The Real 10 Platform

[bond]

it can definitely be that aac is more complex than mp3
the aac guys again claim that vorbis is even more complex than aac etc...
hm i am surely not the right one to ask, what format needs what processor power on what devices aso :D

hm about he-aac i also guess that compared to aac its harder to decode, but he-aac is also used by the digital radio mondial system, which maybe also had "small devices" as recievers in mind...

well whatever, as long as it decodes on my poor p3 866mhz :D

[kamiller42]

Originally posted by bond
before claiming that the test could be biased better read how it was done
Thanks for the information. I barely have time to keep up with this thread much less keep up with every detail of the latest in music compression formats and tests. Your direction helps.

A blind test does help eliminate bias a lot, but I am not sure it can eliminate all bias. If I listen to mp3's all the time, I can probably better pick out an mp3 in a line up than another format because my ears are tuned to listening to mp3s. This "picking out" may not be intentional either. It may be simply subliminal preference.

Despite all of Pepsi's intentions, I still don't think Pepsi tastes better than Coke despite their blind taste tests. I can find the Coke in a Pepsi challenge. :)

I guess I was looking for something a little more precise than a blind listening test. I mean, we can take an image, copy it, and then pass both images through an analysis program and see how they differ. It would seem the copy with the lowest delta would be best.

[kamiller42]

Originally posted by karl_lillevold
kamiller42:

I don't have a system with RealOne V2, so I am not 100% sure I remember this feature. However, it seems in RealPlayer 10, you have both hyperlinks and pulldown-menus to navigate back up the library tree. See this snapshot:

Great!! I am glad to see the feature is still there. The application looks the same in a lot of ways, but I'm still getting used to the subtle differences. Thanks for pointing it out to me.

I would recommend AAC at 192 kbps, which I think there is wide agreement is very close to transparancy, and sounds better than HE-AAC at 96 kbps. This will play fine on devices that support AAC. I know of the iPod right now, others are probably soon arriving. Since RealPlayer 10 currently rips to .ra, you will have to go through some trans-muxing steps to convert to a device friendly format, until RealPlayer 10 Gold ships. See my FAQ in the first post.
That paragraph alone is worth its weight in gold. I've been going back and forth trying to find my "next generation" audio format, and now you've helped me answer that question. :)

You mentioned you're ripping to AAC 192kbps. I assume you're talking about ripping and encoding with the RealPlayer or encoder in .ra format. I'm ready to rip to 192 AAC, but I would like to avoid any rework if things are different when Real 10 Gold releases. Then again, I am not sure how or if things will be radically different when Gold releases. I guess I'll have to wait.

[bond]

Originally posted by kamiller42
A blind test does help eliminate bias a lot, but I am not sure it can eliminate all bias. If I listen to mp3's all the time, I can probably better pick out an mp3 in a line up than another format because my ears are tuned to listening to mp3s. This "picking out" may not be intentional either. It may be simply subliminal preferenceif you have the interest and are fast there is atm such a blind listening test running on mp3 @ 128kbps :)
http://www.rjamorim.com/test/mp3-128/presentation.html

have a look at it but hurry up, today is the last day!

[karl_lillevold]

Originally posted by kamiller42
You mentioned you're ripping to AAC 192kbps. I assume you're talking about ripping and encoding with the RealPlayer or encoder in.ra format. I'm ready to rip to 192 AAC, but I would like to avoid any rework if things are different when Real 10 Gold releases.
Actually, since RealPlayer 10 is still Beta, and to avoid any extra work converting files later, by finding a complex solution to a automate the procedure now :) I am currently using an alternative solution to encode my CDs: Exact Audio Copy (EAC) with its command line encoder interface to do one of two:
1) EAC calls the RealProducer command line, which outputs AAC in .ra, then automatically and losslessly convert those files to .mkv, then too .aac, finally converting .aac to .m4a, and add tags. I described this earlier in the thread, but as you can see, there are many steps involved. Easier is
2) as discussed on hydrogenaudio, EAC calls command line nencode.exe, an interface to the Nero AAC encoder. The output is .m4a directly, and all that is needed now is to add tags.
Neither is all that easy to get started with, especially 1), but works well enough once up and running.

RealPlayer does include a "Convert Media" function which when Gold ships, can probably be used to convert losslessly from your .RA files to .M4A. However, since RealPlayer 10 Gold functionality in this respect is not yet certain and since it is Beta, I would probably recommend waiting. It should not be too long.

[bond]

thanks for your efforts karl :)

i guess the most easiest way to rip and encode to real mp4/m4a files is using itunes!
it includes one of the best (independantly prooven) aac codecs, is available for free and outputs real mp4/m4a files right away

it should be the best/easiest solution for newbies and people who want quick solutions

for other freaks, which maybe like he-aac too, the two ways you described should work best

[Sirber]

maybe someone could write a MP4 outout plugin for producer, like the MKV one :)

[SeeMoreDigital]

Hi everyone,

I've just generated some RealAudio 10 encodes at different bitrates using RealPlayer 10 beta.

Can you please refresh my memory/confirm which codec and AAC complexity was used at following bitrates: -

32kbps
64kbps
96kbps
128kbps
160kbps
192kbps
256kbps
320kbps

Sorry to be a pain!

Cheers

[iwod]

Few points to made across........

1. Yes it is a fact that vobris is more complex than AAC. It was never designed with that in mind. AAC ( hi, bond? a open standard ^_^ ) has huge backing from the industry which means optimazation is faster.

2. Even though HE AAC is complex, by the time these tiny devices are widely avalible to play them, there processing power has gone up and decoder would have been better. Not to mention the new Paramatric Stero mode which suppose to increase quality at low bitrate ~ 24kbps.

[SeeMoreDigital]

Further to my previous post, the encodes sound a bit distorted!

Has anybody else reported this?

Cheers

[karl_lillevold]

SeeMoreDigital: 32 and 64 are cook. 96 and above AAC. Note that you can use D-C's useful RM analyzer/shell extension (http://forum.doom9.org/showthread.php?s=&threadid=55392) and get all the details when right clicking on any RealMedia file (rename to .rm first, D-C did not know about .ra when he wrote his tool).

Re distortion: no, not to my knowledge. I have been playing these files in many different players, and/or after various conversions to other container formats, and not noticed any problem. What kind of distortion, and which of the codec flavors? There are a number of steps you can do to isolate the problem : change player, change decoder, change format (to mkv, then aac).

[SeeMoreDigital]

I've been playing the RealMedia .ra encodes using my Xcard (yep it can play these files too!). I noticed the disortion when comparing the rips ones I made using dbPowerAMP.

I've posted some links to the files here: -

Created_using_RealMedia_AAC@96kbps.ra (http://82.2.167.24/Uploaded_Files/Created_using_RealMedia_AAC@96kbps.ra)
Created_using_dbPowerAmp_AAC@96kbps.mp4 (http://82.2.167.24/Uploaded_Files/Created_using_dbPowerAmp_AAC@96kbps.mp4)

It might well be just an Xcard thing.

Cheers

[karl_lillevold]

Thanks. I think it is the Xcard. Both your samples sound fine here, RA played in RealPlayer and MPC (w 3ivx LC-AAC decoder), MP4 played in RealPlayer, MPC, and foobar2000.

[SeeMoreDigital]

Originally posted by karl_lillevold
Thanks. I think it is the Xcard. Both your samples sound fine here, RA played in RealPlayer and MPC (w 3ivx LC-AAC decoder), MP4 played in RealPlayer, MPC, and foobar2000. That's good to know.

I'll pass this info on to the guys at JovePlayer. After all, if it was'nt for them, I would not be able to watch all my wonderful .rmvb encodes on my TV via RGB/component video!

Thanks again

[phrentec]

For some reason I'm having problem with playing back a realaudio file. Can someone please give me the various names of the sipr dll. Thanks.

[Shinobu]

try to install my decoding pack, it's in frensh but easy to understand and it can decode lots of vidéo file included all real file, mkv, mp4, mpeg4, vp6 .... it's a pack of filter so it won't distrub your codec install ++

http://satsuki.yatoshi.free.fr/
download =>

Satsuki.Decoder.Pack Version 1.0.1.9 (4.29Mo)
give it a try, i think it's a very usefull pack, i might trad it in english someday ^^

++

[karl_lillevold]

File sharing now works on the new helixcommunity.org, but it will take some time to get all the previous binary releases in place. In the meantime, I have made available the DLLs that fix the problem playing back HE-AAC (racp) in RealPlayer 10 Beta:

HE-AAC (racp) fix for RealPlayer 10 Beta (https://helixcommunity.org/project/showfiles.php?group_id=50)

If you have not already signed up, you have sign up to gain access to this. If you have signed up in the past, but not yet accessed the new site, you need to reset your password (https://helixcommunity.org/account/lostpw.php).

I have updated the FAQ in the beginning of this thread.

[DanielSun]

What is the difference between raac-2 and raac-7 ?:confused:

And could somebody kind enough to give a complete list of cook,raac and racp profiles?

like this:

------------------------------------------------------------
64 Kbps Stereo Music - RealAudio 10 racp 0 44.1 kHz
96 Kbps Stereo Music - RealAudio 10 racp 1 44.1 kHz
128 Kbps Stereo Music - RealAudio 10 racp 2 44.1 kHz
------------------------------------------------------------

:rolleyes: :rolleyes: :rolleyes:

[hellfred]

Originally posted by DanielSun
What is the difference between raac-2 and raac-7 ?:confused:

And could somebody kind enough to give a complete list of cook,raac and racp profiles?

like this:

------------------------------------------------------------
64 Kbps Stereo Music - RealAudio 10 racp 0 44.1 kHz
96 Kbps Stereo Music - RealAudio 10 racp 1 44.1 kHz
128 Kbps Stereo Music - RealAudio 10 racp 2 44.1 kHz
------------------------------------------------------------

:rolleyes: :rolleyes: :rolleyes:

The full list of codec flavors can be seen in the documentation that e.g.came with Helix Producer. Have a look at producer/docs/docs/AudienceFile.htm#Audio_Codec_Tables.
The documentation is available at Helixcomnuity, too, but i do not have the URL right away. But i am sure one can make its way through that side.

Hellfred

[Shinobu]

Loock at the documention of the audiance in producer, you'll find all the audio descrition you search

++

[DanielSun]

Originally posted by Shinobu
Loock at the documention of the audiance in producer, you'll find all the audio descrition you search

++

I got it !
https://producerapps.helixcommunity.org/cmdproducer/docs/AudienceFile.htm#Audio_Codec_Tables (RealProducer® 10.0 Audience File Reference)

[dckc]

Originally posted by karl_lillevold

64 Kbps Stereo Music - RealAudio 10 racp 0 44.1 kHz
96 Kbps Stereo Music - RealAudio 10 racp 1 44.1 kHz
128 Kbps Stereo Music - RealAudio 10 racp 2 44.1 kHz



Hi karl_lillevold,

I have tried encoding anime clips using the latest Helix DNA Producer (10.0.0.545) with the aac addon. The codec flavor that I have tried are racc 0, racc 1, racp 0 and racp 1.

However, when playing the resulting rmvb files in Real Player 10 (6.0.12.872), the "clip source" shows that the sampling rate is 22050 Khz for racp 0 and 1, and 32000Khz for racc 0.

Stream: 1 Audio Stream
MIME type: audio/x-pn-realaudio
Max Stream Bit Rate: 64.0 Kbps
Audio Codec: (RealAudio) 22050 Khz

Is this only a bug in Real Player 10 that a wrong value is shown? Or the audio of rmvb clips are actually at 22050 Hz? Or 22050 Khz?
How can I verify the sampling rate of the clips?


Regards,
Daniel

[Shinobu]

the racp is in 22050 hz upsampled to 44100 while playing (like mp3pro).
i'm not sur realplyer fully support racp.
try to play your clip with mpc + coreaac
++

[dckc]

Originally posted by Shinobu
the racp is in 22050 hz upsampled to 44100 while playing (like mp3pro).
i'm not sur realplyer fully support racp.
try to play your clip with mpc + coreaac
++

Thanks for your reply. I have searched this forum using the keyword "mp3pro" and found much more information on this issue. Thanks a lot.

:)

Edit:
I have tried playing the anime clips using MPC with coreaac and mpc's internal aac decoder. The AAC+SBR and upsampling 22050 -> 44100hz info can be found in the coreaac property sheet. Moreover, the audio quality of racp 0/1 (just by my ears :)) when playback in RP10 is about the same as in MPC. So I think RP10 now fully supports HE-AAC playback. Thanks again for your help.

Another question:
Can I assume that racp is better than raac at 96Kpbs, and racp is not useful at 128Kbps? Is this just a subjective opinion and cannot be compared at a technical point of view?

[Shinobu]

Just test with your ears raac@128 vs racp@128, if the 2 are for you equivalents, use raac (lest cpu usage while encoding & decoding)


++

[Rash]

I think the fact Helix is developing RV10 toghether with this community is very appreciated.

Congrats Karl. ;)

[EDIT] Oh, sorry for posting this so late, hehe. :) I've been quite busy.

[karl_lillevold]

Better late than never :) thank you!

[lunaticmoon]

Could Helix Team take measures against hybrid material more and more ?

At the present time, correspondence is WMV9's interlace maintain mode only.
if RV10 deal with hybrid material, I can recommend RV10 for anime freaks in my community.
but most people of them are using WMV9 now :(

[karl_lillevold]

lunaticmoon: Please see my reply in the other thread (http://forum.doom9.org/showthread.php?s=&threadid=78948), where I asked you to explain the problem in other words. Producer's auto de-interlace and IVTC filters are designed to take care of video consisting of both interlaced and telecine'd material mixed together. It can also encode VFR without any problems. This is why the problem is not clear to me, and I need to understand it better, and what hybrid material means. Thanks.

[iradic]

sorry if this has been answered before...

what is difference between 'helixproducer_rmrepair_10_windows.zip' and 'helixproducer_10_windows.zip' at Helix DNA Producer Apps (https://helixcommunity.org/project/showfiles.php?group_id=50) ?

thanks

[Rash]

Can I make a stupid off topic question?

Don't you guys at Real have any interest on competing with WM9 for the HD-DVD format? I think it is a lot more mature than WM9 since there has been a massive development on it. ;)

[Sirber]

Originally posted by jcsston
Here the first public release of such a plugin.
helix-mkvwriter-v0.0.2 (http://corecodec.org/project/showfiles.php?group_id=15&release_id=122)
I have tested it with AAC audio, RealVideo 10, cook audio, and sipr audio. (I couldn't get producer to encode any HE-AAC)

Of course being able to mux RealVideo and RealAudio into Matroska with this plugin is just a coincidence ;) Seems this don,t exist anymore :(

[jcsston]

Yes, corecodec has been down and the new server has only been online. Hopefully we'll have the old projects back in soon, so the link will work again.
Dunno if I still have a binary of this on my hd. :o

[Sirber]

If you have, can you mail it to me please? thanks :D

[jcsston]

http://mkvwriter.corecodec.org/mkvwriter-v0.0.3.zip

[karl_lillevold]

I edited the post (http://forum.doom9.org/showthread.php?s=&postid=447903#post447903) I wrote about mkv-writer in the Containers forum, to reflect the new download location. Thanks again, jcsston!

an old question by iradic (sorry for the delay):
what is difference between 'helixproducer_rmrepair_10_windows.zip' and 'helixproducer_10_windows.zip'
the rmrepair version can read RM files, will either re-code (trans-code) or trans-script (re-package). The RM files can be normal, or broken, that is the result of a crash during the encoding, such that they have not been finalized.

[Sagittaire]

Possible to choose quant interval (I,P,B) for second pass encoding in next version of RV10E ... ?

[karl_lillevold]

Sagittaire: I am afraid I don't quite understand your question.. Could you try to clarify?

EDIT: If what you mean is like set quant 1-3 for I, 2-12 for P, 3-14 for B, then I would have to look into it further, but perhaps that could be added, if someone would find it useful.

[WorBry]

Hi,

Although a longstanding "DivX man", I'd heard so many good reports about RV10 that I thought I'd give it a look. First tried to install the 'freebee' basic RealProcuder but I couldnt open up the program -exe. generated errors. So tried AutoRV10 instead and I'm very impressed with the results. I have several (naive) questions:

1. Are there any standalone DVD-MPEG4 players that can playback RV10 video. If not, is this something that could feasibly be introduced as a firmware upgrade or does it involve a totally different decoding technology (like I said, naive).
2. I'm always on the lookout for alternative ways to play videos on my pocket pc (iPAQ3970). I'd actually removed the Mobile RealOne Player that came with it, mainly because its usefulness seemed to be limited to wireless streaming of the occasional emdedded realmedia clip, which at current GPRS rates in my locality is a very costly luxury. Even so, I thought I would chance my luck and encoded some test clips with AutoRV10 at various resolutions (320x240, 240x180, 320x176) and bitrates (180 - 250) with and without realaudio at 44kbits/s, 48KHz stereo. None would play on the mobile player. Is there any prospect of a RealMedia player for pocketpc that is able to play RV9/10?

Thanks.

[Sagittaire]

Originally posted by karl_lillevold
If what you mean is like set quant 1-3 for I, 2-12 for P, 3-14 for B, then I would have to look into it further, but perhaps that could be added, if someone would find it useful.

Yes ... exactly ... but not necessary for each frame type

rcMinQuant and rcMaxQuant for exemple

[Sirber]

If we are about to have it, to specify by each frame type would be the better option. What if I want I to be around 1-2 an P 2-3?

[ganeshpetkar]

Hi all,
i m doing real video decoder10.
in order 2 get license what r the qualities it has 2 satisfy with standard decoder & also snd link for standard input test vectors.
please reply ASAP.

Regards
Ganesh

[karl_lillevold]

The Helix Community (https://helixcommunity.org/) is a much better place to ask questions related to RealVideo decoder porting. You should already be registered there if you have the decoder source code. Here is the link to the RealAudio and RealVideo Technology Compatibility Kit (https://rarvcode-tck.helixcommunity.org/), which includes test vectors.

[CruNcher]

karl anything new ? ;)

[Rash]

Wow! Helix is up and running! That is great! :D It was a long time since I last heard about it.