Uh, what is your fear of memory regarding the 978 based on? Remember, 2 years is a generation where technology is concerned.
You are misreading my concern. My concern about the 978's memory was based upon PeterT's latest post
Fellow Outlaws:
Please know that we hear your comments and concerns about the lack of an update on the Model 978. ... It will take another few days to fully analyze the data as it impacts the schedule, but the headline for today is that the results are favorable in a key area that means we will soon resolve the toughest part of this product, the implementation of XT-32. Our concerns over memory requirements have been addressed...
Peter's comment was like a warning shot over the bow to me. First it made me remember that the companion product to the 990, the Sherwood 965 eventually included a parametric equalizer that was never implemented on the 990 (supposedly due to memory issues). Second, it brought back memories of design choices I had to personally make when designing applications for hardware systems that were severely memory constrained (including a 48K IBM System 7). In that environment, I knew that including one type of functionality might result in the loss of other types of functionality. Consequently, I tied the two together and speculated that 978 might lose some functionality as a result.
I am made no representations about the XMC other than I liked they posted the chips that were being used and their room TACT room correction software was being implemented via a Linux based system (which is what I recommended Outlaw do 2 years ago).