(What follows is somethig I posted on usenet, and sent to [email protected]
as well -- I've edited it slightly. I received very little feedback, and thought this would be a place to fix that!)
I own two Replays at the moment, and am trying to decide whether to buy a new 4000 series box. Perhaps folks could help me understand some of the tradeoffs.
1. Price/performance. The 160 hour unit I'd probably buy is $1499, less a $100 discount available over the phone. Adjusted for incredibly price/performance gains in disk drives, this is a lot higher cost than my original 2002 a few years ago. Is there something more "open" coming down the pike? Cheaper, better, faster?
2. Fixing the major annoyances of the previous generation. My number one gripe with my two replays is the racket they make. My bedroom unit turns itself on at random times for a few seconds -- you hear the unit spin up the disk, access it a few times, and then go back to sleep. I also live in California, so the standby power consumption (even with no disk activity, the unit is warm) is also a bother. Does the 4000 series do this more intelligently?
3. Incomplete features. Why doesn't the unit have a Dolby Digital input? My digital cable box happily pumps out AC-3 for the movie channels, and I'd love to be able to preserve this good audio on my recorded programs. The only explanation I've heard for this is a reach -- potential future programming going over the broadband connection that might be AC-3 encoded.
4. Not future-proofed. Folks spending this kind of money for time-shifting are also probably starting to play with HDTV. It'd be nice to understand SonicBlue's roadmap for covering this. Given the satellite folks are ramping HD up really quickly, I'd expect Replay's target market to be FUDed (Fear, Uncertainty, Doubt) over whether the 4000 is a good 1-2 year investment. Perhaps they're just victims of the same fear everyone has: copy protection schemes invalidating existing users' hardware investments.
5. Unhappy middle ground. I was picking up hints that the next generation Replay was going to be more of a house server, which would have been a great concept. The idea here is that you'd have a box in a central closet that had broadband and TV connections. You'd invest a bit of money in this server, but could then hook up inexpensive (AND QUIET!) set top boxes anywhere in the house -- all they'd need is an Ethernet link (and perhaps an antenna hookup). The house server would then pump MPEG streams to multiple boxes simultaneously. The 4000 kind of gets there: you can watch programs from another 4000, but where's the diskless, inexpensive model? And what's the roadmap for scalable house servers? How many streams can a 4000 handle?
What if the disks were striped (doubling the bandwidth)?
6. Curious design choices. Why is the only 480p output on a VGA connector? I don't know of any set that has more than one VGA connector (and most of the new HDTV sets have none). And most of these boxes will be integrated into AV receivers, which can switch component video -- but not RGB; not even the lunatic fringe receivers do VGA -- http://www.lexicon.com/MC12/index.html
). It seems as though the engineers didn't correctly guess what a high-end home theatre setup would look like in 2001. And given that all the inputs to the 4000 are interlaced, what's the output look like? Do they do intelligent line doubling like the iScan? ( http://www.dvdo.com/dvdo/iscanv2.asp
Note: there are some aftermarket solutions to convert VGA to component:
They're $100-$200 extra (with yet another bloody power transformer). This seems silly.