off getting a new computer there are also plenty of cases where upgrading your existing one is a much cheaper way to get the performance you require.
you can get 512GB SSDs in 2.5 inch 9.5mm high form factor.
For a laptop that means about a 10% difference in max availble capacity if your laptop only takes the thin drives and about a factor of two difference if your laptop can take the thicker drives.
For desktops that means you can put two of the aforementioned drives in the space of a single 2TB hard drive so again the difference in capacity per bay is a factor of two.
I wouldn't call a factor of two difference in capacity per bay to be "far exceeding"
1: performance: afaict SSDs are already the clear winner here. 2: density: I can put a 2TB drive in a standard 3.5 inch bay. Afaict SSDs are generally the same size as laptop hard drives and you can put two of those in a 3.5 inch bay with readilly available adaptor kits. Afaict the drives go up to 512GB so the density is about half that of HDDs. For laptops the density situation is even closer (especially if the laptop in question only has a 9.5mm high bay). 3: cost: the aforementioned 2TB hard drives cost $150-$300 while a 512GB SSD costs $1400 so the cost per gigabyte is about 20 to 40 times higher for the SSD.
In other words the main issue for SSDs right now is cost.
I live in the UK and most people I know have power like you describe with power cuts being things that happen at intervals measured in years. Most people I know in the UK experiance the same.
However in more rural parts of the UK there is a lot of power carried by groups of three bare 11KV wires on poles and these are rather vulnerable to tree branches, unusually high winds etc. Power problems are common in these areas (mostly it's very short glitches as something briefly touches a power line but such areas get more long outages than urban areas as well)..
Also remember that the downtime caused by a power outage can be considerablly longer than the power outage itself. Without a UPS you have at best a reboot and a disk check and at worst a major corruption issue to deal with after tha power is back on.
You don't want a drive reporting different logical sector sizes under different situations, that's just a recipe for breakage with any format that works in terms of sector counts.
And having a 4096 byte logical sector size would cause major compatibility issues.
IMO the best compromise would be to have the drive allow reading of it's physical sector to logical sector size ratio somehow. Partitioning tools could then warn about misaligned partitions and offer to fix them.
At least in the UK the mifi is marketed as a "mobile broadband" device with "mobile broadband" plans.
What the carriers don't want you doing is using an "unlimited" phone data plan with a PC because PCs can much more easilly use a hell of a lot of data. I doubt they would care too much about you using your phone to browse on a "mobile broadband" plan.
The trouble with exams in my experiance is that people who are good at rote memorisation and acceptable at mathematical methods can often pass them with very good marks without ever really understanding the material. Especially as in my experiance each years exam is like the previous but with the numbers tweaked.
1.) The bad code is making a call into OS code that is affected by the patch and it *used* to work but doesn't. 1a) the bad code is making a call into OS code but for some reason is using byte offsets rather than the files symbol table.
If this is the case you can't really blame MS, it's normal and expected that byte offsets will change when files are modified to fix bugs etc and there is no real way for a patcher to see that bad code is relying on them.
And also IMO an increase in risk if some people but not everyone at opera are malicious.
Building a password stealer into a webapp would require a way to get that information out that didn't arrouse too much suspicion. Getting the passwords from the stealer to the way out without getting noticed in code review would be pretty dificult IMO.
OTOH loading a bad bit of code on one server out of a pool or making a legitimate logging function log a little more than it should could be much easier.
And then there is the issue that routing your data through a jurisdiction may expose it to warrants from that jurisdictions courts.
In fact, they will actually not SELL it to you anymore! No such problems with open source. Not being able to get the version you want can be a problem for individual users and small buisnesses but with MS at least it's not an issue for larger companies government institutions etc.
Well intel can put graphics and processor in the same package without putting them on the same die. Indeed they have already done so with thier latest dual core chips (the current gen quad-core chips don't have any support for shared memory graphics at all)
The thing is intel have failed to make decent graphics soloutions (they are getting better but are not yet up to the standards of even the integrated graphics in nvidia chipsets let alone dedicated graphics cards) and nvidia haven't even tried to make x86 processors which is a bit of a problem for making a decent single-chip soloution.
Maybe you don't think thats a problem. OTOH I think it's a problem that anti-piracy measures make worse by discouraging casual copying and encouraging large pirate networks who can deal with the anti-piracy measures.
Let's not forget just how quickly the popularity of GIF plummeted once Unisys decided it would be a good idea to enforce its patents. I seem to remember hardly at all (heck gifs are STILL widely used despite being inferior to PNG). Do you have any data to the contary?
I'm sure I remember an "air crash investigation" episode about an accident caused by counterfeit parts, I don't remember whether it was a full blown crash though.
Slashdot Mirror
10Base-2 (thinnet) cards use 50 ohm connectors.
10BASE-2 cards are connected to the cable via a straight T connector with no resistance, that is going to be a discontinuity regardless of the supposed impedances of the connector on the card. Lukilly the run from T-Peice to card is WAY shorter than the wavelength (wavelength at 10MHz is 10m) so it doesn't matter much.
When the people you are aware of tried this did they make sure the terminators matched the coax? (using 50 ohm terminators and 75 ohm coax is going to cause problems) and did they use 75 ohm coax for the entire run?
It's a bitch to try to crimp Ethernet without the right tools, and those will set you back a couple of hundred dollars (for the good ones).
Never had a problem using a cheap crimp tool, yeah it's slighly more effort because of the lower leverage but the time and effort to do the actual crimping is tiny compared to that spent getting the wires into the plug right.
Frankly though if you are doing a proper permanant install you should be using wall ports and a patch panel anyway and therefore there is no need to crimp plugs to cable.
A cable's impedance varies directly with its length; if your cable has a characteristic impedance (e.g. ohms/meter) 3/2 higher than the standard, then simply make sure your runs are less than 2/3 of the standard's maximum lengths (185 m for 10Base2).
BULLSHIT
But ultimately, I think you got that "50 ohm" number from the required termination at the ends of the cable, not the characteristics of the cable itself
50 ohm is the "characteristic impedance" of the cable. Characteristic impedance is essentially the impedance of an infinite
By matching the impedance of the terminators to the characteristic impedance of the cable you avoid getting reflections from the end of the cable.
P.S. I strongly suspect that with 75 ohm terminators you would have no problem running 10BASE2 on TV coax.
How is it any different than a watch store that advertises business hours of 9am-9pm? You know that noone is there... You know what they have inside
The watch store owner isn't lying to encourage burglers in. He is merely stating facts about his buisness.
I don't know what if anything the law in various US states has to say on the matter but somehow I think the courts would take a dim view of deliberately luring someone into a deathtrap.
I wonder if 10base2 would work on 75 ohm coax with the right terminators. Matching the terminators to the coax is important because if you don't you will get reflections (and reflections are BAD for a system like ethernet) but given that the terminators and coax are matched I don't see any reason the end stations would be particulally sensitive to the impedance of the system.
Unfortunately I don't have any 10base2 gear handy to try it out.
Lots of things people do on thier computers cause a sequence of short operations on the drive at different locations (one example would be scanning a large folder full of files to generate thumbnails or otherwise assess some property of the files). HDDs are pretty constrained in how many operations they can perform per second due to the mechanics of the drive (at 7200RPM rotational latency alone is going to eat about 8ms per operation, 15000 RPM reduces that to about 4ms).
It also means backing up a HDD full of small files on a file level basis (yes I know you can image the whole drive but that eats a LOT of space) can be very slow.
Afaict the RIAA campaign is not about the money made or lost in the cases that actually go to trial, it's about 1: scaring people into not filesharing in the first place and 2: scaring those who are caught filesharing into settling rather than taking it to court.
If the judgements in court end up capped at a level not much higher than the settlements the RIAA currently offers that would likely result in a lot more people daring to go to trial which would make the litigation campaign much more expensive for the RIAA.
Anyway, it's called a beta for a reason. I'm surprised anyone would even use it in production at this point.
What do you consider to be "in production"? from your description it sounds like only a handfull of users in a given group (whether internal or external) using this software can cause a heck of a lot of overbloated mails.
Microsofts proposed soloution seems to boil down to "just delete the affected mails" which is just NOT going to be acceptable in many cases.
Taking a look myself the ratings of these cards seem to vary from awful to about par for the course for newegg reviews (remember, people with a bad experiance are far more likely to review than those with a good experiance)
Afaict there are a few reasons for buying a lowish end soundcard
1: a requirement for more channels than the onboard audio gives (either for surround sound or for a multiroom setup)
2: broken onboard audio
3: a lot of onboard audio either lacks a mic input (having said that from one of the reviews these cheap cards don't look brilliant in that regard either, at least one poster is complaining of a disfunctional mic input)
After 2000, it got to the point where hardly anyone even buys a sound card, and only specialized IO cards exist.
While i'm sure there is less of a market for many card types than there used to be the cards remain available and there are even PCIe models out there nowadays though the PCIe cards often seem to command a significant premium over PCI ones (PCIe sound cards seem to be particulally expensive).
While there are certainly cases where you are better off getting a new computer there are also plenty of cases where upgrading your existing one is a much cheaper way to get the performance you require.
you can get 512GB SSDs in 2.5 inch 9.5mm high form factor.
For a laptop that means about a 10% difference in max availble capacity if your laptop only takes the thin drives and about a factor of two difference if your laptop can take the thicker drives.
For desktops that means you can put two of the aforementioned drives in the space of a single 2TB hard drive so again the difference in capacity per bay is a factor of two.
I wouldn't call a factor of two difference in capacity per bay to be "far exceeding"
No it's essentially the electronic equivilent of rubber hose cryptography.
Lets look at a few metrics.
1: performance: afaict SSDs are already the clear winner here.
2: density: I can put a 2TB drive in a standard 3.5 inch bay. Afaict SSDs are generally the same size as laptop hard drives and you can put two of those in a 3.5 inch bay with readilly available adaptor kits. Afaict the drives go up to 512GB so the density is about half that of HDDs. For laptops the density situation is even closer (especially if the laptop in question only has a 9.5mm high bay).
3: cost: the aforementioned 2TB hard drives cost $150-$300 while a 512GB SSD costs $1400 so the cost per gigabyte is about 20 to 40 times higher for the SSD.
In other words the main issue for SSDs right now is cost.
I live in the UK and most people I know have power like you describe with power cuts being things that happen at intervals measured in years. Most people I know in the UK experiance the same.
However in more rural parts of the UK there is a lot of power carried by groups of three bare 11KV wires on poles and these are rather vulnerable to tree branches, unusually high winds etc. Power problems are common in these areas (mostly it's very short glitches as something briefly touches a power line but such areas get more long outages than urban areas as well)..
Also remember that the downtime caused by a power outage can be considerablly longer than the power outage itself. Without a UPS you have at best a reboot and a disk check and at worst a major corruption issue to deal with after tha power is back on.
You don't want a drive reporting different logical sector sizes under different situations, that's just a recipe for breakage with any format that works in terms of sector counts.
And having a 4096 byte logical sector size would cause major compatibility issues.
IMO the best compromise would be to have the drive allow reading of it's physical sector to logical sector size ratio somehow. Partitioning tools could then warn about misaligned partitions and offer to fix them.
At least in the UK the mifi is marketed as a "mobile broadband" device with "mobile broadband" plans.
What the carriers don't want you doing is using an "unlimited" phone data plan with a PC because PCs can much more easilly use a hell of a lot of data. I doubt they would care too much about you using your phone to browse on a "mobile broadband" plan.
The trouble with exams in my experiance is that people who are good at rote memorisation and acceptable at mathematical methods can often pass them with very good marks without ever really understanding the material. Especially as in my experiance each years exam is like the previous but with the numbers tweaked.
note: my experiances are from EE not CS.
1.) The bad code is making a call into OS code that is affected by the patch and it *used* to work but doesn't.
1a) the bad code is making a call into OS code but for some reason is using byte offsets rather than the files symbol table.
If this is the case you can't really blame MS, it's normal and expected that byte offsets will change when files are modified to fix bugs etc and there is no real way for a patcher to see that bad code is relying on them.
And also IMO an increase in risk if some people but not everyone at opera are malicious.
Building a password stealer into a webapp would require a way to get that information out that didn't arrouse too much suspicion. Getting the passwords from the stealer to the way out without getting noticed in code review would be pretty dificult IMO.
OTOH loading a bad bit of code on one server out of a pool or making a legitimate logging function log a little more than it should could be much easier.
And then there is the issue that routing your data through a jurisdiction may expose it to warrants from that jurisdictions courts.
In fact, they will actually not SELL it to you anymore! No such problems with open source.
Not being able to get the version you want can be a problem for individual users and small buisnesses but with MS at least it's not an issue for larger companies government institutions etc.
bingo, it doesn't make sense to consider the cost of ownership without also considering the benefit of ownership.
Well intel can put graphics and processor in the same package without putting them on the same die. Indeed they have already done so with thier latest dual core chips (the current gen quad-core chips don't have any support for shared memory graphics at all)
The thing is intel have failed to make decent graphics soloutions (they are getting better but are not yet up to the standards of even the integrated graphics in nvidia chipsets let alone dedicated graphics cards) and nvidia haven't even tried to make x86 processors which is a bit of a problem for making a decent single-chip soloution.
Maybe you don't think thats a problem.
OTOH I think it's a problem that anti-piracy measures make worse by discouraging casual copying and encouraging large pirate networks who can deal with the anti-piracy measures.
Let's not forget just how quickly the popularity of GIF plummeted once Unisys decided it would be a good idea to enforce its patents.
I seem to remember hardly at all (heck gifs are STILL widely used despite being inferior to PNG). Do you have any data to the contary?
I'm sure I remember an "air crash investigation" episode about an accident caused by counterfeit parts, I don't remember whether it was a full blown crash though.