Forum Replies Created
-
AuthorPosts
-
JesseGMember
I don’t think I have it anymore, but I made a custom Lame for going 32kHz joint stereo @ 64kbps that was actually not half bad, and we used it at radioio for some years. It still wasn’t magic, but it was acceptable compared to our 64kbps WMA8 streams at the time. (WMA10 uses SBR and PS so it just destroys mp3 for low bitrate now)
Joint Stereo does really work well though, so it’s not like two 32kbps streams exactly. In fact if you run pure mono through it, you’ll get the exact same quality as a mono stream at the same bitrate.
Overall though, I would have to suggest not even offering an mp3 stream under 96kbps at the lowest. There isn’t even a point to having a 64kbps stream anyways, because it doesn’t work over a modem. Even 32kbps is pushing it really.
What I would do is get rid of any mp3 streams below 128kbps, and then offer at least one PS-AAC stream at 24kbps for modem listeners. Then optionally a second PS-AAC stream at 48kbps. The quality of either of these totally annihilates a joint stereo mp2 at 64kbps, even what I remember about my customized Lame. Sad but true.
JesseGMemberSince 3.94 the focus has been on improving VBR, since that’s really the preferred way to encode an mp3. Unfortunately it doesn’t jive well with streaming, due to most players not having the ability to properly buffer it.
I did extensive blind ABX testing to find that 3.93.1 seems to sound the closest to the original input, at 128kbps cbr. This was done when 3.97 was out, so it’s possible that 3.98.x would beat 3.93.1, but I highly doubt it. Plenty of internal things that they have optimized for vbr have effected the cbr negatively, even to the point where they introduced readily apparent constant audio "bugs" for cbr (even at 320) that later had to be fixed.
Unfortunately "in which ways" isn’t a question that can be answered by me without an extensive analysis of 3.93.1 vs 3.98.4 at a specific bitrate and settings. But again, I would be surprised to find out it sounds anywhere near as transparent as 3.93.1 @ the settings in this topic.
JesseGMemberTrueRTA is highly inaccurate, especially at lower frequencies. There are some simple tests to start off with.
Does an RTA only raise one band for a sine wave at the center of the band?
TrueRTA: fail
BreakawayRTA: passDoes and RTA only raise two bands for a sine wave directly between two bands?
TrueRTA: fail
BreakawayRTA: passIs that level it’s showing -6dB lower:
TrueRTA: n/a
BreakawayRTA: passJesseGMemberYou won’t hear this at FreakDay, and not in all of its "HD" glory, so here it is now.
PFD MIX Final (wave) – Passive Aggressor DAB.flac
I’m still blown away by the consistency of loudness and solidness of spectrum this preset has, despite wide open dynamics. Again, this clip’s processing is ONLY the dynamics core of Breakaway, and also does not have any of the developments since the last public versions.
JesseGMemberlibaac.dll should not be used. (does it actually do he profile now?)
enc_aacplus.dll from Winamp’s plugins folder is the latest encoder from Coding Technologies, and what you should be using.
JesseGMemberPost updated to show that it’s a quote of Leif. Also… I’ve fixed the link to lame 🙂
JesseGMemberyou don’t. you install it into winamp folder, and then copy the files, as suggested in Leif’s original post.
JesseGMemberthe installer checks for winamp.exe in the currently selected directory.
JesseGMemberthere are quite a few winamp plugin loaders for VSTs.
that being said… the plan has been to use VSTs directly in Breakaway for a while now. namely because they allow for floating point audio and there’s basically no change of additional clipping. Winamp API does allow for floating point audio now (v5.14 of the SDK and later, if my memory serves me right), but most Winamp plugins don’t make use of that.
JesseGMemberThere will be a version of Breakaway (possibly called "Home Theater") for 5.1 audio upcoming, with speaker calibration as part of it. But the real awesome thing I think that will blow away people is the built in UPMAX II algorithm, the industry standard up-mixer as used on almost every large tv show/event there is. Including the SuperBowl for 6 years running, which btw had 111 million viewers last weekend. The most watched TV show of all time, running Leif’s audio processing. HUAAH!!
Look for that some time summer-ish (northern hemisphere).
JesseGMemberthose are over-runs and under-runs of the buffers.
JesseGMemberit’s also been listed here since i started it:
viewtopic.php?f=5&t=942JesseGMemberThe latest versions of everything are also updated in a thread named for its purpose, and are not just stickied but announced (higher than stickied) in every forum on the board.
I could have just said "i don’t know", that much was obvious yes? But thought it would be a good idea to point you to the search function as well, since some people really have never used a forum before, and this is their first.
JesseGMemberLeif does say over a bunch of posts up to #5978 what he is adding, including in #5978 itself which is where .05 was linked originally.
Just wondering… *why* you want to know? Is there a problem that was introduced or fixed in either version? That seems more important information than us telling you to search/read, and you suggesting you have and didn’t find information I did in 3 minutes.
JesseGMember[quote author=”silverfox”][quote author=”silverfox”]Can someone explain what is the difference between BDJ 0.09.04 alpha & 0.90.5 alpha assuming there is no RC or change log mentioned that I can see. 0.90.5 alpha is shown as the latest download in the Store..[/quote]
HELLO ANY ONE HOME ?[/quote]
I don’t think anyone knows. So.. if you search this forum, and find out, would you let us know?
-
AuthorPosts