Anyone addicted to burning cd's, recording tapes,...

Post Reply New Topic

...creating mixes, compiling playlists ect...?

Seems like whenever I run out of cd-r's I sure just make a playlist in foobar2000 and bounce it down to a single 80 min image wave file (sort of like a virtual cd).

I like playing around with digital audio, converting my music to rare audio formats like musepack/ogg/wavpack, mixing them with each other, doing all sorts of random stuff with them.

Anyone with a similar fetish?

i like toying with audio also. i have a basketful of apps [soundforge, pristine sounds 2000, sound laundry, cool edit pro, dcart, dartproXP, munoise, virtos noise wizard, PSP audio tools, et al] in addition to a CEDAR DCX declicker [very handy for old phonograph recordings] that i use to restore and enhance various of my voluminous collection of flat friends and old audio tapes. what started me on this route is the fact that i never could stand phonographic crackles and rumble, they were like a big cloud of grotzl covering up the music, and i determined to do all i could to eliminate that nasty aural cloud. i also like digitally reconstructed stereo versions of monophonic originals, i wish i had more skill at doing such myself but i'm in the middle of the learning curve. what i find really fascinating, is that some people are now taking old acoustically recorded [IOW musicians shouting and blasting in front of a huge recording horn which excites a needle in a soft spinning wax disc] 78 rpm recordings, which with modern technology can be rendered into something which sounds very modern. the only thing they can't update is the vocals but they are working on modern digital vocoder technology to do that also.

No, but I do experiment with different equalizer settings from time to time.

I had a 4-5 year obsession once with video encoding (and a little of audio encoding, too). I amassed a collection of (video) CD's and DVD's that I recently got rid of (because H.264 is much easier on my eyes/brain than XVID and all my stuff was encoded pre-H.264). I even got about 80% finished with a homebrew mp3-style encoder, but alas my brain was/is not up to hard-core programming like that.

I have no idea how many hours I spent tweaking settings, encoding, and looking at the results, over and over, but it has to be a lot.

what is the secret [aside from brute force bits or limited rez] to minimizing block and mosquito noise in low bitrate video?

pristine sounds 2005 will let you make any response curve you can imagine, as it has a graphic [curves drawn by mouse pencil] parametric eq of high resolution, with as many individual bands as you can draw. it also has a 3D EQ function that i still have not mastered, but in the hands of somebody who knows what they're doing, it can do incredible things.

It's been a while -- if I'm remembering correctly that the problem with mosquito noise is analogous to ringing in jpegs or any Fourier-transform-with-HF-data-tossed-out, then you can put a low-pass filter on the source (XVID did/does that by default). But that's not very satisfying in the gut since it is a degradation of the original signal.

-- But there was an output 'sharpening filter' that I remember using with windows that was pretty good in conjunction with LP filtering before encoding (or even after decoding, i.e. LP de-blocking filter). I can't remember the name of it, though.

Oh also, I recall that some encoders let you tell it to preferentially allocate more bits to the HF data, but of course too much of that can starve the LF data of bits so there needs to be a good balance, depending on the qualities of the input.

I haven't dug into H.264, but I have wondered if it approaches the problem with something like a video version of SBR in AAC+.

if you starve the LF bits, does that give you coarse color contours ["banding"] or does it give large blocking distortion? just curious.

Ack, I actually can't directly remember but I think it would tend to cause both.

does the compressed video format allow you to conserve bits by limiting overall frame rate in slow-moving sections of video? like if something is just crawling across the screen in one dimension, one could get away with perhaps half the frame rate of anything more complex. how 'bout limiting resolution in visually coarse parts of the frame, is that allowable? does standard DVD encoding do this automatically? this is fascinating.

XVID had/has a variable frame rate which is a good idea in principle but it bugs the hell out of me to have to look at it. (I think it's related to being able to see CRT's and florescent lights and such flicker whereas most people can't.) MPEG-2 (DVD) and below don't use VFR AFAIK. As far as variable block size, MPEG-2 doesn't, but H.264 does (it's apparently pretty computationally expensive to encode that way, so maybe older processors weren't up to it).

But yeah, it is amazing what people with big brains can come up with -- I remember being blown away by MP3 when I realized that 90% of the original signal was missing and it still sounded ok! (on a casual listen, at least)

i'm old enough to remember the analog way of "conserving bits" or at least of conserving real estate on magnetic tape or phonographic vinyl- nakamichi had an excellent cassette deck that would play linear [-20 db] out to 15Kilocycles at 15/16" inches per second tape speed, which was 1/2 of standard compact cassette speed [1&7/8" inches per second], with .08 WRMS wow and flutter. you could put 4 hours of decent audio on one 120 minute cassette, comparable in every way to older open reel decks. there were also minicassette/microcassette decks with similarly souped-up specs. radio shack sold an arthur fiedler/boston pops LP which had 90 minutes of music on it, accomplished by rolling off bass below 100 cycles and rolling off trebles above 10K, but it sounded surprisingly easy on the ears despite this. but other than those two salient examples, it's awfully hard to get good clean fullrange audio onto a compressed analog format. fisher/price "pixelvision" was a toy camcorder that used a high-bias [chromium dioxide, specifically] audio cassette tape [at almost 17 inches per second tape speed] to encode a data-reduced monochrome video signal onto the left channel [the right channel was sparkling full bandwidth audio, due to the high tape speed] which was only 120X90pixels @ 15 frames per second. surviving specimens are very popular among independant filmmakers for some strange reason, as the video quality was very hard on the eyes. but it was pretty much state of the art [in data compression terms] back in 1987.

^^^ MP3 actually also rolls off at 16KHz and below 10Hz for bandwidth reasons.

I just googled that -- pretty cool, especially for 1987. I recall reading somewhere that modern VCR's use a rotating head at an angle -- is that done to in order to allow for a higher bandwidth at a lower tape speed?

at 128 kb/s- higher bitrates have wider bandwidth- for example, the 256 kb/s rate that amazon/com uses for its cloud servers/digital downloads, has 17kilocycle bandwidth, and the 320 kb/s that was the original sony minidisc bitrate, goes out to 19 kilocycles. one of my apps, "Pristine Sounds 2000" has a spectrographic graphic tablet app in it that graphically shows the bandwidth of whatever audio i'm working on, so that is how i found that stuff out.

yep. helical scanning with a quickly scanning video head on a rotating drum, provides high head-to-tape speeds necessary for encoding high-bandwidth video, especially important at VHS tape speeds of 1.31 inches per second. the video info is laid down in the form of diagonal stripes on the tape, @ 60 fields per 1.31 inches of tape, per second. the audio is laid down as an FM signal at a different physical level of the tape, also in diagonal stripes which the VCR electronics "stitch" together with a buffer circuit. as a backup, there is a longitudinal pair of stripes on the side of the tape for a low-fidelity analog soundtrack. before ampex discovered this little expedient, the only other way to encode video was to run tape at speeds in excess of 100 inches per second, which ate up tape and video heads in a hurry.