Pic nabbed from Hanon
Theory can fuck off :: 17/5/10 ~01:30 :: Comments: 1
category: random/tech/uni
Well, I said I'd post something this weekend, so I feel obligated to do so now. Damn you all, all 2 of you.
The sad fact is that my comment about yesterday in my last post was shoe related. Hanon were releasing a new pair of trainers that I really wanted to get, New Balance x UNDFTD Desert Storm CM1500s.
Pic nabbed from Hanon
Anyway, I knew these were going to be released on Saturday. I phoned them up on Wednesday to ask how many pairs they'd have - they told me 1 UK9 which had already been reserved for a member of staff, and 1 UK10, and they were being released at 10am on Saturday. So I had to hope that I could get on their site quick enough to buy that single pair in a UK10 before it sold. To cut a long, tedious and awful story short, I was out at the pub on Friday afternoon and, slightly inebriated, I phoned Hanon again to double check. Turns out that they also had 2 pairs in a UK9.5 which they'd neglected to tell me about previously, and they also let me order one over the phone. I was up at 10am on Saturday anyway and watched the site to see what happened; within 15 minutes everything below a UK11 had sold out. So, glad I called. Drunkenness has its benefits.
I have been attempting to do a bit of uni work, or rather work on my dissertation of late. My primary problem is the lab setup in uni - I need to install Windows XP on a computer, then download a few random bits and pieces onto it, go to a few websites etc, basically emulate what a hard drive that could be taken out of a home PC would look like. My problem is the uni lab I use has an incredibly frustrating setup, whereby connecting a new computer to the internet is an arduous chore. And it's been driving me up the sodding wall, to the extent that I brought a hard drive home from the lab which I'd already installed Windows on, only to find that I don't have any sodding working computers lying around that can take an IDE drive.
But enough of that part. Tonight, I decided I really ought to perform a quick experiment with hard drives. At present, I have pretty much no understanding of how the drive communicates between the heads and the cache. There must be a low level instruction set on that drive which deals with the requests to read and write. What I wondered was if this instruction set differs between different drive types; ie could you take the circuit board from a SATA drive, dump it on an IDE drive, and would it be able to read and write. Well the first problem I've run into is the fact that I don't have a single spare unused SATA drive in my house, something I need to remedy quite quickly. I'm sure I'll encounter various other problems once I actually have some hardware to work on. Instead, I experimented with something I was fairly sure was doomed to fail; I changed the circuit boards between two hard drives, but the drives were quite different, aside from the vendor (both Maxtor). The first drive (model number 86480D6) is a 13328 cylinder drive with 15 heads, it's a 6GB drive with a 256KB buffer. The second drive (model number 4D040H2) is 16383 cylinders with 16 heads, 40GB capacity and a 2MB buffer. The 6GB drive is claimed to be 5200RPM as opposed to the 5400RPM of the 40GB drive, but frankly I'm assuming they're both going to be 5400RPM cause I've never heard of a 5200RPM drive before this.
So, two drives, very very different configurations; the point in exchanging the circuit boards was simply to see if this would work. Now, theoretically at least, there's absolutely no reason why switching the boards between 2 identical drives wouldn't work. But since I don't actually know how the drive communicates internally, I don't know what precisely would have to change the mess with this. As you can probably guess, the drives didn't work with the switched board. Well, that's a mild assumption on my part - the 40GB drive didn't work with the 6GB drive's logic board. The 6GB drive is the one I tore to hell in an earlier post on here, so I didn't even bother trying it. What was interesting though was the noise the 40GB drive made when it was trying to seek on the 6GB board. Clear clicking sound, very very much like what you'd hear when your hard drive is dying. So, a bit more research for me to do there, but I do wonder if there's something related to what was causing the drive to be unreadable, and one of the more common forms of hard drive failure, since that clicking sound is something a lot of people will tell you to look out for. Might just be a complete coincidence, or then again it might be completely unrelated - simply the heads attempting to move to a certain location but being unable to because they're calibrated/configured for something else? I don't have the slightest clue, I don't even know if that's possible come to that. Hence, still some work to do there.
Perhaps also obvious is the fact that different manufacturers use different layouts for their hardware. I have a Seagate drive here that I pulled out one of my old computers, another old IDE drive, 7200RPM 250GB. The location of the connectors for the heads on the Seagate drive is different from the Maxtor drives; it's hardly a massive assumption to say there's probably no industry standard for hard drive hardware layout, but then it is something that I find a bit curious - when you're dealing with response times as small as this with performance gains for drives, does the layout make any significant difference? I'd assume not, but I don't know. I am quite tempted to open the Seagate drive up just to do a comparison with the Maxtor one I took apart earlier, but I've got no clue what's on the drive, so I'll need to back it up before I consider destroying it.
So, next step is to switch the logic boards between SATA and IDE and see how badly that fucks everything up. Then it's switching between 2 identical IDE drives, and being immensely confused if they don't work perfectly. At the same time, I need to actually get that uni system working so I can set up my initial XP disk, which then needs to be copied onto another stack of hard drives. Then I can start to break the sods!
I do have one other nagging worry about the way I'm doing this. To attempt to save myself a bit of time, I've decided not to bother installing Windows manually on 20-odd drives. Instead I'm configuring 1 the way I want it, then making a forensic image of it through the wonders of dcfldd, and then just copying that image onto all the other drives. The problem is I don't know the state of these drives I'm mapping this image onto - will being partitioned/unpartitioned mess with this, if they already have data on them will I find remnants of stuff I shouldn't be finding which will affect my results, etc. In theory every disk I'm using is blank, Systems Support said they were going to run DBAN on them all before handing them over to me. Well, the one disk I've tried to use so far wasn't wiped. So I really do have an unknown factor to deal with now in the contents of the drives, and if I need to start partitioning all these drives then that's just another massive time consuming chore to go through. Pain in the arse.
Still, I'll have new shoes in the post soon.
Stars Of The Lid - Austin Texas Mental Hospital (Part 2)
Comment:
You can wear those when you are hunting deer across the Sahara!