Software dictates component choices even in proprietary editing softeware

[Luke Kuhn]

Of course, for the $2,500 price of something like Adobe CS 5.5 or even higher price of some of the Avid stuff, you could do a lot of mix-and-match experimenting, then copy what you find to work. I don't think every workstation, laptop, and old computer I own would collectively add up to the cost of one of those pro-level packages.  Which is cheaper-three video cards at $80 apiece and a few different soundcards and miscellanious pieces, at worst two different $120 motherboards or a $2500 software package that installs on $200 worth of pay operating system?
When possible, the results of such experimentation should be published. I can say the Radeon 5550 and the Radeon 5570 with open-source gallium drivers from the xorg-edgers ppa are not only able to handle the Blender interface, but can even handle GPU rendering in Blender at least twice as fast as 4 cores of AMD Phenom II.  I do not use nvidia cards for any purpose these days except testing the nouveau driver, and don't have any of them fast enough to give it a fair test anyway.
If I needed more sound input channels than the three on my board's onboard sound, I would buy one cheap soundcard, as I know to have worked well int he past with Linux, take it home, and test it. If I needded more channels than that, I would then go get more of those cards after first proving one to be good. Then I would just need a wrapper for the existing Linux drivers to gang them all together to appear as a single big card!  If anyone does this, they should publish not only the driver wrapper code, but also what exact cards they used.
I do not want to discuss portions of the chipset that handle the signal path from the keyboard or the hard drive, as that info could be used by an enemy to prepare a malicious bios to serve as a keylogger, otherwise I could publish my entire output from lswh. 
If enough people start talking about what works and does not, those blacklists and whitelists will become a starting point, and where possible workarounds for blacklisted hardware will help people with scrounged equipment. When blacklists of drivers are embedded in the OS people need to know exactly where to find them so they can experiment, as I had to do when that old Pentium II laptop had a blacklisted audio driver that still worked on that particular machine.
<snip>
Btw. a whitelist at least will be subjective to quality issues.With a fat purse it's possible to test tailor-made Apple and Microsoftbased machines. Searching the web, I didn't find this for Linux audio,excepted of OEM products for some target groups, but not for all audiousers.