Got a couple of 120GB drives via UPS, had to ship to office so the tivo upgrade for jess would be a surprise, forgot to include suite number.
Place (justdeals.com) refuses to acknowledge any communications, so I end up pulling our UPS guy (Stan, really nice guy) off to the side the day before it’s scheduled to arrive and telling him that a package will show up at the building with only my name and no suite number on it. The rest of it goes well, I get the drives yesterday.
Jess actually turns in relatively early (10:30-ish) so I get a decent amount of time to do the upgrades.
Long-story-short, the main issues were:
1) third screw to release the big drive tray in the SAT-T60’s is basically *in* the power supply. I shoved a napkin over top of the capacitors and wedged the T-10 torx screwdriver in enough to get it out. I didn’t tempt on either of them to get it back in, they’re in there enough as-is.
2) used 2 of the 4 existing T-10 screws mounting the existing drive to mount the new drives. No real big deal, though.
3) serious lack of IDE cables, combined with the new knowledge that the ATA-66 (I think, I should look this up) cables have a missing pin (for males) and covered hole (females). The IDE connector’s (known, AFAIK, to be UDMA-33 for these tivo’s) on the motherboards have all their pins, so I have to scavenge cables from elsewhere. As it turns out, the cable attaching the CD-RW drive in my P3/733 box got upgraded as the existing one got salvaged to route in Jessica’s tivo. No K, V, or M attached and I hadn’t checked the bootloader on that box in awhile so I did the cable swapping on the fly (the hard drives were on a diff controller and nothing in the drive, this wasn’t a huge risk). It feels so wrong to do things like that on x86 machines, I’ve just got too much history limiting my “mess with the hardware while the machine’s under load” events to old AIX machines I think.
4) putting in the new drives as “cable select” (just like the existing drives) led to some weird results – the drives were kinda recognized but the cache (2048KiB) was recognized as 64KiB – put the new drive jumper over to slave from cable select and it works fine. That’s kinda odd, I had the existing drive (CS jumpered, still) at the end of the IDE chain, not sure why the maxtor 120 wasn’t happy being CS in the middle, but jumpering to slave’s not a big deal. There’s a chance the BP6 (dual celery, where I do my tivo backups and upgrades) doesn’t support CS on the controller, but it *is* the HPT366, it really oughta.
not really much drama outside of the shipping, actually – I take 2 mfsbackup’s of each drive before doing anything, each of the 2 backups with different cables, and only proceed when they “cmp” fine. The DirecTiVo compressed backups were a lot bigger (maybe 195MB) compared to the standalone’s I had done in the past (maybe 112MB)
Anyway, so far so good. I could have put dual 120’s or dual 160’s or whatever, but 1) the hinsdale howto says if you go much over 150GB you should consider more swap space, and I’m lazy enough to avoid the -s 127 and 2) in my experience, I can only really find use out of around 100 hours or so – the 143 (variable, of course) resulting from this should work out fine.