Why do music modules sound different on anything other than Amiga?

[Foebane]

Very interesting, Roondar. Very interesting indeed...

Over to you, Hewitson.

[Hewitson]

Compact Disc vs Vinyl:



CD and quality don't even belong in the same sentence. Look at the waveform, it's almost constantly clipped. The record doesn't clip one single time.

As I said, a giant step backwards in sound quality. Which admittedly, has more to do than brainless fuckwits "mastering" the CD rather than differences between the actual formats themselves.

[Foebane]

Originally Posted by Hewitson

Compact Disc vs Vinyl:



CD and quality don't even belong in the same sentence. Look at the waveform, it's almost constantly clipped. The record doesn't clip one single time.

As I said, a giant step backwards in sound quality. Which admittedly, has more to do than brainless fuckwits "mastering" the CD rather than differences between the actual formats themselves.

I have seen waveforms for songs that are almost always maxed out on the amplitude, but I think modern records do the same thing. Why shouldn't they? Such a waveform would be easier to cut into the master. Why would they use a more natural waveform for records?

[roondar]

Originally Posted by Hewitson

Compact Disc vs Vinyl:



CD and quality don't even belong in the same sentence. Look at the waveform, it's almost constantly clipped. The record doesn't clip one single time.

As I said, a giant step backwards in sound quality. Which admittedly, has more to do than brainless fuckwits "mastering" the CD rather than differences between the actual formats themselves.

I don't see any waveform here??
Did you mean to include an example?



As is, I can't agree with that statement.

First off, I'm 100% sure that, say, classical music on CD (or even older (say 1980's-early 1990's) pop recordings) has plenty of dynamic range and equally sure that pop music re-released on vinyl these days is likely to also be heavily compressed.

Secondly, you didn't counter any of the points I made on the elements of vinyl audio reproduction that are objectively (as in you can simply measure it) worse than CD's.

And thirdly, the only reason that CD's can have such a 'loud' mixes to begin with is that they don't suffer from the non-flat frequency response curve, poor signal to noise ratio and variable dynamic range* that vinyl does. Or in other words, the crap mixes (which I also dislike!) are a consequence of CD's ability to play back audio with far fewer compromises than vinyl can.

*) DR on vinyl is strongly dependent on what frequencies you record and where on the record the stylus is. High DR music does best at the beginning of the record, as dynamic range reduces the closer you get to the center of the record due to the smaller tracks and constant velocity of the record (which is analogous to slowly lowering the sampling rate as you go from one side of a CD to the other).
--

Originally Posted by Foebane

I have seen waveforms for songs that are almost always maxed out on the amplitude, but I think modern records do the same thing. Why shouldn't they? Such a waveform would be easier to cut into the master. Why would they use a more natural waveform for records?

As I understand it, part of this whole 'loudness war' originates in how the average listener tends to find music that sounds 'louder' (i.e. has less dynamic range) to be sounding better. Crazy but apparently true.

Interestingly, some smaller CD labels/producers are 'fighting back' against this loudness war. Then there are hybrid SACDs, which usually have a very good plain CD layer as they too let the dynamic range be as big as needed rather than compressing it to hell.

[chb]

Originally Posted by Hewitson

CD and quality don't even belong in the same sentence. Look at the waveform, it's almost constantly clipped. The record doesn't clip one single time.

As I said, a giant step backwards in sound quality. Which admittedly, has more to do than brainless fuckwits "mastering" the CD rather than differences between the actual formats themselves.

Well, as you state yourself, clipping on a CD has nothing to do with the limitations of the medium itself (there's actually plenty of dynamic range) but with mastering/mixing habit (aka "The Loudness War"). There are indeed albums that are more sensible mixed on vinyl. But a lot of CD records exists that are not compressed/brick wall limited to death - be assured, you would have a hard time finding a classic or jazz record showing that kind (or even any kind) of clipping.

As much as I like vinyl for various reasons, it is technically clearly inferior to CD audio.

[roondar]

I should add here that I love the experience of playing a record and feel it's a wonderful way to get 'closer' to the act of playing back music (the sheer physicality of seeing the needle move and being able to hear the sounds coming from the actual stylus even when no amplification is used just 'work' for me). My points have nothing to do with me being 'against' vinyl. It's just that I disagree with pretending it's 'better' than everything else regardless of what technical or scientific facts actually tell us.

[daxb]

The two waveforms of a song show a normal one and a ramped (clipping). 16bit and 32bit. That is all the information I can extrapolate.

[oundfire99]

Originally Posted by Foebane

I can take any demo or game and run in on WinUAE, or even real hardware, and it will sound as intended, but when I take the .mod file and run it on something like XMPlay, it will invariably sound different - why?

I'm specifically talking about some of the samples sounding like they have a low-pass filter on, but not all of them. Why is that? As far as I'm aware, music modules don't have a filter on/off for each sample.

So what's going on?

Not an expert but I would say WINUAE does a very good job of emulating the Paula chip, obvs. not the same as Paula but still close

[chip]

WinUAE does a PERFECT job with audio emulation

Just try this by yourself :

Take a module which you like

Make a recording in .WAV with WinUAE option

Create a .MP3 from the resulting .WAV

Then go to this site

http://www.paula8364.com/

Download the "same" .MP3 you created with WinUAE and make comparison

I can ensure you that i tried this by myself with some different modules

Well, result is that the real thing sounds exactly like the emulated thing

Or, at least, my ears are not able to spot the difference

[robinsonb5]

Originally Posted by roondar

As I understand it, part of this whole 'loudness war' originates in how the average listener tends to find music that sounds 'louder' (i.e. has less dynamic range) to be sounding better. Crazy but apparently true.

I think part of it is also that people want to be able to listen to music in their cars and other noisy environments. The low dynamic range means that more of the song is audible over road noise.

[zipper]

Radio senders (commercial) were competing of listeners - louder one got more attention.

[oundfire99]

As someone said on here, answered my thought about analog and 8bit merging together (ie. Octamed) better than modern digital, means you don't need to over compress the shit out of everything to make it louder and lose dynamic range. Really interesting subject though as modern methods require a lot of EQ and work to make mix work where-as everything sounds amazing in Octamed or similar (8 bit tracker) Analog vs Digital argument proven right here ...

[robinsonb5]

Originally Posted by oundfire99

Really interesting subject though as modern methods require a lot of EQ and work to make mix work

That's mainly because modern mixes have many more channels to deal with. A naive method of mixing channels is to add the signals together and divide by the number of channels. That will obviously result in each channel getting quieter as more channels are added, requiring more advanced EQ, compression, etc.


The Amiga has it easy by only having to mix 2 channels for each side of the stereo image.

[Mrz]

I have an old version of winamp which can play Amiga modules using an old plugin called in_mod.dll
this is very precise and the sound obtained playing modules is identical to my Amiga 1200

[brett71]

Originally Posted by Mrz

I have an old version of winamp which can play Amiga modules using an old plugin called in_mod.dll
this is very precise and the sound obtained playing modules is identical to my Amiga 1200

Yeah, I used to use that too, but there is one MOD that it never was able to play correctly, i.e. same as an actual Amiga or with Deliplayer. The mod is called "Dirt.MOD". Something it did with the tempo setting just wasn't handled by that WinAmp module correctly.

[Hewitson]

Originally Posted by Mrz

I have an old version of winamp which can play Amiga modules using an old plugin called in_mod.dll
this is very precise and the sound obtained playing modules is identical to my Amiga 1200

I'm sorry, but that's simply not correct.

[lordofchaos]

To the posters original question... Back in the day I had a decent HI-FI system that the Amiga was hooked up to, it sounded great. These days I`m running WinUAE through a similar HI-FI, and it sounds as if it's lacking some bass frequency along with high end frequencies? It's quite subtle, just sounds weaker.

To remedy this (maybe) I record the sound module in WinUAE using NVIDIA capture (MP4 at highest bitrate, dump the video file into Sony Vegas, then use post processing on PC. Adjusting the EQ, restoring the bass and higher range frequencies. Bit of a pain but can yield better results, to my ear anyway.

I guess it comes down to how you want to consume Amiga audio in the modern era. Real hardware, emulated hardware, PC mod players, YouTube, MP3s, CD's, cassette tapes etc.. Too much choice!!!

[mc6809e]

A problem with digital audio, IMHO, is with the belief that because we have all the samples necessary for perfect reconstruction we get perfect reconstruction when those samples are output through some simple filter.

But this is incorrect.

Absolutely perfect reconstruction requires that all samples be read before the first value is output. Anything less will not give perfect reconstruction. Practically, perfect reconstruction isn't usually necessary, but any filter shorter than the entire length of the sample is going to introduce some amount of interpolation error. How this error manifests itself depends on the playback hardware.

Of course all the typical abuses, clipping, etc, don't help either.

[KONEY]

Because Paula DACs have an unique sound texture.

[FromWithin]

Originally Posted by mc6809e

A problem with digital audio, IMHO, is with the belief that because we have all the samples necessary for perfect reconstruction we get perfect reconstruction when those samples are output through some simple filter.

It's not a belief. It's a mathematical certainty.

Absolutely perfect reconstruction requires that all samples be read before the first value is output. Anything less will not give perfect reconstruction. Practically, perfect reconstruction isn't usually necessary, but any filter shorter than the entire length of the sample is going to introduce some amount of interpolation error. How this error manifests itself depends on the playback hardware.

A filter shorter than the length of the sample? What does that even mean? In fact, that whole paragraph makes no sense.

I've just been reading back at some of the other comments from the past few weeks and there is some extraordinary misinformation in there. "everything sounds amazing in Octamed or similar". You've got to be joking. Most of it sounds like it came out of someone's arse. If you think it all sounds amazing it's because your level of expectation for how it should sound is so low.