The great Dreambox recording experiment

Here's a log file of a longer recording I did overnight. There are some errors in it? Maybe someone can say what the errors might be. I briefly looked at the recording and don't see any glitches on screen.
 
Ok I will run them later after work sorry got mixed up as bit did not read thread correctly
 
Also attached a pic of the mod I used to guide me - I could photo mine but it's exactly the same. Oops its 94K zipped - I can't post it. If I can upload it to the files section I will

I used Sanyo Os_Con caps but 3.3uF 25V but I'm sure any electrolytic cap will do (with + & - leads). Down at this value the leads are about the right distance apart to match the SMD cap already on the board. The pic gives the correct for + & -.

What I did was to cut leads down to about half inch long - dip them in flux and put a light coat of solder on the ends (tinning). Then dip the ends again in flux (you want the lead tip to attach quickly to the SMD cap). Then just position one cap lead on the end of the SMD cap and touch the lead with the soldering iron for a couple of secs - this should be enough to solder it.

Then solder the other lead to SMD cap - I used a thin jewellers srewdriver to push this lead against the SMD cap (Gently) while soldering to ensure contact was made. I just checked soldering by gently trying to move cap with my finger. First cap was solid - second one I forgot to dip in flux after tinning and it wouldn't take to the SMD cap - kept coming loose when I did the finger test.

Be careful when doing the finger test - you don't want to put too much sideways pressure on the cap - remember it's soldered to a SMD cap which has little contact with the underlying trace and therfore the point of contact is small & easy to detach the trace, I would imagine

Hope this helps - maybe others will post their results and/or techniques
 
Last edited by a moderator:
how about hosting them here: imageshack.us

and then copy and paste the link via the Picture Icon at the top
 
dweeb4 said:
Here's a log file of a longer recording I did overnight. There are some errors in it? Maybe someone can say what the errors might be. I briefly looked at the recording and don't see any glitches on screen.
Click to expand...
Thanks Dweeb. I hate to have to rain on your parade, as I had hoped this would have worked, but I'm afraid the mod hasn't solved the problem. You've still got 102 errors in 42 minutes, which is lower than some of the worst recordings I have seen, but still much too high, and I've seen better from unmodded boxes. Interestingly enough, the errors don't appear until about 7 minutes in, but then they come thick and fast.
 
Ah, I was a bit premature posting this in the downloads section then - oops

I will check the soldering to make sure it's good

OH DEAR
 
Nice one Kidder - I just tried another recording and just as many errors - still ooks good on screen though. I haven't checked the soldering yet but will do so tonight.

I fell into the age old trap of wanting it to be better after doing a mod and rushed to conclusion.
 
dweeb4 said:
Nice one Kidder - I just tried another recording and just as many errors - still ooks good on screen though. I haven't checked the soldering yet but will do so tonight.

I fell into the age old trap of wanting it to be better after doing a mod and rushed to conclusion.
Click to expand...
It's an easy trap to fall into, though, so don't feel too bad. ;)

I started looking into the recording problems some time ago, and I was initially encouraged when people I knew said they had perfect recordings, but after a quick check with MPEG2repair, it turned out their recordings were full of errors.

For many people, I think one area of confusion has been that the recordings might often have seemed ok(ish) on the PC, but caused playback problems on the box, leading some to think network problems were to blame.

Depending on what software and codec you use, the PC is generally fairly indiscriminate in its playback, and will just plow through the file, and if the bytecount on the specific errors isn't too high, it often won't be that noticeable, and any minor image corruption will by many people be put down to "poor image quality", which is a concept of digital recordings that seems to be greatly misunderstood, because all things being even, there aren't really any quality variations in digital recordings the way we're used to with analog formats like VHS or even DVD recorders.

On the Dreambox, however, the file has to conform to certain broadcast standards for correct playback, and depending on the size and position of the errors, problems can range from audio sync, macroblock corruption and dropped frames to outright lock-up.

If PT-1's Dream guy's comments about the demuxer not being powerful enough were true, and there's certainly no reason to think otherwise, no amount of network mods is going to help. I don't know if it's possible to switch the demuxer chip, but it certainly sounds like a considerably bigger ordeal than adding an extra capacitator. What might be worth looking into, though, is how the Neutrino images process/demux the stream. Neutrino TS files have slightly different specs than Enigma ones, and my own experience is that Neutrino images produce far less errors than Enigma images. Last night I recorded just over an hours worth of tennis from Sky Sports, and only had 11 errors with Cryo's Neutrino 0.9b umage. I recorded the exact same channel on my Dbox as well, just to rule out errors in the original stream, as has been happening on some channels recently, and I had one of my friends record on his Dreambox running an Enigma image too. My Dbox produced 0 errors, and his Enigma image coughed up 138. Perhaps the Neutrino method puts less of a strain on the demux chip?
 
If PT-1's Dream guy's comments about the demuxer not being powerful enough were true
Click to expand...

He is on IRC on the Dbox2Gui Channel ;-)

and he is only the one half of the Enigma Inventors ;-)
 
pt-1 said:
He is on IRC on the Dbox2Gui Channel ;-)

and he is only the one half of the Enigma Inventors ;-)
Click to expand...
So you're saying there's a 50% chance he's right? :)

Kidding, of course. Like I said, I have no reason to question your sources.
 
I tried testing a recording last night but everytime i tried to feed mpegrepair a ts file i had recorded it scans it ok and then comes up with an error box at the end. I cannot recall exactly what it said now. something about choosing another channel possibly. Anyway mpegrepairs options were all greyed out so i could not do anything else other than try and load another ts file. I dont think it was finding the PID's.
I am in work just now so will need to have another look at it when i get home.

Any suggestions as to why its not working. I am not getting any kind of log or anything. Dont know if its relevant but the file only seems to load to 99%.
 
johnbhoy said:
I tried testing a recording last night but everytime i tried to feed mpegrepair a ts file i had recorded it scans it ok and then comes up with an error box at the end. I cannot recall exactly what it said now. something about choosing another channel possibly. Anyway mpegrepairs options were all greyed out so i could not do anything else other than try and load another ts file. I dont think it was finding the PID's.
I am in work just now so will need to have another look at it when i get home.

Any suggestions as to why its not working. I am not getting any kind of log or anything. Dont know if its relevant but the file only seems to load to 99%.
Click to expand...
If you're recording from an Enigma image using the Direct Record method, you should just be able to select another audio or program track if the default scan doesn't work.

If you're using Neutrino or server based recordings, the recordings will not be correct TS specs, and MPEG2repair won't be able to read them, until you have run them through ProjectX and converted them to TS files that are on spec.
 
I'm going to try a Nuetrino image on my box as well, I assume they have same functions as the DW image I'm currently using i.e. timed recordings from epg ?

What do you think is currently the best Neutrino image ?
 
I've been reading this thread with great interest and as a relative newbie, can I ask a simple/obvious question?

Does a tool such as MPEG2Repair or VideoReDo fix the files so that they can play with little or no obvious playback problems? Or is the file so screwed up that it's still pretty bad after "fixing" them?
 
Don't know what others have found but the one I repaired wouldn't play at all.

Frenchman
 
simonk1969 said:
I'm going to try a Nuetrino image on my box as well, I assume they have same functions as the DW image I'm currently using i.e. timed recordings from epg ?

What do you think is currently the best Neutrino image ?
Click to expand...
The Neutrino Dream 0.9b is very good, and seems to be the most popular one.

grim00 said:
I've been reading this thread with great interest and as a relative newbie, can I ask a simple/obvious question?

Does a tool such as MPEG2Repair or VideoReDo fix the files so that they can play with little or no obvious playback problems? Or is the file so screwed up that it's still pretty bad after "fixing" them?
Click to expand...
MPEG2repair was created specifically to repair files so they could play back correctly. VideoRedo just fixes minor timestamp gaps and that sort of thing, I don't think it repairs actual errors, unless they've upgraded it recently.
 
You must log in or register to reply here.

Similar threads

O
    • Like
BT and Sky users may finally get ‘turbo-charged’ broadband boost - check your postcode
Replies
3
Views
421
leemoo
L
janobi
    • Like
Awoo installer
Replies
1
Views
478
spud1966
spud1966
alimac
    • Haha
    • Like
ohoh
Replies
0
Views
223
alimac
alimac
O
    • Like
Virgin Media offer discount to price of bundles
Replies
0
Views
400
Oggiman
O
D 8 RCS
    • Like
Have you done any Car Modding ?
Replies
6
Views
461
cokeaddict
cokeaddict
Back
Top