return of the studio jedi(maybe it's time to retire the star wars references in the titles, huh?)
so when we last checked in, we had finished setting up the drums, replacing the heads and conditioning the drums for this particular project...and we'd spent a little more cash than we'd wanted to on building custom-sized baffles around them to deaden the leakage into other parts of the house.
we stayed up all night running cables, wiring outboard gear into the patchbay, and snaking ADAT connectors behind the console...then subsequently, we sat for some time peeling little, tiny stickers and labelling the patchbay connections.
we put up this kickass shelf that holds the computer monitors and the studio monitors, that has the added benefit of making the desk that the console sits on portable to the extent that it can be pulled out for access to the back of the console - this is a huge benefit, in that as much as we'd like to believe that we think of everything when we set things up initially, there's always something to be changed or repaired over time.
we ran an eight channel snake from the mic inputs on the console back to the drum room, to facilitate mic'ing the drums and recording them through the console to either the ADATs or directly to the computer.
oh, yeah - the computer.
the computer in the studio was custom built by yours truly for audio recording. i bought the ASUS A7N8X deluxe motherboard and set it up with an AMD XP3200 processor. the motherboard i chose uses the NVIDIA chipset, which had gotten pretty high marks for recording in a number of user forums that i frequent, and i chose it for that particular reason. it also supported serial ATA, and i bought two serial ATA drives and set them up in a RAID configuration, striped, for maximum speed. i put in a DVD burner and an external IDE hard drive bay, to facilitate moving data between the studio computer and other machines without having to transfer the data across my home network. i built the whole system into a 4U rackmount case, and put it into the same rack as the ADATs, on the left of the console.
the issue of what soundcard to use is something that most recording engineers labor and languish over - and certainly, i was no different. i approached it a little differently than maybe a lot of people do, though, because my needs aren't typical...at least i'm not sure that they are.
since i had history with my ADAT machines, and they were likely to factor into future projects, i opted for a card that would interface with them. ADATs utilize what alesis calls lightpipe technology, which is essentially a digital connection that carries all eight channels through a single connector. all ADAT machines came with a lightpipe input and a lightpipe output connector that uses fiber cable to transmit the data digitally to whatever it's connected to. so, if properly equipped, you could connect a digital mixer to an ADAT machine with two cables, instead of the analog method of running a single cable for every track, both in and out, to the machine.
frontier design sells a card called the dakota that utilizes the lightpipe technology to interface ADAT machines with the computer with two sets of ins and outs, a sync connection, and MIDI connectors as well. the way the dakota works, you could use the ADAT machines as analog-to-digital converters (or A/D converters for short), interface the ADATs with the dakota card to get the audio into the PC, and then record the audio directly to disk once it's in. the beauty of this card is that with dual lightpipe ins and outs, you could record or playback as many as sixteen tracks at a time. for me, that could easily be enough for an entire mix - so i could mix completed songs "in the box" (everything inside the computer itself, with no outside processing) or "out of the box" (bringing all the audio tracks out of the computer, through the ADATs, into the console - with full access to all my outboard gear).
whether i went with the dakota card or not, that method seemed like the only way to go for my particular operation. as soon as i saw a dakota card on ebay, i jumped on it...16 bit critics be damned.
now, it could be said that i made the mistake of not choosing some of the components as carefully as others, as i came to realize as i booted it up and brought it to life...i had removed a couple of case fans that the manufacturer had included, simply because they were mounted in an odd place, and they were cramping the PCI cards i had installed in the box - so out they came. i used the stock heatsink and fan that came with my processor, and the power supply that came with the case...so that, combined with the fan in the removable hard drive bay, created an unacceptable noise floor. the removable hard drive bay was an optional component - it only needed to be operational when i was using that particular drive - so it wasn't as much of a concern as the other two components. but still i knew that once i started recording anything other than the scratch tracks i'd been laying down for the poco album project, that the noise was going to be an issue. i'd been working thus far by putting a click track down and recording a scratch acoustic guitar and vocal to work to as a foundation, so it hadn't really bitten me yet.
nonetheless, i ordered a fanless power supply, and started researching low-noise CPU fans. the one i settled on had the lowest dB rating of any of the ones i'd seen - a whisper quiet 16 dB!
an yet -s it was on backorder when i placed the order and didn't show up for almost two weeks. so i worked on various stuff while i waited for it that didn't require complete quiet - including importing some of the unreleased tracks from the blake allen record from ADAT into the computer so that i could edit them and do some rough mixes in the box of things i'd played on that i wanted to preserve. i even did a couple of rough mixes of some of the songs - there were two in particular that i slaved over, in terms of coming up with parts that i wanted to take a swing at...specifically to use for an edit mix of instrumental parts that i'm putting together as a new audition clip for the website.
i remain quite proud of the work that i did on that album....but i digress.
once the fan showed up, i rushed it home from work, eager to tear into the box and install my new components and spread a hush over the room - the power supply was a piece of cake, but the fan on the new heatsink assembly was so large, there was no room to install it - not because of the depth of the case, but because of the proximity of the CPU socket to the power supply - there was no way it was going to mount flush on the socket with that gargantuan fan.
so i sent it back and went and bought a fan at a local outlet for $20 that came with an additional 4 dB of noise at no charge. it did, however, have an attenuator built in to it that allows the user to set the fan speed.
after i installed the fan and the power supply, i booted up the system with the case reassembled - and marvelled at how quiet it was for a few minutes before i took the occupants of the house out to dinner - it was sunday, so we went to cousins' in shoemakersville for some football and the best burgers in the county....
...and when i came home, the PC was dead.
the light on the front of the case was on, but it was completely unresponsive. i tried restarting it - nada.
i couldn't believe it - first i was pissed at myself for leaving it on and leaving the house and not monitoring it after installing all this new stuff...i had assumed that my CPU was barbeque, although it later occured to me that the mobo had a COP feature (Cpu Overheating Protection) that may have saved me from having to invest in another chip - but i still wasn't sure what the source of my problem was.
until i opened the case.
the heatsink/fan combo that i'd selected screwed into a backplate that had to be mounted on the back of the board (which involved completely removing the mobo from the case - or, in other words, taking the entire computer apart. roughly the cyber-equivalent of a head gasket job on your car) and attached to the backplate via four long screws that pass down through the fan, through the holes in the mobo, and into the threads on the backplate.
two of the screws hadn't been threaded into the backplate as tightly as i thought they had, and they'd popped loose from the backplate - so the heatsink wasn't making proper contact with the CPU chip.
i took out the CPU and examined it for scorch marks or any other visible signs of damage, and saw none...so i decided to use the standoffs that had come with the fan, run a short screw through the back into one side of the standoff, and then run the screws down through the fan into the other side of the standoff. at least that way, i was assured of a solid connection - provided all the screws were the proper length.
and there was no reason i shouldn't be able to count on that, right?
well, maybe not.