14.57731 bit audio - Page 2

ross · 09 February 2021, 13:04

Wow, this thread escalated very quickly.

I can't wait to hear the sounds coming out with this method from nikosidis' A1200.
Then will sure follow tests by physicists, but curiosity in the meantime is high

nikosidis · 09 February 2021, 13:27

Ross: It is not CD quality at all since normal PAL screenmode.

I have not joined the files so here are the files form A1200 output, Left and Right. First two files are from the player provided L,R. Next 2 mono L,R from Songplayer, using the supplied wav file.

http://s000.tinyupload.com/index.php...14481464936341

I noticed the output level was higher with the .exe player so I had to adjust the input of the recorder to not get any clipping. It is therefore a little difference in recording levels but should not matter.

ross · 09 February 2021, 13:35

Quote:

Originally Posted by nikosidis

Ross: It is not CD quality at all since normal PAL screenmode.

I have not joined the files so here are the files form A1200 output, Left and Right. First two files are from the player provided L,R. Next 2 mono L,R from Songplayer, using the supplied wav file.

http://s000.tinyupload.com/index.php...14481464936341

I noticed the output level was higher with the first 2 files so I had to adjust the input of the recorder to not get any clipping.

Just not to misunderstand: ExeL/R.wav is from this new player (converted from the original WAV to 28Kz then upsampled to 44.1), and SongplayerL/R.WAV the same but with 'normal 14 bit' player?

Mo matter it stays at 28Khz, I was more interested in understanding how it sounds than normal 14bit play ..

EDIT: in fact ExeL/R doesn't sound good, but we have to compare on the same conditions

orangespider · 09 February 2021, 13:38

Quote:

Originally Posted by nikosidis

Ross: It is not CD quality at all since normal PAL screenmode.

I have not joined the files so here are the files form A1200 output, Left and Right. First two files are from the player provided L,R. Next 2 mono L,R from Songplayer, using the supplied wav file.

http://s000.tinyupload.com/index.php...14481464936341

I noticed the output level was higher with the .exe player so I had to adjust the input of the recorder to not get any clipping. It is therefore a little difference in recording levels but should not matter.

For some reason they aren't played back at the same rate. I have converted the sample to 27965 Hz, which should be at 128 period, I wonder if I am playing it at the wrong pitch or the player is.

btw about PAL screenmode: Can someone explain what period should I put to get 44100 or close to it on a VGA screen? I really have no experience with that. I could easily convert to 44100 too.

Anyway, we definitely have a proof of concept.

edit: As for the higher output level, that is the feature of this system compared to the 14bit modes, the audio output level is 1.49 times higher than that of a 14 bit audio (but still only 0.75 compared to an 8bit sample playing at max volume at both channels)

ross · 09 February 2021, 13:41

Quote:

Originally Posted by orangespider

For some reason they aren't played back at the same rate. I have converted the sample to 27965 Hz, which should be at 128 period, I wonder if I am playing it at the wrong pitch or the player is.

btw about PAL screenmode: Can someone explain what period should I put to get 44100 or close to it on a VGA screen? I really have no experience with that. I could easily convert to 44100 too.

Anyway, we definitely have a proof of concept.

27710, this is the right frequency for period 128 (half the PWM rate)

Period 80 = 44336,19 Hz
Period 81 = 43788,83 Hz

nikosidis · 09 February 2021, 13:44

Yes, we have to use same player and as you say the .exe files does not sound good at all. Productivity screenmode is what I use for 44.1kHz audio, but DoublePAL is also good.

orangespider · 09 February 2021, 13:46

Quote:

Originally Posted by nikosidis

Yes, we have to use same player and as you say the .exe files does not sound good at all.

It sounds to me that when my exe was playing the system lowpass filter was still on, and it wasnt on when the player was playing it. Anyone knows how can I disable it?

robinsonb5 · 09 February 2021, 13:47

Quote:

Originally Posted by orangespider

For some reason they aren't played back at the same rate. I have converted the sample to 27965 Hz, which should be at 128 period

That's the right frequency for NTSC systems - on PAL systems it's 27710 as ross says. (Due to the slightly different base clocks.)

orangespider · 09 February 2021, 13:51

Quote:

Originally Posted by robinsonb5

That's the right frequency for NTSC systems - on PAL systems it's 27710 as ross says. (Due to the slightly different base clocks.)

Anyways, the player he used was playing them back at a higher pitch, I have just compared them with the source wav and my program is closer to the actual pitch than the player is. Looks like I am playing a 27965 Hz sample at 27710 rate while the player is playing it at 28600 or something close.

Anyone knows if I can disable that lowpass filter?

nikosidis · 09 February 2021, 13:53

Quote:

Originally Posted by ross

Just not to misunderstand: ExeL/R.wav is from this new player (converted from the original WAV to 28Kz then upsampled to 44.1), and SongplayerL/R.WAV the same but with 'normal 14 bit' player?

Mo matter it stays at 28Khz, I was more interested in understanding how it sounds than normal 14bit play ..

EDIT: in fact ExeL/R doesn't sound good, but we have to compare on the same conditions

I played back the files in normal PAL screenmode. That is what was recommended for the player.

ross · 09 February 2021, 13:57

Quote:

Originally Posted by orangespider

Anyone knows if I can disable that lowpass filter?

bset #1,$bfe001

Thorham · 09 February 2021, 14:02

Quote:

Originally Posted by orangespider

But you can easily make your own player just use the lookup tables attached to my original post.

Got it to work, indeed very easy. I'm content with just playing 128k samples because that's more than enough for testing. The audio is very noisy (no filter). Does this need calibration?

orangespider · 09 February 2021, 14:16

Quote:

Originally Posted by nikosidis

I played back the files in normal PAL screenmode. That is what was recommended for the player.

I have turned off the filter and fixed the frequency issues, can you re-record it, please?

here is the fixed version:
http://s000.tinyupload.com/index.php...21693371953375

Quote:

Originally Posted by Thorham

Got it to work, indeed very easy. I'm content with just playing 128k samples because that's more than enough for testing. The audio is very noisy (no filter). Does this need calibration?

It shouldn't be noisy, altho I was only testing with filter on, so I'm not sure how it works on a real amiga atm. Are you sure you're using the correct tables? If you switch the 64 and 63 tables you will get very high noise.

nikosidis · 09 February 2021, 14:25

Quote:

Originally Posted by orangespider

I have turned off the filter and fixed the frequency issues, can you re-record it, please?

here is the fixed version:
http://s000.tinyupload.com/index.php...21693371953375

It shouldn't be noisy, altho I was only testing with filter on, so I'm not sure how it works on a real amiga atm. Are you sure you're using the correct tables? If you switch the 64 and 63 tables you will get very high noise.

I can record again but I listen to it and it is to much hiss or noise with your player. Is calibration needed?

orangespider · 09 February 2021, 14:35

Quote:

Originally Posted by nikosidis

I can record again but I listen to it and it is to much hiss or noise with your player. Is calibration needed?

Nevermind then. Apparently it is the same noise Thorham is getting as well. I will have to do more test on the real amiga. If/when I get the noise under control I'll post a new exe. Thanks.

pandy71 · 09 February 2021, 14:35

Quote:

Originally Posted by nikosidis

I can record again but I listen to it and it is to much hiss or noise with your player. Is calibration needed?

Try attached one - 5 seconds, single channel, 16 bit wav, 997Hz sine with sample rate 27710.1171875Hz with almost 0dBFS level quantized at 15 bit with HPTPDF dither. Should be ok with 14 bit playout routines.

Thorham · 09 February 2021, 14:40

Quote:

Originally Posted by orangespider

Are you sure you're using the correct tables? If you switch the 64 and 63 tables you will get very high noise.

Yes, I'm sure, and when I swap the tables it makes no real difference.

orangespider · 09 February 2021, 14:40

update: It seems that most of the noise is coming from the channel 0 and channel 3 mix. Channel 1 and channel 2 mix seems much much better. Is there any chance that channel 3 playback is delayed for some reason?

orangespider · 09 February 2021, 14:41

Quote:

Originally Posted by Thorham

Yes, I'm sure, and when I swap the tables it makes no real difference.

Can you try to disable channels 0 and 3 on your player and just listen to 1 and 2? Does it make any difference?

Thorham · 09 February 2021, 14:51

Quote:

Originally Posted by orangespider

Can you try to disable channels 0 and 3 on your player and just listen to 1 and 2? Does it make any difference?

Channels 1 + 2 are indeed quite a bit less noisy.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
15 bit 44 khz audio idea.	Thorham	Coders. General	33	15 September 2021 06:22
24 or 32 bit audio capture within WinUAE	EAUniW	support.WinUAE	7	17 September 2018 22:22
Questions about 14 bit audio playback	xxxxx	Coders. Asm / Hardware	16	22 December 2014 19:30
High Quality reproduction of Audio on 8 bit.	pandy71	Amiga scene	0	01 July 2013 15:08
Simple 14 bit audio question...	Thorham	Coders. General	7	06 June 2010 10:55

09 February 2021, 13:04	#21
ross Defendit numerus Join Date: Mar 2017 Location: Crossing the Rubicon Age: 54 Posts: 4,488	Wow, this thread escalated very quickly. I can't wait to hear the sounds coming out with this method from nikosidis' A1200. Then will sure follow tests by physicists, but curiosity in the meantime is high

09 February 2021, 13:27	#22
nikosidis Registered User Join Date: Jan 2020 Location: oslo/norway Posts: 1,613	Ross: It is not CD quality at all since normal PAL screenmode. I have not joined the files so here are the files form A1200 output, Left and Right. First two files are from the player provided L,R. Next 2 mono L,R from Songplayer, using the supplied wav file. http://s000.tinyupload.com/index.php...14481464936341 I noticed the output level was higher with the .exe player so I had to adjust the input of the recorder to not get any clipping. It is therefore a little difference in recording levels but should not matter.

09 February 2021, 13:44	#26
nikosidis Registered User Join Date: Jan 2020 Location: oslo/norway Posts: 1,613	Yes, we have to use same player and as you say the .exe files does not sound good at all. Productivity screenmode is what I use for 44.1kHz audio, but DoublePAL is also good.

09 February 2021, 14:40	#38
orangespider Registered User Join Date: Feb 2021 Location: Becej / Serbia Posts: 120	update: It seems that most of the noise is coming from the channel 0 and channel 3 mix. Channel 1 and channel 2 mix seems much much better. Is there any chance that channel 3 playback is delayed for some reason?

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)