samplehunter

So let's resume: There are units which store all the data with encryption in one single file There's virtually no way to "log on" to the unit to bypass SS and get the decrypted data directly from disc This all was done due to DRM reasons. Now SS3.4 allows you to upload any Track from the disc, even already transferred tracks, unlimited times. Well, sony, wouldnt it have been cheaper without the whole DRM crap at all? Why not simply storing the files as FILES on the disc (unencrypted)? This would even make disc backups possible and would enable Mac and even Linux users to use ALL features of the HiMD. they would just need the codecs. Ok, the Hardware was built before the great change of Sonys DRM Policies So, lets think about firmware upgrades via usb... But thanks although for finally understanding the users wishes.

If you have a NetMD recorder, it won't even have the code in the firmware to upload atrac data over usb. If you have a HiMD unit, it MIGHT be possible to hack it, that it transfers a track "accidentally" but the resulting file will be useless if it is a SP track since the Atrac1 (SP) codec does not exist in software. There is not one single SP Codec available. the initial Atrac algorithm seems to be one of the most protected secrets at sony. I think, they used a few patents from others for this algorithm and have only a licence to use it in hardware. the later algos (Atrac3/3+) might be completely invented by sony and therefore they can deliver software codecs for this. So a transfer of a LP2/4 Track MIGHT be possible. But sony wants to sell their new HiMD units...

i think mp3 frames are normally filled up with zeros or junk data to match sector size better or to be more compatible when used as live stream. But it's not necessary. some mp3 encoders allow you to save just the raw stream without these fillings. i suppose that if you upload a transferred mp3, transfer it to the unit and back to the pc the size will not decrease further. Otherwise they must drop some of the actual mpeg data and then the signal would not be identical anymore. btw are there any audio demos of the "bad" mp3 playback quality available here?

I think it's possible to re-upload the mp3s from the unit to the pc, right? If not, you couldn't haave done crc checks on both versions :-) ok, to check if the data itself differs you can do the following: decompress 1.mp3 to 1.WAV transfer it to the unit and back to the pc as 2.mp3 check, if 1.mp3 and 2.mp3 have exactly the same bytesize decompress 2.mp3 to 2.WAV in an audio editor you can now substract one file from the other. (invert the phase of one file and mix them) If they are identical, there will be nothing left but silence. else you'll get (hear) the difference. This would be unlikely because then the file must have been transcoded which costs cpu power. IF the decompressed files are identical, the mp3s may differ only in the kind of padding the data or different flag bits like "original" or "pre emphasis". mp3s can store the frames either with no padding at all or with iso compatible or cd-rom compatible padding. You can also try: transfer a mp3 with ID3v1 or v2 tag. does it have the tag after transfer back? transfer one without any tags. give it a title on the md. does it have a tag after transfer back?

This "Conversion" to pcm isn't really one if you use uncompressed WAV files. There is a Header at the beginning which is lost at DL and reconstructed at UL. Try saving a WAV with additional Chunks with META-Data or Sampler Infos like Loop-Points or Cuelists etc. You can do this with e.g. CoolEdit. If you transfer the File and get it back to convert to wav again, I'm sure you will lose the Meta-Data and therefor get a different Hash. Maybe mp3s are re-padded, or the ID3 Tags are stripped out before transfer. So the file might be physically different after back transfer even if they contain the same mpeg stream.

I have a nh900 which was euro capped too. I first set it also to the Hongkong Setting (25) but got problems in sonic stage with titling. I officially had an Hongkong MD unit with a european disc inserted (because the disc label and the track titles were set before the mod) and a US-Sonicstage. Then I set it to the US Settings (21 for nh900) without these Problems. I recommend the US settings because it seems to use the same character set as the european setting. The Hongkong setting seems to enable Kanji input which causes the non-Japanese Sonic Stage to "protect" the titles. But due to annoying Menu and Stop Button Problems i may have to "re-europize" the unit for the Support...

how did you get this odd "line numbers" or sectors? this seems to be the beginning of the visible "partition" of the HiMD. You can read it just with a disk editor like dskprobe. This is logical sector 0. Or have you managed to read physical sectors? Then the range 0-00D35D4 would be quite interesting...

You're right, it will take ages (at least with the defrag tool from win2k). So here's my way to defrag a HiMD / MD in HiMD mode: (Works at best with disks containing only music. with additional PC Files it will take longer) -Start defrag, but press "check" instead of "defrag". You will see if defragmentation is necessary. -Move the ATDATAxx.HMA (the big file in HIMDHIFI) from MD to the harddisk -if you have additional files on the MD, it's a good idea to move them also. -defrag the remaining files (They are small, it should take only a few minutes) -move the additional files back to the disk (if there were some) -At last, move the ATDATAxx.HMA from the Harddisk back to the MD. The file will now be on the MD in one single Fragment and hopefully at the end of the disk. Defragmentation is usually caused when you erase a track in the middle. CAUTION: If the big file is once moved to the HD, don't disconnect the unit until you have moved or copied it back to MD. Otherwise the unit tries to read the MD in "music mode", will fail to find it's HMA file and will probably ask you to format the Disk or will generate new keys, which makes the file unusable. If you do it exactly this way, the disk should be playable anymore. I've tested it with a not so important disk and it worked very well for me. It even works if the File is bigger than half the size of the disk. In this case the defrag tool won't even touch the file. So at last we CAN get a defragged version of our recordings saving battery power. Something we missed sometimes often the old MDs.

i have just tried but unfortunately this allows me not to batch-convert (which is faster than convert>wait for unit>transfer>wait for unit>convert>...) and then transfer a bunch of files. if i do batch convert before a transfer, the oma files will remain on disk. then i have to right click every file, go to properties, select the atrac file and press the delete button - for EVERY f**ing single file. The old menu was much more intelligent here. It was good, so why did they remove this? instead we get an annoying, un-turn off-able jingle after each transfer.

... which is quite silly because the same content on a CD could be easily transferred. another problem is that there is no legacy atrac codec on the pc. I think sony has never written a software codec for this. there was a modificated MDH-10 data drive which was able to grab and to put atrac data on a md but it cost about $10000 (the mod with the software). It was meant for rather professional use...

Some annoying changes were done in SS 3.2 I have no option to delete only the oma converted file anymore! I had SS 2.2 before and used to import wav or mp3 files into an album and converted it at a bunch and then transferred it to my nh900. After that I deleted this .oma-crap from the pc. don't want to store everything twice. I marked the entries in the album and pressed delete which brought up a nice menu where you could choose to delete the entries only or to delete also the pysical files. If you checked "delete files from disk" you had the choice to delete converted files (omg/oma) or to delete the original file too. With SS 3.2 you have only the choice to remove only the title from database or to delete "the file". As we all know, sony has much fun in deleting something and therefore "the file" means ALL versions of the file including the original. Is there any way to change this behaviour to the old style? I want to delete the converted files after transferring because i have no use for them but logically I want to behold my original files. The second thing which is quite annoying is this sound which is played after a transfer and there seems to be no option in the preferences to turn it off.

try to play a HiMD via usb in sonicstage when connected to the pc. if there are no dropouts then, the problem could be the remote cable or the headphones. if there are dropouts too, it will most likely be the unit. I luckily hadn't this problem yet with my nh900. It seems that sony made far better units in the past. my mz-r30 doesn't have one single software or hardware problem but the firmware of the new models seem to be programmed quick and dirty...

Hmm, could it be the reason, that Sony Music was bought by BMG ? Or is it the new Sony Boss? They seem to cancel more and more of their silly restrictions. at last they will release firmware upgrades for the first gen Hi-MDs

It depends on the signal (and if you encode it in your unit or with the pc i think) when i got my nh900 i wanted to test all possible codecs, so I've downloaded these EBU Test-Signals somewhere linked at minidisc.org . These EBU Tests are instrument and speech samples with which you can test a codec e.g. as a developer. The speech sample sounded VERY different to the original in Hi-SP with the PC. It sound somewhat grainy or so. It even sounded better in Hi-LP. the heaviest thing seems to be the Harpsichord sample. It was clearly different from the source with all codecs (Hi-SP/LP, LP2, LP4) i've tested. But on the other side I recorded a Concert via Line-In in Hi-SP and this recording had an excellent quality. But these test signals were carefully sorted to stress a lossy audio codec as much as possible and the Hi-SP encoding in the unit seems a bit better than the codec on the pc.

This was the first thing we all have tried i think. The files are changed indeed. Bit this doesn't really help. Just do the following: record some audio, make an image of the disk, transfer the file, restore the old image to the disk. It's now unplayable! There are kind of two partitions on the disk. One for the user data which is FAT formatted and shown as a removable drive. And [mystic sounds here] THE OTHER! The "dark side" of your HiMD. It's where the Content keys and the storage keys and other system data is stored. This part is not accessible from the pc. this part is maintained by the unit only i think. you can request a session key and a content key to encrypt atrac or pcm data to send it to the unit. So if you transfer something to the pc, there is something changed in the protected area and a flag in the file. If you restore the old image of the user area the checksums/keys/whatever wont match anymore and it cannot be played. I think it works as follows: You have to log in to the unit and request new keys which are used to decrypt the data from the usb. i think it's the unit that marks these tracks as transferred. So the second time the function will fail. the only thing you can do unlimited times is requesting to stream a track as pcm in realtime. this is encrypted too, but as it is not the original file there is a virtual content key generated which enables Sonicstage to "play" the HiMD via the soundcard. So we first have to do authentification to the unit: -Request a key or passphrase from the unit. -send the right answer to this phrase to the unit. -the unit recognizes you as a valid application and sends a session key. This would be the most difficult part. You'll have to find out where and how the authentification process is done in sonicstage.(Any success with debugging yet?) all further keys transferred from and to the unit are encrypted with this session key. To read a file, the unit sends the encrypted content key which you can use to decrypt the data. the 1:1 transfer of a file would still be possible only once. alternatively you can request (with another command) a pcm stream of a track. you get the valid content key for THIS stream, not for the file. If the track was recorded in pcm this doesn't matter, because it will be a "pcm->pcm" conversion which is identical. an atrac file would be first decoded to pcm. Unfortunately this is done only in realtime i think. If this is wrong and faster speeds are possible this would be interesting nevertheless. Otherwise Total recorder would be much easier to use. BTW: is there a freeware alternative to Total recorder?

Yes, right. This seems to be a quite valid source. OK, so it's AES 128bit. Now that we know the "enemy", do we know any weaknesses? Any chances for brute force in a time shorter than a human life? Or could it be done with more sophisticated methods?

I just have found this document: http://www.sony.net/Products/SC-HP/cx_news/vol20/pdf/tw.pdf describing the encryption technology of the MG Memorystick. I think it would be similar to the HiMD encryption. From other sites i've read i guess that the algorithm could be a kind of DES. Hope that helps. (Sony can be proud... the DVD encryption was cracked after a few months and this "Open"MG thing goes about since 2002)

if you own a HiMD unit and can borrow a MD unit with optical out, you can connect the opt. out of the deck with the opt. in of your HiMD and then do a pcm-recording on a 1GB disk. You are making a 1:1 copy of the PCM output from the MD deck. Next, you can transfer the tracks from the 1GB disk to your pc via usb. As you were recording from the digital in, the wav-converter will not work but himdrenderer would do it without problems. The result are "drm-free" wav-files from which you can burn a cd or edit in your favourite audio software. The only thing you can't do is cloning atrac encoded data directly to another MD as the digital out is always PCM, but the quality loss from an atrac->pcm->atrac chain is very marginal at the first generations. Take a look at the minidisc.org faq where some guy has made a test over 100 generations of recompressing atrac with audio samples on the site. I would say the first 20 generations are quite unhearable.

I thought via digital in, trackmaks are only set if there is a track change at the source. But as i recorded from a harddisk recorder (studio equipment) i noticed that there are trackmarks at the silent parts too. So if i set this auto timemark to the highest value, will this turn off the silence detection? would be great. And what's about recording from CD with silent parts via spdif. Will there be additional trackmarks at the silent parts too? (never tried)

Hi there, I have formatted an old MD in HiMD mode with my NH900. with my MZ-R30 the disc will show these tracks labelled "HiMD Disk". this is normal. In the faq it's said that the first track contains silence. (actually it's a 1khz sine at about -12dB for me) The frightening thing is when you try to erase the disk in the R30. It's just impossible! I would understand that you cannot delete a single track solely. But if i want to execute an "Erase All" i normally know what I'm doing. The only thing the R30 shows is "TrPROTECT". So if this behaviour is the same on all old MD equipment you could be a bad guy and render someones MD blanks useless for him. Only you, with your sophisticated HiMD unit can allow him to store music again on his MD blanks (and only if he really begs you for it :-) ) So what do you think about this "feature"?

My dream of a hiMD would be: -recording in all modes at the unit (not only A3+/PCM on HiMD) -customizable bitrate (not only 256 OR 64kbit) -using a portion of anti-shock memory alternatively as storage for custom expansions (other codecs, like ogg, wma, real or effect plugins like reverb, delay, loudness maximizer) would be funny to route these effects from line in to line out. -Oh, a true line out with SPDIF out was nice, too. -UNLIMITED TRANSFERS from self recorded material to the pc! -instead of crappy software with high restrictions (DRM) it would be nice to have a simple ATRAC ACM or DirectX Codec at the PC/Mac/Whatever to encode and decode atrac files on the pc and transfer them as simple files from and to the unit. -Backlit display on the portables would be nice. (Already done on some 2nd gen models) -And finally: We want home decks!

The low contrast of the display is caused due to the semi-transparent metallic surface of this platic part. I had to polish the display of my nh900 because there were some scratches on it. after polishing i noticed, that this metallic thing was also polished away at the display. the positive effect is, that the display is now far brighter (about 30-40% i guess). So for field recordings it's more usable now.

on the unit side, you can always play every disc. the whole security thing is related to the transfers to a pc. you can transfer some audio to the himd, some other should also be able to transfer some audio to the disc from his pc, send you the disc back and you will be able to play the whole disc on your unit. But transfers FROM the unit to the pc are heavily restricted. The only thing that could be transferred to the pc are recordings made with the unit and this only once per track. So if you have made a recording with the unit, then give the disc to someone, he would be able to transfer this track to his pc. If he gives the disc back to you, the track will be marked on the disc as being transferred already and you won't be able anymore to transfer it to your pc.

I also don't think they do a simple xor. (that was too easy :-) ) But if you put in a well defined pcm signal (e.g. bunch of zeros, don't know if it's eqal to dig. silence) then you will get kond of pseude random noise in the encrypted file on the disc. And maybe there is a chance to recognize a pattern or the "noise" will somewhere repeat (an indication for the keylength maybe). Then reformat the disc and copy the same file (the silence) again and look at the "noise". if it is different you could repeat this a few times and look at the differences between the encrypted versions. As a next step it could be tested if there is a difference between recording zeros via SPDIF in and via SS. If the "noise" is really always the same (this would be almost too good to be true) the next step would be recording a bunch of 0001x, 0101x or FFFFx or some other word value and then a 32236 words long ramp from 0000 to FFFF and so on... I think there must be a kind of repeatness in the encrypted zeros as the algorithm may have only a limited number of keys. The algorithm may also have a limited complexity as the decoding in the unit has to be done in realtime and it surely has to be not too expensive. Maybe Sony also decided to use an encryption algorithm that is free for comercial use rather than spending huge licence fees. Anyway, I think it's just worth another try and the cracking of an algorithm would surely be easier if you exactly know the plain data you put in.

just another idea: if you record a long period of digital silence in pcm with the unit and look at the encrypted HMA file, along with the other infos you've got. Wouldn't this help to learn about the encryption? if you know every byte of the soure stream (which is possible with pcm) you probably get the key...