Frame loss - Mpeg capturing

[ardor]

Hi,

I am trying to convert analogue VHS video casettes to digital MPEG2 video in order to make a DVD. Unfortunately I have a frames dropping problem.
If I capture on a low resolution (350..) I have almost no frame-loss. But if I'm capturing on a higher resolution I have a frame-loss of 50% and even more sometime.
I have searched the internet to find a solution to this problem. I have changed some hardware like changing the harddisk's channels etc... I have a separate disc dedicated to capture.
The disks are defragmented and nothing else is running on it. When capture is in progress I close down all applications including Norton. I allow no preview of the picture. I have tried to capture with different software Pinnacle Studio 8 + 9, Windows Movie maker, Ulead products, VirtualDub, Windows Media Encoder, iuVCR along with Huffyuv compression etc...
But nothing helps, the problem of 50% frame-loss remains. This is very frustrating...
I think maybe there is a hardware device malfunctioning but which one? I am desparate.

Who can help me, please? Any help is welcome.

Thanks a lot.

Configuration:
Win XP Home SP 1
CPU: 1.7 gigahertz Intel Pentium 4 with Asus motherboard.
RAM: 256 MB
Display: NVIDIA GeForce
Asus TV/tuner capture card

[jggimi]

From your description, it sounds like you are capturing direct to MPEG-2. When capturing to MPEG-2 with a standard, low-cost capture card, that encoding is done by the CPU during capture.

To reduce drops, you can lower your resolution, as you've discovered. Less pixels per frame is less work for the CPU to do. Of course, you end up with less image that way. In the same fashion, you can lower the bitrate, which may provide some additional CPU resource for encoding on-the-fly.

When I was first doing analog capture, I did the same thing ... I captured to MPEG-2 directly. I discovered that if I set the GOP to 100% I-frames, that consumed the minimum resources. Unfortunately, I could not eliminate frame drops.

You might prefer capturing to .avi with a lossless or nearly lossless codec first. Then, you convert to MPEG-2 after the capture has completed. It's what most of us with inexpensive cards do, and is the method outlined in the Capture Guide.