Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.

Poll

I am...

interested to see an MP3 (~128 kbps) test first and I will contribute
[ 8 ] (11%)
interested to see an MP3 (~128 kbps) test first, but I won't contribute
[ 11 ] (15.1%)
interested to see a multiformat (~48 kbps) test first and I will contribute
[ 35 ] (47.9%)
interested to see a multiformat (~48 kbps) test first, but I won't contribute
[ 16 ] (21.9%)
not interested or I don't care what is tested first
[ 3 ] (4.1%)

Total Members Voted: 88

Topic: Last Listening Test in 2006 (Read 10146 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Last Listening Test in 2006

The lack of enthusiasm for the planned MP3 test at ~128 kbps and the arrival of the new Windows Media Player 11 which is shipped with a new Windows Media Audio encoder makes me wonder what to do next: continue with the MP3 test and conduct the multiformat test at ~48 kbps next year or the other way round.

My main concern is that the MP3 listening test will not have enough participants meaning that I will either have to extend the testing period or cancel the test. Organizing the long-awaited ~48 kbps multiformat test could bring some fresh wind and might turn out to be the better idea. I imagine discussing encoders and settings for about one week, including creating of bitrate tables for all contenders and discussing the choice of samples for about three or four days. If everything runs smoothly, I expect the test to finish in mid-December. After this, all participants can relax, enjoy Christmas and New Year (or whatever) and we could start the MP3 test next year. The preparations then shouldn't take more than two or three days because everything was discussed already.

This poll will be open until tomorrow evening. Based on the results, I will either organize the MP3 test or the multiformat test.

Last Listening Test in 2006

Reply #1
Generally speaking I would be much more interested in a multiformat listening test. But the recent events (new WMA encoders, aoTuV updates) makes my interest even bigger. I could evaluate myself the quality but a collective test would open a wider perspective and give more validity to the result. Furthemore, such test would also be easier to perform (48 kbps whereas 128 kbps is known to be much more difficult to ABX even if competitors aren't strong encoders).
By testing MP3 now we're closing the door for a collective evaluation of WMA, WMAPro and aoTuV beta 5 up to next year. It means: beginning of the test in january and publication of the result somewhere between the 15th and the 30th january. I'd rather see the first results of up-to-date encoders on the end of november... On the other side, I don't see any urgency to test a bunch of old MP3 encoders.

Moreover, the idea of testing MP3 came from the impossibility of starting new multiformats tests including latest WMA encoders. The MP3 listening test was suggested in order to pass the time until WMA final availability. The time has passed, WMA is now available... the main reason of testing different MP3 implementation is gone.

Last Listening Test in 2006

Reply #2
I for my part am very interested in a multiformat test, but not at 48 kbps. For me, using modern codecs like Vorbis and AAC, 96 kbps have turned out being a good compromise between very listenable quality for PC playback and file sizes which are bearable being used in conjunction with today's portable players, especially those which feature the common 512 MB of RAM.

Most people would suggest encoding two different collections, something like -V2 LAME MP3 for PC playback and -q0 Vorbis for portable use; but honestly, why should one have to manage two audio collections if you could also have decent -q2 Vorbis files which work pleasantly for both kinds of use? A listening test close to this bitrate would bring the answer whether it's really an alternative to a second collection for most people. I wouldn't ever want to claim that it is since I'm aware of the fact that there're enough people out there whose hearing's better than mine, which has partially been suffering extremely loud noise during work, causing some noticeable permanent derogation.

Personally I even vote for 80 kbps since I was already using 96 kbps with older Vorbis and AAC encoders. I'd like to see whether there've been some noticeable improvements in these bitrate regions during this year. My own ABX tests with the latest Nero encoder and aoTuV beta 5 couldn't be called very representative and reliable, due to my hearing already having a lot of trouble ABXing many 64 kbps VBR samples.

Last Listening Test in 2006

Reply #3
80 or 96 kbps is not an option now. Decision should be made between VBR MP3 at ~128 kbps (actually it's more like 135 kbps) and VBR multiformat at ~48 kbps. Other tests at other bitrates might follow after.

 

Last Listening Test in 2006

Reply #4
I agree with guruboolez: "By testing MP3 now we're closing the door for a collective evaluation of WMA, WMAPro and aoTuV beta 5 up to next year."

I'm also looking forward to a test of the brand-new AoTuVb5 and WMA10Pro against the yet-untested Nero 1.0.0.2 HE-AAC.

Now that all 3 major contesters are new and rearin' to go, I suggest a test be performed at first opportunity.  I, for one, can hardly wait! 
Copy Restriction, Annulment, & Protection = C.R.A.P. -Supacon

Last Listening Test in 2006

Reply #5
80 or 96 kbps is not an option now. Decision should be made between VBR MP3 at ~128 kbps (actually it's more like 135 kbps) and VBR multiformat at ~48 kbps. Other tests at other bitrates might follow after.

Indeed  testing AoTuVb5, WMA10Pro, and Nero 1.0.0.2 AAC is more interesting than an mp3 test now, but I must agree with Junon: 48kbps is not a realistic bitrate, even with these great encoders. You have to be hard pressed for space if you use it. With my 8GB (flash mem) Nano, I have already space for 3200 4-minutes songs when using 80kbps (using Apples estimate of 2000 4-minute songs @ 128kbps).

Edit: Testing Nero 1.0.0.2 @ 48kb means using HE-AAC, which cuts away a large portion of interested people (including me), as only LC-AAC is supported by iPods.

Last Listening Test in 2006

Reply #6
but I must agree with Junon: 48kbps is not a realistic bitrate, even with these great encoders.

I agree with both of you 


Last Listening Test in 2006

Reply #8
The 48 kbps multiformat test was planned 6 months ago (at least) and was postponed due to Microsoft licensing restrictions.
On one hand I agree that 48 kbps is not the most enjoying bitrate (I guess it's what you meant by 'realistic'); on the other hand I must say that quality could be really decent (for DVDrip, streaming, listening on poor speakers, etc...). I just encoded one track with WMAPro at 48 kbps and quality was anything but bad. And IIRC some people using Vorbis with their DAP already stated on this board that quality is perfectly acceptable.
HE-AAC; WMAPro, Vorbis: we got at least three formats that may offer a quality that was unreachable at such low bitrate one or two years ago.

A 80 or 96 kbps listening test may follow this one (I'm personally in favor of planning multiformats tests at higher bitrate -- 80, 96, 128 kbps -- instead of comparing gogo and fastenc) but it's another debate which should only start next year.

Last Listening Test in 2006

Reply #9
Well I was 18th user who voted for "multiformat test at 48 kbps and will contribute". I will contribute my ears (one is bigger than another  ) for this test.
The results aren't final yet. But it is already enough to begining test, right?

Last Listening Test in 2006

Reply #10
The results aren't final yet. But it is already enough to begining test, right?


From Sebastian's introduction:
This poll will be open until tomorrow evening. Based on the results, I will either organize the MP3 test or the multiformat test.


Last Listening Test in 2006

Reply #12
I know we're not supposed to mention the 80 kbps range anymore, but what was the motivation for testing such low bitrates? I haven't been very active here on HA in the last weeks... Are we that interested in ringtone bitrates? I share the opinion a test in the common DAP bitrates would be far more useful, with all the new developments in aotuv, lame, and nero aac.

Last Listening Test in 2006

Reply #13
I think Guru already mentioned some possible fields of usage, such as DVD ripping, streaming (maybe this is not really ideal because of VBR) and listening on portable phones or in noisy environments. Some developers claim that current 48 kbps encoders are as good as or better than some 64 kbps encoders a few years ago (mp3PRO for instance). Don't worry, other tests will follow. I also want to see if Microsoft's claims regarding the new WMA encoder are true (better than HE-AAC).

Last Listening Test in 2006

Reply #14
From my side (I did vote), I think the multiformat test would be most interesting. As Guru and Sebastian said, there are lots of interesting uses for a bitrate this low.