Very nice and informative board - I have learned alot in a short time.
I like to experiment with different encoders in search of the one that my ears really enjoyed. I found an encoder called PlusV created by a company called Ionific from Finland.

The encoder is a vbr encoder that reads the wav file and creates a plusv stream. The original wave is then sent to lame and is encoded at 22050 hz with an average bit rate of 64 kbs. After both streams are created, they are put together and the resulting file is an .mv3 for mp3 plusv.

I encoded several files from rock to country and electronic. I only used the mp3pro from the free encoder at 64 and cdex with lameplusv as an external encoder. I used 1by1 with both decoders, mp3pro most recent, and in plusv from their website.

To me the .mv3 file was noticably better. My question is has anyone else messed with .mv3 files or are my ears really wacked?? I think this could a rival to mp3pro. The bit rate and file size is small but the sound is big.

Any response or suggestions would be greatly appreciated.

Thank you
xbenchman

PS the website is located here h**p://www.plusv.org/ in case anyone is interested

I reed about it, and found this thead on Xiph.org (the makers of Ogg Vorbis):

Read this entirely if you have further interest (I warn you: is a loooong thread)

so is this plusv thing inspired by the ars technica thread that stated 2 128kbit mp3s == 256 kbits mp3? it just mixes two encodes automatically?

uh? Not really tried, but AFAIK with superficial reading of www.PlusV.org, no, i don't think so...

No, it looks like another form of SBR (spectral band replication) technology, similar to Mp3Pro except that it uses LAME instead of FhG's mp3 encoder. In either case, when a regular mp3 decoder is used, the SBR band (HF content) cannot be decoded.

ah.. interesting. i thought lame was generally considered to be inferior to FhG's implementation at lower-bitrates, however.

You require the PlusV plugin in order for the MP3 stream to be decoded correctly as it has been intended. All the available information is available at the PlusV website. I played around with this format, and I was surprised at how well it performed.

From playing around with it and some of what the forum on the plusv website says, that it is not spectral band replication. nothing is created in the decode process. the first stream created covers a range of bands something like 11k to 19k...don't quote me on this as I am still learning how it works. and this stream is added to the avg bit stream that lame creates. so nothing is created, just decoded from the mv3 file.

Yes the decoder is needed to enjoy the file but, simply rename mv3 to mp3 and it will decode with the normal mp3 decoder. the quality will not be there it be just an ordinary 64kbs file and it will sound like one.

If anyone wants a sample just let me know where to send it.

I was really impressed with it and no I have no connection with them. Just wanted to share something I found. I am somewhat unknowledged on these
tests that are being performed to determine how well the encoder actually performs. If anyone could point in the right direction I am curious to see how the digital ear hears it. My ears like it alot.

Thanks for the comments and let me know if anyone performs any tests

It puts the 2 different encodes together after each one is performed. The plusv layer gets encoded firtst, from the original wav file, then the original wav file is handed off to lame. After both are encoded, the plusv "wrapper" is added to the lame file.

I hope this helped - if not I can try it again

If this stuff was GPL'd or BSD'd then it would be worth investing time in, IMO.

The code is gpl.

How to get the help screen? "plusv.exe -h" doesn't seem to work. I would like to know what options can be used. If I just do " plusv.exe infile.wav outfile.mv3" I get a DOS screen full of strange characters and a mv3 file that doesn't play.

Speek, the exe you want to use for normal encoding is "lamepv.exe". "plusv.exe" generates a downsampled wave file, that needs to be lame encoded, and a separate file containing the plusv data which then has to be interleaved with the lame encode!! And, yes the help doesn't print correctly from "plusv.exe". I have corrected it and can post it all if anyone is interested.

This is the help:-

a little off topic

woulde tecnology like plusV or mp3pro enhance sound quality at
lame --alt-preset 128 ?
or ogg.vorbis at -4.99 ?

these are typical used in my group's video releases

does anybody know what the "maximum bitrate" these technology can enhance.

sven_Bent,
from playing with plusv in dos if you pushed it to a constant bit rate of 80, the interleaver would drop bits because there were too many or they were too high. I don't exactly understand all this but i do know that if the encode rate is too high not all of the bits are kept.

in the forums on the website is a front end for lameplusv. i played with this in the advanced mode to try and learn just what was going on. the frontend sends an average bit rate with the maximum being 72. When I configured cdex 1.50 b7 to use lameplusv as the external encoder i sent the --abr 74 and it encoded all the bits with no loss (none that I could tell the screens flash by quickly). this little tweak of higher gave the high notes more resolution and bass a little more depth (at least that is my perception and why I posted about lameplusv)

as far as mp3pro you can vbr a file with nero but I am not sure how that will turn out exactly

Ogg vorbis is one codec that i hav not messed with in any depth

Thank you for all the input.
Which lame version should I use. Currently I have stuck with v3.92 but from reading in the forum it is recommended that 3.90.2 ???. Which version has the best encoder for mpeg2 layer3 or does it not really matter?? The one thing that cathes me as peculiar is the frequency mp3pro uses 80kbit and plusv uses 16kbit. To me that looks like they are worlds apart. both have stuck wtih mpeg2 layer 3

Any ideas why on is at 16 and the other at 80kbit?

where can i find the frontend ? I am too stupid to encode a testsample

Benjamin Lebsanft,

lame plusv frontend can be found on this page...
http://www.sunpoint.net/~zaha/
A graphical frontend for lamepv, an mp3+v encoder. Just click the highlighted graphical frontend and it should download approx 230kb and a readme is included.

I played with this frontend and it is pretty nice. there are 2 screens a basic screen and an advanced screen. the basic screen takes your wav file and encodes it at a bit rate rate of 64 (standard). the next screen is the advanced screen which allows some modifications as to quality and bit rate.

Just unzip the frontend in the same folder as the encoder from the plusv website. You will also need a version on lame. I put both the dll and exe in just to be sure.
The version of lame I used was 3.92. I am not sure if this version is the best or not...still waiting for someone in the know to specify which version has best results.

Any other problems, just post and I will try to help
Enjoy

wow, if someone would work on this one it could certainly compete with mp3pro !

Lame 3.92 is fine, relax

Dibrom recommends using his 3.90.2. However, any differences between that and 3.92 are probably almost indiscernable.

You only need the '.exe', not the '.dll'.

Benjamin Lebsanft,
Impressive isn't it - went through and reripped several songs in .mv3

Kblood and john33,
thank you both - i really appreciate the input. I am still learning. Just trying to understand how all different versions fit together. I assumed I only needed the exe because that is what the program calls for when the handoff occurs

thank you

PS I am presently using cdex and the external encoder option to rip directly to .mv3 instead of to wave and then to .mv3. It is working but I am not sure I have all the correct parameters input.
this is my parameter string - any comments would be appreciated
-q1 -ms --abr 74 %1 %2
quality level 1 - default is -h or -q2 takes a little longer but the concentration is higher around the avg bit rate set
stereo mode - default is joint stereo - should i stick with stereo or chang to js
average bit rate of 74 - my experimental trials with higher bit rates - seems to give the high notes more clarity and bass more depth
%1 infile
%2 outfile

-q1 is experimental, -q 2 should be used and why pure stereo at such low bitrates ??? Use joint stereo at all bitrates!

impressive thats all, vorbis is much better!

Benjamin Lebsanft,
Thank you for your help...but I disagree on ogg vorbis at the 64 to 80 level.
If you please explain to me how to encode a vorbis file in the 64 to 80 kbs area, I will encode the same song with several encoders - mp3pro plusv and ogg and then listen to them. I have read alot about ogg vorbis but need help to encode at a similar bit rate to compare apples with apples.

I don't disagree ogg vorbis sounds good at the higher bit rates but when I last encoded an ogg file in the 64 80 96 area it sounded less than desirable. I could have done something wrong and am willing to redo it as I don't want to discount any encoder due to operator error.

thank you

you have to use version 1.0 and -q 0 as commandline, it sounds much better than plusv and is comparable to mp3pro at 64 kbps, above 64 kbps (> -q 0 ) it is better than every other format! only mpc is better at very high bitrates!

for an easy start you could use oggdrop:
http://homepage.ntlworld.com/jfe1205/oggdropXPd.zip

Ogg Vorbis @ ~ 80kbps -q1, Stereo, 44100Hz
Ogg Vorbis @ ~ 64kbps -q-1, Stereo, 44100Hz
Ogg Vorbis @ ~ 40kbps -q-1, Stereo, 32000Hz
Ogg Vorbis @ ~ 32kbps -q-1, Mono , 44100Hz
Ogg Vorbis @ ~ 30kbps -q-1, Stereo, 22050Hz
Ogg Vorbis @ ~ 24kbps -q-1, Stereo, 16000Hz
Ogg Vorbis @ ~ 24kbps -q-1, Mono , 32000Hz
Ogg Vorbis @ ~ 16kbps -q-1, Stereo, 11025Hz
Ogg Vorbis @ ~ 16kbps -q-1, Mono , 22050Hz
Ogg Vorbis @ ~ 12kbps -q-1, Mono , 11025Hz
Ogg Vorbis @ ~ 8kbps -q-1, Mono , 8000Hz

The quality settings can only be used in VBR mode and VBR does make it a little difficult to get an average bitrate. So the displayed bitrate is the nominal bitrate.

MINz

Yes, but is patented as well. It seems that it is in conflict with the GPL.

Read the thread i linked before to find more...
Edit: I mean this thread...
Read this entirely if you have further interest

RC55: ...you too, as well, for the same reason...

A search in this forum will give you the answer many times over.

-q 0 ~ 64kbps
-q 1 ~ 80kbps
-q 2 ~ 96kbps
-q 3 ~ 112kbps

You can also use 0.5 to get something ~ 72kbps, for example

On the quality of Ogg Vorbis compared with other codecs at low bitrates:

http://ff123.net/64test/results.html
http://www.heise.de/newsticker/data/anw-08.09.02-000/

Ogg may sound less than desirable at 64kbps, but have you really compared? Coz using something else, odds are you find that something else less than less than desirable.

My fault on the settings - only been here 2 days.

the first article showed "Mp3pro wins overall ranking points, but it's not significantly better than oggq0 or ogg64." But my question now is how do they compare to the plusv file?? I am in the process of doing my own 'ear tests' - too much industrial noise so there is some hearing loss I am dealing with. I really do not care what codec I use as long as my ears think it is good. I tried ogg out a while back (months several of them) and was somewhat disappointed. My disappointment was with the low bit rate - at higher bit rates (I used a 6 in cdex) it sounded like the cd. Hence my search for a low bit rate good sounding (not necessarily perfect) encoder. I want to get a portable player (mp3 ogg mp3pro) which ever. The encoded music must sound good to my ears. If the latest updates to ogg (July 2002) make it the better encoder to my ears then I will search for a portable ogg player if not then it will most likely be a pro player.

Thank you for the input

The code beeing gpl and implementing a patended algo doesn't comflict.

This is because there is an automatic and explicit patent license for software release under gpl by the plusV patent owner (VLSI).

PlusV is "old news" and has been discussed before here at HA. See these threads:

http://www.hydrogenaudio.org/forums/index....b3468c9bb8e3237

http://www.hydrogenaudio.org/forums/index....b3468c9bb8e3237

http://www.hydrogenaudio.org/forums/index....b3468c9bb8e3237

http://www.hydrogenaudio.org/forums/index....b3468c9bb8e3237