Is it possible to create a lossless DVD at all? Not talking about motion pictures, just private videos. The software I have seems to want to compress the video (MPEG2 as far as I can tell) into hideousness even though the movie doesn't fill the entire DVD..
What is the source of the video data? Another file? A camera? A stream? Do you know the format of the source? It sounds like you know that you're starting with a lossless format. Are you familiar with codecs? It might be possible for you to install a lossless codec and then somehow configure your system to use the new codec. Another thought that comes to mind is that you may find a downloadable DVD ripper which gives you more options.
Someone I know made a family video, but they did it in portait mode. So the video is now on it's side. They want it turned 90 degrees, which is no problem. But then I have to deal with a 1080x1920 video file (vertical) that needs to be burned to DVD (720x576 PAL). Everything I've tried resulted in atrocious quality. The original file was 1920x1080 (on its side). I'm using AVS Video Editor and Format Factory. I think the source is a Smart Phone of sorts (MP4 30FPS)
@ Enmos, VSO-software has a program named "ConvertX to DVD" which converts many different 'file' types to (PAL or NTSC) DVD. - find it at : http://www.vso-software.fr/
Google AVC FREE DOWNLOAD for a good program to convert the format and rotate the video. In AVC click Add Video(s) and add the video, select the required output format, click Add Effects, click rotate left or right 90, click OK then click Convert Now to start the process. You will probably have to fine tune the settings through a process of trial and error.
no. the only true lossless format is analog. DVD doesn't use analog technology. although digital isn't "lossless", it can reproduce very high quality recordings depending on the bit rate.
That's not quite right leo. Every analog rendition contains noise, hence there is always some loss of fidelity. The digital sampler tends to be less noisy (principally in the storage process) than analog; and at a sample rate greater than human detection, the thing you are concerned about (quantization noise) is imperceptible. But the term of art "lossless" purely describes the type of compression algorithm involved. One kind (run length encoding / Huffman encoding) is no different in quality of the rendered signal from raw sampled data. It merely takes up less space and bandwidth by replacing long strings of repeated bits with a number indicating the string length. But it doesn't give great compression ratios. The other kind (MPEG and all popular formats) breaks the movie into frames, then breaks each frame in blocks (e.g. 16 x 16 pixels) then computes something like the Fourier Transform on each block (typically the DCT - discrete cosine transform) row by row, then column by column -- e.g. a 2D transform which produces a frequency spectrum (sometimes called "sequency"). Once the spectra for each block are known, it's possible to track spectral variation in each block frame by frame, and only record updates in the blocks which changed. This is one reason low bandwidth video is possible whenever jumpy motion is acceptable. This is quite effective when averaged over the typical movie, which contains many pauses in action, still shots, still backgrounds, etc. So effective that you see huge compression ratios never before achieved when processors were simply too slow to do the heavy number crunching in real time. Let's say there are 100 compression techniques out there. It's just more convenient to call them "lossy" or "lossless" than to remember which ones add noise and which ones don't. I think it's more likely that the rotation itself is producing the distortion. You can notice this even with still shots like JPG. The term of art that comes to mind is aspect ratio. You want to try to preserve the aspect ratio from the get go. Let me outline the procedure as I think it will produce the best results: (1) Determine the aspect ratio of the original: a = 1920 / 1080 = 16:9 (2) Determine the aspect ratio of the target device: a = 720 / 576 = 1.25 --> 4:3 (PAL displays as 4:3) (3) Load the movie in an editor which allows rotation and resizing. Rotate the frame 90 degrees. If it asks you to specify the aspect ratio, try 1.25 (PAL). (4) After rotation the viewable area will be too tall, leaving black bars on either side. If not, then the result will be stretched across the horizontal, looking both weird and crappy. (5) if you want to get rid of the bars, you can crop but it will chop off the top and/or bottom of the frame. (6) select PAL output format and generate a test version. Play it. Same problem? Go back to (3) and this time select 4:3 aspect ratio before choosing PAL, if your editor allows this. (7) if you still get sucky results, keep searching for a better editor. If the ones suggested here don't solve the problem, I would look on sites (probably close to your home) where there are media-philes accustomed to working in PAL. The last idea I had was to convert from your file to lossless format (RLE/Huffman etc) first before you do any more editing. Play it in that format to be sure the quality is acceptable. Then import the lossless format file into the editor as above. Theoretically, this should override any sources of distortion that may be built into the editor. My guess is that this is painfully slow. That at least explains why editors may not include this step. I hope this helps.
yes, correct. i should have said: the only true lossless format is live. next would be analog. digital is more "lossy" than either. edit: it might be a good idea to define what we mean by "lossless" i can think one anything that isn't an exact reproduction of the original. this would include distortion by amplifier circuits for low voltages, this can be eliminated. digital, however, suffers what AI mentioned, namely quantization errors. this can be reduced by increasing the bit rate, but not entirely eliminated. propagation delay (the time it takes for a signal to pass through a gate) is the major reason. if you sample faster than you can process, you drop data points. there are various ways digital sampling can be implemented. one way would be with an up counter. the other would be direct to digital with an R2R network. i would assume video would use the network because of its speed. meh, more ramblings from yours truly.
Enmos, you can also HDMI it out Leopold, digital is MUCH better than analog. There is no longer any comparison. Even for major theatrical release i would shoot digital.
the question is about losses. TV is a typical example. the NTSC standard was implemented to compete with 16mm film, and does so, but it still suffers losses. TV cannot reproduce all colors, but 16mm film can digital is the same. the quality of digital is great, but it still suffers losses
Dude...i a m lpost here. Digital is better...period. Your eyes are digital and your voice is analog. Really. What more can i say? You are wrong? Everyone is wrong? Everyone is like a neophyte? What...what u tell me?
Whooops. I apologize there guys.Blame it on the vodka.Please Register or Log in to view the hidden image! But, i think the question hath been answered. Now, i wonder if there is some merit to what i said. The eys (as in human) are better suited to digital....Or, specifically in the digital realm! and you voice...better suited to analog. I think back to the old SB16 and how hard it was to generate suitable audio, drivers and what not and how....visually just better suited. More of an understanding...an enmeshment. A digital handshake is...idk, evident to me. Like when a hand slips into a glove. mpeg2 is antiquated though.