[haiku-development] Re: Laptop HDA audio

  • From: Simon Taylor <simontaylor1@xxxxxxxxxxxx>
  • To: haiku-development@xxxxxxxxxxxxx
  • Date: Fri, 22 Jun 2012 09:41:29 +0100

Hi Dane,

On 21/06/2012 15:59, Dane wrote:
The bigee right now (I mean REALLY big bigee) for TT Systems right now is HDA.  
I don't want to say money is no object, but that's almost true.  We've GOT to 
get HDA (input and output) in good shape, and fast.  We'd be willing to also 
donate a motherboard with HDA to the developer.  Getting the HDA driver healthy 
will keep us going, and of course will benefit lots of other users too!

This is might be an unpopular thing to say here(!) but have you considered switching to Linux for TT? I'm clearly a big Haiku fan and want the project to succeed, but if I was in the market for a radio automation system then Haiku it is not a platform I would choose to use, regardless of how excellent your software may be on top.

Your original decision to go with BeOS was a very good one at the time; the sales messages (preemptive multitasking, excellent media performance) really were advantages of BeOS over Windows 9x and Mac OS 7 and I'm sure made a very positive difference to the radio automation experience at that time. However times move on and that is no longer the case. Other operating systems have improved their performance (I still find OS X pretty dodgy at keeping media playing whilst doing other things, but Linux doesn't seem to have issues), and hardware has massively increased in performance to the point where playing songs solidly shouldn't really stress any modern system.

My advice would be to investigate the cost of porting your UI code to something like Qt and weigh that up against the cost and probability of getting Haiku into suitable shape for your customers. It is far easier to find people with Qt skills than to find someone capable of fixing the HDA driver on Haiku. Even when (or if) the HDA driver is fixed, you're still lacking touch drivers and you're still running on an alpha OS with a number of stability problems and occasional freezes (scheduler issues). Linux would give you the benefit of a solid underlying platform with strong hardware support, allowing you to focus on adding real value for your customers. Qt would also mean you could sell versions for Windows and OS X if you wanted a software-only model.

A benefit of Qt is there is already a Haiku port, and when Haiku matures it would be possible to easily switch back to Haiku as an underlying OS when it starts to win on the key performance metrics (solid media performance, stability, ease-of-use).

I'm a massive fan of all you've done for BeOS and Haiku over the years (I was a regular BeOSRadio listener around 2000) but I'd hate to see your business run into difficulty due to the current alpha state of Haiku. Obviously no-one knows your business better than you but I hope at least I provided a little food for thought.


Simon

Other related posts: