[hroeder]

I'm convinced. I recorded a number of films last week. Something I normally don't do. (They usually go to the DVD recorder if I'm going to save them.)


But after watching what I recorded I'm convinced that on the 4504, at least, the recorded image is better and the recorded sound is better than the live buffer.


This was NOT the case on my old machine. (A 20xx). The High record and High buffer seemed almost identical.


I know this has been discussed before, but it would seem there should be an answer, and a correction. Hey, we are the people who've kept RePlay alive. Why has the product quality not remained consistently good?

 

[Dallben]

I really haven't noticed much of a difference on either my 4000 or my 4500...

 

[mikea28]

i actually have a theory on how this *could* happen, although I've not observed it myself. When playing back the live buffer, CPU utilization is a good bit higher as it's helping encode and decode simultaneously. More than likely, encoding takes a higher priority (as observed when rewinding to a segment that had just stuttered). Whereas with playback only of a recorded show, it's only decoding, therefore less likely to drop frames or have other decoding errors. With that said, I've not noticed the difference myself except that there's less stuttering with playback alone. Quality seems identical on my 32" Sony.


I will also say that I've noticed on both 4xxx and 5xxx units that CPU utilization is way more variable than it ever was on my Showstopper. The Showstopper stayed at a solid 60% while watching live tv. It would only jump up if i started going through menu's or otherwise using the unit beyond live tv. The 4xxx howevered around 30%, but would jump around from 25-40% regularly, just on live tv this time. The 5xxx is also usually around 60%, but can be from 50-75% it seems. again just live tv. Not coincidently, my Showstopper virtually never skipped, while both the 4xxx and 5xxx skip entirely too often for an embedded device.


There's either more background processes going on with the newer units or the software is just not programmed as well. I'm hoping both of these issues could be taken care of with future updates.

 

[hroeder]

I'd say your experience probably mimics mine. I don't remember my old machine changing much, ever. But something is certainly different on the new machines. Especially when there is quick motion.


My own "feeling", and I have absolutely no proof, is that when I'm watching the live output I'm not actually seeing High definition. I'm seeing somewhat less than High. But when i record something at High it definitely gets encoded that way, and plays back that way.


I remember a setup screen from the old machine that said whatever I chose as customary record mode would be the normal play through mode. I don't remember that same statement on the new machine.

 

[mikea28]

i'm fairly confident all replay's always record at high. an easy way to see this fact is to hit pause. Notice that the amount of pause time is equivelent to the amount of high quality record time left. further you can certainly see the difference between a high/med/low if you're watching the recording as its in progress (ie sort of like watching live tv)


i agree that the new replays encoding seems more skewed toward low motion/higher detail but stumbles a bit on faster motion.

 

[franc11s]

Quote:

Originally posted by mikea28 Not coincidently, my Showstopper virtually never skipped, while both the 4xxx and 5xxx skip entirely too often for an embedded device.

That is also what I'm seeing. I'd say sometimes if I'm just watching, I'd get skips, subtle but they are there. If the unit is recording or streaming to another device the incease in skips is significant.


Definately I never noticed it at all on the Show-stopper, like you...