Originally Posted by JHBrandt
That's correct; the raw data rate is 19.4 Mbps. The transmitted data rate is about 32.3 Mbps, but that includes things like FEC and trellis coding that don't make it past the tuner, so 19.4 Mbps is the figure you need to use.
The iView's .mts recording format causes a bit of "inflation:" it adds 4 bytes to each 188-byte TS packet. But that only brings the total up to 19.8 Mbps or so. And an actual recording would usually be less than that, because packets for subchannels other than the one you're recording don't get written out.
There's also file system (FAT32 or NTFS) overhead, but that's hard to calculate. You just have to guess. From your experience, I'd say 22 Mbps is close but not quite enough for the most demanding channels. I'd think 30 Mbps would be enough, but as you say, who knows?
Yeah, who knows? There seems to be more to this than meets the bit-rate statistics...
About a week ago, I decided to try to pause/fast forward/etc. a 720p channel, didn't work well, I was expecting that. This was using a flash drive independently rated at a 22Mbps write speed, a drive that couldn't record a single-subchannel 1080i program without stuttering every 10 seconds or so, with presumably only a 19.4Mbps rate or so.
Then right now, I tried to live pause another 720p channel...WORKED PERFECTLY. I paused 18 minutes of the program, fast-forwarded through the commercials, watched the program, paused it some more, NO PROBLEM.
The channel that didn't work had four SD subchannels, the one that worked has one other 720p subchannel. GO FIGURE.
So I looked at what I had on the flash drive. Good thing I did, I had a SD program I DVRd several weeks ago that I want to keep forever, so I want to transfer it to my media server.
I also had something else I had forgotten about, a FHD program I DVRd not really caring about it because I did not really expect the recording to be perfect...BUT IT WAS.
In looking at the independent USB flash drive test site, one thing I can note is there IS a big deviation in results between different testers, sometimes REALLY big. I can also note that by reading some of the reviews on Amazon for flash drives, some of them show HUGE differences at different times with the same drive, and on different machines. There are always big differences between file sizes, and also significant differences depending on file system format.
I don't think I'll get the Sandisk Ultra since a lot of the Amazon reviews single out really bad write speeds as a deal-breaking negative. But who knows, maybe it's great with writing videos from an iView, some weird serendipitous match between drive hardware and iView hardware...
I now am thinking of getting a Lexar Jump Drive P20, which has some of fastest write speed benchmarks of any drive on the planet, and the Amazon reviews specifically mention being able to write like 170GB of videos to the thing in like 10 minutes, and unlike certain other super-fast drives, it didn't melt or destroy the USB port with excessive heat.
So we'll see how that works in the iView...