ConiKost wrote:
Thank You for your answers! I will try the Homebrew FW (I guess you mean the WDLXTV-Live Firmware?) But I don’t get it, why is samba so slow on the WDLiveTV? I’ve at home here Gigabit an get over Samba about ~ 35MB/s… much faster than the WDLiveTV can get…
That’s gigabit, there currently is no mediaplayer chipset on the market that features gigabit (although Sigma has 2 in the pipeline). And yep, I meant the wdlxtv FW.
Sigma has 4 chipsets that have dual Gig-E interfaces.
The 8656 and 57
The 8646 and 47
They all have dual Gig interfaces.
TonyPh12345 wrote:
Sigma has 4 chipsets that have dual Gig-E interfaces.
The 8656 and 57
The 8646 and 47
They all have dual Gig interfaces.
Yeah, well I said two since these are two with or without SoC (even numbers with Soc, odd without). Anyway, these indeed will be the first 2/4 with gbit. Although I really don’t expect them to get you 30 MB/s. And they’re not on the market yet, still in their pipeline, should hit the market in september this year.
Ah sorry, I’ve forget to say, I know that the WD-Live-TV has only 10/100. But it seems that is can’t get near the 100mbit. Because, all of my 10/100 only devices here get about 10-11mb/s. But the WD seems not to achieve this when playing avatar. And this is, what I don’t understand why. Because IMHO its not the smb-protocol, as I can get higher speeds… Anyway, I will try today the homebrew firmware.
ConiKost wrote:
Ah sorry, I’ve forget to say, I know that the WD-Live-TV has only 10/100. But it seems that is can’t get near the 100mbit. Because, all of my 10/100 only devices here get about 10-11mb/s. But the WD seems not to achieve this when playing avatar. And this is, what I don’t understand why. Because IMHO its not the smb-protocol, as I can get higher speeds… Anyway, I will try today the homebrew firmware.
Not sure why Samba only gets around 50 Mbit on the WDTV live, I know overhead is high, not sure if it’s all due to overhead though. Anyway, NFS will get you around 80 Mbit, which is more than enough (both in theory as in practice: i haven’t found a single file that stuttered via NFS).
Man, that handbrake prog is kinda buggy, isn’t it ? I tried to re-encode that famous bird scene from planet earth -pole to pole. The ‘original’ file (h264/mkv) plays fine but when I re-encode with h264/mkv again, it introduces audio stuttering, both on my pc and wdtv. I tried re-encoding another cut out scene from pole-to-pole and then handbrake just kept crashing on me and ended up deleting my source file, LOL. Luckily I had a backup.
Anyway, the file that I *could* re-encode looks quite a lot like the original but if you compare them side by side, you can spot the differences. Here you can see the stills from the same frame: http://www.megaupload.com/?d=T3O3EDVE
If you flip through them you’ll easily spot the difference. That being said, of course these differences are only very minor and probably won’t bother the average user. However, I don’t consider myself being an average user 
You probably didn’t download one of the nighlies, did you? The “official” release hasn’t been the right one to use for over seven months.
In any case, you must be doing something wrong, because Handbrake works – all the time, every time, on any video source you throw at it. But I’d have to come over to your setup to see what’s going on (or you’d have to post the relevent logs on the Handbrake forum). It’s sure to be user error somewhere.
And as for you seeing the differences – nope, that’s definitely impossible. I can promise you you either didn’t encode properly or you didn’t have someone give you a blind test. Trying to look at stills on a monitor won’t cut it – unless that’s the way you always view your movies.
Let’s do this – if you are EVER in the central Florida area (number one tourist destination spot in the U.S.) drop me a PM here and I’ll give you my phone number. You can come over (I’ll even show you around the town) and I’ll run the test on you. You can even pick the movie – I have hundreds of them, and I’ll go to the attic and get out the blu-ray original for you. If you can spot the Handbraked one even SIX times out of 10 I’ll buy you dinner (and, if so, you’ll be the only person out of dozens I’ve done this with. Most give up after two tries and say they are just guessing anyhow, as there IS no difference).
Afterwards we can watch the movie on my 9’ HD screen in the theater room (and I’ll even make the popcorn :>)
I found out that the reason it was crashing was because the source file, that contained an error (although it would have been nice if handbrake just reported that error instead of crashing, like Avidemux did). Anyway, I just recoded a video and I think I see the differences, but since the file sizes are not identical I can’t perform a good blind test (and I know how the brain can trick you into seeing differences where there aren’t any, so the only real way to perform a test like this is doing a true blind test). Ok, going to mess around a bit to setup the perfect blind test and will report later (and also post the video’s I’ve used).
Hehe, sorry to say it Mike but the first test I did, i had 10 out of 10 times good
I’m uploading the testfiles now, I’ll post the links in a bit. Here’s what I did:> I re-encoded a H264 MKV with handbrake to H264 MKV, high profile, RF set to 20. The file size of the original was about 30% bigger (I chose a scene that obviously compressed well), so I padded the shorter files with zero’s, so that the file sizes are absolutely identical. Next, I padded the filenames with 0’s, so that on screen there’s no info at all that could hint which file I was playing. Next, I muted audio (since one file is a few frames shorter, the audio would give it away). Then, you play both files for a few seconds, so the next time you’ll start any of the files it starts with the ‘resume ok-cancel’ dialog. > Now, you put both files in 1 directory,go to the menu and hold the up or down button on your remote pressed, so it starts flipping from one filename to the other. Next, look away from the TV and press enter after a while. Now, the last step is pretty important as well: look NEXT to your TV, so you can see from the corners of your eyes that the movies is started and THEN look at the screen and guess what movie is playing. That last step is needed, because the WDTV will show the first frame for a (half ?) second as a still shot and that still shot shows the differences even much better, so it wouldn’t be fair to use that (we already agreed that still shots are different)
As said, I’m uploading the files now, will post the links once that’s done and also explain (later on) how it was (at least for me) pretty easy to see which file was playing.
I used a 52" SONY XBR9 to test on.
And here’s the recoded version:
http://www.megaupload.com/?d=4ZAIAM5O
I only needed the first 7 seconds in the video’s to find out which one is which.
Did you use 0.9.4 again, or one of the nightly SVN builds? They’ve made changes to the x.264 library a few times since 0.9.4’s release.
TonyPh12345 wrote:
Did you use 0.9.4 again, or one of the nightly SVN builds? They’ve made changes to the x.264 library a few times since 0.9.4’s release.
0.94 (2009112300)
It’s up to you, but you could try one of the later versions…
Hi! It worked fine with the hacked firmware and nfs I’ve also tried with stock firmware and ntfs usb disk. This worked also perfectly fine with avatar m2ts.
ConiKost wrote:
Hi! It worked fine with the hacked firmware and nfs
Glad to hear it worked ! Enjoy.
| Forums | News and Announcements | Register | Help | Forum Guidelines |
Copyright © 2021 Western Digital Technologies, Inc. All rights
reserved.
|
Trademarks
|
Privacy
|
Community Terms of Service
|
Site Terms of Use
|
Copyright
|
Contact WD
|