The Cell would have been nice were it not for the surprisingly slow DMA read problem.
I still don't see why it's a problem. Let the SPU do a DMA write and fetch the data when they're written. PCI is pretty much the same; you want the hard drive controller to do DMA write instead of the CPU fetching across the bus.
Just don't try to use your wind turbine. In our location, to run my small car on its current, fairly low usage cycle, I would need a 6M diameter turbine on a 40M pylon, and I suspect the neighbours would object.
That would be a very small turbine. Why not go for something like this one, 82m diameter with a 59m tower? That should power a few cars, and it's a good mid-size turbine for use in areas where you cannot depend on a strong, regular wind. Place it half a mile or more from houses, of course, otherwise the sound can be a bit tiring. If the neighbours object, offer them a part in the project -- people are much more receptive to such things if they make money from them.
Why no IRC? It seems interesting that IRC would be the only thing they would disallow. Is there something about the traffic patterns or legality of IRC that would make it a problem? What about a server that does the same function, but isn't specifically IRC?
IRC servers attract DOS attacks. It's better or worse depending on which particular IRC network they connect to, but it's no wonder that the providers find it easier to just blanket ban IRC.
An interesting side note, my CRT takes less power when displaying a primarily black screen, my LCD's don't.
That is expected, actually. LCD's work by blocking out the backlight. Their "natural" state is transparent, so in theory they should even use slightly more power when showing black. In practice I doubt you can measure the difference.
Sure, light mostly ends up as heat too, but you can't be entirely sure that the light doesn't go through a window and maybe, just maybe, ends up going into space. Fairly theoretical for a hard drive in an enclosure.
I read the articles. I didn't see anything about heat and noise output. Can anyone fill me in? I would guess it would be minimal and none, respectively.
Well, based on an energy consumption of 0.5W and an educated guess that they probably aren't emitting much light, I'd say that the heat output is 0.5W.
Even as a home user I populate all of my DIMM slots, and I guarantee you the revCG Opteron cannot drive 4 slots at full speed without registered memory!
Then you'll just have to live with being a generation behind, speed-wise.
Many chipsets/BIOS'es don't know how to remap memory above the 4GB barrier. Linux could in theory do the remapping itself, but that would require very intimate knowledge of the chipset and motherboard -- LinuxBIOS can probably do it, but that is only available for very few boards.
[Please don't waste time trying to convince me I don't need ECC. SGIs taught me otherwise and soft error rates really are on the rise. Just answer the question thanks.]
Can I convince you that you don't need registered RAM? It isn't the ECC that is killing speed, it's the buffers.
If Yellowstone is going to erupt, it might be worth making sure that money you're putting aside includes a bar or two of gold.
What makes you believe that gold would be valuable in a global depression? Gold is a rather useless metal, with a few exceptions. We have lots more of it than we need. Like diamonds, it's something which has because people believe it has value.
The truth is that it is really hard to find something durable which will actually keep some value in a depression. Weapons perhaps, if the depression is bad enough.
Unfortunately, I think BIND (in its infinite brokenness) didn't handle that very well or something.
BIND seems to handle it just fine; I don't know of any problems with UTF-8 in BIND. I still don't get why punycode was invented, and this is the one issue where I agree with Daniel J. Bernstein. See his page on the issue.
That's true, but a seek to read the same track on the next platter should be very quick
On some disks the track-to-track seek time for a single platter is shorter than the time to switch to the next platter. Switching platter means you need to find the track again, and you don't know how far you're off to start with. Switching track on the same platter is sometimes easier, because you know exactly how far you are going.
IPPlan is what we use. It is by far not perfect, and we have basically switched to doing most modifications directly into the (Postgres) database. IPPlan was developed for MySQL, so it doesn't use the IP address features of Postgres. We have added a few stored procedures which keep an extra column in ip4r format, for easier manipulation by other tools.
Why IPPlan? Because the other free alternatives are even worse.
You should never use --force. The option should just be removed. Use the individual overrides instead; that way you will at least think about what it is you override. Anyway, you shouldn't even use rpm by hand if you're a regular user, yum can do everything you need.
In any event, it should be difficult if not impossible to fundamentally break a system as an end user unless you're putting it to some extreme abuse.
rpm -e --force e2fsprogs-libs will do it nicely. Of course it's trivially easy to fix with a rescue CD, but apparently ESR hasn't noticed the rescue option when installing.
I want to know WTF Cox is talking about when he says that "The moment Fedora includes non-free stuff it becomes a problem for all the people who redistribute and respin it". The people who respin it aren't your problem. You're not obligated to support them.
Fedora cares about people who respin it, because letting people respin it is an explicit goal for Fedora.
DC is much more troublesome to transport because of waste, not because it's a hazard.
Why do you keep believing that DC wastes energy when transported? DC is a tiny bit MORE efficient to transport at the same voltage, since there is no skin effect. So use the same voltages as you did before with AC.
At that point, we still have to convert down to lower voltage, and probably wouldn't be able to do it any more efficiently than we already convert AC power... so where, exactly, is the huge advantage that would make it worth it?
It is significantly more efficient to transform DC than AC at relatively low voltages. That is the whole point. Power supplies today have to smooth the output because of the fluctuating nature of the AC input, and they also have to do Power Factor Correction on the input. DC-to-DC has no power factor to correct and it is trivial to smooth the output at a typical switch frequency of 1MHz.
Slashdot Mirror
The Cell would have been nice were it not for the surprisingly slow DMA read problem.
I still don't see why it's a problem. Let the SPU do a DMA write and fetch the data when they're written. PCI is pretty much the same; you want the hard drive controller to do DMA write instead of the CPU fetching across the bus.
Just don't try to use your wind turbine. In our location, to run my small car on its current, fairly low usage cycle, I would need a 6M diameter turbine on a 40M pylon, and I suspect the neighbours would object.
That would be a very small turbine. Why not go for something like this one, 82m diameter with a 59m tower? That should power a few cars, and it's a good mid-size turbine for use in areas where you cannot depend on a strong, regular wind. Place it half a mile or more from houses, of course, otherwise the sound can be a bit tiring. If the neighbours object, offer them a part in the project -- people are much more receptive to such things if they make money from them.
i'm not so sure why everyone bans irc clients though. it frustrates the living @#$@ out of me.
Same deal -- worries about attacks. Why not run the irc client from home? (Actually I run my IRC client on my server, so I'm one of the weird ones.)
Why no IRC? It seems interesting that IRC would be the only thing they would disallow. Is there something about the traffic patterns or legality of IRC that would make it a problem? What about a server that does the same function, but isn't specifically IRC?
IRC servers attract DOS attacks. It's better or worse depending on which particular IRC network they connect to, but it's no wonder that the providers find it easier to just blanket ban IRC.
An interesting side note, my CRT takes less power when displaying a primarily black screen, my LCD's don't.
That is expected, actually. LCD's work by blocking out the backlight. Their "natural" state is transparent, so in theory they should even use slightly more power when showing black. In practice I doubt you can measure the difference.
Sure, light mostly ends up as heat too, but you can't be entirely sure that the light doesn't go through a window and maybe, just maybe, ends up going into space. Fairly theoretical for a hard drive in an enclosure.
I read the articles. I didn't see anything about heat and noise output. Can anyone fill me in? I would guess it would be minimal and none, respectively.
Well, based on an energy consumption of 0.5W and an educated guess that they probably aren't emitting much light, I'd say that the heat output is 0.5W.
Duh.
Even as a home user I populate all of my DIMM slots, and I guarantee you the revCG Opteron cannot drive 4 slots at full speed without registered memory!
Then you'll just have to live with being a generation behind, speed-wise.
Bit flips used to be a big problem. Redesign of the individual memory cells as well as picking non-radioactive casing (duh!) has helped a lot.
Many chipsets/BIOS'es don't know how to remap memory above the 4GB barrier. Linux could in theory do the remapping itself, but that would require very intimate knowledge of the chipset and motherboard -- LinuxBIOS can probably do it, but that is only available for very few boards.
[Please don't waste time trying to convince me I don't need ECC.
SGIs taught me otherwise and soft error rates really are on the rise. Just answer the question thanks.]
Can I convince you that you don't need registered RAM? It isn't the ECC that is killing speed, it's the buffers.
If Yellowstone is going to erupt, it might be worth making sure that money you're putting aside includes a bar or two of gold.
What makes you believe that gold would be valuable in a global depression? Gold is a rather useless metal, with a few exceptions. We have lots more of it than we need. Like diamonds, it's something which has because people believe it has value.
The truth is that it is really hard to find something durable which will actually keep some value in a depression. Weapons perhaps, if the depression is bad enough.
Unfortunately, I think BIND (in its infinite brokenness) didn't handle that very well or something.
BIND seems to handle it just fine; I don't know of any problems with UTF-8 in BIND. I still don't get why punycode was invented, and this is the one issue where I agree with Daniel J. Bernstein. See his page on the issue.
That's true, but a seek to read the same track on the next platter should be very quick
On some disks the track-to-track seek time for a single platter is shorter than the time to switch to the next platter. Switching platter means you need to find the track again, and you don't know how far you're off to start with. Switching track on the same platter is sometimes easier, because you know exactly how far you are going.
IPPlan is what we use. It is by far not perfect, and we have basically switched to doing most modifications directly into the (Postgres) database. IPPlan was developed for MySQL, so it doesn't use the IP address features of Postgres. We have added a few stored procedures which keep an extra column in ip4r format, for easier manipulation by other tools.
Why IPPlan? Because the other free alternatives are even worse.
You should never use --force. The option should just be removed. Use the individual overrides instead; that way you will at least think about what it is you override. Anyway, you shouldn't even use rpm by hand if you're a regular user, yum can do everything you need.
Indeed, someone replied to his rant with that one. Unfortunately he chose to not follow his own advice.
( we're not talking use here, but actual contribution ).
We are? Like what?
In any event, it should be difficult if not impossible to fundamentally break a system as an end user unless you're putting it to some extreme abuse.
rpm -e --force e2fsprogs-libs will do it nicely. Of course it's trivially easy to fix with a rescue CD, but apparently ESR hasn't noticed the rescue option when installing.
One would think Raymond is among the few people who have earned the right to say "wow, this sucks and needs to change".
Why would one think that? What exactly has he done to earn that right?
I want to know WTF Cox is talking about when he says that "The moment Fedora includes non-free stuff it becomes a problem for all the people who redistribute and respin it". The people who respin it aren't your problem. You're not obligated to support them.
Fedora cares about people who respin it, because letting people respin it is an explicit goal for Fedora.
More seriously, broadcast radio is probably a blip
Quite possibly. However, radar seems likely to be around for a long time. Incredibly powerful signals, very easy to detect.
DC is much more troublesome to transport because of waste, not because it's a hazard.
Why do you keep believing that DC wastes energy when transported? DC is a tiny bit MORE efficient to transport at the same voltage, since there is no skin effect. So use the same voltages as you did before with AC.
At that point, we still have to convert down to lower voltage, and probably wouldn't be able to do it any more efficiently than we already convert AC power... so where, exactly, is the huge advantage that would make it worth it?
It is significantly more efficient to transform DC than AC at relatively low voltages. That is the whole point. Power supplies today have to smooth the output because of the fluctuating nature of the AC input, and they also have to do Power Factor Correction on the input. DC-to-DC has no power factor to correct and it is trivial to smooth the output at a typical switch frequency of 1MHz.
Well, AC is much easier to transport, for a start. DC is, in general, horrible for anything of size.
What is horrible about 400V DC? If that's too dangerous, go with 240V DC instead.