Which bitrate do you use most often?

[Arr-Nine-Hundred]

I think the simplicity of just having a "limited" set of very good encoding options is a plus of MiniDisc. If I was faced with loads of tweakable settings like in the LAME encoder I'd be forever changing things or wondering if I could have done better by using a value of 100 rather than 99 for some factor. So it's easy - if recording from mp3, wma, mp4 or another format OR a double CD album then LP2, if recording from an outstanding CD album then SP (eg. RaTM's "The Battle of Los Angelos" lends itself to SP so one can hear every snarl in de la Rocha's lyrics).

So to answer the question: I use LP2 most often, followed by SP. LP4 I never use.

I always encode with Type-R and decode with Type-S.

[Avrin]

SimpleBurner encoding uses the same encoding algorithm as SonicStage does in Normal (Faster), and not High, Recording Quality. But this definitely preserves more high frequencies in LP2.

[Azureal]

SimpleBurner encoding uses the same encoding algorithm as SonicStage does in Normal (Faster), and not High, Recording Quality. But this definitely preserves more high frequencies in LP2.

Arvin, why would the Normal (Faster) recording quality produce better results than the High (presumably slower ...) recording quality, preserving more high frequencies in LP2? I am curious as I am trying to discern what bit rate/codec to use for some recent recordings.

I used to love LP2 even in the pre-Type S days of playback, but I find it a bit harsh now. This is likely due to all of the real-time optical SP recording I have done over the past few years. SP is hard to beat according to my ears, but I am trying to fit more music on one disc ...

Is it possible to record SP real-time, via optical in, onto an RH1 with a standard MD disc formatted as Hi-MD? This could allow nearly three hours of SP recording on one disc ...

[freddyjollo]

I use Lp2 or Lp4. Lp4 for recording spoken radio shows. I did buy a HiMd portable but now only use it as a good NetMd device. I (used to mainly) use the 940/980 brand of deck of which I still have a few, & had Sony brought out a HiMd deck that was readily available I would probably have changed my choice - but they didnt. I would have preferred they had made a deck rather than the MZ-RH1

[Avrin]

I'm not sure why this happens, but frequency analysis in Adobe Audition shows that more higher frequencies are preserved when encoding to ATRAC3 @ 132 kbit/s using the Normal (Faster) mode or SimpleBurner, than when using High Recording Quality. And I'm still not sure about the actual quality of encoding (more higher frequencies do not automatically mean better quality).

Also, it is not possible to record in SP/MONO/LP2/LP3/LP4 on the unit in Hi-MD mode (only PCM, Hi-SP, and Hi-LP are available for recording). But you may transfer ATRAC3 @ 66, 105, or 132 kbit/s from the PC.

[sfbp]

I use whatever bitrate I won't have to transcode.

For CD->MD this means compressing using SonicStage to Atrac Advanced Lossless (which will generate "essentially" untouched PCM, OR Atrac3+ just as easily). In addition my MXD-D400 CD/MD deck does a nice job making both SP and LP2 digitally.

For recording off the radio or internet, I use either LP2 or Hi-SP. In the case of LP2 I use a deck which has Type-R and this (and the advanced A->D built into the deck) makes sure the optimum use of the data is made to produce a full, rich sound despite compression. In addition I capture long radio compilations at LP4, and the sound is surprisingly reliable after upload.

In the case of Hi-SP, I upload to the PC and never change its format again even if transferring back to MD for a compilation.

In the case of LP2, I use the RH1 to upload it to the PC, and if I want to make a compilation, I transfer the LP2 **unchanged** to HiMD or MD (both work). Note that it's easy to put LARGE amounts of LP2 and LP4 onto HiMD media, almost twice as much (compared to NetMD) on the same physical medium, and up to 16.5 or 33 hours respectively on a single 1GB disk.

I have a feeling that many of the reports comparing the different bit rates used flawed methodology, starting with a particular format (I already drew attention to what an appalling job SonicStage does at ripping PCM to the computer in the interests of speed) and transcoding that file to different rates. This doesn't work. I abandoned LP modes for the first 2 years after I got a portable because the quality was, frankly, crap. Once I worked out to leave the data alone, the LP modes started to shine.

For live recording I use the RH1 and a 1GB disk, at PCM rate (1411Khz), 94 minutes max. This means that I have lots of headroom if I need to play with the signal (eg if it is underrecorded) after uploading.

For transcription (re-mastering) of old analogue recordings, I capture at SP and then upload, either via optical or using the RH1 (and Sonic Stage's automatic WAV conversion), to the PC as a WAV file, which I can edit using one of the many tools (eg. CoolEdit or Audition) available. This avoids bad stuff getting into the signal because the inputs to the MD deck are really clean, and SP->WAV looks pretty good as long as you don't transfer *via* MD more than once.

[Arr-Nine-Hundred]

Is it possible to record SP real-time, via optical in, onto an RH1 with a standard MD disc formatted as Hi-MD? This could allow nearly three hours of SP recording on one disc ...

That would be very cool indeed but according to this chart it is not possible: http://www.minidisc.org/hi-md_faq.html#_q93

[Arr-Nine-Hundred]

For recording off the radio or internet, I use either LP2 or Hi-SP. In the case of LP2 I use a deck which has Type-R and this (and the advanced A->D built into the deck) makes sure the optimum use of the data is made to produce a full, rich sound despite compression. In addition I capture long radio compilations at LP4, and the sound is surprisingly reliable after upload.

I was about to jump in say that Type-R only benefits SP, not

LP2, recordings but it seems you are correct:

http://www.minidisc.org/type_r_atrac.html

Some user comments suggest otherwise:

http://www.minidisc.org/brian_youn/MDLP_usercomments.html

[sfbp]

I was about to jump in say that Type-R only benefits SP, not

LP2, recordings but it seems you are correct:

http://www.minidisc.org/type_r_atrac.html

Some user comments suggest otherwise:

http://www.minidisc.org/brian_youn/MDLP_usercomments.html

I don't think the recordings made with my deck would sound as good as they do 'twere it not for Type-R.

It may be that the transcoding in software doesn't always match the Type-R which is presumably done in the DSP chip(s). That would account for what I have observed, if so. Looks like SOME means in software work ok, but it's a bit hit-and-miss. SimpleBurner seems agreed to do a good job, for whatever reason.

[md user]

I know a little off-topic (just) but has anybody compared the quality of Type R SP@292 to [Atrac3+ @256 or @353kbps]?

I'm considering buying one of those NW-A ranges - but they don't offer SP@292.. just wanted to know what to expect (I'm very pleased with SP).

If this has already been discussed, please point me in the right direction ...

Regards All,

mdmad.

BTW, I would have voted for SP@292 - but obviously not an option. Is it correct to presume that Sony considered SP superseded by the above, and so did not include it as native with SS on HiMD, hence the lack of this option in the poll?

[Arr-Nine-Hundred]

I don't think the recordings made with my deck would sound as good as they do 'twere it not for Type-R.

The Type-R will definitely help but remember that it's the Type-S feature that also "gets extra" out of LP2/4 recordings on playback.

I maintain this http://opticalgarbage.com/minidisc/type-sr.html and have updated the "Enhanced Recording (encoding)" column for Type-R to show it supports LP2 and LP4.

It may be that the transcoding in software doesn't always match the Type-R which is presumably done in the DSP chip(s). That would account for what I have observed, if so. Looks like SOME means in software work ok, but it's a bit hit-and-miss. SimpleBurner seems agreed to do a good job, for whatever reason.

As I understand it, the LP2 encoding in SonicStage software is technically inferior to the hardware DSP method. I can't remember the specifics now but there is a technical reason. This is why I never use SonicStage or any NetMD functions.

But.

Transfer of track names would be great though but I can refer back to the playlist I recorded them from or http://www.freedb.org/ if I was desperate.

[sfbp]

The Type-R will definitely help but remember that it's the Type-S feature that also "gets extra" out of LP2/4 recordings on playback.

However, I have looked as carefully as I can at those devices for which I have schematics. The JB980 and MXD-D400 being honorable exceptions (don't have the diagrams), I came to the conclusion that possibly the Type-S is invoked in the D->A. Many (if not all) of the units with Type-S, instead of having a separate D->A, have a so-called "D-class" output for the headphones.

The implication being, that perhaps optical out is not affected. I recently got a receiver with Optical in made by Onkyo, and so far I don't hear the difference between the JE640 deck (Type R) and the MXD-D400 deck (Type-S). Prior to that, LP2 was almost not worth listening to on the analog output of the 640, and the MXD-D400 was a real step up. If Type-S is part of the analogue signal stream processing, then these observations would make sense. I *do* know that on the 640 the output of the DSP chip goes through NO circuitry on its way to the optical transmitter.

I maintain this http://opticalgarbage.com/minidisc/type-sr.html and have updated the "Enhanced Recording (encoding)" column for Type-R to show it supports LP2 and LP4.

I will wait to be proved wrong about my observation

As I understand it, the LP2 encoding in SonicStage software is technically inferior to the hardware DSP method. I can't remember the specifics now but there is a technical reason. This is why I never use SonicStage or any NetMD functions.

Opinions differ. Obviously subjective. You can see that if a given algorithm does a better job if it can actually use more CPU, then the possibility of some new processors "beating" the 10-year old DSP hardware might exist. But that's all speculation on my part. Avrin seems to be quite actively looking into this.

Stephen

[Arr-Nine-Hundred]

However, I have looked as carefully as I can at those devices for which I have schematics. The JB980 and MXD-D400 being honorable exceptions (don't have the diagrams), I came to the conclusion that possibly the Type-S is invoked in the D->A. Many (if not all) of the units with Type-S, instead of having a separate D->A, have a so-called "D-class" output for the headphones.

The implication being, that perhaps optical out is not affected. I recently got a receiver with Optical in made by Onkyo, and so far I don't hear the difference between the JE640 deck (Type R) and the MXD-D400 deck (Type-S). Prior to that, LP2 was almost not worth listening to on the analog output of the 640, and the MXD-D400 was a real step up. If Type-S is part of the analogue signal stream processing, then these observations would make sense. I *do* know that on the 640 the output of the DSP chip goes through NO circuitry on its way to the optical transmitter.

Very interesting stuff Stephen, thanks. 99% of the time I listen with headphones on a portable to MP3's recorded digitally in LP2 in a quiet environment so the Type-S is definitely something suited to me.

I have no deck yet but I do have a Logitech 5.1 surround [1] system with optical-in and 3.5" minijack headphone-in and might do a listening test between a source MD I know well from one of my portables.

Opinions differ. Obviously subjective. You can see that if a given algorithm does a better job if it can actually use more CPU, then the possibility of some new processors "beating" the 10-year old DSP hardware might exist. But that's all speculation on my part. Avrin seems to be quite actively looking into this.

Stephen

Agreed. This is somewhat analogous to software versus hardware RAID in the IT world perhaps. In this era of cheap GHz CPU, one could say that a dedicated hardware RAID card will not perform as efficiently as the CPU.

About the encoding algorithm improving with the CPU being used, hmm, I doubt it. You will see an improvement in speed but I always regarded the encoding algorithm as "fixed" (eg. a crude analogy is that 2 + 2 will compute as 4 on every CPU but will compute some nanoseconds faster on a Quad-core than a Pentium III 800Mhz). Of course, I am happy to be corrected!

Anyhow, you've compelled me to give SonicStage LP2 another go as it will be fun to compare it to digitally recorded LP2. I'm going to use SonicStage 4.3 "Ultimate" release 2 for this as referenced here: http://forums.sonyinsider.com/index.php?showtopic=24405&st=20&gopid=172325&#entry172325

I will let you know :-)

[1] http://www.logitech.com/en-us/speakers_audio/home_pc_speakers/devices/224

[sfbp]

Agreed. This is somewhat analogous to software versus hardware RAID in the IT world perhaps. In this era of cheap GHz CPU, one could say that a dedicated hardware RAID card will not perform as efficiently as the CPU.

I actually suggested this to one of my profs (I went back to school for a late Master's degree in my chosen calling, in which till then I had no formal training), and he was quite supportive. On one of these Quad core machines, simply devote an entire CPU to compression (and decompression and encryption). Would probably lead to some radical hardware/software designs.

[Arr-Nine-Hundred]

I have no deck yet but I do have a Logitech 5.1 surround [1] system with optical-in and 3.5" minijack headphone-in and might do a listening test between a source MD I know well from one of my portables.

Realised that I cannot do this as the portable has no optical out. Now I understand your /rant Stephen in another post ;-)

Wonder why didn't add it as a default, I would guess space limits, cost perhaps and that most people would have a deck for optical-out purposes. The smallest device I have with optical out is a slimline PS2.

[sfbp]

Realised that I cannot do this as the portable has no optical out. Now I understand your /rant Stephen in another post ;-)

Wonder why didn't add it as a default, I would guess space limits, cost perhaps and that most people would have a deck for optical-out purposes. The smallest device I have with optical out is a slimline PS2.

I still have a project queued to attempt to add optical out to HiMD. It got a bit set back (I got all the bits I needed, I think) because my e-guru friend with the scope has been way too busy to come over and help me tinker for a half a day on it.

I have been intrigued by a PS3. However I have in the meantime discovered another way to get optical from HiMD to my stereo with the decent speakers - a HTPC with HDMI (or if I needed it, optical) out. If they had made a PS3 with a HiMD drive I would have led the stampede (at least on Boxing Day) to Curry's/Dixons or its equivalents over here.

(There's one remaining problem which has nothing to do with anything we care about, and seems to occur randomly on different sources and sinks on the entire system. I can get rid of it by turning off the TV, so it's probably a ground loop problem, which will eventually go away if I ever upgrade my CRT to something more modern).

[JG98258]

CDs and vinyl albums I record get done with HiSP 256 kBPS, either out of of my vintage 30 yr old Tandberg receiver for vinyl or directly out of my not so vintage analog only Denon CD player. I use a Monster Interlink 400 Mk 2 RCA to miniplug cable into the RH1. My really favorite stuff I record at PCM, but it uses up the discs quickly. . .

I did a test with several tracks from one of my favorite Pat Metheny albums recorded onto the minidisc at PCM, 352, and 256. End result of the test was 352 and 256 were indistinguishable and PCM was only slightly better than those two. So- 256 it is. Works well for recording time- 8 hrs worth on a Hi MD, 2 hr 20 min for a regular minidisc.

Good headphones and a headphone amp help a lot too.

[Guest Talisin]

Hi, mdmad. I know this isnnt a very scientist answer, but I can tell you that after many tries, for me SP292 sounds better than 256 and even 352. Definitively, cleaner and clearer than 256, and maybe a tie with 352, but 292 sounds more warmer, more...real? 352 sounds very very good, and also 256, but a bit...artificial?mechanichal?I can't find right word, but I have the feeling that SP sounds more natural for me.

Of course, this is only a personal opinion. But I can tell you that I have had some friends and family doing blind tests, and I have gotten same answers, even in people that don't know anything about bitrates, digital audio and the same. They all say SP sounds "real, warmer, natural..."

I recognized 256 and 352 are very very good. Indeed, when out of home, I used 256 for having a lot of music in a few Hi-MD. But at home, when I really can enjoy music, I always use SP. Indeed, my MD collection are all in SP (with some minor tracks or spoken passages in LP2 in order to get them into the disk). HI-MD bitrates are used only in my tablet pc to transfer them to HI-MD when out.

Hope this helps.

I know a little off-topic (just) but has anybody compared the quality of Type R SP@292 to [Atrac3+ @256 or @353kbps]?

I'm considering buying one of those NW-A ranges - but they don't offer SP@292.. just wanted to know what to expect (I'm very pleased with SP).

If this has already been discussed, please point me in the right direction ...

Regards All,

mdmad.

BTW, I would have voted for SP@292 - but obviously not an option. Is it correct to presume that Sony considered SP superseded by the above, and so did not include it as native with SS on HiMD, hence the lack of this option in the poll?

[sfbp]

Just a small note: I was "virtually" present when the SP codec got added to ffmpeg, and the new format .aea got created. So linux players that use ffmpeg library should be able to play such files once people have figured out how to upload them. Currently the only way is with the RH1 and the QHiMDTransfer program.

[MD4ME]

Hi folks, Just looking to upload some of my MiniDiscs onto my computer using a MZ-RH1, and aren't too familiar with all the options and the technicalities, so should i go for the 'PCM' option or the '256kbs' ? Thanks in advance.

[sfbp]

It depends what you will do with them and what the source represents.

I am assuming that you already recorded SP, because MDLP (LP2 and LP4) will be uploaded "as is" meaning they have had the minimum done to them. As such, MDLP will be transferred and played back without further modification (transcoding). Even if you convert them to WAV files, it doesn't matter what you have the setting you ask about set to, since the conversion will take place AFTER the transfer.

So, to SP. If you plan on converting SP to WAV files then you should use the PCM option. It won't make a huge difference (and it will fill your disk real quick), but this is the least amount of messing with the bit patterns of your music.

Don't forget to decrypt everything using the File Conversion Tool, or those PCM files stored in Sonic Stage will be completely useless (if you have the automatic conversion to WAV turned on, then you can actually delete the sonic stage files, of course).

If you "only" want to listen to them (either from puter using SonicStage/Windows Media Player, or by transferring to HiMD) then by all means upload them to 256kbps (Hi-SP) since this will now represent the least transcoding of your music. But keep them in this format after that, if possible.

Every time you change the format you risk significantly degrading the sound. The only way to be sure is to try it and run A-B comparisons to make sure you can live with the result.

[MD4ME]

Many thanks for the response, i've a load of pre-recorded discs as well my own recordings, what would be best for tranfering these ? considering i might want to burn some of them onto CD at a later date, or send them to i-tunes .

[sfbp]

Upload to PCM, then. You can export to WAV if it makes you feel safer, but you won't have the titling metadata in the file any more (WAV predates ID3 tags). Don't forget to de-encrypt as soon as you finished uploading.

One more thing, you can't do much with PCM format, it cannot be compressed to AAL (my preference) although you CAN edit it with Sound Forge, which knows about the metadata.

In your shoes, I think I would upload to PCM, export to WAV, delete the songs from SS, reimport as WAV files. A trick here is to keep one song so that the folder stays alive, and then delete that 1 (PCM) song and import its WAV file last. This route sounds crazy but it will be the minimum of typing once you get it figured out.

I would then burn to CD (using SS if you prefer, or probably better Nero) or even DVD as archive format (do you have a DVD player that will play back WAV files?). THEN and only then would I compress to AAL, which leaves you perfectly set up to transfer them back to MD or HiMD in a hurry any time you choose to get out the ol' MD.

[MD4ME]

Many thanks for your replies sfbp, i'll spend some time experimenting. BTW whats 'AAL' (excuse my ignorance ) ?

[sfbp]

Many thanks for your replies sfbp, i'll spend some time experimenting. BTW whats 'AAL' (excuse my ignorance ) ?

Sorry. Atrac Advanced Lossless. You will notice it in SS if you

a. import a CD, it's one of the choices (different bit rates but 256K is probably enough)

b. right click on WAV files in SS display... lossless compression.

I use it because it's half way to regular ATRAC (transfer FROM this format to HiMD or NetMD is fast and reliable-sound) but takes half the space of 1411kHz WAV (PCM) file. Its other nice attribute is that it is NEVER encrypted.

[StevenQBosell]

In the past, i've used MDLP to rrecord old talk radio, (Art Bell, Phil Hendrie, etc), and found that LP4 was absolutely fine for that.

LOL back in those days, compared to cassette, LP4 was a Godsend for talk radio junkies - 320 mins on 1 disc? You could recored the whole 4 hour show on 1 disc... that was astonishing at the time, and to some degree still is, considering it's such a small device...

[mmp64]

Sorry. Atrac Advanced Lossless. You will notice it in SS if you

a. import a CD, it's one of the choices (different bit rates but 256K is probably enough)

b. right click on WAV files in SS display... lossless compression.

I use it because it's half way to regular ATRAC (transfer FROM this format to HiMD or NetMD is fast and reliable-sound) but takes half the space of 1411kHz WAV (PCM) file. Its other nice attribute is that it is NEVER encrypted.

Just so I'm still straight on AAL - you cannot export in AAL format (to HiMD) - it is only for archiving purposes. Correct?

[sfbp]

Right. Of all the formats I have seen, it's the only one that seems to convert flawlessly both up and down. There exists no device that actually plays this format, however (other than a PC). It consists of a lossy (ATRAC) part and the "lossless" part which is what it needs to remember to rebuild (presumably) PCM that it was compressed from.

In addition the transcoding to NetMD speeds seems fast, IIRC. Certainly if your "lossy" part is 256k (probably the best choice) then AAL->HiSP is very easy indeed.

[darrencouch]

Lately I have been using a lot of hi-SP through simple burner or direct recording. Seems very decent.

[Stuge]

These days I only prefer 320kpbs in general for any portable device .

[sfbp]

That would be my attitude too, IF I was only dealing with MP3. (and you do say "these days")

However I would be swamped in data if I couldn't have lots of LP2, LP4, and Hi-SP (256k).

There's one more reason for NOT preferring huge data rates (apart from size of stores required) - ease of transmission. It's very easy to set up network storage and stream 66,128,132, even 256 kbps from it. But the higher data rates are a disaster, expensive to transmit and often refusing to stream.

Cheers

Stephen

[sportster44]

For HI-MD I use only ATRAC lossless, and was using LP2 for net MD, but based on what I've been reading here, I'm going to try SP on net MD to see how it sounds compared to AAL.

[Chickenvenus]

For Hi MD I go for lossless or sometimes LP2

For MD Decks, I record radio programes in LP2, and occasionally SP (Most of the shows I am interested in are typically 2 hours in duration).

(For MP3 or other formats I tend to seek the highest bit rate, 320kbs, but increasingly I'm becoming interested in lossless compression formats (such as Flac) and finding better ways to use these now and in the future).

[THIS SUCKS]

i use 192k usually and sometimes 256k. there are some things i listen too that only sound right if they are transfered at 352k. 128k lp2 sounds not good enough for me. lots of artifacts and too much of the music is missing.

[KJ_Palmer]

Well I've changed (from 352/256kbps) since my earlier post. Now everything's 132kbps LP2 (both standard and Hi-MD) for listening purposes. Seems to strike the right balance between quality and quantity.

[sfbp]

Well I've changed (from 352/256kbps) since my earlier post. Now everything's 132kbps LP2 (both standard and Hi-MD) for listening purposes. Seems to strike the right balance between quality and quantity.

Agree, especially when I don't need to post-process the sound in any way. When I do (e.g. noise reduction from an analogue source), then I will record in SP and upload, or LPCM and upload (both are good) to WAV file in either case. WAV can then be sucked back in to Sonic Stage, and compressed to AAL; but usually I make a permanent (CD) version before throwing away the WAV files.

If I need to edit only using delete/divide/combine on the PC, then the (Sony) Sound Forge 9 allows me to do that, provided I decrypt the files first. Even AAL can be edited with this tool.

In addition, anything over 161 minutes (80m disk at LP2) I will use a 1GB disk at 256kbps in a HiMD unit, since this gives almost 8 hours in a stretch. But Kevin is right, once you have "good" sound in LP2, this is a great format for playback - on an "80m" disk you can get 4hr 50 minutes by transferring such tracks to HiMD.

[PhilippeC]

SP if the album is in 24bit-96kHz and close to 74 / 80 mn.

LP2 if I can put 2 albums in one MD, specially in 16bit-44.1kHz quality.

Real time recording only.

[Juan22]

Unable to vote...292 kbps SP mode [CD -> PS2 (optical) -> MD]

[Guest Talisin]

Neither me. I also use 292 kbps Sp mode the most, recording cds from my laptop or hi-fi using optical. This way is how I record my 80' Md, 1 disk=1 album plus some bonus tracks I choose (usually from live albums). These disks are what I use daily, at home and while going by bicycle to my job, but I also have all my music in 256kpbs in (in a external) to transfer it to Hi-Md disks I take when doing long travels.

[trott3r]

Not sure what bit rate Full PCM/WAV is, but that is what i use to record vinyl on to my rh1