miscellaneous notes
=========================================

2011-02-21
--16:30ish: channels 22 and 76 consistently have bad contacts; i think this is due to lack of stiffness in pre-amp pins; removing gold rubber addressed this problem
--20:36: channels 41 and 43 look pretty noisy; mostly uncorrelated spiking
--21:40ish: 2011-02-21-23-25.spike is larger than neighboring files, likely because I was opening and closing the incubator a zillion times to try to make adjustments for the culture on the Annihilatrix rig (in the same incubator)--so probably a lot of garbage in that recording

2011-02-22
--00:09: channels 41 and 43 look okay again; less uncorrelated spiking, and bursting about once every few minutes
--00:17: 41 and 43 looking sketchy again
--10:51: definitely bursting with a higher amplitude (in ASDR) than before; immediately after burst there's little spiking, then slowly becomes more frequent until eventually a large burst occurs again (a lot like chick spinal cord mEPSCs and 'episodes'); burst frequency about once per minute
--12:45: channel 22 is really noisy--the large file size for 2011-02-22-12-22 is probably because of falsely-detected spikes
--12:52: stopped recording cycle...
--15:00: no bursting during the 5min i've been standing here, but still good about of uncorrelated spiking; file sizes have definitely gone down (but still same order of magnitude)
--18:25: 41, 43, 58 have slightly higher RMS (esp 58)
--19:30: all channels look okay!

2011-02-23
--15:09: I came in to find that there have been no spikes recorded since 22:48 last night.  There is definitely spiking on the array so this means that there is something wrong with the recording.  It also means that I've lost all data from the hypothesized critical period in homeostatic recovery.
	phenomena: all spike file sizes are 0kb, 650kb, 1.3mb.  Jon says these are probably just the headers.  the detected spikes are appearing on the scope just fine.
--16:30: active channels include... 71, 53, 87, 68, 61
--16:55: Jon helped fix a problem in the version of repeat-record that is running, apparently since file sizes got so small, and the different meabench modules were closing before dumping the spikes to a file.  (he added some interrupt statements to fix this).  even those files with larger file sizes will have the last chunk of time clipped off since those weren't dumped to a file either.  it's running properly now.  for the larger file sizes, firing rate data can be salvaged, but it's kind of a bummer for the small files... still, i should be able to see what the minimum file size saved was and use this to extrapolate a "maximum firing rate" for those lost epochs.  in the end, it's a good lesson for me to pay better attention to this sort of stuff (it took me 2 days to realize!) and learn more about the scripts that jon has been writing for me.
--23:23: changed media in both cultures (regular Jimbo's in 363, Jimbo's + 20uM CNQX in 359); spiking a lot after the media change; channel 22 has slightly higher noise, but definitely tolerable given the size of the spikes
--23:52: previous file size is HUGE.  it looks like noise levels on 22 are pretty variable... be sure to exclude that channel in 2011-02-23-23-23.spike file

2011-02-27
--22:09: spikes on 15 and 85 are not real... looks like there was just an increase in noise on these channels; meanwhile, there's something wrong with 21, 12, 22, and 33 (~0rms on these guys)
--11:46: wash (3x w/ media)

2011-02-28
--00:38: it doesn't look like the wash made any difference in activity at all (e.g. no bursting at all)... i'm going to attempt one more wash before going home.  also, ignore all the stuff on 21, 12, 22, and 33
--00:46: after the next wash, it still look like only the lower left quadrant is spiking.
--01:55: spiking just hasn't rebounded (it's actually less than right before the change).  something doesn;t seem right about this...