I'm not going to buy a mono stylus for my cartridge. I don't even know if they make one. My question however is about the software.
Should I bother setting my software (I have Sound Forge and Audacity) to mono recording? Or should I just record my mono vinyl in stereo and then decide what to do with the 2 channels? There has to be some standard way of recording mono via a stereo stylus.
Thanks for any input!
Or should I just record my mono vinyl in stereo and then decide what to do with the 2 channels?
I'd recommend this.
One channel, i.e. side of the groove, may turn out to be cleaner than the other.
Ok, that's what I thought. Then do I double it to both sides? Leave it as just one channel? I'd like to be able to keep it digital and burn to CD as well. I heard mono CDs don't exist. Is that true? I have several mono CDs so how do they burn them then?
If you plan to declick, record in stereo. Clicks are often much more prominent in one channel than the other, and if you record in mono they are less well detected.
If you plan to declick, record in stereo. Clicks are often much more prominent in one channel than the other, and if you record in mono they are less well detected.
Thanks! Great point! Will do.
The surface noise is stereo too. If I record a mono record in stereo, the program material is in the middle while noise is distributed across the soundstage. Prob'ly you could use some filter, similar to pro-logic or a reverse karaoke filter, to remove the non-centered signal.
Ok, that's what I thought. Then do I double it to both sides?
Yep, that's it, and yes, there is no such thing as a single-channel audio CD - Red Book (standardized) audio CDs have to be 2-channel, 16-bit, 44.1 kHz.
and yes, there is no such thing as a single-channel audio CD - Red Book (standardized) audio CDs have to be 2-channel, 16-bit, 44.1 kHz.
Wow. That's so interesting. So, the double channel doesn't affect volume or resonance? It's exactly the same as a single channel? I'm comparing for example to a film on DVD. If you play it as 2.0 Mono, it sounds hotter than 1.0 Mono but that could be due to the strength of the center speaker.... ok, nevermind, silly point.
There is a definite advantage to recording "stereo." It is true that sometimes one channel is better than the other (and this isn't necessarily constant; you may need to pick and choose from each channel, using copy and paste to make one best) but most often you can make use of the fact that you have two, out of phase with each other. Automated declicking works better against both of the channels (rather than first converting to mono, then declicking), as does decrackling (decrackling is more likely to be desirable with the older mono disks than with newer, probably in better condition, LPs).
After declicking, convert to mono. This in itself will remove, or greatly reduce, a large number of clicks . They are partly, or more or less completely, out of phase with each other and cancel. The same is true with a lot of the background broadband noise. Its level will be decreased significantly by converting to mono. All this assumes the phono cartridge is properly aligned to allow it.
I don't know what Sound Forge provides. In CoolEdit/Audition there is a channel mixer that, properly used, gives the net mono in one channel and the removed difference in the other. I find this frequently helpful during the manual declicking stage. Canceled clicks (partially or fully) are in the removed channel and align with their former position in the music. Those that are only partially removed can be seen much easier, telling you where more work needs to be done (if the click isn't completely removed from the music). If you are going to do only manual declicking, this arraignment may be even more helpful.
I use this two channel form for the rest of the processing, right through converting to 16 bit. Then I convert the reject channel to silence and copy-paste the good channel into it. Both channels are now identical, even to the dither. It is ready to write to CD-R and it gives the smallest possible file size when converted to FLAC.
I'm comparing for example to a film on DVD. If you play it as 2.0 Mono, it sounds hotter than 1.0 Mono but that could be due to the strength of the center speaker.... ok, nevermind, silly point.
The compressed audio formats that DVD supports (Dolby Digital (AC3) and DTS) can have only one channel.
2.0 Mono is equivalent to a "mono" CD: two channels with the exact same content - the original mono track. As such, they would play out of the front LR speakers in a surround configuration, while 1.0 Mono would only play from the center channel.
Whether or not the 2.0 will sound louder or quieter than the 1.0 in a properly-balanced system will depend on how the 2.0 track was mastered, and also on your room acoustics and speaker placement.
I guess what I'm getting at, and perhaps this is conceptually incorrect, is the question of whether 2 mono channels one on top of the other wouldnt sound louder than just the one channel? because there's two of them I mean. Or does it not work like that?
"converting to 16-bit" - is that the same as saving at 16-bit? Or is it necessary to use the actual bit-depth converter?
Also, with the bit-depth converter in Sound Forge, I have the option to do dither and noise shaping right away. It says, "In general, Highpass Triangular with noise shaping produces the most favorable results." Is that true for our purposes?
It defines Highpass Triangular as:
"Eliminates distortion caused by conversion to a lower bit depth and eliminates noise floor modulation by producing a slightly higher noise level. Noise is shifted to higher frequencies than standard triangular dithering."
If you play a mono track from the computer it will normally go to both output channels. At any given volume setting, playing two channels is going to be louder than playing only one. However, duplicating the mono track in each of the tracks of a "stereo" file will still put out the same signal to each output channel, therefore it will not be louder: playing a one track file into both channels of a stereo system is the same as playing a two tracks identical file into a stereo system (if all are at the same dBfs value).
It is essentially holy doctrine than any reduction of bit depth should be done with dithering (to avoid quantization distortion). It does avoid quantization distortion that will otherwise be there without the dithering. However, actually being able to hear that distortion from a single bit reduction conversion is difficult, and probably impossible with any real music.
Noise shaping is another perfectionist measure. Dither is added noise. The unavoidable quantization error is additional noise. Noise shaping takes most of that noise out of the audible range. Again, hearing it, noise shaped or not, from a single application, is rather unlikely with music.
Probably any noise dither/shaping option offered is good enough. There are patented formulas and extravagant advertising claims, but probably no positive ABX results for any reasonable approach. If is possible to do something worse if you really try but that isn't likely to be an offered option with any reasonable software package.
You can test, to an extent, by generating silence in 24 bit or floating point, then converting to 16 bit with different dither/noise shaping options. This will show you what gets added to the music.
Also, in the more specific case of an LP or cassette transfer, the broadband background noise provides adequate dither already (as does the noise from most microphone preamps, etc. But, more doesn't hurt in the slightest.
I guess what I'm getting at, and perhaps this is conceptually incorrect, is the question of whether 2 mono channels one on top of the other wouldnt sound louder than just the one channel? because there's two of them I mean. Or does it not work like that?
I would expect the downmixing to work like this: C = (L + R)/2 or C = 0.5*L + 0.5*R
Which means you should get the average volume of both.
I'm not going to buy a mono stylus for my cartridge.
There are neither mono nor stereo styli. This depends on if a cartridge is one channel horizontal groove modulation transducer, or two channel 45°×45° to vertical, groove modulation transducer.
There are neither mono nor stereo styli.
Oh, I see... my mistake. Thanks!
I'm not going to buy a mono stylus for my cartridge. I don't even know if they make one. My question however is about the software.
Should I bother setting my software (I have Sound Forge and Audacity) to mono recording? Or should I just record my mono vinyl in stereo and then decide what to do with the 2 channels? There has to be some standard way of recording mono via a stereo stylus.
Thanks for any input!
hmm- the thread didn't expand, so I'm afraid I repeated a lot of what others have written...sorry, couldn't see how to delete the post, so here it is anyway:
The best way is to record your mono record in stereo, then analyze the two channels. Often you'll find one or the other is "cleaner" in some way, less distortion caused by wear, less noise, etc. There are three ways to compare: !. Left only 2. Right only 3. Left plus Right. Listen to all three, pick the best way. For example, if you find that the Left channel sounds less distorted, just delete the right. If a mono sum sounds best, then do that.
Also, if you do any de-clicking via software, those systems often use a Left minus Right signal on which to base their click detection, so a stereo input signal is important to have, even with a mono record. De-clicking also benefits from additional bandwidth, so if you plan to use one, a higher sample rate is of some benefit.
If you are trying to record 78 RPM records, you have a whole different issue though. The groove is bigger, so you do need a real 78 RPM stylus, the standard stylus will just ride in the bottom of the groove and rattle around in there sounding terrible. 78's also used several different EQ curves (not RIAA), so you need to deal with that.
Here's something about that relating to Audacity: http://forum.audacityteam.org/viewtopic.php?f=39&t=35665 (http://forum.audacityteam.org/viewtopic.php?f=39&t=35665)
Still record them in stereo, pick the best channel or sum of both.
I'm not going to buy a mono stylus for my cartridge.
There are neither mono nor stereo styli.
Yes there are. Mono records (from the mono-only era) have larger grooves. Ideal tracking is with a larger stylus. They play OK with a stereo stylus, which is what most people do - but then can sound better when played with a larger mono stylus. Depends on record wear.
Mono records from the stereo era (e.g. re-mastered re-issues, compilations etc) usually have grooves identical in size to those of a normal stereo recording, and would be damaged if played with a "mono" stylus.
You
do not record mono records, played in stereo, in mono on a PC - because this will usually only capture the left channel. You, at least, want to be able to sum both channels together to cancel out all the out-of-phase noise. You can't do that if you only have the left channel . You can do this electronically before digitising, but (as has been amply commented on) after is better - it leaves far more possibilities for digital restoration.
Cheers,
David.
If you are trying to record 78 RPM records, you have a whole different issue though. The groove is bigger, so you do need a real 78 RPM stylus, the standard stylus will just ride in the bottom of the groove and rattle around in there sounding terrible. 78's also used several different EQ curves (not RIAA), so you need to deal with that.
Yes, I know, I already have a 78 stylus for my cartridge.
Still record them in stereo, pick the best channel or sum of both.
The sum of both thing is what I don't understand. I get the idea of taking the best sounding channel, duplicating it to both left and right and then outputting. But I don't get the idea of taking the left and right and somehow getting them to cancel each other's noise. It just doesn't make sense to me. Both sides of the groove are going to have different amounts of wear no matter what. How can their sum actually cancel out noise?
Think of it this way...
A stereo cartridge generates a left signal when the needle moves 45 degrees to vertical in one difection, and a right signal when it moves 45 degrees to vertical in the other direction. Since these motions are perpendicular to each other, there is no interference between them.
If you have the same signal in both channels, they are phased in such a way that the needle moves horizontally, just as it would for a mono record, i.e. push against the needle from one side, and pull equally from the other side.
You don't normall get a purely vertical motion, This would require out-of-phase signal in the left and right channels. When you combine the left and right to mono, vertical motion cancles out. Therefore, any dust or dirt or scratches that produces a vertical motion of the needle cancels at least partially when you mix down to mono.
100% of the wanted signal is horizontal motion.
50% of the unwanted signal is vertical motion, 50% is horizontal motion.
Therefore by throwing away the vertical motion (adding the two channels and dividing by two), you can double the signal to noise ratio (halve the noise).
Actual numbers can be a bit different. The noise isn't perfectly uncorrelated, so the gains aren't as much as you'd expect. But the quantity of noise in vertical domain is sometimes higher than in the horizontal, so the gains are greater than you'd expect.
Lots of supposedly wonderful turntables from the mono era weren't really suitable in the stereo era because they had lots of rumble in the vertical plane which was inaudible in mono, but caused easily audible rumble through stereo speakers.
Cheers,
David.
Ok, I'm going to use Audacity to record my mono record in stereo at 32-bit float point.
Then I'm going to listen to both channels to see if the right really is cleaner and whichever is cleaner I'm going to dupe to be on both channels.
Then I'm going to do a click clean up and then I'll finish by saving as 24-bit wav but in stereo instead of converting to mono.
Does that sound ok?
Record at 24-bit, make sure your editing software is processing with at least that much precision and then save the end result in 16 bit (dithered).
Ok, I'm going to use Audacity to record my mono record in stereo at 32-bit float point.
Then I'm going to listen to both channels to see if the right really is cleaner and whichever is cleaner I'm going to dupe to be on both channels.
Then I'm going to do a click clean up and then I'll finish by saving as 24-bit wav but in stereo instead of converting to mono.
Does that sound ok?
Most of the advice has said the noise removal and declicking work better on the stereo signal. So no.
Also, from the fairly large number of mono LPs I done, only a couple have not done better summing the two channels rather than choosing one channel. I believe both exceptions required choosing parts from each channel to make one best.
The summing does need to be done as the proper step, after declicking and (when needed) after decrackling.
If you don't know about the best way to decrackly (possibly barring some very expensive software), you need to find out about the "younglove script."
Sorry to keep beating a dead horse over here but what exactly did those "mono" buttons do on old receivers? I remember growing up in the 80s, there was always a mono button that could be clicked on during playback.
There was always a mono-FM switch, which would reduce noise on distant stations. If there was another mono button, I would assume it blends the channels.
I have a couple of older receivers at home. I'll have to check to see if either of them have an overall mono switch. (I don't have the owner's manual for either one.)
Mono, usually adding the channels together, making right and left identical, were common in the preamp stage, and in stand alone preamps. Internally they might have just selected one channel and put it out on both channels, but mixing was simple to achieve and usually the desired result.
If it is an old mono recording, you do not want to use a stereo stylus, it will ride the bottom of the groove, and make the whole thing a lot more noisy than it would be if it were traced by a real mono stylus.
Ellipticals and such are even worse than conical stylii in this regard.
While that may be true, few recent cartridges have a mono stylus option. Should one be available, one needs to consider whether that cost, and the probably necessary readjustment of tracking parameters in order to switch back and forth between styli, is worth the cost for a few old LPs. Once on the computer, much noise can be significantly reduced. That is probably a different consideration for a professional doing conversions for hire.
I was buying LPs in the mid 60's when they sold both mono and stereo versions of the same record. You had to buy the version that matched your hardware or else you could cause some damage.
My recollection is that by the late 60's they no longer sold the mono versions and all LPs were stereo, even if their content was mono.
Thankfully, by the early 80's I was able to dispose of my entire LP collection and replace it with CDs.
Hey guys,
Ok, so I ran some tests. Very elementary, but still. The thing is that I'm in the process of transferring music for a short film I've made and I'm using two records I have to get the songs. (Yes, I've acquired all the necessary licences )
I'll call the two samples A and B. Before starting, I should say that the only "processing" done was using a little noise reduction and click removal before proceeding with the steps below...
In A, I took the 32-bit float Stereo transfer I got using Audacity and then switched it to Mono using Sound Forge. It became 1-channel. I then saved it as a 16-bit PCM WAV in stereo.
In B, I took the same 32-bit float Stereo transfer, copied the right channel, deleted the left, pasted the right channel in its place and saved, again, as a 16-bit PCM WAV in stereo.
I've listened to both thoroughly with three different headphones of varying quality. All have good frequency range. I personally couldn't hear a difference. Does this make you say, "Ok, well, sounds like you finally did it!" or "No, man, you did it wrong."?? Don't be shy. I gotta learn. And it's for my film so I have to do it right even if nobody but me will notice.
If the LP is good enough condition, and your cartridge and following circuitry are adequately balanced, both channels are likely to sound the same. Summing the two channels to mono can have two advantages.
[] It will remove, or significantly reduce, many clicks. If you intend to do all declicking manually, this may be less of an advantage. If you do auto declicking, summing to mono is done only after the auto declicking. It is like another step of auto declicking. Especially if there are a large number of clicks, this can make things significantly easier.
[] It will usually reduce the background noise by about 3dB, requiring less NR processing. How significant this is depends on the material. If there is much in the way of low level passages, the noise difference from converting to mono may be easily noticeable.
I'm dealing with dithering now and realised something...
Why is it that I'm allowed to save a WAV as a 16-bit file but I'm also allowed to control the bit depth via the bit depth converter tool? Why are there two options? It seems that even when I saved as a 16-bit WAV, the file properties still said 32-bit float.??? So the function in the Save File options aren't doing anything? Is it me or is this weird?
Ok, I'm going to use Audacity to record my mono record in stereo at 32-bit float point.
Then I'm going to listen to both channels to see if the right really is cleaner and whichever is cleaner I'm going to dupe to be on both channels.
Then I'm going to do a click clean up and then I'll finish by saving as 24-bit wav but in stereo instead of converting to mono.
Does that sound ok?
If you're going to record in Windows system with Audacity, to get >16-bit resolution recorded from your soundcard, you need to compile ASIO compatible version of Audacity by yourself.
http://forum.audacityteam.org/viewtopic.ph...53&p=152019 (http://forum.audacityteam.org/viewtopic.php?f=12&t=8653&p=152019)
http://wiki.audacityteam.org/index.php?tit...Audio_Interface (http://wiki.audacityteam.org/index.php?title=ASIO_Audio_Interface)
Juha