Sorry, am I full of shit? I was basing that on what Dell charges for 4gb modules, which apparently is ocompletely out of step with the market?
Ok, so if you're building a >12gb system you can save a shitload of money by not buying from Dell.
But you still can't do it for $250.
So I read up on i7 today, sounds frikkin awesome. The cheapest chip was $279.
I'll wait until it becomes old school, but I want one of those someday.
So I'm back to core2 duo/quad.
Yeah, I should check memory prices before I go whole hog.
I underestimated newegg's search tools. Very impressive. At first it was way overwhelming, now it is only moderately overwhelming as I start to learn all this crap I hoped I'd never need to know.
I'll let you know what I come up with.
There are boards with 6 slots, I think mebbe I go with one of those.
Odd things like that are good, they cut the options from around 250 to 6. :-)
What I can gather about the memory market:
DDR2: available in unregistered (desktop memory) form in capacities of up to 4GB per DIMM at reasonable prices.
DDR3: still only available in 2GB/DIMM capacities unless you want to move to registered (server) memory and corresponding motherboards. Companies like Dell seem to have access to higher capacities of unregistered memory, but they charge an arm and a leg for them. This situation should change in a year or two.
Once again it still seems good to live 6 months behind the curve.
Watch out for those 6-dimm-slot socket 775 boards... usually they're a mix of ddr2 and ddr3 and you have to pick just one
Ford: Asus P5QL-E
Ford: Asus P5QL-E
Sounds nice, but doesn't seem to be available anywhere...
Newegg has it unless I mistyped the name.
Looks like they sold out overnight.
I'm using an ASUS P5N-E SLI motherboard and have been for about 4 years now.. hasnt given up. might want to try that one if u want expandability.. it's not tri-channel but it'll get the job done with a quad core intel.
Stephen D King
The Kings Photography
"When the rich wage war,
it's the poor that die..."
- Linkin Park "Hands Held High"
On a web forum where people are discussing consumer-grade routers, one person said "<provider> will replace your router for free; mine got killed by lightning and they replaced it for me"
I'm wondering whether "killed by lightning" has already become a sort of folk myth. Most of us have seen our share of dialup modems that met an early demise after taking a hit from a lightning bolt. I've opened a few up and seen the roasty-toasty components near the line jack. But in this case, the person is talking about a device that does not directly attach to any outdoor wiring -- and his outdoor wiring is fiber anyway.
The idea of a device like a router getting damaged by power drops and surges during a storm is somewhat more believable, but I really think people are remembering the "killed by lightning" routine from the days of dialup and inappropriately carrying it forward. It's an interesting observation of human behavior.
Are POTS lines grounded at all? What about DSL-ified POTS lines? I know for CATV or an antenna drop, you need to ground the coax sheath. This protects you from surges on the sheath. It doesn't protect you from surges on the center conductor, except to the extent that your equipment can shunt surges to ground.
We had some lightning strikes very close by last night, causing a brief power brownout or two. I don't have an inline surge suppressor on the antenna drop, but it is grounded. Not grounded to code, as we're not using #6 copper, but it is grounded. I guess it's holding up so far.
However, the person who made this comment has fiber. The cable stock that Verizon uses is completely non-metallic -- it doesn't even have a messenger wire. That means 100% of his copper wiring, for all services, is indoors.
My comms gear isn't "grounded to code" either. It's "grounded to common sense" which is probably something I ought to revisit.
Aug 19 2009 1:39pm from IGnatius T Foobar @uncnsrd
The POTS pair itself is not grounded, but the messenger wire usually
doubles as a ground. I'd definitely believe it if a DSL modem were
fried by lightning, or even if a strike made it through the modem and
also took out equipment downstream.
Ok, I just looked it up. The telephone signal pair uses differential signalling, so it's not possible to simply ground one conductor in the pair like you ground the coax sheath. The messenger wire doesn't go to your premises and although it may itself be grounded, as said it's not possible to ground signalling wires to the messenger wire. So with a few miles of local loop, there are lots of opportunities for induced current to come in.
If you want surge suppression, you need to ground to your electrical system.
Regardless of that, though, you're absolutely correct; there are lots of opportunities for induced current to come in.
So what's the point, exactly? The messenger wire is grounded, but it's not connected to anything else in the phone system. They're gambling that some percentage of the overcurrent from a lightning strike will just "jump" to the messenger wire and follow that to ground? Doesn't seem like it will help much - lightning can jump a lot of things and a messenger wire hardly seems like the shortest path to ground from the perspective of a lightning bolt.
Meanwhile, the path to ground created by the messenger wire is not connected to your handset/modem/whatever, and this stuff is all typically connected without any shunt to your electrical ground.