source lenth ?
This is a 5 chapter DVD recruitment video for a local college with each video being 3-5 minutes.
content type?
Not sure what you mean but they are primarily interviews (not necessarily a lot of movement when the pixelation happens) shot with a Canon XL1S, captured via firewire.
encoder settings? My last was using the (batch render) Main Concept MPEG2 customized to DVDA vieo stream 2 pass, 8 max, 7 avg, and 6 minimum but it renders random (different spots every time) pixelation.
source type and codec ? mini DV...I used vidcap to capture (is that what you mean?)
Thanks very much,
Randy
Thanks B_JM, the .jpg was grabbed in V5 from the many different Main Concept mpegs (all settings , high and low and in between). The AVI when sitting on the timeline does not have the pixelation. My deadline is tomorrow so I had to go ahead and start burning (and hope that the client doesn't complain) but I'd like to learn what the heck is going on. When I get a chance I will post a .png of the mpeg and the .avi (I no longer have the mini-dv tape).
Pelladon, is that MS keyboard/mouse drivers you're talking about?
Thanks all,
Randy
I asked for the version number, if the intellipoint software was installed.
The problem was that XP SP2 installs intellipoint ver 5 by default. There were problems with ver 5 that not only affected mouse performance, but even caused problems not related to the mouse. Video problems for instance.
Try installing the updated version 5.2 drivers. If that doesn't help, uninstall it and download the older 4.12 drivers, see if that makes a difference.
Although I've been considering installing SP2 recently, I'm still using SP1.
When I do update I'll be sure to get the 5.2 drivers though.
Thanks pelladon,
Randy
Thanks Bob,
So the captured tape (avi) could feasibly play flawlessly on the timeline but not be able to (play flawlessly) after encoding to mpeg? That would make sense because as I recall one of them happened on the same clip every time.
I would have recaptured but I no longer had that particular tape ( long story) but I thought I was safe since everything played great on the timeline (during the edit) and also afterward when rendered to AVI.
What did I learn? Make sure to always keep the source tape until at least 30 days after delivery ....okay for anyone still wondering why I no longer had access to the tape.... here's that long (and embarrassing) story: I usually order 10-20 tapes at a time but l let myself run out and used the source tape in a "pinch" (I know, I know...)
BTW Bob, please verify that I understood your statement correctly...thanks again!
Thanks Bob and everyone else for your help!
Randy
Good video should encode way better than that. Problem is there can be things wrong with video that are not that easy to see. Tape dropouts may only last one frame, you need a decent sized monitor and to keep staring at it for a long time to see them. From what I've seen they are not handled well by mpeg-2 encoders, I guess 2 pass and high bitrates should handle them reasonably well.
Thing to do just to be certain is to drop the encoded mpeg file onto the same T/L as the source and see if you can find anything wrong in the matching frame of the source. It might be only one frame long so you have to look carefully.
Bob.
I guess 2 pass and high bitrates should handle them reasonably well.
Not in this case. Thing to do just to be certain is to drop the encoded mpeg file onto the same T/L as the source and see if you can find anything wrong in the matching frame of the source. It might be only one frame long so you have to look carefully.
That's exactly what I did Bob, I highlighted the (avi) clip (about 8 seconds) and rendered several settings all the way up to the highest (9,800,000 CBR). I then put the mpeg above the avi perfectly lined up. The pixelation was there until I muted the upper track (with the mpeg) to reveal the avi. why i suggested you post proper screen shots of both source and target - to avoid a lot of guess work
Which is what I was going to do B_JM until I realized I no longer had the tape ( I know, I know...)
I thought you meant from the recaptured tape. I didn't think you meant from the already captured avi. I should have gotten back with you to let you know I no longer had the tape and you would have probably told me the already captured avi would work...saving a lot of "guess work"...my bad, sorry.
Anyway, if there's anyone that hasn't lost their patience with me I guess could post the before and afters if you still want to see them but I can assure you that the exact frame of the avi is fine (but I may not have time until Friday).
But back to Bob's theory; I have the JVC (SR "pro" model) deck that has the bad reputation and did notice pixelation upon capture at times. If I had a problem with footage I was going to need, I would recapture (it's a random problem and never does it in exactly the same frame).
My question is, could there be "weak" spots on a frame that look fine on the timeline until having to render them? If this is possible then I guess my only solution would be to capture directly from my camera (XL1S). It's a bit of a pain but certainly worth it if these "weak" spots can go unoticed. Of course the smart thing would be to keep the tape for recapture (I know, I know...)
Thanks for your patience,
Randy
Having a look at the captured frame(s) would be very useful, after all that's what the encoder is trying to encode!
However the trap with mpeg-2 is it doesn't encode each frame as an entity, it's effectively encoding the difference between frames. If you've got a sequence of frames with every pixel changing then it's going to have big problems and give a pretty aweful output. Sorry that's not a very technically precise description, there's plenty on the web that's more revealing, but hopefully good enough for you to grasp the issues.
So when you've got a lot of noise or blocks of pixels doing wierd things due to tape dropouts then the encoder can have problems. If you've been having issues with the deck I'd certainly start by suspecting what's going into the encoder as being the cause of the problem. If you really cannot see anything wrong with the AVI source why not try running some known good video through the encoder?
When I've got wierd things going on I resort to using generated media as a test signal, I know it's clean and should encode very easily. Just moving a grid pattern slowly around the frame and then encode that, see how it looks.
Bob.
Having a look at the captured frame(s) would be very useful, after all that's what the encoder is trying to encode!
This morning I have created a full size .png of the avi from the timeline but apparently my website will not accept .png because I kept getting errors; so I created a full size .jpg and it uploaded but it is showing the pixelation!
So, at your request I'll send the full size .png (w/ no pixelation (487kb)) along to you if you're interested. If you really cannot see anything wrong with the AVI source why not try running some known good video through the encoder?
I'll try that whenever I get a chance although this is the first time I've ever had this problem and it only happened in 3 frames out of 14 total minutes.
Thanks,
Randy
Apparently the site is not dial-up friendly with 3 different tries, after only about 15-20 seconds, I got "Problem: Unknown Error -1."
Can I not e-mail it to you?
Thanks Brian,
Randy
I'm thinking you're right Bob...which is actually a relief in that my problem is solved as long as I don't capture using my PIECE OF CRAP JVC decks!
Thanks,
Randy