Ac­tu­al­ly, I pre­fer Ap­ple’s Quick­Time ov­er eith­er Win­dows Me­dia Play­er or Real­Me­dia. The rea­sons are pret­ty tech­nic­al and get in­to the sheer power for mul­ti­me­dia in the var­i­ous plat­forms, and Quick­Time is just by far the most pow­er­ful.

But I re­al­ize that the main point of in­ter­est to this Top­ic is in the qual­i­ty of the au­dio co­decs. In this re­gard, Win­dows Me­dia 7 (and the forth­com­ing 8) are in fact quite good, bet­ter than Real 8 for the most part. But they’re both up against Q-De­sign’s Q-Mu­sic co­dec, which has been in­cluded in va­ri­ous ver­sions in Quick­Time since ver­sion 3 (now, Quick­Time 5 in­cludes Q-Mu­sic 3, the la­test and great­est ver­sion of this co­dec). The ol­der Ver­sion 4.2 of the Win­dows Me­dia En­co­der (re­leased around the same time as Win­dows Me­dia Play­er 6.3/6.4) in­cluded some co­decs that Win­dows Media 7 and 8 have dropped, and I found them the best of all: Vox­Ware’s mu­sic and speech co­decs. I used their mu­sic codec, via Win­dows Me­dia En­co­der 4.2, to pro­duce the .ASF ver­sion of the demo I did for the “Search for the Ul­tim­ate Sax” Top­ic in this For­um last week. Of all the codecs I had tried for en­co­ding a short VL trom­bone sam­ple I had made as a demo of Dig­it­al Ear’s out­stand­ing au­dio-to-MIDI tech­nol­o­gy, only Vox­Ware had a high-qual­ity mon­o­phon­ic co­dec (both the Win­dows and Real co­decs seemed to as­sume that one would only want high qual­ity if one also want­ed ster­eo, and since I was doing a mon­o­phon­ic sample, why should I waste half my band­width on two chan­nels when I only need­ed one?), and only Vox­Ware would let me get to a rea­son­ab­ly low samp­ling rate (as it turns out, about 64k/sec for 16-bit mono) without hav­ing a weird “wind chimes” type dis­tor­tion ar­tif­act­ing show up on the breathy at­tacks of the VL trom­bone sound. I had to go to at least 96k/sec. to get rid of that using any other co­dec, or else could not get rid of it at all, in­clu­ding with my pre­vi­ous fa­vo­rite, Ya­ma­ha’s own Sound­VQ (a li­censed ver­sion of Twin­VQ, the first of these “su­per-co­decs” that are at least twice as good as .MP3 in com­pres­sion at the same qual­i­ty, or qual­i­ty at the same com­pres­sion — Twin­VQ pre-dates Real G2, Win­dows Me­dia 6, Vox­Ware, etc. by sev­er­al years at least!). I later learned that the “wind chimes” ar­tif­act­ing was part­ly caused by too high sound lev­els in my source au­dio — such co­decs need extra “head­room” with which to work their mag­ic ef­fec­tive­ly, so one should not do the usual prac­tice of nor­mal­iz­ing the gain to 100% on audio in­tend­ed to be com­pressed with mod­ern co­decs. Nor­mal­ize to about 90% in­stead. Some­time I will have to com­pare them again us­ing the same source clip but nor­mal­ized to 90% to el­im­in­ate that cause of ar­tif­act­ing, and see which co­dec wins out.

I do like how ev­en Ver­sion 6.4 of Win­dows Me­dia Play­er will au­to­mat­ic­al­ly down­load the co­decs need­ed to play a clip en­co­ded in Win­dows Me­dia 7 or ev­en 8 (so no need to set­tle for an ol­der co­dec to av­oid lock­ing out users with ol­der play­ers, nor do you need to down­load the la­test and great­est play­er just to be sure you can play all the new­est con­tent). Real also has a sim­il­ar ab­il­i­ty now. I like Win­dows Me­dia Play­er 7’s SRS Wow! (a sort of ster­eo-wide) and TruBass (a Bass en­han­cer that fools the ear into think­ing it’s hear­ing bet­ter bass than the speak­ers are phys­ic­al­ly ca­pab­le of re­pro­du­cing [mere­ly up­ping the bass gain, like typ­ic­al Loud­ness or Bass Boost sys­tems do, just re­sults in dis­tor­tion when the bass cap­ac­i­ty of the speak­ers is ex­ceed­ed], using psy­cho­ac­ous­tics) tech­nol­o­gies, which make just about an­y­thing sound bet­ter.