hi everybody I've got this question concerning archiving.

I ripped all my Cds to TAK (before TAK it was FLAC, but I've transcoded all FLACs to TAK then), but now my external 500 gb recovery drive is getting close to overload

When I want to listen to music but don't use the drive I transcode the TAKs I want to NeroAAc @ q0.31 (Just for listening, not archival use ^^) on my Computer.

So now I am thinking all the time about... maybe transcoding all my TAKs to a high bitrate lossy codec, to save space but keep archival quality status. I've already tried following options:

- NeroAAC @ q 1.0 = ~ 400 kbps (nero's dowload page says this bitrate is for perfect quality...)
- Musepack @ q 10.0 = ~ 350 kbps
- AoTuV @ q 10.0 = ~ 500 kbps

all of them are used with the latest stable version.
Concerning quality there's no question: all of them are more than excellent. But useful for archiving?
I also thought about using a codec without a psychoacoustic model, like

-wv lossy @ ~ 400 kbps
-lossy wav (? - I haven't read much about it yet)

Does anybody know which option qould be best for achiving, or is it senseless.... i mean transparency is arrived at around 175 with nero, mpc @ 170 and og with 160, but would moving the bitrate up be useful for my purpose? Are the known problem samples on HA for lossy codecs gone away in this bitrate area at around 350 - 500 kbps?

Or should I stay on lossless and grab another 100 euros to buy a 2nd external drive? (I don't like wasting money... but if it's neccessary... )

Many many thanks for your answers,
Greetings from Germany

They are all options . They all have / had 1 or 2 samples abxable at 400k or so, but at those bitrates it should be close to impossible to find problems. Wv has no known samples in 480 k + ditto lossywav. My pick would be Wv or lossywav around 400k and 500..550k as an overkill option. In reality you can get good quality at much lower bitrate. I am using Wv currently @ 270 k and find that it plays and transcodes nicely for those DAP's. I'll keep the correction files offline.

You can experiment on a few albums before making big decisions.

Ok thanks for this answer. So you mean a lossy codec without psychoacoustic model would be better than a codec with psychoacoustics?

I'll try wv lossy and check out lossyWAV.´
This will be a hard decision

In my opinion you would be making a big mistake. In time storage will likely become cheaper and 500 Gigs won't seem that much at all.

A true archive or backup is in no case a temporary solution and should be able to completely replace any redbook CDs or DVDs.


This is why people choose not to use lossy.

Of course HD price is a good argument for the - I guess - most usual approach: a lossless archive + a productive transcoded collection for DAPs using mp3 etc.
But it's not THE solution.

As shadowking pointed out we can exchange lossless encoding (with a bitrate of ~800-900 kbps for pop music) for an extremely high quality lossy encoding (with a bitrate of ~500-550 kbps) which in practice is equivalent to lossless. And we can choose to lower this bitrate down to 350 kbps or lower (depending a bit on the chosen codec) without really having to fear a remarkable quality decrease. Or choose a very high quality compromise in the 450 kbps range.

Moreover we have the choice to throw away the 2 way process of having an archive + a productive collection. This is what I decided to do (using lossyWAV
--extreme as a preprocessor for FLAC).
The details how to do it depend a bit on the DAP used, especially the capacities and the playable codecs.
When buying a new DAP restrictions on codecs or capacity go more and more into the background. But it needn't be lossyWAV + FLAC or wavPack lossy. That's a solution for the extremely quality aware people like me (and is a bit paranoid I'm afraid). It's also fine to use very high quality encodings with the more popular codecs like mp3, vorbis, aac for which a bitrate in the 250 kbps area should be enough.

To say hdd is cheap is like saying throw out old PC's cause new ones are cheap, or run any bloated app you like cause new systems have enough ram & hdd. Sure adding external hdd is easy enough for most, But what about those proprietry jukeboxes ? are they as easy to upgrade ? Do you just throw them away ? I think that is another case for losswav / WV. personally I don't want 900k files on my main drives for albums that sound like FM-turd (most modern cd's). For me wavpack 270 k is x 3~4 times less than lossless. I am getting high quality playback and even for transcoding . I am keeping correction files on DVD. Why do I need so much bloat for what I don't need ?. I don't like changing pc every 3 years. I've kept the P3-550 with minor upgrades since 99' and will upgrade it soon. I remember 3 years ago I tried lossless, bought 2 x 120gb drives only to realise i'd need another 1 or 2 for music backups only.

/If/ you are going lossy, I'd always pick MP3. It plays everywhere, so you're absolutely sure you don't have to transcode at any time in the future ever again, and at v0 or CBR320, who's going to notice the difference?

Yep, you're right. But I use NeroAAC for lossy (for PC listening and my DAP), just because I get very small files and good quality, for a portable player that only has 1 gb of memory it's more useful than mp3.

I'd like to keep all my archives being an archival quality. MP3 @ 320 wouldn't be a good solution I think (CBR and no higher bitrate available), therefore MPC AAC OGG gives a better solution with VBR in that very high bitrate range. I love Lame @ -V 2, but it's "just" transparent, not archival quality.

Oh man, it would take such a lot time transcoding around 400 gig of TAKs to an lossy codec... is there a very fast one?

Maybe I have to buy a 2nd HDD.... in 2 years a 3rd... then a 4th... it doesn't look nice anymore when my table is full of HDDs.

I think mp3 isn't a big quality issue but one can feel 'locked in' to the format. Even with lossless freedom most are still locked into mp3 transcoding. One of the advantages is that one can transcode to their AAC / OGG / MPC player of choice from lossless without fear of lossy-lossy issues. This approarch takes much more space than a single normal bitrate lossy approach.

I suppose the question has haunted the minds of a few members here inc mine: Can there be a near-lossless substitute giving us the advantages of lossless at reduced bitrate ? Its can be a controversial topic, I am not saying its 100 % correct but I don't think its easy to dismiss the need for it either.

@ memomai:

So if I understand things would be better if the archive was using 150-250gb instead ? What kind of saving are you after ?

About speed, WV without -x is insanely fast even on old PC's. With higher bitrates its no big issue.

Yeah in this way, I just want to save space by keeping archival quality, so reducing the filesize of all my taks from 400 gig to around 200 - 250, mainly to avoid buying me a second external drive... It took about 2 years until I reached 400 gb of lossless music.

Can't you sell your 400gb and buy a new 1TB? I've seen 1TB externals for 170 euro (so the price can't be the problem) and it saves you from having multiple drives on your desk. Rince and repeat every two years for a drive double the size

Personally, I wouldn't dump the lossless archive for anything.

I would say if you are going to go lossy, then use a format with correction files and save them somewhere, on DVD's or whatever.

One of the lamest things ever, and I know because I've done it, is having to re-rip all your CDs again and again and again. If there is even the slightest chance you are going to suffer transcoder's remorse and not be completely satisfied with your lossy copies, then keep your originals somehow.

You should ask yourself, how much did you pay for all of those CDs that are stored in 400 GB (I am assuming that you paid for them) compared to the cost of the HDD?

You might also ask yourself how many weeks it would take to listen to every song in your collection one time each, and are there any songs in there that are not worth keeping?

It might sound odd, but I would pick musepack @ braindead. A 320 kbit MP3 is not an all-round archiving solution, for some genres, it can be easilly ABXed (you want an example? Dimmu Borgir - Spiritual Black Dimensions or Enthrone Darkness Triumphant - their aggressive hi-hats and overall distorted sound with many higher harmonics are just real nightmare for MP3). Vorbis provides excellent quality for a chosen bitrate but transcoding from it usually produces heavy artifacts (my own experience), it's good for terminal bitrate but I wouldn't use it for archiving with potential transcode in mind.
This said, I haven't encountered any of the mentioned shortcomings with Musepack. It is able to encode even some problematic stuff and transcodes real good (way better than Vorbis or even MP3) as it uses a different internal base (MPEG-1 Layer 2 based, IIRC) and, when transcoded to MP3 or Vorbis, mostly avoids typical transcoding artifacts. Being based on MP2, Musepack should also be pre-echo resistant as opposed to MP3 (at least, this was claimed by MPEGPLUS documentation several years ago).

If you will transcode them to lossy - there was a little test by Guruboolez regarding transcoding. As I remember the best codec to transcode from was Ogg Vorbis. I think using Ogg Vorbis over q7 is insane. ABXable artifacts at q8 will be ABXable at q10 too (IMHO - no test is done).

But why don't you want to keep your favorite music lossless and 3 and 4 star music transode to lossy?

No, I want to keep everything I have I love(d) every music on this storage, and when I listen to music that I've listened several years ago, then there's that "The good times" feeling in me ^^.

You might also know it, some people threw their music away and years later they want to listen to it again and can't find 'em... I want to avoid it.






One says musepack is good for transcoding, the other one says vorbis is the best one. I think there will be no clear winner, *cause every lossy codec has its problems with transcoding, and I also think this is unavoidable.


@ pdg: good point, I haven't about it until now, it's possible that I'm kinda stupid
A 1TB recovery drive would be a nice alternative to convert TAK to lossy.


But it's still a good question which lossy codec would be the best compatible for archiving.

There is a bit of risk involved in using correction files. For your "archival" backup to do any good, you must have the "playback" directory undamaged. Kinda defies the meaning of the word backup. And what are the gains of this inconvienent storage method? About 192 kBit/s less data that you'd otherwise use for lossy meant for playback.

I would suggest buying another HDD. I prefer to buy HDDs and external cases separately because it allows me to find components best fitting my needs and I think it’s cheaper.

Like 750 GB for € 75 or 1000 GB for € 115 plus a case (depending on your needs, I’d buy one with eSATA) starting at about 10 € (with eSATA about 20 €)
If you want to use other interfaces than (e)SATA you should look at some benchmarks since the USB (and probably Firewire) speed depends on the chipset. I use eSATA because it’s about 4 to 6 times faster than the USB interface of my case.

Now, this by itself is an interesting question.

What are the functions of an archive: it should be as high quality of a backup as possible, that can be listened to when desired, but especially able to transcode from to a lossy(er) format for use on portables. For discussions sake lets say that an additional transcode will be done from the archive before going on to the portable - this way there will always be two separate libraries, portable and archival.

As far as the high quality part goes: There are lots of options. Many tests have been done to show where given codecs become transparent, so from this CBR bitrate/VBR quality setting or anywhere above you should be safe. How much overhead you want to give is up to you - I would base this on your personal music tastes, namely what the rate of lossy problem samples that are in your particular type of music/audio, as well as your particular level of paranoia that an error may slip through.

When it comes to transcoding lossy to lossy, not as much research has been done due to the fact that this transcoding will invariably introduce some distortion, whether audible or not, and as such is not recommended. However, my personal trials have suggested that transcoding between different kinds of formats is a better idea than transcoding between the same one, eg: if the first generation is Vorbis, make the second generation MP3. The theory behind this is there is a smaller chance of errors being audible, because two different encoders processed the information (which will choke up at different points and produce different errors) instead of the same encoder running twice (since the same encoder would likely choke up on the same error with each generation) which will produce more pronounced/audible errors. I've done some modest testing, and this seems to be the case - overall files that were encoded between two different encoders sound better than files encoded continually with the same one. Because of this, I would recommend you use a format/encoder for your archival library that is different than the one you plan to use on your portable libraries.

So the conclusion: use a lossy format that you are unlikely to use on your portable, encode it at a high enough bitrate that has been proven transparent with added headroom for your particular music's problem sample rate as well as whatever paranoia you might have.

A lot of my opinions here seem to be in the area of theoretics, though, and probably go way beyond the level of practicality. At these high bitrates it may take a couple of generations before audible errors show up, if at all. Only by doing lots and lots of listening tests could you guarantee that the music is error free. However, I feel that these guidelines should provide you with the best possible lossy archive.

Transcoding to mp3, Try Vorbis > Vorbis. I got nasty artifacts even with high bitrates like 350 k .

One day, the day when 500gb is only 1/1000 of typical hard drives, you'll regret you've encoded to lossy and wasted valuable information. Keep all information!

I believe the test to which de Mon refers is this one. The test scope was limited, so its results should be considered accordingly. And you're right to say there's no clear winner, other than the LAME MP3 version used in the test not generally being a good transcode source. Vorbis had the highest average score, but it looks to me like it's probably within the error margin with at least the MPC encoder. The other person's claim is skirting a TOS#8 violation.

Hard disks may be cheaper now than they once were, but that will never justify using lossless when you don't need to. $100 is still a lot of money that I'd rather not spend - especially when I know I don't need to. Doesn't anyone else think this mantra of "just buy more drives and go lossless" is basically FUD when you really think about it? How many of us remember participating in the last 128 kbps listening test? I still haven't done my transcoding abx tests but I think I'll end up with either lossywav or back with wavpack lossy no correction file. I'm sick of the term "archive" like it's some sacred thing that just deserves lossless without any thought. What I want out of my "archive" is a convenient way to listen to music that sounds perfect and can perfectly transcode to a format if I need it.

Of course you should use lossless for irreplaceable favorites, or masters if you are involved in producing music. I wrote the above about the usual music collection.

If you want to be able to "perfectly transcode", then lossless will be your only solution.

But I agree with you about preaching "buy more HDDs". Any solution should fit a particular person's needs and pocketbook.

I see the biggest concern as avoiding any significant future regret. Think through where you'll be in, say, five years, what the state of audio encoding might be at that time, and how often and to what formats you might want to transcode. For some people Vorbis or MPC or AAC may be fine, while other's may legitimately need a lossless format.

Also consider that if you don't have the money to buy more hard drive space now, do you reasonably think you might within a year or so? If yes, then it may be best to stick to lossless and "overflow" onto DVD-Rs for the next several months, then if/when you can get more HDD space you can pour the overflow onto it along with the music on your old hard drive. Even if you're not sure, then you can always transcode your "archives" to lossy at that time just like you could now.

If you don't reasonably foresee being able to get more space within that time, then it might be best for you to consider a lossy format that's decently rated (or at least not badly rated) as a transcoding source.

I'd just think long and hard before going to lossy for archival with that much music.

I meant perfectly transcode "perceptibly". I guess I should always stick that word in there. But also like I said, fear, uncertainty, and doubt. There's a lot of those vapors in that post. I don't fear, doubt, or feel so uncertain what my hearing situation will be in one year or five. The files that I fail to abx this year have a DECREASING chance of being successfully abx'd in the coming years, not an increasing chance. I'm not going to think long and hard before my decision -- I'm going to listen and test long and hard... I've already participated in the 128 kbps test (and failed on most of those files). And I always wonder how many of the hard disk salesmen have done the same...

If you're sure you won't regret it in the future, then you've already outpaced the biggest concern, making the rest of the recommendation moot. Now it's just a matter of picking the format. According to the results of Guru's "short transcode test", either Vorbis or MPC would probably serve equally well, so it'd just be a matter of personal preference.

I can only back the other person's claims about vorbis->vorbis transcodes. It produces a real mess whatever the bitrate! I "archived" several albums I borrowed from a friend using Vorbis and I regretted the decission later since I didn't have the possibility of transcoding them properly into a lower bitrate Vorbis for my portable.
But if the person claims that he has problems recognizing a 128 kbit file, then he perhaps won't notice these (hi-freq) artifacts as well. But you should also think about the possibility of "tape trading" with someone else who could be annoyed about your archiving solution decission.

I'm very curious to see your test results on that. Not as a challenge of any kind, but the more such results we have, the more informed people can be making these very kinds of decisions. Can you post them for us?

That's 15 years away. This person wants a solution now.

Cheers,
David.

I'm sorry, I already ditched everything related to vorbis (including my test files) since the other shortcoming it had was that the battery life on my portable was half that of MP3. I switched to MP3 since the quality was not worth the trouble.
I also lack motivation to try and prove something to the HA audience as I already tried several times in the past but failed utterly, the problem obviously being my favorite genre of black and death metal, and majority of the people not used to such genre is not able to pick out/confirm the artifacts even if I point them out. It's kind of uphill battle.
Perhaps the other person complaining about the same problem could provide more "suitable" samples.

I completely agree with you. People who compile "archives" carelessly will later post how great redbook CD was: "I lost my correction files, damaged my princo CD-R written at 40x, dropped the hdd on the ground". Yet the original is still CD readable; what a great technology. A true archive is made for years to come, not today.

9 months ago a TB cost about € 300 (in Germany)
6 months ago € 240
today less than € 120
If 750 GB is sufficient, that would be € 75, or € 65 for 640 GB or € 53 for 500 GB
Conclusions:
- If that’s still too expensive, it could be affordable in a few months
- Used drives are probably cheap enough
- Burning to DVD and buying a large HDD later sounded good to me until I looked at the prices of DVDs which would eat a lot of the saving. Additionally it would be very time-consuming.

I think the problem here is that many lossy users understand why other people might want to go lossless, but recognise that it isn't for them.

Whereas many lossless users seems to think that all lossy users are somehow misguided, and need educating, because they've clearly made the wrong choice.


The cost of lossless is time (5x longer to back up and move about), money (storage costs), and convenience (fewer HDDs sitting around; mp3 plays nearly everywhere without re-conversion).

Taking drive prices - they may be getting cheaper, but you'd be silly to keep only one copy - so you need at least two drives.

A _good_ 1TB drive isn't as cheap as you suggest in the UK...
http://www.dabs.com/productview.aspx?Quick...376770000,50410

Even the lower quality non-warranty version costs more than 120 Euros...
http://www.dabs.com/productview.aspx?Quick...376770000,50410


I have photos, more video that I could fit on several of these drives, and audio. I have decided that I don't need bit perfect lossless audio. The photos and video are lossy too. All are more than good enough for my needs.

I only need lossless for things that I have generated myself, that I may re-edit and transform through multiple generations. Clearly lossy would be a silly choice here.

For everything else, it's fine. mp3 isn't going away any time soon.

Cheers,
David.

When transcoding, mp3 (or aac) is most likely the end result. There's not much point transcoding from vorbis to mpc.

If the lossless is too expensive or time consuming, I suggest either high bitrate mp3 or aac to save yourself the extra effort on transcoding--not to mention the effort on abxing which is the better lossy transcode source.

You can try any lossy solution around, have ABXed every lossy codec for transparency once you tested the safety & freedom of lossless, lossy will always leave a bittertaste in your mouth.

after having tested fhg/blade/gogo/lame/wma/vqf/mpc/vorbis/megamix/aotuv/nero aac/faac ... lately I have even been thinking of using lossy as CDImage or lossywav+tak ... just to gain some space ... in the end I think I will just buy a new 1 To hard drive ... the 11th !!! (Pata120: 40/120 Go Pata133: 120/120 Go Sata150: 160/160/200/200 Go Sata300 320/320 Go) ... nowadays I only use lossy for video because 1: the source is already lossy AC3, 2:because I don't like 5.1, 3: because I only listen to it once when I watch the movie, 4: because you don't have to worry about gaplessness with video ... lossy audio is for videophile & streaming ... if you're an audiophile there is no escape: buy a new hdd ...

HDDs sold in Germany usually come with 3 years manufacturer’s warranty, with the exception of Seagate’s drives which usually come with 5 years. Besides manufacturer’s warranty, the retailer must give 2 years warranty (required by law).

You can have a look at the prices in Germany here, or sorted by price, sorted by price per GB. The cheapest TB drive is the Samsung Spinpoint F1 which is available for € 118,60 (free shipping). I think it’s also the fastest mainstream HDD.

as far as I am aware in the 1 To/334Go by platter/3 platters competition, the samsung spinpoint F1 is cheaper, but the western digital WD10EACS-00D6B0 (be careful it is very new & there is a 4x250go=1to model too, not as good) is better ... personnaly I will take a WD when it will be available, I dunno why but the samsung is too cheap too be honest & I don't want to lost 1 To !!! 4 years ago samsung wasn't even a top HDD sealer ... & suddenly they have the best cheapest HDD ... I may be paranoid but I find it too suspect where have the research & development cost disapeared for samsung ? I used to buy maxtor/seagate in the past ... but in between wd have made the raptors ... so I will stick with wd personnaly.

You must realise the bitter taste is in your imagination?

First you mention "freedom", then you mention 11 HDDs - that's not the kind of freedom I'm looking for!

Cheers,
David.

well sometimes I wish I was dumb & deaf but splitability problem, untranscodability, no gaplessness (mostly solved with time) & DCT artefact below 128Kbps are very real I fear ... it will be the 11th HDD I buy but I only have the 6 sata remaining, I already dropped pata several years ago. Yes I feel more free with lossless, I only wish I didn't felt poorer too ... it's not a problem of quality if I could losslessly split/join lossy I would like lossy much more ... I store as CDImage & I will not go back to separate tracks as tagging is a waste of time for me ... but I could go back to lossy, if it were more friendly for single file+cue, I am aware of placebo, I just don't feel free with lossy. I prefer losing money buying HDD than losing time tagging & splitting all my single files. Maybe my first post was unclear it's not a quality problem for me, it's a usability problem.

*counting the TOS#8 violations in this thread*

*ran out of fingers and toes*

(Or has HA dropped that rule? I've been busy for several months and may have missed that.)


That's an important one for me, too, which I forgot to mention in my earlier posts.

What do you mean "solved with time", though? I've never foreseen that being changed in any non-gapless format (MP3, AAC, etc.)

you forget the "mostly" ... read: "improved" if you prefer ... last time I checked the major lossy formats (nero aac/vorbis/lame) were gapless in software (sometimes by tagging tricks) but it was the lack of hardware support for their gaplessness in DAP that was faulty.

There's a Report button below each and every post.

I guess it's more effort for you to click it than whine about how the thread isn't being moderated.

With the title "lossy archiving, makes it sense?" I'm not the least bit surprised to see that this thread isn't getting any serious attention.

Whoa, calm yourself down, there. No one's whining, and no one's targeting moderation. I was just honestly curious.

And yes, the title is a bit off-putting, isn't it.

Why? I thought HA is a forum discussion site where some very high skilled members can help me with my problem, and I'm sure I'm not the only one who has the same problem as I have with space and finding another solution besides lossless archiving.

Lossy archiving is not an ironic question, it's an interesting point, that's all.

I've never met a user who left so many posts which attack other user's abilities and individuality as you do, greynol.



do you mean lossy archiving is not worth talking about?


-------------------------

What about AAC @ very high bitrate to transcode from? Has anyone made experience with it?

Next weekend I'll take the time and try out lossyWAV, maybe I can get some "lossyTAKs"

No, I think for people with a large amount of music, limited hard drive space and limited resources for expanding said hard drive space, that it's quite worthwhile to consider non-lossless encoding solutions at least to cover system fault tolerance and disaster recovery while providing a perceptually acceptable transcode source. Folks without enough hard drive to accommodate 35MB music files deserve some form of archival, too. That's why I've been making serious recommendations all along. I was just foreseeing that people of somewhat limited maturity might poke fun at it.


Guruboolez did in this test. Although it was a fairly limited sample set, only one tester, and it was done three years ago, the results should still be somewhat relevant as it did focus on high bitrates, the guy who performed the test is trusted as having some of the best ears in this place, and three years isn't very long in the field of audio encoding development compared to the speed at which the rest of the technology industry moves.

(I hope it helps you a bit in your decision - and please forgive me if it doesn't.)

If you're curious about TOS #8, read the Terms of Service (there's a link at the top of this page), but limited maturity???

@Nick E:
It wasn't a dig on the poor English (now that would have shown limited maturity). It's the fact that lossy archiving is an oxymoron. Making a backup of your lossy collection is one thing, but to downgrade from lossless to lossy and call it an archive is something entirely different.

I make no apologies for what I've written here.

Perhaps he merely meant the title sounded odd -- "makes it sense?" is not idiomatic English. On the other hand, when he read the thread I suppose he'd have seen you're not a native speaker, so perhaps not.

True, with a higher than average percentage of international participation here, people don't usually concern themselves with picking apart English grammar or spelling.


That says it all, doesn't it?

Yes, I meant exactly this test and I had mentioned it was 'little' test.

Anyway, I think the rational way would be lossless or MP3 V2 (despite you have player which plays other codecs). Despite I use Ogg Vorbis as lossy, the quality of sound is not the main gain for me.

I like Ogg Vorbis's for:

1. replay gain tagging possibility
2. encoding speed (Lancer's compile)
3. tags
4. true gapless playback
5. FREE and open source
6. quality per kbs ratio

Yes, quality at the last place despite it is very very good codec.

Anyway I can listen my Ogg Vorbis files without transoding everywhere except my car.

But if someone wants compatability - better use MP3 (LAME at V2) and don't bother with spending time on transcoding.

Back on the original topic, there is only 3 options:

-lossless 100%quality/100%freedom
-transparent lossywav 99%quality/99%freedom/50%spacegain (no warm & fuzzy feeling)
-transparent lossy 98%quality/10%freedom/90%spacegain

pure lossy is very risky as an archive format, each time I tried I always ended deleting all the archive, from my limited experience, recoding from lossy whatever the bitrate is always a bad idea because whatever the bitrate I always heard added quantization noise if I transcoded at ABXable bitrate ... many people think transcoding from an higher bitrate is better ... but from my limited testing they foul their own mind ... because they transcode lossy at overkill bitrates to lossy at transparent bitrates ... so yes indeed most often it's still tranparent ... but if you transcode below transparency ... you quickly realize that transcoding lossy to lossy is really a no-no due to exponential additionnal degradation ... so if you know that you won't buy a new HD & at the same time wants to transcode to lossy then I would go lossywav ... I have the same problem actually but I ended buying a new hdd. You alone can decide for yourself.