The reason your svhs looks better is because its at a lower resolution than your 1394. The lower resolution results in fewer noticable artifacts.
your svhs will look better than your composit analog, but nearly the same.
FYI all 1394 data will be compressed whether it orginating from ccd or tape. The data has to comply with DV std in order for your pc to process it.
MovieStuff wrote:
crimsonson wrote:
DV by definition has more chroma bandwidth than S-Video. It can also resolve more lines [a little over 400 lines vs a little over 500].
Agreed. I never said that using SVHS direct resulted in higher resolution. That's why I said "sort of" in answer to Andreas's question. I was relaying that it LOOKS better to me directly off the SVHS prior to compression, which it does.
crimsonson wrote:
So even if you do get signal prior to DV compression, the conversion to S-Video will result in less quality than DV.
Then why does the SVHS out of my miniDV look better than the firewire signal?
A most interesting discussion. Any thoughts on taking a telecine feed from mini dv cam phono outputs directly into a dvd writer with mpeg 2 encoding and capturing directly onto dvd? In my view (and I am only a novice in this area!) there are 2 limiting factors the capture abilities of the camera and the encoding to any intermediate/final data storage used.
Cel
I believe all DV25 camcorder's S-video outputs are merely feeds of the same signal that ends up on tape. However even if they are not the quality is limited to less than the capability of DV. As Crimsonson has said DV is 5:1 compression with true component (4:2:0/4:1:1) encoding, yielding a little over 500 lines of horizontal resolution, whereas S-video is merely y-c or luminance chrominance only, yielding 400 lines rez - in other words it is not even capable of realising the full DV luminance or chrominance signal let alone surpassing it.
I think part of the confusion arises with Roger's original explanation:
"When I take the SVHS out of my miniDV and record it directly via SVHS inputs to my Pinnacle Real Time NLE system (which is not DV codec based) at 13MPS, the results are far superior than taking the firewire out of the same camera to the same NLE system"
This seems to suggest that the reason for the discrepancy in the two images has to do with the digitising method used - which I would suspect is Mpeg based, and the way his particular NLE handles the DV input. I promise you if I take the same material into my NLE - Avid xpress DV - from both S-Video (transcoded via the tape deck) and firewire, DV is by far the hands down winner!
Anonymous wrote:
I think part of the confusion arises with Roger's original explanation...
This seems to suggest that the reason for the discrepancy in the two images has to do with the digitising method used - which I would suspect is Mpeg based, and the way his particular NLE handles the DV input.
Nah. The visual difference was present even when I took the firewire out of the camera to a Canopus convertor that fed directly to a monitor. After looking at my set up more, I realized that the problem is just a crappy camcorder and nothing more. While the DV codec isn't great IMHO, when I feed my Ikegami or Hitachi broadcast cam through the Canopus and convert to firewire, it looks fine.
Still, it is interesting that on some cheap miniDV camcorders, the analog out "looks" better than the DV out, which looks obviously compressed compared so the corresponding analog signal run through the Canopus to a monitor. I suspect that the internal digital conversion of my Panasonic dinky-cam simply isn't on par with the Canopus, even IF you lose 100 lines of resolution coming from the SVHS out.
