USB CD/DVDW/R external drives

I’ve been experimenting with a cheap external DVDRW (Samsung SE-208DB/TSRS, TSSTcorpCDDVDW) and getting mixed results. This is using the most recent beta of CDVDBurn, and other software inc some test progs I’ve written. I can list more results below, but first I’ll raise my central questions.

1) Is it possible via getting the USB responses to identify the settings, transfer modes, etc, USB optical drives require to operate? At present trying to get CDVDBurn to work seems to be a matter of guessing at a whole series of possible variable values. Making it hard to stumble over just the right set.

2) Might some of the problems be due to incorrect USB transfer modes, etc? I ask this for reasons I give below. But it makes me wonder if it isn’t even as simple as guessing the right details for CDVDBurn’s setup.

I got the device working as a reader quite quickly. Simply connected it via a 4-way hub into a rear USB socket of my ARMiniX. Used !Boot to set 2 CDROM drives. Then Robert was my father’s brother! MusicMan2 and my own test progs could find the drive and rip audio from audio CDs, find the track start and end points, etc. Also CDFS, CDROMFS, and CDVDBurn’s filers would show files on data discs and let me access them.

So far so good. But trying to write anything fails. I can ‘burn’ a data CD and although CDVDBurn goes though the motions none of the data gets burnt to the disc. It is finalised, though. So shows a root window with the items listed. Just that none are accessible, and when I look at the disc there is no ‘darkened’ area showing where the data was burnt to the disc. And during ‘burning’ the access LED on the drive stays dark during writing, but lights during finalising.

Then I tried the drive on a 7-way hub connected to a front USB socket. If I do this the drive simply doesn’t work. No reading at all. Still no writing. My test prog for listing drives picked up by CDFS lists the drive responses and its CDFS number. But no attempts to read or write work.

BTW cdparanoia doesn’t work, regardless of which hub/port I use. That’s a shame as it is useful for assessing the reliability of audio ripping. But it implies it is using a method that differs from the CD Audio swis I’m using in my test programs.

So might I get any data that would help sort this out by looking at the USB device responses? i.e. similar to trying to sort out the settings of a USB Audio device?

Jim

It won’t help you at all, but I picked up a Samsung SE-S084D USB DVD writer to use with my eeePC.

Mmm… As a data file reader, it was plug’n’play.

As a DVD player, both SMPlayer and VLC work happily with it (though sometimes SMPlayer jumps directly to the film which is annoying if you have to tell it “not French!” and the audio settings won’t permit you to change them on the fly because the DVD was authored by a paranoid asshat who believes in taking away choices).

As a music player, it fails utterly. I can select tracks, the mechanism moves, the disc spins, but there’s no how no way no sound.
Yet it’s a breeze to rip the music using AudioGrabber. That works. Go figure.

As a writer – the supplied Nero failed to install. Kept trying to install some VC Runtime and choking. Examining the install log suggests it was correctly installed multiple times, but the installer is rubbish (one time it told me that the program no longer works on Win32 – WTF? This is XP!). Additionally, Windows own burning thing appears to work, but then kindly offers me an “unspecified error” (useful!) when it gets to try to actually do anything with a disc.

So for burning DVDs, I’m afraid I’m still using the old NAS PC, a length of ethernet cable, and VNC. <sigh>

Just had pointed out to me that the current version of cdparanoia isn’t arm v7 safe. Just tried it, and it works when run using AemulorPro and lets me rip tracks using the Samsung.

FWIW I plan to also experiment with using the device with my Linux boxes. But at present simply having it as an alternative disc reader is useful. Would be nice to crack having it as a writer,too, though. And if I get a chance I may try writing a simple prog to read tracks and play them out via USB. :-)

Jim

I use also this version of CDVDBurn and it works with all my devices on all RISC OS5-computer (xM with a slimline Sony Optiarc, PandaES with a fullsize Sony Optiarc both internal with a USBtoSATA-Adapter and a external LG with my Pandora and RPi).
I use CDRip from Chris Johnson with AcornCD. It works fine. The cdparanoia inside I have try. It works with alignment exceptions off.
Also working in MusicMan from rcomp. It can play the AudioCD “direct” (open a temp. pipe to RAMDisc and play this).
For both (CDRip and MusicMan) I use the A9 lame from Chris Gransden to mp3-encode).

Just to emphasise that my queries are about being able to extract USB info from an optical drive to aid trying to know how to get it to work as a writer as well as a reader. That said,

1) I can’t get lame to work with my copy of MM2, so wonder how you do that. With MM2 I have to rip as wave, then convert to flac as a separate process.

2) Not tried CDRip. Does it make repeat / overlapped reads to check for consistency and alignment like cdparanoia? This is important for me as I’ve found that CDRs may be hard to read accurately and cdparanoia shows this whilst simple ‘read and assume OK’ may fail to either indicate or fix problems. Hence the value of cdparanoia. MM2 is much faster, but only useful for ‘reliable’ disk+drive combinations. Hence probably fine for most commercial audio discs, but needs using with caution for CDRs.

3) What settings do you have for the Sony drive? Here I can’t get CDVDBurn to write ‘disc at once’ audio CDs. And data DVDs seem to have a ‘broken’ filing system. I can see them with CDVDBurn’s filer (and with Linux) but not with CDFS or CDROMFS!

Jim

Just had pointed out to me that the current version of cdparanoia isn’t arm v7 safe. Just tried it, and it works when run using AemulorPro and lets me rip tracks using the Samsung.

I’ve rebuilt the current version of cdparanoia so that it should be armv7 safe. I’ve not tested it yet. If you want to try it you can download it from here

I’ve rebuilt the current version of cdparanoia so that it should be armv7 safe.

Just tried it, and it seems fine here on a quick test! :-)

I used the same audio CD as yesterday, in the external Samsung drive. Your new version gives the same disc info as the old one using the -vQ switch. I then ripped the same track with the new and old versions. The results are identical. Works without needing AemulorPro.

When I get a chance I’ll test it a bit more, inc. with some ‘difficult’ CDRs. But it looks good.

Thanks for that! :-) I still have about 50 home-recorded audio CDRs to process. Linux is faster, but my experience is that with difficult discs RO works better for whatever reason. Maybe simply because the rip is slower and gives the drive more chance.

Jim

1) I can’t get lame to work with my copy of MM2, so wonder how you do that. With MM2 I have to rip as wave, then convert to flac as a separate process.

The A9 lame needs Chris Gransdens SharedUnixLibrary 1.13 to work.
I not use flac because my “car” can only play MP3.
With MP3 enc via lame I have no problems.

…CDRipEnc (sorry my mistake)…cdparanoia…

CDRipEnc use cjrip as default. It is much faster then cdparanoia (thats the reason I never use this).
Cjrip works with all my AudioCDs (selfmade and commercial). No problems here.

3) What settings do you have for the Sony drive?

CDVDBurn detect a Optiarc DVD RW AD-7261S. It works with Track- and Disc-at-once.
In 2011 I try a Freecom/NEC DVD Writer. It needs a dirty (non official) firmwareflash to do so. Before no AudioCD possible.
Witch DVD (+ or -) you burn. I mean Steffen says it burns only + without problems.

Does it make repeat / overlapped reads to check for consistency and alignment like cdparanoia?

No – as I said in a reply to you in another maillist – it is a simple ripper that does the job when required. One advantage it has over CDParanoia (apart from being much faster) is that it will rip tracks from ‘protected’ CDs where CDParanoia gives up.

My original CDRip application allowed the choice of ripper and encoder utilities.

@Raik. Now that Chris (G) has produced an ARMv7 CDParanoia – you just need to replace the older version in CDRip if you want to give it a try.

I will try later today with !CDRip on my PandaBoard.

CDRipEnc use cjrip as default. It is much faster then cdparanoia (thats the reason I never use this).

I use CDparanoia because it detects and handles the problems with discs that don’t read cleanly. Other rippers may simply give erronious sample values or have gaps or stutters. Accuracy matters more than speed from my POV. That said, this rarely matters for commercial Audio CDs as they usually read cleanly. So there I’m using CDparanoia to have it confirm a clean read.

But I need CDparanoia mainly to read the CDRs I used to record onto in past years. Most are fine, but some are not.

For speed I’d use CDparanoia on a Linux box. However that may simply stall or fail for some CDRs which a RO system will read, albeit with some struggle.

CDVDBurn detect a Optiarc DVD RW AD-7261S. It works with Track- and Disc-at-once.

Alas the drive in my ARMiniX is a 7740, not 7261. However I’d still be interested to know all the internal settings you have for your copy of CDVDBurn. These will be in the relevant ‘Drivers’ file as well as in the ‘Settings’ file. Here DAO simply won’t work with the 7740 with any settings I’ve tried. And TAO isn’t much use for classical music, etc. Fortunately I still have my Iyonix and that writes DAO AudioCDs fairly well. Not that I need this very often nowdays. However it is clearly unsatisfactory that we have problems ensuring we can get writers that can be made to work OK without such bother.

Jim

One advantage it has over CDParanoia (apart from being much faster) is that it will rip tracks from ‘protected’ CDs where CDParanoia gives up.

What kind of protection, how does it do this, and how do you know the result is accurate? Curious about this.

So far as I know I only have one CD that openly claims to have an anti-copy protection. I don’t normally rip commercial disks, quite happy to just play them. But purely to check, I captured this disc’s content via spdif and that seemed OK. So much for ‘protection’ schemes! Since I always use an external DAC I captured what I’d hear anyway. If it fouls that up, the disc is no use to me.

Despite that I haven’t bought any other discs of that kind as I object in principle to being sold a disc under the pretense that it is Red Book when it isn’t.

Jim

What kind of protection, how does it do this, and how do you know the result is accurate?

The result is accurate if the number of songs, and their length, roughly matches what it suggests on the inlay.

Protection? I can think of four I have encountered:

1. Playing with the order of audio/data tracks. Normally, audio tracks come first and data, if present, is added afterwards. Some CDs put the things the wrong way around so computers tend to only see the data part and not the audio part.
2. Messing with the TOC or having multiple TOCs. CD players will read the first (correct) TOC and computers will read the last (garbled) TOC.
3. Messing HARD with the TOC, so much so that you’ll need a program capable of analysing the CD subcodes or, worse, stepping through the CD data looking for inter-track gaps.
4. Sony rootkit – some Sony CDs from the – what was it, mid ‘00s? – included an autorun.bat that installed software on your machine that attempted to ’vet’ what you could and could not do with your CD drive. I believe I have such a CD around someplace, but it didn’t work as the first thing I do upon receiving a Windows setup is turn off the bloody autorun. Talk about a security hole the size of a wooly mammoth just to cater for people too stupid to work out how to use Start→Run ! The downside of this is… well… read this: http://en.wikipedia.org/wiki/Extended_Copy_Protection – in particular, read the “Description” part. If the security flaws were not bad enough, the software itself is quite nasty. And while this is OS-specific, it counts as a “protection” (though, given how it works, maybe “protection” as in “racket”).

I use Audiograbber under Windows.

There’s only been one CD it completely failed with. Classical (piano) music, of all things! I could have done an analogue audio capture, but to be honest it wasn’t worth the hassle. I’m trying to find this piece – http://www.youtube.com/watch?v=DFf31th_f_A – but don’t seem to be able to find a piano arrangement, only full orchestral. Perhaps I should be looking under Franz Lisz instead? Any ideas?

The result is accurate if the number of songs, and their length, roughly matches what it suggests on the inlay.

I’d not personally be confident that ensures accuracy TBH. Seen too many examples of what seems like the correct lengths, etc, but dodgy results in terms of the samples… Mostly from CDRs, but also from commercial CDs. Some of which are oddly made these days now some may not bother with Red Book compliance. Indeed, I’ve now encountered ‘audio CDs’ that won’t play on any conventional Audio CD hifi player I have. Only on computer drives! Had to rip one of them simply to make a CDR I could play on a normal player! Mad, or what?!

The ‘protected’ disc I’d referred to changed the actual data ripped. On the face of it, the tracks were the right length, etc. They did it, I think, by playing games with the correction redundancy in the channel stream.

I’m aware of the systems you described. But surprised by the assumption that if the results seem the right duration, etc, they are OK. Usually when I get error problems, etc, the lengths are ‘correct’ but samples are wrong. (Inconsistent or show clear signs like repeats or discontinuities.)

Afraid I’m using NetSurf on RO so the youtube page link didn’t tell me enough to know what specific work/performance you are after. Looks there more like an item by Beethoven, not Liszt.

Similarly, I have no idea how “Audiograbber” works so can’t comment on that.

The Sony ‘rootkit’ scheme was ruled illegal in the USA and they had to pay damages to people who were affected. I presume there is a list somewhere of affected discs. But since I don’t try to play most CDs on a Windows box it isn’t something I’ve worried about from a personal POV. Just as a civil rights infringement.

Jim

They did it, I think, by playing games with the correction redundancy in the channel stream.

Oh, that’s nasty. And could easily mess up playback on anything smarter than a dumb ’90s-era CD player.

But surprised by the assumption that if the results seem the right duration, etc, they are OK.

No, it’s a start. The methods I described, asides from the Sony rootkit, tend to play with the index markers or session information so the computer either can’t find any music, or it is broken in all the wrong places; effectively thwarting “easy ripping” while leaving the disc mostly playable in a “dumb” CD player.

I’d consider a CD with mucked up correction bits and such to be “defective”. I certain wouldn’t want random loud noise, clicks, etc either coming out of decent speakers or going in to my ears via headphones. As my portable CD player has a fast-read mode to buffer skipping and idle-run to save battery (plus MP3 playback), I wonder if I have anything that is dumb enough not to run afoul of these various forms of protection.

To put it in context – I rip all of my DVDs now. Why? Two reasons. The first, I tend to watch films on my iPad these days. Second reason, my TV hardware is a USB video capture card for watching via the computer, and an analogue TV so old the tuner is a turn-dial (no presets, no CVBS input). Neither device survives modern Macrovision so in order to be able to even watch the content for which I have paid, I have to strip out the so-called protection.
I think protection like that applied to music CDs is also risking causing more problems for end-users than it is worth.

I quite like Amazon’s idea. Bury some sort of watermark into the downloaded MP3s. Then I can put the music on my phones, iPad, RPi, etc etc without grief or restriction (!DigitalCD copes just fine), it only becomes an issue if I were dump the content onto filesharing sites, when the watermark can be used to trace the origin back to me. “Protection” without the undesired limitations usually associated with DRM.

FWIW So far I’ve not bothered with ripping commercial DVD Videos, although I’ve checked and confirmed I can play them OK. The main reasons that might lead me eventually to rip them would be:

1) VLC lets me skip all the ‘nag screen’ crap.

2) Save space for storing the discs as the content would be easier to find on a HD or few.

3) When I finally get around to getting an HDTV with HDMI inputs. Any day now… as I’ve been saying for years. :-)

In reality, though video doesn’t matter as much to me as music.

Jim

I’d consider a CD with mucked up correction bits and such to be “defective”.

So do I. That’s why, having bought one, I’ve avoided other discs from the same company/artist. Which is a shame as I like the music, but don’t want defective goods palmed off on me.

One of the problems of music retailers vanishing from our streets is you can’t now examine carefully the disc before you buy. So you can easily buy something which isn’t quite what you expect. Alas the ‘popular’ music biz has form as long as yer arm in terms of selling defective goods. Excessive compression, clipping, ‘HDCDs’ that aren’t, and ‘CDs’ that are actually ‘HDCD’ with or without the flags codes, etc, etc. As they say in the movie biz, no-one knows anything. Nor cares, it would seem. :-/

Jim

I will try later today with !CDRip on my PandaBoard.

@Raik: !CDRipEnc seems quite happy to use the new CDParanoia on the PandaRo, and looks quite a bit quicker than on the Iyonix. If you are still using !CDRipEnc and have any comments, let me know – I might look at updating it, as I haven’t done anything to it for a few years.

Why me? It was only an information and a notice of your very good program. ;-)
I also tested last night. Unfortunately, my post with netsurf is “gone” (maybe it shouldt be a notice in the bug area) and I did not want to write the whole again.
Now again briefly my results. I tried some new and old CDs (bought and self-made) and had no problems with cdparanoia and not with cjrip but cjrip reads round about 4x faster the tracks with my hardware.
I think R10 was the last version you sent me. I can not say much about it ;-), except it works very well. With AcornCD it is a very good ripping package. Before, I used RipperX on Linux, since I use yours with my Panda I ripped nothing with Linux.
I had been the luck that cdparanoia I not need ;-)

@ Jim
Here you can find my CDVDBurn settings. But I change it if required.